Here’s a preview of what we’ll cover this week: 

• Macro: Costco Seconds A Healthy Economy

• Markets: Strategic vs. Tactical; Time To Leave Oilfield Services; No One Left To Sell; Cheap, Cash-Generative, Ignored; Oil and The Geopolitical Spike; THE S&P: Magical or No?

• Lumida Curations: Software Selloff or Buying Opportunity?; The Energy Race Powering the AI Boom; The $1 Trillion AI Question

Spotlight

This week, I sat down with Alex Wilson for a Lumida Non-consensus investing podcast— serial founder, and co-creator of The Giving Block, which he later sold to Shift4. 

We covered the payments category: why it's been in a bear market, where the value is hiding, and why Stripe's $130B private valuation doesn't hold up against public market comps. 

We also got into Alex's new venture Cyclops — a picks-and-shovels infrastructure play helping payment processors close the gap on stable coins and crypto rails. 

Watch the full Lumida Non-consensus investing podcast on Youtube here.

We are also on Apple podcast and Spotify.

Strategic vs. Tactical 

Over the last few weeks, since our ‘Markets at a Crossroads’ piece we’ve taken a bearish stance on markets.

I did an FSD stream this week, titled ‘tactical vs strategic’, where I laid out the current standing of equity markets from both a strategic and tactical lens.

In short, the strategic view is bearish. The tactical view suggests we are prone to a sharp rally on the whiff of almost any good news.  

Let me explain the difference.

The Strategic View: Multiple Bubbles, One Market

The big picture for markets isn't great.

There isn't one problem. 

It's several problems stacked on top of each other — each with its own unwind timeline, each connected to the others in ways the market hasn't fully priced yet. 

Here’s the issues list:

• OpenAI and Anthropic have $1 Tn revenue forecasts, and committed obligations (e.g., debt) across the datacenter ecosystem. They face capital constraints.

• Private Credit. We’re seeing redemptions and gating. That will continue. Private credit finances, in part, the datacenter expansion

• Regional Banks are providing back leverage to Private Credit.

• Industrial stocks are in a bubble, and started breaking down this week. The leading industrial names (e.g,. Caterpillar, GE Vernova, Corning, etc.) are all linked back to the OpenAI theme.

• Elevated Positioning

  and add to this lift the rising inflation concerns from the spike of oil which creates inflation risk. There’s nothing worse for markets than elevated inflation concerns.

We've flagged most of these individually over the past few weeks. 

1. Private Credit

The pain in private credit isn't done.

This week, BlackRock told investors it would not redeem more than 5% of their capital. Shares dropped on the news. 

That 5% ceiling isn't a BlackRock-specific policy — it's the structural limit baked into most 40-Act private credit funds.

Cash flow generation on these vehicles runs 5–7%, enough to service normal outflows. Not a rush to the exit.

More gating events are coming across the category.

The deeper issue is loan quality. 

The 2021 vintage is where the stress is concentrated — cheap money met aggressive underwriting, and those loans are now seasoning in a different rate environment. 

The tell isn't non-accrual rates, which can look clean on paper. The tell is loan modifications. 

Modifications are the private credit version of pretend and extend — they restructure terms to avoid a technical default without fixing the underlying business. They just delay recognition.

If you have private credit exposure, ask three questions: how much is 2021 vintage, what percentage of the book has been modified, and what's the non-accrual rate. 

The answers will tell you more than any headline return figure.

To be clear — the economy is still fine. 

An elevated default picture in private credit doesn't mean zero returns. It means you make 5% instead of 10 to 12%. 

The ripple effect is worth watching though. 

Several regional banks have financed these private credit vehicles. If default rates tick up, we might see impacts there.

The valuations are still high for Blackstone and KKR, which add to their vulnerability. 

Apollo looks relatively cleaner. But I've been burned before finding the right name inside a broken theme. 

The theme has issues. You can't fully outrun that.

2. The Semiconductor and Datacenter Complex

This week, Oracle exited its commitment under the Stargate joint venture, citing free cash flow concerns.

Oracle spent the last several quarters positioning itself as a core infrastructure partner for OpenAI. 

They signed large revenue performance obligations, raising $25B in debt, $25B in equity to fund the buildout. 

That capital raise was a tell in itself. 

You don't raise that kind of money unless you believe your customer can’t pay your bills. Oracle gets an up-front payment, but then must undertake a multi-year capex build out. They have credit risk on OpenAI.

Now they're unwinding the Stargate commitment and cutting headcount. Meta assumed that obligation. But, there are over $100 Bn+ in obligations…

The implied reason is simpler: Oracle doesn’t have confidence in OpenAI's ability to pay, and neither does the market.

We wrote about this months ago in our newsletter on ‘How Sam Altman broke the world?’.

By the way, Oracle senior executives including the Co-CEOs have sold over $4 Bn in stock including in recent months.

OpenAI has roughly $1 trillion in committed spending obligations across its datacenter and semiconductor partners. 

Those obligations don't appear as debt on OpenAI's balance sheet — they're structured as contractual arrangements.

But they show up as assets on their suppliers' books as accounts receivables (asset).

The question markets are now being forced to ask is what those assets are actually worth if OpenAI can't fund them.

Oracle is in a structurally similar position: assets on the book whose face value depends on a counterparty that is fundamentally short of cash.

Meta assuming the agreement changes the risk profile of that specific contract — Zuckerberg has the cash flow to back it. 

But every other provider still holding OpenAI-linked obligations has to run the same math Oracle just ran. 

The results won’t be good for earnings, and thus valuations. 

Markets are repricing the datacenter complex.

Micron's chart, for what it's worth, confirms the pressure. It's broken below the 50-day moving average. Not something I'm touching here.

That is a very bad open for the Semiconductor complex.

We believe the Micron trade is over.

3. Industrials

This is a bubble that hasn't finished popping.

Look at Caterpillar's trailing P/E. 

CAT's core datacenter link is its industrial generators — hyperscale facilities need massive backup (and increasingly primary) power, and grid interconnection delays mean on-site CAT generation runs longer than originally planned. The Energy & Transportation segment (~40% of revenues) is where this shows up, specifically large-bore reciprocating gas and diesel engines with 18–24 month lead times.

It's a capex-cycle play, not recurring revenue, so the risk is order book deterioration if hyperscaler build rates slow.

Now, if CapEx is slowing due to capital dependancy issues and cancelled contracts, what does that mean for Caterpillar? Not good. (We are short CAT as of this past week - always wait for the “right shoulder”…)

Caterpillar went into bubble territory, and we think it comes down to earth.

Look at Corning (GLW) — where the entire executive team has been selling stock for months.

(Note: You can use the www.lumidainvest.com app to track insider selling and buying for your stocks. We’ll soon be adding AI tech that automatically scans your portfolio to detect hidden risks like this and others. If you want get on the waitlist to to invest in Lumida ahread of our public launch, visit www.lumidatribe.com. )

Look at Teradyne (TER) trading at 70x earnings on the back of an Amazon contract.

So what if you have a robotis contract with Amazon? Intuitive Surgical has a robotic story. It’s down over the past year. People fell in love with their ideas way too much.

What if Amazon pulls back capex? They are also laying off workers and issuing debt to shore up free cashflow.

That's the question no one in those names is asking.

We're short Caterpillar. 

The broader industrial theme is expensive, and it's starting to crack. Industrials are a large part of the S&P 500 - this is a major concern point.

4. Oil and the Strait of Hormuz

This is the newest entry on the list.

The market got rattled this week by a Bloomberg piece featuring a retired admiral who laid out Iran's asymmetric capabilities — thousands of mines and speedboats in the Strait. 

The article went viral. Markets felt it immediately.

We believe the U.S. has an overwhelming advantage over Iran.

In Hegseth’s presser last Thursday, there was no Q&A. Minimal messaging from the White House on the SPR or tanker logistics.

Here's the reality: oil moving through the Strait is the jugular of global energy supply. 

You don't need Iran to "win" a conflict to disrupt it.

Until tankers start moving again, oil prices stay elevated. 

Elevated oil is an inflation input, a consumer spending headwind, and a Fed rate cut killer — all at once.

This is what makes the strategic view uncomfortable. 

None of these four issues exist in isolation. 

Private credit stress links to regional bank exposure. 

OpenAI counterparty risk links to the datacenter and semiconductor complex. 

Industrials valuations were inflated by the same AI capex narrative. 

And oil — the newest entrant — is the macro wildcard that tightens everything simultaneously.

It's worth drawing the parallel here with 2022.

We had war-related shock, Inflation pressure, Midterm year, Nonsense in private markets, and Crowded valuations in certain categories.

Back then, the bubble was in software and payments. Those categories spent three years decompressing. 

Now the bubble has migrated: into industrials, datacenter-linked names, private venture capital, and hot thematics.

Crypto was the first ‘hot thematic’ to crack - immediately after the peak euphoria of the Genius Act.

We’re seeing marginal liquidity markets an thematics under-perform now, and give way to value and international stocks.

We believe that trend will continue. It’s a great time to be a long / short investor.

The Tactical View

Here's where the picture changes.

VIX at 29 going into a Monday is panic-level fear. 

That's historically unsustainable. 

You almost always get a mean reversion — a multi-day, multi-week bounce — when fear reaches this pitch.

Often times you see a Monday gap down open. You will see a lot of volatility, include a large open to high print that can get faded and move to new lows. And possibly a Turnaround Tuesday type rally.

We are seeing large nasty red candles across many chart and sectors selling off in unison. That’s weirdly a good sign to see. Any whiff of good news on tankers would send many stocks soaring.

However, we think such rallies would ultimately be faded.

We see staples rolling over (see XLP) below.

Staples no longer offer a safe haven. Who is buying Walmart or Costco at 40x PE. (We are short Walmart). And the Walmart family is selling their stock en masse.

We saw a similar set of conditions in the dotcom era. There would be punctuated sharp and furious multi-week rallies when the Vix hit 30. Then markets would roll over again.

If capex does pull back, and we think it does due to capital markets forcing the issue - that would have slower growth implications for GDP and corporate earnings that are not priced into stocks. Analysts estimates would be too bullish. Markets are re-discounting those assumptions.

That doesn’t mean you don’t get a sharp rally on Tuesday, let’s say, due to tankers flowing and the release of oil.

Incidentally, if that happens, energy names that have rallied would likely decline. There’s a lot of ‘whipsaw’ risk in the market.

Where I'm Finding Ideas

The chess term for being in an optimal but constrained position — where every move deteriorates your situation — is Zugzwang. 

That's what markets feel like right now in most sectors.

But there are a few pockets that look interesting through that fog.

The best approach is to buy good businesses, priced well, that sold off hard due to present geopolitics that don’t impact the long-term earnings power of the company.

United Airlines. 

I put a toe in UAL at the close. It’s a partial position. Half of me thinks we may be 1 day early. But, there’s a chance Trump says something constructive like releasing the SPR, so need to have some exposure.

6.9x forward P/E with FCF yield at ~9%.

When oil prices drop — and they will — the stock recovers.

Airlines also have the chance to hedge their oil prices, which will help produce better results than the market is expecting at current valuations.

And, oil futures prices looking a few months out is still in the low 70s range. Even today, airlines can hedge energy risks. So, the sell-off seems overblow.

The demand is Boomer-driven, who are still going to travel once geopolitics recede. 

Overall, UAL is a real business at a dislocated valuation. 

Interestingly, going into this week last Sunday evening I noted that the worst thing you can own are airlines and we were short the $JETS ETF (since covered).

Markets require mental flexibility.

China and Emerging Markets. 

China imports heavily from the Strait of Hormuz.

So does Japan. So does South Korea, which has already taken an 18% drawdown from peak. 

When the oil disruption resolves, these are the beneficiaries. 

We noticed South Korea and China went up on Friday despite a rising oil price.

When you see positive news despite negative headlines, that’s constructive.

Still, it could be early or a relief bounce. That’s entirely possible. At the same time, valuations are attractive again. So, taking an initial investor position makes sense.

We added some PDD and already own Tencent, and added EWY.

Here’s a chart of PDD.

China is getting hurt due to stronger dollar. Dollar will weaken as conflict abates. We want to start accumulating at favorable levels and be patient. If we buy businesses priced right, we expect to do well. 

The long-term earnings power of these businesses is not determined by the short-term price of a barrel of oil.

One More Indicator to Watch: Insider Activity

When management teams sell, listen.

I've found two names this week where the insider selling picture is alarming. 

Corning (GLW) was one. 

The other I won't name, but the pattern was worse: not just the CEO or CFO, but the Chief People Officer were selling too. 

When everyone from the C-suite to HR is liquidating, that's not confidence.

Now, how do you check insider activity for all stocks quickly?

This is exactly the kind of edge the Lumida Invest app is built for.

We track Insider activity on all stocks, and help you filter stocks with highest insider sales and buys. 

But that's one piece. 

The app also gives you AI-driven bull and bear analysis on individual stocks, insights from hedge fund positioning, and theme and factor analysis for your portfolio.

The goal is simple: give everyday investors the same quality of information that institutional desks get.

If you'd like to try it, sign up at lumidainvest.com.

Lumida is Scaling. Join Us Now and Own the Future of Investing.

Lumida has about 9 developers feverishly building the wealth manager for the next generation. We aim to take on RobinHood - but with an AI-native, mass affluent focus.

We want to invite our community to be a part of the journey.

$84 trillion is transferring to the next generation. And the wealth management industry isn't ready for them.

The next generation doesn't want to attend the US Open with an old-school banker. They're not watching CNBC.

They live online. They find their news on TikTok, learn on YouTube, and trust personalities over institutions.

Traditional wealth managers have spent decades building businesses for a generation that is handing the keys over.

That's the opportunity.

Lumida is building the wealth management firm for this generation — digital-native, AI-powered, and built from the ground up to serve investors who think differently.

We envision a world where an AI advisor understands your family's goals as clearly as you do.

One that speaks your language, operates at your pace, and never stops working. You can talk to your advisor at 2AM in your pajamas. On your terms.

We are already executing on that vision.

The Lumida Invest SuperApp puts institutional-grade research in your pocket — AI bull/bear analysis, hedge fund 13F tracking, insider activity signals, and private market deal access.

Join our investor community and be part of our growth. We will launch an equity crowdfunding soon.

This is the waitlist. We already have 100 sign-ups, and we’ll prioritize allocations as best as we can. We’d rather have as much participation as possible to amp up social media impact and our events strategy.

You’ll see our strong revenue growth (with third party audited financials) and valuation details soon to follow before you need to make your final decision.

The waitlist will get you access to these communications.

Macro

Costco Seconds A Healthy Economy

Costco reported Q2 2026 earnings this week. Net sales grew 9.1% to $68.2B. Net income came in at $2.03B, up nearly 14% year over year. 

The transcript provides a clean read on the state of the economy.

Management highlights the consumer is still spending.

CFO Gary Millerchip confirmed that members "are willing to and have the capacity to spend" — and it's showing up in the data. 

Traffic increased 3.1% worldwide. Average transaction was up 4.2%. People are visiting more and spending more per trip.

On inflation, the picture improved. 

Millerchip noted overall inflation is "trending towards low single digits" down from the low-to-mid single digit range of prior quarters. 

That's a meaningful tailwind for real purchasing power, and it supports continued spending without requiring the Fed to move.

However, the management flagged inflation risk: "The situation in the Middle East could impact fuel costs and shipping schedules if there is instability in the region for a sustained period of time." 

Gas prices were down mid-single digits in Q2. If that reverses, the inflation picture changes quickly.

Overall, the transcript highlighted  a healthy economy with solid customer spending.

We have a short on Costco.

It’s funny how a growing business can sometimes be a terrible investment.

How do you justify a P/E NTM of 47.0x with single digit revenue growth?

Markets

Time To Leave Oilfield Services

Energy was one of our biggest exposures heading into this week. 

It had a solid run since the start of the year.

The setup favored it - Venezuela, Iran, defensive positioning, and reasonable valuations. 

‘What can go wrong for this category?’- Everyone started to think the same over the last month. 

And, when everyone agrees on an idea, it gets crowded, and that is the best time to exit.

That’s what we did this week. 

The OIH was up 43% YTD last week, and it feels like it had ran beyond the fundamentals.

We are now seeing the unwind.

FTI, WFRD, and the entire oilfield services complex broke down this week. 

We sold FTI. We love the business. But it feels like the recent run is complete. The stock broke down below its trend.

We highlighted FTI back in our newsletter on 11th Jan. SPY traded at ~6950 compared to 6700 today.

This is what active management looks like in action.

The setup that made oilfield services attractive — under-owned, ignored, with a credible energy demand thesis behind it — is no longer the setup. 

Everyone got in. The upside is fully priced.

We're happy with the gains. We're moving on.

No One Left To Sell

Markets are mean-reverting hard post-Iran.

The worst-performing decile YTD rallied an average of 4.4% this week. 

Every other decile finished in the red, with the best-performing stocks actually falling the most, down 6.1%.

The Iran shock forced institutional deleveraging, and the unwind hit the crowded longs hardest.

Software was the primary beneficiary of this mean reversion.

IGV had already fallen over 22% in the first two months of the year, putting its total decline from peak above 30%.

Citrini’s article marked the bottom in IGV. Investors spent the weekend reading about the death of software at the hands of AI. 

Monday the 23rd opened into low liquidity and an exaggerated flush.

IGV has rallied 13.9% since the close on 2/23, trading higher in eight of the last nine sessions — the sixth-longest open-to-close winning streak in the ETF's history. 

The Citrini report, depite its dim intellectual merits, was the sentiment clearing event the category needed. 

Maximum bearish narrative plus forced selling plus low liquidity. That's what bottoms look like.

The AI Apocalypse thesis was always overblown. 

Enterprise software doesn't get disrupted overnight. Workflows are already embedded with SAAS, and switching costs are high. 

Management finds a way  to integrate AI within their platforms, which turns AI from a disruption threat into a new opportunity. 

HubSpot's earnings proved it. They saw an increase in customers, retention, and average revenue per user by integrating AI within existing products. 

We added Hubspot(HUBS), Godaddy (GDDY), and Adobe (ADBE) in the prior weeks. 

The thesis was simple: high-quality businesses with durable earnings, trading at valuations that had fully priced in an apocalypse that wasn't coming. 

The markets seems to be realizing the thesis. 

HUBS was up double digits. The broader cohort followed.

We're staying in our positions.

The earnings are there. The valuations are near all-time lows. 

And the crowd is only now starting to return to a category it abandoned too aggressively. 

That's usually a good place to be.

MetLife — Cheap, Cash-Generative, Ignored

MetLife is the anti-theme stock we’re considering this weekend. 

It is one of the largest insurers in the world, providing services across the US, Asia, Latin America, and EMEA. 

MetLife trades at a FCF yield of 36.5%, at highest levels of 3Y history, and leading peers.

Forward P/E is 7.3x, which has compressed from its usual 8.5x level.

We believe buying real businesses like this is what you want to do in an environment like this.

Q4 earnings were solid. Adjusted EPS came in at $2.58, up 24% year over year — the strongest single quarter in recent history. 

The business is executing. Revenue growth, expense control, and higher variable investment income all contributed.

Across geographies, EMEA earnings grew 64% YoY. Latin America was up 13%.

The capital return picture is equally compelling. 

MetLife repurchased $2.9B of stock in 2025 and paid $1.5B in dividends — a combined shareholder yield of roughly 11%. 

The dividend alone yields 3.16% with a sustainable 50% payout ratio and a 5-year CAGR of 4.3%.

The stock is down 10% over the last year despite decent fundamental performance. 

That disconnect — between earnings delivery and price action — is the setup.

The risk? 

Insurance earnings are sensitive to interest rates, equity market volatility, and derivative mark-to-market swings. 

Q4 net income was clipped by derivative losses from rising rates, even as the adjusted earnings picture was clean.

But at 7.3x forward earnings and a 36% free cash flow yield, the market is pricing in a lot of bad outcomes that haven't materialized. 

We still plan on researching it further - especially the balance sheet and how sensitive it is to rates.

But, we like it here from a ‘buy and research further’ perspective.

Oil and The Geopolitical Spike

Oil is trading four standard deviations above its 50-day moving average — the most overbought reading since 1985. 

These levels suggest we are closer to a mean reversion today than we were yesterday. 

Every prior instance of this extreme move in oil has been followed by weakness over the next six months, with an average six month drop of 15%.

Any drop in oil here would be good news for markets.

We are now trading in a deep fear territory. The VIX spread confirms the panic. 

It is in the same territory as the tariff crash in early 2025 and the AI worries spike last fall. 

Both resolved into rallies within days.

But, with geopolitical conflicts, we have a lot more moving parts.

An extension in the conflict can mean the market’s bottom is still some time ahead. 

Here’s how markets have performed previously with geopolitical tensions. 

The recent mentions on this list suggest we are close to a bottom here. 

THE S&P: Magical or No?

A lot of discussion on how the S&P is only down a few points from ATHs despite violent rotations under the surface.

Here's the thing...

The Materials and Staples categories moved into bubble territory to achieve that outcome.

And that support is no longer there as their safe haven status was lost on Friday.

The point is if you think the S&P is a magical index you are mistaken.

What you had was the 'hot ball of money' - a term I see increasingly used on X along with Non-Consensus - flood into Staples and Materials.

That's a temporary phenomenon.

Take a look at the attached valuation multiples for the XLP ETF and XLB ETF.

See how they are trading in their all-time high range. 

The trade is crowded, and we might see an unwinding here. This will reflect in the index’s performance.

Lumida Curations

Software Selloff or Buying Opportunity?

Wedbush’s Dan Ives argues the recent market selloff has indiscriminately punished software stocks, creating potential long-term buying opportunities in names like Microsoft, CrowdStrike, and Salesforce for investors with a multi-year horizon.

View Curation

The Energy Race Powering the AI Boom

The surge in data centers and AI infrastructure is igniting a massive energy demand cycle, where even small breakthroughs in batteries, geothermal, or nuclear efficiency could create billion-dollar opportunities.

View Curation 

The $1 Trillion AI Question

While AI narratives dominate headlines, the discussion highlights a contrast between real capital flows in private credit and the still-speculative assumptions behind trillion-dollar AI forecasts.

View Curation

Meme

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: How's The Economy?

• Markets: The AI Apocalypse Trade; Morningstar: The AI Fear Is Wrong Here; Nvidia And Datacenters; Druckenmiller Is Non-consensus; The Two Most Important Charts

• Lumida Curations: Why Technical Analysis Isn’t the Edge It Used to Be; The AI Arms Race Is Already Global; Markets Aren’t Confused About AI

Spotlight

This week, I had a bits + bips podcast discussing the unwind of crowded thematic trades and what comes next.

The core view: Anything with a KOL behind it — Bitcoin with Michael Saylor, Ethereum with Tom Lee, quantum with Cathie Wood — is getting unwound. 

The right move is to rotate into what sounds boring: Small cap Industrials, waste management, oilfield services, anything with free cash flow. 

Watch our discussion here. 

US ATTACKS IRAN

Back on Feb 19th, Trump said Iran had 10-15 days to make a deal, and if a deal isn’t reached, “bad things will happen”.

On Saturday, the US began major combat operations in Iran. Israel joined with missile strikes simultaneously. 

Iran retaliated, targeting US assets and Israel, the UAE, Kuwait, Bahrain, and Arabic countries. (The fact lost on most major media is that no one in the Middle East actually likes Iran.) 

Our view is that any Iranian counter-response will be limited and ineffectual. The leadership is gone. You can’t mount a couner-response without command and control.

The US & Israel be able to fly with impunity within one week according to former Defense Secretary Epster.

Iran cannot block the Strait of Hormuz. We believe an oil price stock will be sticky. In fact, Saudi Arabia - an enemy of Iran - has plenty of reason to pump production.

Iran’s leadership is meeting in bunkers because they believe their digital comms are compromised. It’s already over.

And, there’s no credible blockage of the Straits of Hormuz.

The longer-term question for Iran is what institutions and leaders step in to fill the void.

Today, Iranians around the world and in Iran are rejoicing. Let’s hope they join the West.

Source: Telegraph

But, none of that has much to do with corporate earnings. A short-term spike in oil is possible. But, that should recede within weeks. We found it suprising that the WH has no plans to release oil from the Strategic Petroleum Reserves.

We talked about the “Iran risk” in our newsletter on 1st Feb, titled Markets at Crossroads and advised caution around risk assets in our 12/31 podcast titled ‘Is Goldilocks Over?’

We discussed 10 other signs that suggested the road ahead for equities wasn’t rosy. 

And, it played out as we expected. 

We also shared a post op-ex risk-off signal in that newsletter worked out well. SPY has lost ~0.6% in the last 1M.

The latest Iran attack will likely lead to a large Monday morning gap down. 

But, history tells markets bounce from the immediate bottom, and the drawdown might get absorbed within the week. 

I agree with a lot of what Alex says here below.

The harder question is — what assets and themes should you own when markets bottom?

And, will the 88-member council of Clerics choose another Supreme leader, or will the Iranians take back control of their government?

We Hit Peak Goldilocks In January

I've been bullish since the start of this newsletter in April 2023.

Goldilocks was my view. 

Disinflation, productivity growth, earnings growth.

That view was correct. 

And now it's fully priced in. 

Everyone owns that position.

That's the problem.

Somewhere between the end of last year and mid-January, we hit peak Goldilocks. 

The sentiment has peaked.  

And now that Consensus crowdedness is unwinding.

Who caused the crowdedness? 

The KOLs. Cathie Wood. Michael Saylor. Tom Lee. Dan Ives. You name it.

The tribal leaders pushing quantum stocks, Robinhood, and Palantir.

They built a thesis. It worked for a year or two. The tribe grew. The trade got crowded. It got expensive.

And now the capital is leaking out.

Here's a useful mental model.

What's the average duration of a relationship between two people before things start to fall apart? 

About 1.9 years.

What's the average duration of a market trend? 

About one and a half to two years.

3D printing. Cannabis. Potash. Quantum. 

The biology is the same. 

After two years, people start to notice the flaws. 

The attention wanders. Their brain wants fresh stimulus. The dopamine doesn’t fire as much.

Then the ‘tribe’ behind the investment theme fractures.

Our own biology may be the reason why trends last 1.5 to 2 years before they get tired and roll over.

People got tired of Internet Stocks in 2000, why can’t the same happen for Memory and Datacenter stocks now?

AI is a real trend — a big secular one. 

But even real trends get crowded and expensive. 

We're at that inflection point now.

The old themes continue to wash away lifting under-owned names to the surface.

Old guard high beta is done. Look at Palantir, SOFI, Duolingo, or Robinhood — all in deep bear markets.

What new themes may rise to the surace?

Main Street.

What's working?

Anti-themes are working.

• Infrastructure ETFs

• Oilfield services: see our pick in FTI, but also see Oil Field Services

• Small cap industrials

• Metals and mining

• Commodities…but seems to be crowded

• Equal weight healthcare names (not Eli Lilly which hit peak sentiment at $1 Tn market cap_

These are names that growth-focused, thematic investors won't think of. 

That's exactly why they're not crowded.

Since Sam Altman’s October interview, we have been shifting from thematic stories to bond proxies– stocks with durable cash flows, and reasonable valuations. 

Markets are now noticing them.

It pays to be a contrarian.

AI MEETS INVESTING

Here’s one way AI can transform investing. Faster insight.

Can you front run Charles Schwab?

Charles Schwab is selling shares in SCHW.

The Walton family is selling Walmart's shares

The President of Caterpillar is also selling his shares.

I figured this all out in 30 seconds with the latest killer feature added to the Lumida Invest app. 

We also found some undiscovered gems with CEOs buying their own shares. 

Insiders sell for all sorts of reasons: tax, liquidity, diversification. 

They buy for only 1 reason…

We are now integrating this data feed into the existing Lumida Bull Bear scorecard and charting AI…

We’re coming for Robinhood. 

Want to try the app? Sign up here.

Lumida is Scaling. Join Us Now And Own The Future Of Investing.

I have some exciting news to share.

Lumida is building the future of wealth management. 

It’s time to move beyond Robinhood’s gamified “Trade, Trade, Trade” model.


It hurts investors and their own data shows it.

(See this tweet from James Chanos on Robinhood’s retail investors - they have barely gained despite S&P 500 producing double digits gains in the last few years). 

The way we look at it, apps like Robinhood, E*Trade, and WeBull are primarily geared to “reward” you for trading. 

They make money on the backend, getting paid for order flow every time you make a move. 

We are investors, and we only do well if you succeed. 

Join our investor community and be part of our growth. We will launch an equity crowdfunding soon.

This is the waitlist. We already have dozens of names, and we’ll prioritize allocations as best as we can.

You’ll see our strong revenue growth (with third party audited financials) soon to follow before you need to make your final decision.

Our strategy is to grow with our community. The community and social media has helped us grow. We believe community is a powerful growth driver.

We’d like you, our valued community, to be a part of that as we seek to disrupt the trillion dollar wealth management industry.

AI Avatars are coming…and I expect we’ll be among the first to disrupt boomer advisors with AI Avatars that can meet you anytime, anywhere and have your best interest at heart.

Macro

HOW'S THE ECONOMY?

The US economy has expanded meaningfully over the last year, and this week’s data show the economy has more room to run. 

Initial jobless claims came in healthy at 212,000, down from last year’s readings. 

Continuing claims also fell to 1.83 million, lower year-over-year. 

We are in a healthy job environment with lower layoffs, and the unemployed are finding it easier to get into newer jobs.

This employment data also reflects positively on higher business confidence.

When businesses are uncertain, the first thing they cut is hiring. 

However, this isn’t the case. Businesses are adding headcount because their forward order books justify it. 

Earnings season confirmed it. 

~75% of the companies beat their revenues estimates, while ~70% of companies performed better than expected on earnings. 

The outperformance came despite higher revisions to estimates after Q3 outperformance.  

This earnings season outperformed the vast majority of quarters over the past two decades, including several that came during genuine economic booms. 

This shows an economy running hot.

The consumer is holding up their end too. 

The Redbook Retail Sales index rose 6.7% YoY — well above trend. 

And, we have more fuel in the tank. 

Tax refunds this season are expected to rise by roughly $1,000, bringing the average check to nearly $4,000, up from $3,100 last year. 

This gives consumers greater room to increase spending, leading to a surge in aggregate demand across the economy.

More jobs. More sales. Better earnings. The economy has no reason to not grow.

Markets

THE AI APOCALYPSE TRADE 

I did an FSD video, titled AI apocalypse, discussing how the AI scare is overblown – it is my rebuttal to Citrini’s article, and where they got it wrong.

Markets are looking at AI as an asteroid. 

Every industry in its path: extinct.

It started with software. 

Then cybersecurity. 

Then payments and financial services. 

Then wealth management. 

Then office real estate.

Now consulting, logistics, and travel. 

Bespoke put together an "AI Doom Basket" spanning all these categories — the basket is down 19% year-to-date.

Doomers think: Anthropic's LLMs can write code better than coders. 

So who needs coders? And Who needs software companies? 

From there, the fear spreads. 

If AI eliminates white-collar headcount, who needs office space? 

If AI automates financial advice, who needs advisors? 

If AI handles freight routing, who needs brokers?

If AI does everything, who needs employees?

The narrative has a certain viral logic to it. It's clean. It's dramatic. It sells.

Citrini’s article is an example.

However, Citrini's article failed to understand how labor markets, business owners, and consumers actually behave. 

Having real world experience matters as an investor.

The actual data tells the other side of Citrini's view.

Indeed job postings for software engineers are up 11% year-over-year. 

Companies are bringing on more engineers to expedite product development. 

AI is helping boost productivity for each engineer, and companies are, in turn, paying their employees more for higher output.

See this tweet by Seth Golden. 

Every time, productivity jumps, we see both margins and wages grow.

This is how it has always worked. 

In 1900, one in two Americans was a farmer. 

Agricultural technology came. 

Farming employment reduced. 

What happened to those workers? 

The invisible hand doesn't move labor from A to B. 

It diffuses it — to B, C, D, E, F, and places no one predicted. 

AI productivity is the same. 

It compounds in millions of small increments across the economy.

That's a harder story to tell than "AI-kills-economy." 

But it's the true one. 

Talking about how the world is ending sells more clicks than a story about incremental enhancements compounding quietly across industries. 

The former is just easier to write.

The jobs at real risk are the menial ones. 

Data entry. Rote monitoring. Basic logistics tasks no one wanted anyway. 

The jobs that survive are judgment-intensive and trust-intensive — chef, architect, business owner, advisor, anyone where accountability matters.

You still need a human in the loop when the cost of being wrong is high.

Did you see how an AI at Amazon AI pushed a bad update and took down servers?

That cost tens of millions. At least.

How much supervision is that worth? A lot.

The AI apocalypse is a crowded trade idea, and sounds ‘easy’. 

Don't confuse an easy story with a true one.

Not all of SaaS is cheap (avoid cybersecurity for instance). But there are bargains emerging.

MORNINGSTAR: THE AI FEAR IS WRONG HERE

Morningstar is down 41% over the last year. 

It landed in the AI Doom Basket alongside financial advisors, SaaS companies, and wealth managers. 

We saw this - and some other names - and said ‘AI can’t kill this business.’

The fear is that AI replaces analysts, kills data subscriptions, and hollows out the business.

That fear misunderstands what Morningstar actually is.

Morningstar's moat isn't its analysts. 

It's its data. And data moats don't weaken with AI — they deepen. 

CEO Kunal Kapoor: "It's hard to build databases. There's the depth, the breadth, the quality, the timeliness, all of that adds up. We've been building our equity database for more than two decades, and every time I think we're done, there's the next thing you have to add." 

A language model can replicate a summary. It cannot replicate 20 years of structured, verified, institutionally trusted data with proprietary collection relationships.

Morningstar isn't defending against AI. It's using it. 

Kapoor has reframed the entire AI opportunity around removing friction: "We know the answers our clients want to get to, so why are we making them click six to eight times? With AI, the idea is remove the friction. Get people what they need quickly." 

PitchBook has already launched AI profile summaries, AI screener construction, and AI earnings call transcripts. 

Data collection is being automated, which means more database coverage without proportional headcount growth. 

AI is making the moat wider, not narrower.

The growth vectors have nothing to do with the disrupted parts of the market. 

PitchBook sits at the center of the private markets boom, providing the essential data layer for the private-assets world. 

Margins for the segment have expanded from 15.9% to 30% in the last two years with earnings compounding meaningfully.

The Direct Platform — institutional analytics and data — continues to grow with a stable, subscription-based client base of asset managers and advisors. 

Similarly, the Indexes business is the fastest-growing index franchise of any size globally.

None of these are getting replaced by AI. 

All of them are growing because of structural shifts in capital markets.

Now the numbers.

Q1 2026 revenue is tracking at $626M, up 7.5% year-over-year.

EBITDA margins are expanding toward 30%. 

The forward P/E sits at 16.1x — the lowest of its 3-year range.

FCF yield is 6.11% - highest in 3-years with a buyback yield of 11.17%.

Management is buying back shares at the fastest pace in company history. That shows their conviction in the business’s fundamentals.

The bear case?

• AI fear could keep MORN down for way longer. 

• PitchBook is exposed to deal activity, and PE fundraising remains challenged. 

• The Direct Platform has decelerated from 8% organic growth to the low single digits. 

But at current growth and valuations, the margin of safety is real. 

Morningstar has compounded uninterrupted for 20 years. The market has handed you a discount to own it.

We bought it on Wednesday after the bespoke report, and it feels like everyone else followed. 

The stock had five consecutive green days in over 6 months.

We bought Morningstar the morning we saw it on an AI doom basket. It rallied every day since.

We are stalking other names on that list too… a credit bureau and an enterprise data provider.

NVIDIA and Datacenters

NVIDIA just printed another monster quarter. $68.1B in revenue, up 73% year-over-year. 

Net Income rose 94%. Guidance for the next quarter came in strong as well – $78.0B expected, 65%+ YoY growth. 

By every conventional measure, it was a clean beat.

The stock still dropped 5.5% on the day, then kept sliding the next session.

The market had its reasons.

The issue isn't NVIDIA's business. 

NVIDIA is an extraordinary company, and Jensen Huang is playing the game at a different level than almost anyone in tech.

But the question the market is now asking is who keeps paying for the datacenters NVIDIA supplies.

On the earnings call, an analyst asked exactly the right question: 

How are your customers going to keep affording these chips when their free cash flow is collapsing? 

Jensen's answer was that token consumption is revenue — that GenAI is driving inference, inference is driving prompts, and that demand loop will generate the returns that justify continued spending. 

That's partially true. 

Cloud spending growth remains strong and is going up every quarter. We don't dispute that.

But a significant portion of that token consumption is internal — coding tools, models being built in-house, experimental workloads that aren't attached to any external revenue line. 

It consumes cost without generating a corresponding return. 

And free cash flow is a real constraint. 

Mag-7 free cash flow is down 90% as capex has consumed nearly all of it. 

You can't indefinitely spend on capex beyond the FCF generation of the customer business.

There's also a structural shift happening that markets are beginning to price in. 

The first phase of AI capex was the one-step jump — companies woke up, realized they needed GPUs at scale, and ramped spending aggressively. 

That happened over the last two years. 

The next phase is slower and more dependent on the actual payback yield from those investments recycling into incremental cloud revenue. 

That's a different growth profile than what drove NVIDIA's last two years.

Think about where each player sits on the value chain.

At the end of the whip, you have the LLMs — OpenAI, Anthropic — which have no real pricing power, are burning cash, and are projecting trillion-dollar revenue figures that the market is increasingly skeptical of. 

Move up the chain and you get the hyperscalers, whose stock prices have been volatile as their margins compress under capex pressure. 

Then NVIDIA. 

Then Micron. 

Then silver, which tracks Micron so closely right now that one of them is probably wrong.

The farther up the supply chain you are, the more insulated your stock has been from this repricing. 

The closer you are to the end demand — the LLMs with uncertain revenue — the more you've felt it. 

We are skeptical of capex receivers and capex payers going forward.

We still own Taiwan Semiconductor and may buy names for tactical opportunities.

DRUCKENMILLER IS NON-CONSENSUS

Druck is a contrarian, even though he says he is not. 

You are all missing this. 

Druck says ‘Never Invest in the Present’. 

He says, ‘Always look out two years’. 

Guess what?

The Present is Mainstream. 

Two years out?

That’s different than the Mainstream. 

That’s vision. That’s ‘what might be’. 

That’s contrarian. 

1. Druck sold the Nasdaq in January 2000.  

Contrarian. 

2.  Druck bought Nvidia in 2022 when ETH mining was dead and gaming demand was weak. He saw AI early. 

Contrarian. 

3. Druck sold Nvidia at the April 2024 local top. 

Contrarian. 

4. He sold AI stocks in the Fall around when Nvidia hit a $5 Tn market cap. 

Contrarian. 

5. He is buying Copper when everyone is buying Gold. 

Contrarian. 

Druck also said technicals are 20% of what he does down from 80% in the 80s. 

Nope. 

I don’t buy that. 

Druck also has said he likes to hold a few positions and ‘watch his eggs closely’. 

Then why 80+ positions. 

Druck is a clever guy. 

He’s not going to give you the magic formula. 

So, what’s the role of Morgan Stanley? 

Why this relationship?

Morgan Stanley is the ‘idea diffusion’ engine. 

If you are early to a trend you need others to follow your thesis. 

That’s the job of Morgan Stanley. 

They market trade ideas. 

It’s copy trading for institutions. 

It all becomes a self-fulfilling prophecy. 

Druck bought into biotech… biotech goes up.  

It has nothing to do with AI adjacency (even though that’s what he says). 

My point is this…

Druck wants to preserve his edge. 

Markets are highly competitive. He is a competitor. 

You need to critically engage with the subject matter. 

Study what he actually does.  

Assess the ideas more rigorously. 

Listen to this FSD livestream, titled Druckenmiller, where I discuss his latest interview, and what to make of it.

The Two Most Important Charts

This is the S&P 500 weight on Technology.

This is free cashflow from Mag 7, the largest weight in information technology, drying up.

The crowdedness in Mag 7 names will take time to unwind.

So... breadth expansion and international names remains the path forward.

Lumida Curations

Why Technical Analysis Isn’t the Edge It Used to Be

Druckenmiller explains why markets today react more to news, flows, and macro narratives than chart patterns — and why strategies that everyone knows stop working.

View Curation

The AI Arms Race Is Already Global

Bill Gurley explains why the real competition in AI isn’t just about who builds the best model — it’s about who captures the users, the ecosystem, and the world.

View Curation

Markets Aren’t Confused About AI

The gap between AI hype and AI profits is showing up in prices, and the market is starting to treat long-term disruption as real while questioning the near-term winners.

View Curation

Meme

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: Walmart Stays Constructive On Economy; Trump Wants Tariffs

• Markets: Midterm Risk; Software Is Value?; Antero Resources: The Energy Play; CLAUDE: Sovereign AI Is Here; Robinhood Ventures

• Lumida Curations: White House Shifts to Section 122; AI as Your Life Operating System; Compute Is the Catalyst

Spotlight

I did a Lumida Non-consensus investing podcast this week, titled ‘Hedge Fund 13F Deep Dive’.

We discuss hedge funds’ latest moves, thematic rotation, and where the next opportunity lies. 

You can watch the podcast here.

(The marketing team really needs to update that 10 years old photo of me, geez! :)

Is AI a Bubble?

We had Michael Parekh this week on our podcast to answer the hottest question of this decade — ‘Is AI a bubble?’.

Michael is one of my favorite analysts in the space. Glad to call him a friend.

We discuss the valuations of hyperscalers, the reality of future returns, and whether the expectations are overoptimistic.

Watch it here.

Why do most managers lag the S&P 500?

One word: trends.

Humans don’t buy stocks. 

They buy stories.

AI.

3D printing.

Genomics.

Crypto.

A trend is ‘story diffusion’ over time.

Phase 1: Non-Consensus, Hated

Every major trend starts out uncomfortable.

In 2023, consensus said Nvidia was a bubble.

The S&P was headed for recession.

China was ‘uninvestable’. 

Crypto was a ‘scam.’

Mag 7 stocks are a bubble. 

That’s the early phase of a powerful move:

Low ownership. High skepticism. 

Strong upside reflexivity.

Phase 2: Momentum

The story spreads. Price confirms the narrative.

More allocators pile in. Career risk flips.

You no longer look reckless owning it.

You look reckless not owning it.

That’s when diffusion accelerates.

Phase 3: Late Momentum

Everyone is in.

Valuation stops mattering.

Quality names detach from gravity.

Think of “can’t lose” stocks trading at 40–50x earnings.

The safety of the crowd feels rational.

Phase 4: Break

The trend rolls over.

At first, investors buy the dip — because that worked in Phase 2 and 3.

But this time it doesn’t.

They add.

They lever.

They rationalize.

Then comes recognition.

The story no longer explains the price action.

That’s when fear replaces comfort.

Most managers lag because they:

Underweight Phase 1 (too uncomfortable)

Chase Phase 3 (too comfortable)

Overstay Phase 4 (too hopeful)

Alpha lives in understanding where you are in the diffusion cycle. 

Trends don’t end because the story sounds wrong.

They end because everyone already believes it.

Should You Buy The Dip?

The old guard of high beta is done. 

Duolingo is down 60% in the last 6 months. 

Robinhood is down 30% from the highs even as the S&P 500 is near record highs. 

SoFi has hit its highs and it has also rolled over. 

Bitcoin is well off the highs. 

Palantir is not near highs. 

Quantum stocks are toast. 

The vast majority of crowd favorite names that worked in 2023 to 2024 are in a bear market.

Retail sentiment drove these stocks beyond their actual valuations, but reality is striking now. 

These crowded stock holdings are drifting back to earth, and are taking investors’ portfolios with them.

The best performing ideas are those that are ignored and under-owned: oil field services, pharma stocks, midstream energy, select staples, and international, small cap value.

I see no reason why that trend won’t continue.

The under-performance of the QQQ vs the S&P and vs the Dow (in that order) is akin to a pinata getting smashed — the capital from the most crowded categories is flowing into its twin opposite.

The problem: retail investors made money in those categories, and now their instinct is to buy the dip because they feel they "know" the category.

Even worse, they are doing it with leverage. This means the losses are proliferated. 

It is time to pivot to high-quality, value-oriented, cash generating assets – a theme we have been flagging since the start of 2026, and hedge fund filings second.

When to Buy the Dip?

Buying the dip has not worked recently. Our recent buy of Coupang and Software stocks are examples of that.

The market will always give you feedback… It’s important to listen to it. We have highly consistent processes. So, when we see that, we treat it as a form of signal.

I created a simple algorithm that helps us know when to Buy the Dip.

You can see it on this chart below.

Buying the dip strategy has absolutely crushed the S&P.

It works.

However, in corrections or bear markets, this strategy gets destroyed.

When the equity curve line is red, the strategy return under-performs the S&P.

When the equity curve line is green, the strategy return beats the S&P.

My thought is to use this as a Regime Indicator potentially in two ways:

(1) When the equity curve line is red, avoid or short high beta.

(2) When the equity curve line is green, own beta. Buy dips, run bull market strategies.

I haven't really seen anyone talk about how to use backtests, and the performance of backtests, to give insight into the current market regime.

I think it makes a lot of sense...

...and you're seeing a lot of accounts blow up that could have avoided that with this handy indicator.

Lastly, this is the type of strategy I want to bake into the Lumida Invest app.

Imagine having an AI that beats the S&P 500 in your pocket.

Still need to test it more, but quite promising isn't it?

We are hiring a few more quants so we can stay on offense.?

Lumida is Scaling. Join Us Now And Own The Future Of Investing.

I have some exciting news to share.

Lumida is building the future of wealth management. 

It’s time to move beyond Robinhood’s gamified “Trade, Trade, Trade” model.


It’s hurt investors and their own data shows it.

(See this tweet from James Chanos on Robinhood’s retail investors - they have barely gained despite S&P 500 producing double digits gains in last few years).

The way we look at it, apps like Robinhood, E*Trade, and WeBull are primarily geared to “reward” you for trading.

They make money on the backend, getting paid for order flow every time you make a move.

We are investors, and we only do well if you succeed.

Join our investor community and be part of our growth. We will launch an equity crowdfunding. This is the waitlist.

You’ll see our strong revenue growth (with third party audited financials) soon to follow.

Our strategy is to grow with our community. The community and social media has helped us grow. We’d like you, our valued community, to be a part of that as we seek to disrupt the trillion dollar wealth management industry.

AI Avatars are coming…

13F Analysis

13F filings came out this week. we have these automatically updated in the Lumida Invest app. 

You can sort to see which manager performed the best or worst. 

Our AI also scans all 13F holdings to look for a setup validated by the trader agent. 

Our LLM agent also does a fundamental bull bear case. 

So, all bases for informed investments are covered. 

Consider testing the Lumida Invest app, sign up here. 

Animal Spirits Are Being De-Risked

Back in Nov, when the S&P 500 sank from its then all-time high of 7,000, we wrote markets were moving towards quality and away from animal spirits. 

The Q4 13F filings second the move. 

We are seeing systematic de-risking in the “animal spirits” bucket.

Capital is rotating away from high-risk, hype-driven names and toward underowned, value-oriented sectors.

Renaissance Technologies is the clearest signal.

A few quarters ago, they were long animal spirits names such as Reddit, Palantir, Hims, GameStop. 

These names traded on narrative velocity and retail participation. 

That exposure has now been sold down.

Renaissance does not trade stories. It trades momentum. 

When momentum works, they lean in. When it turns parabolic or starts fading, they exit without hesitation. 

The same pattern shows up across other quantitative managers.

Quant firm Voloridge reduced Robinhood and SoFi.

Citadel cut exposure to MicroStrategy and Coinbase.

These hedge funds stepping away from animal spirits shows a broader de-risking taking place.

The funny thing is we’ve done the exact same thing - and we did it not know other sophisticated hedge funds were doing the same.

That makes us feel smart, I guess. We want to make sure we’re swimming in the same currents as the best hedge fund investors in the world. No, not Bill Ackman, I mean the quant funds!

It doesn’t mean they are always right - but, they usually are - and we should test our view of the world against theirs constantly.

The Dow outrunning the QQQ exemplifies this move away from animal spirits.

We have been moving away from Animal spirits and into quality stocks since last Nov, and it has paid back well. 

Leaning Away From Mag 7

The 13F filings show a consistent pattern: hedge funds reduced exposure to the Mag 7 complex. 

Meta was the most common exit.

Renaissance and Druck sold Meta completely. Two Sigma trimmed their Meta position.

Meta was a compelling story thanks to its advertising engine. However, the issue is capital intensity.

Meta’s capex has ramped materially to over $100B expected in 2026.

This elevated capex means Meta is burning cash today for a possible return in the future. 

Markets don’t like sitting through these periods of uncertainty, and this is why the stock continues to under perform. 

We also exited our position in Meta in the week after earnings.

Google also saw cuts - Multiple hedge funds, including Renaissance, and Paul Tudor, trimmed their positions.

Over the past year, Google moved from trading near 20x forward earnings to closer to 30x. That re-rating was driven by a shift from “lagging in AI” to “AI leadership.” 

At 30x forward earnings, the debate changes. 

Investors are no longer only paying for a cash-flow compounder trading at a reasonable valuation.

They are paying for future AI monetization (which the market is discounting heavily now). 

More and more, these Mag 7 companies are transitioning from cashflow rich, capital light models to debt heavy and capex intensive models with the prospect of heightened regulation from state, federal, and EU.

Mag 7 names are starting to look like utility companies.

Take a look at this chart.

The days of $1 Tn in S&P buybacks fuelled by the hyperscalars are behind us.

It’s important to recognize these shifts quickly, as they happen, rather than wait to respond. 

This heavy spending might eventually cause the returns to shrink.

We aren’t buyers of capex payers here (although we have a tactical weight on Amazon).

Surprisingly, despite it being on the other side of capex spend, Nvidia also saw outflows. 

We are also of the view that Capex Receivers may be at risk as well.

The Blue Owl news means financing for Datacenters in the debt market is drying up.

Capex payers, like OpenAI and Anthropic, have a capital markets dependancy.

It’s inevitable that datacenter slowdown will impact capex receivers.

Besides, aren’t capex receivers also a crowded concept?

Hedge Funds are going for Dividend and Defensives 

The rotation away from crowded growth has not gone into cash.

It has gone into cash generators.

The 13Fs show hedge funds moving up the quality curve and into cash-flow certainty.

Dividend payers and defensives keep showing up as incremental adds.

Point72 added utilities exposure, including names like Ameren (AEE) and Black Hills (BKH).

Voloridge added AT&T (T) and reduced exposure to higher beta financial-technology names.

Overall, across the quant cohort, we keep seeing the same preference: higher quality businesses, more dividend support, less sensitivity to risk appetite.

We flagged this shift last week on how dividend paying stocks were outperforming the markets, while non-dividend stocks were lagging. 

The 13Fs confirmed the move.

Utilities sit at the intersection of defensives and dividends.

They offer yield support, earnings visibility, and benefit from structural AI demand. And, Hedge funds noticed this. 

Most hedge funds bought utility names during Q4.

Southern Energy is one of our picks in utilities – we bought it at lows in Dec. 

Southern energy is a leading utility provider based in the Southeastern United States, serving approximately 9 million customers. 

They have a growing natural gas distribution segment, and are strategically positioned to benefit from rising energy demands, particularly due to investments in data centers. 

Southern Company reported an impressive 11% YoY growth in electric sales to data centers in the latest quarter. 

They have a robust pipeline of industrial projects totaling over 50 gigawatts. 

Dan Tucker, CFO, stated, "Data centers are a focal point of our growth strategy, indicating strong future electricity consumption."

Southern Energy also has an attract dividend yield of 3.4% which is higher than the industry average.

We sold SO after the rally on Feb 13, taking the gains. The stock continued its move upwards post our sale. We highlighted in green and red vertical lines below so you can see our buy and sell point for education purposes.

However, the technicals don’t look compelling right now for a re-entry.

Both Staples and Utlities have run-up quite a bit on increased anxiety in the market.  

Macro

Walmart Stays Constructive On Economy

Walmart reported earnings this week with revenue up 4.9% (constant currency) and adjusted operating income up 10.5%. 

Management’s tone was constructive on demand, with the key message that consumers are spending.

John Furner (CEO) said “spending continues to be resilient.”

Walmart sits at the center of everyday spend. When they describe resilience, it reflects real basket behavior.

Higher-income households are driving the demand.

Furner: “the majority of our share gains came from households making more than $100,000.”

Lower-income stress is present, but it has not broken demand.

Furner: “For households earning below $50,000… wallets are stretched… paycheck to paycheck.”

That is a real pressure point. 

But it coexists with continued purchasing activity. 

This supports a “two-speed” consumer: higher-income remains healthy; lower-income remains constrained; aggregate spend holds up.

Disinflation is a tailwind.

John David Rainey (CFO) said like-for-like inflation is running “a little bit above 1%.”

That helps real purchasing power. It also reduces the probability of a broad demand shock. 

The consumer does not need wage acceleration to keep spending if inflation stays contained.

Management’s macro posture remained constructive with optimism reflected in guidance.

Rainey: “we are… constructive on the economy.”

They have guided towards 4% midpoint growth in revenues, and 7% in EPS. 

Walmart earnings confirm our thesis on a stable and growing economy. 

As a stock investment, we would still avoid Walmart. It trades at a P/E NTM of 45.1x and a PEG over 3.5. 

Investors’ recent defensive positioning drove a rally in the stock, but we believe it is overpriced. And, we have better opportunities. 

Trump Wants Tariffs

On Friday, the Supreme Court struck down Trump’s tariffs, stating he overstepped his power.

Trump did exactly what markets expected him to do.

He reinstated a blanket 10% global tariff as the replacement. Then he upped that to 15%.

If a U.S. importer was paying ~20% to bring goods out of a country and now pays 10%, that is a direct margin release.

Importers can take it as profit, or pass it through as price cuts to defend unit volumes.

Either way, demand elasticity improves when the tariff wedge shrinks.

Vietnam is a spotlight.

They export over ~$143B, 30% of their GDP to the USA (their largest export country). 

Most of the exports are consumer items and electronics. So, the price sensitivity is usually high.

Lower tariffs improves relative competitiveness for Vietnamese products in US markets. 

It also makes Vietnamese products more lucrative for importers, since lower tariffs give more headroom for pricing flexibility and profits. 

If this tariff reset holds, we believe Vietnam is poised to do good here. 

Markets

Midterm Election Risk

2026 is a midterm election year.

History says volatility rises and returns compress during these years.

Also, we believe the botched handling of the Epstein files creates material risk that Trump will lose the mid-terms.

Trump is directly linked to many themes - AI, nuclear, rare earth materials, space, crypto, energy, financial deregulation, and more. An unwind of the ‘Trump Bump’ should cause volatility as we get closer to the summer.

Since 1945, the S&P 500 has averaged roughly +9.3% annually.

In midterm years, that drops to about +3%.

More important than the average return is the path.

Midterm years also have a higher likeliness for a sharp drawdown in first half of the year.

The average peak to drawdown in midterm years is roughly -17% compared to ~13.7% in regular years.

Half of the 20%+ intra-year drawdowns since 1945 occurred in midterm years.

We are seeing a similar tape play out this year.

SPY has traded in its narrowest range since 1966, and has practically stayed in the same 6,900 region for the last four months.

However, within the index, we are seeing rotation from high growth sectors to value and cash oriented companies.

This shows investors are positioning defensively.

We have been making a similar move in our portfolio by adding bond proxies.

CMCSA is an example. We wrote about it back in Dec, and the stock has rallied nicely from its lows.

We have been taking the gains, and exiting a chunk of our position after the ~35% move.

But, we still think it has legs as it ties to various factors that are doing well.

You can read our thesis here.

Software Is Value?

Software’s forward P/E relative to the S&P 500 is now ~20% below its historical average premium. 

IGV now trades at a lower forward multiple than consumer discretionary and roughly in line with industrials.

We are seeing the worst of the AI scare priced into software.

The IGV ETF has lost roughly ~35% of valuation premium over the last year.

Multiples have compressed hard, but earnings haven’t.

Q4 software earnings have kept pace with expectations, and broadly, we saw no signs of fundamental deterioration. 

That said, the “AI Apocalype” is still in motion. It has spread to cybersecurity stocks, a subset of SaaS, and hurt names like Crowdstrike and Palo Alto Networks. (We were short both going into earnings and remain short as a partial hedge against our other software longs which sport cheaper valuations and farther along the bubble decompression.)

It’s also true that the AI Apocalypse story is not real.

The market priced in an “AI apocalypse” scenario: legacy software disrupted overnight, customers canceling contracts, entire categories obsoleted.

That framing misunderstands enterprise software.

AI implementation is not flipping a switch.

Large companies run embedded workflows across CRM, ERP, HR, finance, logistics, and compliance systems. 

Replacing that infrastructure requires retraining users, migrating data, rebuilding integrations, and reconfiguring internal controls.

That creates inertia.

Users have switching costs not just in dollars, but in learning curves and operational risk. Enterprises do not rip out mission-critical systems because a new AI model demos well.

AI will get implemented. But it will likely be integrated into existing platforms rather than replacing them wholesale.

We are already seeing this.

Software providers are embedding AI inside their primary offerings. 

AI becomes a feature enhancement: better pricing, automation, productivity gains, cost reduction. 

The provider improves ARPU and stickiness rather than losing customers.

That dynamic supports incumbents more than the market initially assumed.

We discussed Hubspot’s earnings in our last newsletter, and they tell us exactly what happens to software after SAAS. Read here.

This is what makes markets complex.

As my friend and hedge fund manager Neal Berger says, “Would you rather be right or rather be rich?”.

I believe the AI Apocalypse story is wrong… but, I’m not going to back up the truck here until I see the market start to shift.

Antero Resources: The Energy Play

The 13Fs show hedge funds are reallocating toward energy exposure, with a clear bias toward cash-flowing U.S. domestic barrels.

Hedge funds are positioning for a power-constrained economy where AI, and LNG exports act as structural demand drivers.

Antero resources was a popular feature across the 13Fs. 

We also bought it recently, alongside AQR, Point72, Paul Tudor and other hedge funds.

AR is a U.S. Oil + natural gas producer in Appalachia (Marcellus/Utica) with leverage to LNG and power demand.

The setup is simple: the market wants free cash flow, inventory quality, and cost discipline — and it wants it at a reasonable multiple.

AR screens that way.

At ~10.9x forward P/E, AR is priced like a commodity stock. 

But, LNG exports and data centers are acting as credible growth levers to drive sharp demand for years to come.

The company generated >$750M of free cash flow in 2025, used it to reduce debt by ~$300M, and still repurchased ~$136M of stock (2% of MCap). 

The fundamental business is also solid. 

Revenues grew 20% in Q4 with improvements in production, and better oil prices.

EBIT also increased 4x from last year, indicating the operating leverage. 

Management guides an increase in production from 3.4 Bcfe/d in 2025 to about 4.1 Bcfe/d in 2026, with optional growth toward 4.5 Bcfe/d in 2027.

It will support strong capital efficiency and margin expansion.

AR has also hedged ~50% of 2026 volumes at reasonably higher price, which reduces left-tail risk while keeping upside if gas tightens.

They also extended their runway via the HG Energy acquisition (adding ~385k net acres and 400+ drilling locations), which is how you compound inventory without blowing up the balance sheet.

The risk is similar to other energy producers. If oil prices decrease sharply, it might hurt revenues and earnings. 

Execution of HG energy also needs to be on point to ensure cost synergies are realized. 

However, at a ~10x P/E multiple, most of the risk is already priced in. 

Technically, we believe this is in a bull market. Wait for a good entry, don’t think it’s this Monday or Tuesday.

Typically there is equity index weakness immediately after quarterly options expiration that lasts about 2 days or so.

CLAUDE: Sovereign AI Is Here

The WSJ reports that the Pentagon Used Anthropic's Claude to arrest Maduro.

LLMs are moving from 'enterprise SaaS' to statecraft.

Consider that Pete Hegseth wants to increase Defense spending by 50% to $1.5 Tn.

How much of this is going towards AI?

The single most important datapoint to underwrite datacenter capex is this:

What % of the backlog and future demand is represented by the U.S. Government?

Whichever sovereign achieves "omniscient and omnipresent" battlefield awareness has an extraordinary competitive advantage.

The Department of Defense is price insensitive. 

They can afford to pay for a massive capex build out as compared to more frugal S&P 500 CFOs.

Note: The FY 2026 Pentagon budget has ~$14 Bn ear marked for AI spending.  

That's not a big number.

But, that number doesn't include classified line items. 

And, the Pentagon still can't account for a trillion dollars in spending on failed audit tests.

A lot of this 'Is Datacenter Capex a bubble?' question rides on how much of the $1 Tn revenue projections is linked to the U.S. government.

This might be the single most important question to underwrite 'Is AI a bubble?'

Robinhood Ventures

1) Databricks is SaaS. Its private mark is absurdly high vs public comps. 

I passed on Databricks multiple times. 

2) They are offering Ramp. 

Ramp’s competitor, Brex, was just bought at $5 Bn by Capital One. That’s 50% lower than their price in 2021. 

These are direct comps. Ramp will have the same issue. 

3) Stripe just raised at $130 Bn

But, every payments company in the world is in a bear market - including darlings like Adyen. 

4) Robinhood’s first pre-IPO deal Figma is down 80%

Contrast this to Lumida

1) Our first deal Coreweave is up multiples

Everyone made money. 

2) Another deal, Brad Jacobs QXO, also made money

3) We are investing in Defense Tech - front-running the Pentagon…

Fundamentally, the difference is this:

Robinhood is a broker. They are pushing products. 

We are investors. We want to make money.

Lumida Curations

White House Shifts to Section 122

After the Supreme Court blocked Trump’s reciprocal tariffs under IEEPA, the administration is pivoting to Section 122 authority to impose a 10% global tariff—aiming to offset lost revenue and keep 2026 tariff collections largely intact.

View Curation

AI as Your Life Operating System

Alexandr Wang describes personal super intelligence as an AI extension of you—understanding your goals, tracking progress, and proactively managing health, work, and daily tasks so you can focus on what matters most.

View Curation

Compute Is the Catalyst

AI training compute has grown nearly a trillion-fold in 15 years—with another ~1000x expected soon—driving models that already outperform most humans at coding and accelerating the redefinition of machine intelligence.

View Curation

Meme

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: How’s The Economy?

• Markets: There Were Signs; The AI Apocalypse Trade; Hubspot Earnings; Investors Are Positioning For Cash; More Trouble For OpenAI; WSJ is the contrarian Indicator; The Bubble In Venture Capital; Trading As Gambling?

• Lumida Curations: The AI Ads Debate Goes Public; AI Agent “Taught Itself” a New Skill; Olive Oil Is the Anti-Inflammation Hack

Spotlight

This week, I met Angelo Robles to talk about the end of the KOL-driven trade and why high-conviction tech and crypto portfolios are getting repriced. 

We also discuss the rotation out of U.S. mega-cap leadership and into international markets, banks, and value. 

Here’s what we cover:

• The decline of U.S markets versus rest of the world.

• Banking as an attractive investment sector. 

• Why Bitcoin trades like the high beta, and reasons behind near-term weakness

• Investment themes - nuclear renaissance, biotech, semiconductors/AI.

• Public market opportunities related to AI advancement.

• Our predictions on the top 10 global companies in 5-10 years.

Watch the full podcast here.

ANTHROPIC CAPEX BUST?

The CEO of Anthropic is as delusional as Sam Altman.

He is projecting $1 Tn in revenue by end of the decade and 10% GDP growth.

The promise of AI is real… but this thought process betrays a lack of understanding.

S&P 500 CFOs are not going to tank their earnings and fork over $1 Tn to capex spend.

They see what happens to their stock price. And, Dario ignores the power of price competition in these LLMs that have no real moat.

I thought maybe it was Sam Altman having an ego complex.

No, this is industry wide delusion. And, it squares with our view that private venture capital markets are in a major bubble.

I did a 3 part twitter thread discussing Dario’s interview, and its implication on the broader market. I recommend reading it here. I

Mag-7 Go Capital Intensive

This week, Alphabet raised $20 Bin its biggest-ever U.S. dollar bond sale. The issue was upsized from $15B and drew over $100B of orders. 

Google is also lining up issuance in Switzerland and the UK, including talk of 100-year bonds. 

Google needs this financing to fund its $180B capex plan in 2026, and ahead.

Overall, the Mag 7 are going from capital light model to capital-intensive.

The balance sheets that used to be “pristine” are getting levered to buy compute. 

Hyperscalers went from using around 30% of their OCF in capex to an expected 92% in 2026. 

Capex is tricky for stock performance. 

You have an outflow today that will have a possible return in the future.

Even if the ROI is good, the market has to sit through the cash burn first.

The uncertainty is discounted heavily if markets fear the returns will be lower than expected (as the case is today). 

Market sentiment has turned weaker for the Mag-7 in the last couple of months.

They have underperformed the benchmark, and are down 4.25% in the last 3M.

But, from hyperscalers’ perspective, the capex is fully rational. 

You can’t “wait and see” in the AI model race. 

Your target is to be #1 or #2. Falling to #3 puts the core business at risk. 

Hence, spending $150B in CapEx to protect or grow a couple trillion of market cap is rational. 

It is a positive expected value decision. You spend to stay relevant.

However, the risk is overinvestment – a movie we have seen before.

Housing in 2006–08. Dark fiber in dot-com. 

When everyone builds at once, you can create excess capacity. 

Even the “obvious” winners can correct when the trade gets crowded. Fundamentals can be right and the stock can still be wrong on timing.

This is why we are seeing the broadening reaction in markets. The capital is rotating from Mag-7 into value and international plays. 

Take a look at Google’s chart… showing weakness after winning the Consensus AI crown.

Side note: this corporate re-leveraging is going to keep rates higher for longer. 

Sovereigns are already highly leveraged post-2008. Inflation is sticky. Long rates are already elevated. 

We remain of the view it’s time rotate away from bonds, and into high value, cash generating stocks.

I did a FSD stream this week, titled ‘Mistakes were made’, discussing the near-term markets volatility, the outperformance of bond proxies, vix decay and how to position in current market regime. Watch it here.

You Only Eat After Tax Returns

This was a beautiful email for a client.

Wealth management is not unlike a physician practice. Instead of tending to health, you are tending to wealth. We take great satisfaction in helping our clients acheive their goals.

Raymond is a Lumida licensed advisor who left UBS and Fidelity. He wanted to be a part of the trend to modernize wealth management.

I love client emails like this. 

‘I felt trapped in the position… I didn’t want to sell… high marginal taxes. Thank you. You saved my ass.’

We had a new client join recently. 

They had a boatload of unrealized gains. 

We helped them exit and keep more of their hard earned wealth. 

Remember, you only eat ‘after tax’ returns. 

Now, this client has more to save for his retirement, his kids’ education, and his family’s future. 

They are one step closer to living the life they want to live.

You can sign up for our wealth management services, where we optimize for after-tax returns. Schedule a free consulting call here. 

Macro

How’s The Economy?

This week’s data releases show US economy is on a solid footing. 

Labor market data came in stronger than expected. 

• January payrolls rose by 130,000 versus expectations of 65,000

• The unemployment rate ticked down from 4.4% to 4.28%

Interestingly, this happened despite the highest labor force participation since 2000. This shows hiring is gaining momentum. 

Businesses are confident in the economy, and are ready to invest more. 

Q4 earnings support the thesis. 

84% of companies have performed better on revenues than analysts' estimates, and 70% have beaten on earnings. 

This outperformance is even more remarkable considering it came after upward revisions during Q4’25. 

The other major economic gauge is inflation - it also came in better than expected. 

Inflation was 2.4% YoY in January, coming in lower than Dec’s 2.7%. This was the lowest inflation print since 2021.

Productivity growth through AI and automation are helping higher output at a lower cost. This reflects in net margins reaching all time highs. 

We can expect productivity to go higher from here, as AI is implemented on a wider scale.  

Overall, we have a resilient labor market, positive earnings surprise, and lower inflation - what else would you expect from a healthy economy?

Markets Have Now Priced in Goldilocks

At the same time, Markets have now come around to our view that we have a Goldilocks economy. Take a look at one of my favorites, the Citi Economic Surprise Index.

That spike? That’s when markets realized the economy is good. And, it started to price in inflation concerns.

What’s rallying? Commodities, energy linked names, and consumer staples.

Markets

There Were Signs

SPY had three red days this week. Did you see it coming?

There were signs

1. Shay Boloor took the opportunity to dump on his followers citing paying for his marriage as rationale

These stocks are down massively.

2. A prominent SoFi booster got a job at SoFi and went dark. 

This is ‘narrative cover’. 

A convenient way to dump shares and not disclose. 

SoFi is in a bear market.

3. Fortress and Citadel invest in XRP venture deal at a stupid valuation. 

4. A bunch of non-bankers with no experience in Lending start a bank called Erebor at a stupid valuation. 

5. Dozens and dozens of crypto DATs went public. 

The vast majority of these tokens solve no real world problem. 

6. IPO Season Fatigue

A deluge of IPOs hit the market. 

Many drop below IPO price, burning investors and dampening animal spirits. 

7. Robinhood offers a first of its kind pre-IPO deal (Figma) at a stupid valuation. 

If promptly tanks 80%. (Figma still has a 100X PE ratio.)

8. Competing ‘thematics’ - drones, robotics, AI, crypto - fragment attention markets. 

SPACs proliferate fragmenting focus in a marketing frenzy. 

9. Economic populism - on the banks, on pharma, on housing, on private equity, on defense stocks - hurts confidence. 

10. Sam Altman states that he has $1 Tn in capex spending obligations… and he’s got hundreds of billions in RPOs (eg, debt) and equity all over the tech system.

If you read our newsletter two weeks ago, you’d remember I flagged the incremental negative news.

The Dow Jones is making new highs - atoms and materials - not bits and bips. Value is beating growth.

The Nasdaq continues to lag vs Small Caps and the Dow Jones:

I also continue to find people that are trying to ‘buy the dip’ on two to three year old themes… when these themes appear to have broken in unison.

They are going to give back their unrealized gains.

The key suggestion here is : Rotate, Rotate, Rotate.

Another… I noticed we bought Coupang and Software too early. They failed to bounce where they are supposed to.

And, plenty of names are doing the same: Amazon is below the 200 DMA, Microsoft is under pressure, Netflix is in a bear market, Apple’s chart looks terrible, and Tesla is in a downtrend. (We have a short on Tesla now.)

The general point: when stocks fail to bounce at key technical levels, that’s the market giving you a key signal.

It’s not that we were “right or wrong” to buy Coupang and Software at those times. The process said it made sense. But, now we are getting market feedback that we are in a new regime.

The last time we saw this phenomenon was in March of last year during the tariffmageddon.

I do think the risk of a potential correction is heightened.

The S&P we believe will struggle to get over 7,000.

What’s working? Commodities and memory stocks. But, aren’t they the same trade? Compare these two charts (MU and Silver).

Now, doesn’t DRAM use Silver as an input? How can both of these continue to go up?

Isn’t it the case that these are both rallying on momentum? And, perhaps that momentum is getting crowded?

Remember what happened to Hershey’s when Cocoa prices went thru the roof?

How can both of these go up together?

All to say, there are imbalances in markets, and markets will correct these in my view.

The fundamental imbalance is ‘Crowdedness’.

That’s why I believe we’ll see many KOLs lose reputation this year. It’s not that they are ‘right or wrong’ on some fundamental thesis.

It’s that they have caused crowdedness. How do you flip on a thesis you’ve been pounding the table on for so long?

I don’t think these KOLs can step out of their own bubble and do so. And the pressure from their ‘tribe’ for doing so would be like a scalping.

Look at the pressure Nic Carter faces from the bitcoin community.

Who are these KOLs? Cathie Wood, Michael Saylor, Tom Lee, and several tribal leaders on twitter who are pushing quantum stocks, Robinhood, and Palantir.

The AI Apocalypse Trade

Markets are looking at AI as an asteroid that will kill any industry that comes remotely close to AI implementation. 

We saw several examples this week. 

1) The first was trucking companies. 

C.H. Robinson (CHRW) fell 15% and Expeditors (EXPD) fell 12.5% in a single session. 

It occurred after Algorhythm Holdings announced they have developed an AI tool which can help scale Freight without increasing headcount. 

Investors rushed out from legacy players into the the AI product, causing Algorhythm Holdings (RIME) to gain 79% on announcement. 

2) We saw a similar scare in Office REITs (notice how far the effects are spreading).

The sector hasn’t caught a break since COVID, but the fall narrative has evolved from “work-from-home” to “AI reduces white-collar headcount.” 

The category is now trading at 2009 lows. 

3) Financial services also got the hit from a similar scare. 

Altruist introduced ‘Hazel’, an AI platform to automate and streamline traditional advisor services. 

This caused a sharp sell-off in any financial company that had a large team of advisors- LPL Financial was down 12%. Charles Schwab dipped 7%, and even Morgan Stanley lost ~3%. 

Financial Services ETF (IYG) slipped below its 200-day moving average after the news. 

We are already seeing the worst scare in software names, with IGV having lost over ~35% of its valuations in the last 1Y. 

Overall, we believe the sentiment around the AI scare is overblown now. 

AI implementation isn’t like “flipping a switch”. Companies have embedded workflows that will need to be replaced.

Users have inertia in learning and execution, which leads to higher switching costs.

AI will get implemented eventually, but it would involve legacy systems integrating AI within existing processes. 

This will mean existing providers improve prices, costs and productivity rather than them losing customers.

We are already seeing it in software names, where AI is now actively integrated in the primary offering.

Hubspot Earnings: AI fears are Overblown

HubSpot’s latest quarter came in strong: sales and earnings grew ~20% YoY,  and margins held firm at 18%. Customers count also grew by 16%

The earnings tell us exactly what happens to SAAS after AI. 

Hubspot’s net revenue retention rose to 105%. This means existing customers are spending more with the company.

Unlike market fears, customers are not leaving Hubspot for AI providers.

Instead, businesses are adding more seats and more products to their existing contracts. 

The Reason?

HubSpot is providing the AI features customers want inside the products they already use. 

Yamini Rangan (CEO) emphasized that they have “embed AI across the platform,” making it part of daily user workflows rather than a standalone tool.

Customer usage data supports that view. 

HubSpot said “50% of Core Seat users have tried and are using [the AI assistant].” 

This shows AI is becoming part of SAAS rather than replacing it. 

We believe this is exactly how AI gets adopted at a larger scale – legacy providers will integrate AI within their offerings, so customers can easily use AI offerings without any switching costs.

We added to Hubspot post-earnings, and they were up 6.5% on Friday.

The fundamental business is solid, growing double digits with operating margins at ~20%. 

They are gaining share against Salesforce. Lumida has also recently transitioned to Hubspot for our marketing activities. 

The valuation metrics on HUBS have also never been cheaper. 

Investors Are Positioning For Cash

Since Nov, we have been talking about rotation out of growth, and into cash-rich, high value assets. 

The data makes it clear.

Dividend-paying stocks are decisively outperforming year-to-date, while companies with no dividends are lagging. 

The same pattern appears across valuation metrics: stocks with lower price-to-sales ratios are outperforming the highest-multiple names. 

This highlights investors are moving away from expensive growth exposure to better valued companies. 

High-dividend and value-oriented names offer earnings visibility, free cash flow support, and downside protection at a time. 

PAGS was a Lumida 2026 pick with a dividend yield of ~11%. The stock’s up about 14% in 2026. We added to it on Friday. 

PAGS has evolved from a payments processor into a full-stack fintech platform serving Brazilian SMBs and consumers through PagBank. 

Payments remain the entry point, but the real value sits in the attached banking services: deposits, secured credit, insurance, and investments. 

That ecosystem has allowed PagSeguro to grow earnings while many global fintech peers have struggled with funding costs and margin compression.

PAGS trades at roughly 6.4x forward P/E, and under 1x sales. It generates exceptionally high free cash flow, with FCF yields north of 20%. 

Margins remain strong, with EBITDA margins in the mid-40s, and high operating leverage. 

Management has been conservative, focusing on secured lending and prioritizing profitability over aggressive market share gains. 

That has kept losses contained even as Brazilian rates stayed high. 

As rates are expected to normalize, operating leverage could surprise to the upside.

The risk is clear: Brazil remains competitive and macro-sensitive. The competition from Nubank and StoneCo can also hurt long-term growth. 

But at today’s valuation, the market is already pricing in a lot of bad news.

More Trouble For OpenAI

OpenAI’s troubles have become a regular feature for this newsletter, and this week wasn't any better for them. 

Anthropic has planned to raise $30B at a valuation of ~$380B in Series G. 

While this raise will further enhance their ammo against OpenAI, the more concerning report was how they are already winning market share.

Business adoption is no longer moving in OpenAI’s favor. 

Ramp data shows overall business AI adoption hit 46.8% in January, but Anthropic penetration rose to 19.5% (from 16.7%) while OpenAI dipped to 35.9%. 

Roughly 79% of Anthropic customers also pay for OpenAI, and 16% of businesses now use both platforms, up from 8% a year ago. 

That points to a buyer mindset change. They are no longer just using OpenAI, but are exploring other options alongside it.

That weakens OpenAI’s leverage in renewals and pricing, because it is easier for customers to shift workloads when outcomes look comparable.

On the consumer side, Gemini has a structural distribution edge. 

Users do not need a migration plan. They just click what is already embedded in Google’s workflow. 

That kind of default placement reduces inertia and can quietly drain usage, even if OpenAI’s product is strong.

Then, there is DeepSeek. 

OpenAI reportedly warned U.S. lawmakers that DeepSeek is attempting to reverse engineer its models. 

It says they are using distillation methods to bypass access restrictions through masked routing. 

If that dynamic persists, we might see more competitors with similar capabilities flooding the market. This will accelerate commoditization, and hurt pricing power. 

OpenAI now faces a twin squeeze: softer demand power from multi-vendor adoption, and higher competitive pressure from lower-cost models.

Things aren’t looking good for OpenAI. 

The Bubble In Venture Capital

Stripe is expected to raise at $140 Bn

Stripe raised at ~$86 Bn valuation at a top of the market round in 2021.

Adyen, a fintech darling, peaked at $32 in 2021. 

It dropped 24% this week alone and is now at ~$11.

Adyen is down 65% over the last 5 years in public markets - along with every single payment stock.

Stripe is up 62% over this time frame... even as competition intensifies from Tempo, Circle, etc.

Do you really think Stripe is worth $140 Bn?

Maybe it's worth $40 Bn?

Trading As Gambling?

Robinhood users have made very little money since the meme-stock era ended.

We see some spikes during the risk-on phase, but sharper drawdowns every time. 

It’s usually the same for other retail trading platforms. 

These apps are optimized for trading - the more traders trade, the more money the app makes. 

The AI apocalypse is coming for RobinHood, SoFi, and the legacy wealth management complex.

And, Lumida is leading the charge.

Lumida Copilot is not designed to maximize clicks or trades. It is designed to help investors think. 

Users get AI bull/bear indicators, structured theme views, and institutional-style positioning signals — tools that show when the wind is at your back or in your face. 

The goal is simple: have an informed AI partner for your everyday investment decisions. 

Check out this feedback from users testing the Lumida Invest App

Sometimes You Can Just Do Things.

If you’d like to get on our external testers list, sign up here. (It’s free!)

Lumida Curations

The AI Ads Debate Goes Public

A fresh exchange between Sam Altman and Anthropic highlights a growing divide over whether advertising belongs inside AI products — and what that means for user trust, privacy, and future monetization.

View Curation

AI Agent “Taught Itself” a New Skill

An anecdote from OpenClaw’s founder describes an AI agent independently figuring out how to process voice input. This shows how fast autonomous AI capabilities are evolving beyond their original design.

View Curation

Olive Oil Is the Anti-Inflammation Hack

New evidence suggests that moderate daily intake of extra virgin olive oil can produce mild anti-inflammatory effects, supporting joint health and recovery through mechanisms similar to common pain-relief pathways.

View Curation

Meme

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Markets: Zoom Out; Growth To Value Continues; The Bubble In Venture Capital; Databricks Wants To IPO; GS: Software Is The New Newspaper; How Does OpenAI Compete With Google?; 2026: The Trough For Managed Care; An International Value Stock Pick

• Lumida Curations: The Digital Twin Stack Goes Real-Time; Dollar Weakness Is Optics; Dry Sauna Is the Longevity Baseline

Spotlight

I did a FSD live stream, this week, titled ‘How Sam Altman Broke the World’. I recommend giving it a listen. I’ll identify the high level ideas below.

We are in a new market regime, and it started after Sam Altman’s disastrous interview in Oct where he noted OpenAI has $1 Tn in committed spend obligations across the datacenter and semiconductor ecosystem.

I explain how you should position the current market regime in this FSD stream. Watch here. 

How Sam Altman Broke The World?

OpenAI has $1 trillion in committed spending. 

A more common name for committed spending is ‘Debt’.

Now the reason these spending obligations—what OpenAI owes to Oracle and Microsoft—don’t show up as debt on the balance sheet is because they are a form of off-balance-sheet financing. 

It’s not a loan in the accounting sense.

But functionally, OpenAI has an obligation to pay out roughly a trillion dollars over the next couple of years. 

So how does OpenAI fund it? 

OpenAI can fund the capex from cash flow from operations (CFO), or they can fund it through raising capital (CFF).  

Let’s look at the revenues first. 

OpenAI generated $23 billion in revenue last year.

Let’s say somehow their growth curve inflects higher than Gemini. (Gemini is growing faster than OpenAI, November usage of GPT shows a decline[!!]).

Let’s say OpenAI does a miraculous $100 Bn next year. 

Let’s assume that’s pure profit. (Hardly, these are massive cash burn enterprises).

It’s still not enough money to fund a trillion-dollar obligation. Not even close.

OpenAI will close a $100 Bn round soon — led by its suppliers and SoftBank. That’s still not enough.

OpenAI will go public and raise $100 Bn. That’s still not enough.

Add to the picture the competitive landscape.

Freemium pricing models from Google and Meta are pushing pricing power lower.

XAI is priced at zero. Gemini is priced at zero. You also have Microsoft Copilot, which is taking enterprise business away from OpenAI. 

OpenAI wants to roll-out ads, but Anthropic is making fun of them. Now, Sam Altman is walking back ads a bit.

Competition is the enemy of high margins and rapid growth.

So, OpenAI is going to have a significant gap to fund these obligations. 

So, if OpenAI hits revenue and raise gets you to $200 billion, now you have $800 billion more to go. 

OpenAI is forced to go public to raise money. 

This is the recognition moment markets are going through. 

Microsoft dropped 10% on earnings because half of their backlog is tied to OpenAI. 

Broadcom and Oracle also have over $100 billion in RPOs tied to OpenAI. 

Those firms have downstream providers in the datacenter space linked to revenues from OpenAI.

Many firms made capex investment decisions based on trusting Sam Altman.

So here’s my theory: 

OpenAI went one by one by one to different providers and said, “Hey, look, I have massive compute demand. Can you help me build a data center?”

One a one-off basis, each decision to invest makes since. In aggregate, however, it looks like madness.

So, multipe providers committed $100B+..

OpenAI was worth $500 billion in September and $86 billion a year before. 

The providers would have looked at OpenAI’s growth, and equity, and thought OpenAI will have the capital and customer demand to pay the bills.

But, then Gemini started getting traction. And, equity values in Mag 7, Nvidia, and the datacenter space experienced a correction.

(OpenAI’s stock price went ‘higher’ because they pick their valuations - it’s not real.)

My hypothesis is no one knew OpenAI was in bed with everybody. 

It’s because If you look at their balance sheet during counterparty diligence, you wouldn’t see that off-balance-sheet liability show up as debt, because it’s not debt. 

It’s a contractual obligation. 

We saw some of this with Enron, by the way. 

Enron had a lot of off-balance-sheet financing. 

I’m not saying OpenAI is Enron. 

I’m showing you how investors can be misled. 

So, cumulatively, OpenAI signed up hundreds of billions of dollars in revenue performance obligations across the datacenters. 

For compute providers, these obligations show up as an asset – accounts receivable – on their balance sheet. 

As these providers deliver the services, they’d recognize the revenue. 

That’s how these contracts are supposed to work. 

But here’s the problem. 

OpenAI doesn’t have enough cash flow—by any of the methods we discussed—to pay those obligations. 

The datacenter providers are now realizing this. 

What did Oracle do? 

They had to raise $25 billion in debt and $25 billion in equity.

Why? 

Because they have the potential of bad debt expense vis-à-vis their commercial arrangement with OpenAI. 

The U.S. equity market is about $60 trillion. A trillion dollars in OpenAI spending is big. 

OpenAI’s debt is a trillion dollars socialized across the ecosystem. 

And how much of that is actually worth its face value? Those commercial obligations have real risk of default.

Each OpenAI contract corresponds to an accounts receivable with its customer. The debt does not show up on OpenAI’s balance sheet, but the asset does show up on the statements of its suppliers as Revenue Performance Obligations (or accounts receivables).

Why did Microsoft stock drop 10%?

The hundreds of billions in ‘assets’ linked to OpenAI aren’t in fact worth face value. They are worth a fraction of face value. That causes an impairment to equity.

Markets wiped over $300 Bn in Microsoft equity due to these RPOs.

Similarities to 2008

I remember seeing this back in the 2008 crisis.

Big investment banks - I was the youngest VP at Merrill at the time - had hundreds of billions of ‘assets’ backed by suprime loans.

Those loans weren’t worth par value. Writedowns and impairments to equity started.

That hurt equity values.

This is why Oracle is raising equity capital.

I would be cautious on the datacenter space right now. We still have highly targetted ‘clean’ exposure - names like Taiwan Semiconductor, and Flextronics. But, we’ve reduced our exposures quite a bit to names that directly touch OpenAI.

We’re keeping this sector on a tight leash.

That doesn’t meant that we aren’t in a massive and early cycle of investing in thousands of new GPU led datacenters.

Both statements can be true.

How much Market Cap Is Linked to OpenAI?

A lot.

If we look at OpenAI's spending, most of the money flows back to NVIDIA. They capture the lion’s share in semiconductors and data centers. 

It’s a well-run business, incredible CEO, dominant competitive advantage—75% margins. 

I assume 75% of OpenAI’s $100B raise will flow back to NVIDIA, and with 75% profit margins, you’re talking roughly $50 billion in earnings. 

Put a 40x multiple on the $50B earnings and that’s $2 trillion in market cap from Nvidia’s 4.5T.

If the market starts discounting the probability of those earnings, Nvidia gets discounted.

And, it’s not only Nvidia, it’s across the AI flywheel – all the way from chip producers to power companies. 

Nvidia wants to make sure OpenAI meets its commitment, and it has decided to invest in the current round. 

But, this isn’t a Nvidia-like decision. It would have been smart for NVIDIA to invest in OpenAI when it had a cheap valuation. 

NVIDIA is buying an asset that’s not actually worth $860 billion. 

So what’s it actually worth? Maybe $100 billion. It’s not worth 80 times price-to-sales. 

Especially with a growth rate that’s declining and they’re no longer the leader because Gemini is taking share. 

Maybe it’s 20 times if you’re generous. Maybe it’s 15. Maybe it’s 10. 

So if NVIDIA is showing up at an $860 billion valuation, the true value they’re acquiring is closer to $200 billion. 

They’re losing on that transaction, but they do make sure the commitment spending is met, and they get their earnings at the back end. 

But net net, what’s happening is NVIDIA is accepting lower margins. 

They’re paying for the future revenue by buying inflated company stock at a premium.

Investing equity stakes made sense when they were cheap because you got the double benefit: the AI companies grow, you mark up the position internally, you win both ways. 

But when the asset is overvalued, you get the opposite effect and it starts to hurt margins. 

That’s the issue.

I’ve been on this for the last two months. 

The bubble is OpenAI.

What If You Had A Copilot?

Markets are tricky - you need to be accurate and quick in every decision.

But, there are thousands of stocks in hundreds of themes, and sectors.

Keeping a constant track on all of them seems impossible.

But, what if you had a copilot.

That’s what Lumida has been cooking.

Lumida Copilot is an investment super app that is your one-stop shop to investing.

It provides you AI fundamental and technical analysis, regular news updates, themes tracking, hedge fund monitoring and more.

Our portfolio connect feature lets you connect your retail portfolio to the Lumida app, and see which factors and themes you’re most exposed to.

You can sign up here to become one of the first few external testers for the Lumida Copilot.

Markets

This is your friendly reminder that 'old guard high beta' names are in a bear market.

That includes darling social media stocks that did quite well and have since given up 30 to 40% of their returns in a few months.

The vast majority of crowd favorite names that worked in 2023 to 2024 are now in a bear market.

If you made money in those categories, your instinct is to buy the dip because you "know" the category or have some unique insight and feel that confers an edge.

Now is not the time to be ideological and fall in love with your thesis.

You should be making, or have already made, hard pivots in your portfolio to new themes you likely have had zero exposure to before.

That means:

- value stocks

- international stocks

- commodities and materials

- select consumer staples

- energy 

I bought a British Tobacco company (BTI) recently for example.

You need a 180 degree flip in thinking.

Growth To Value Continues

Back in Nov, we noted markets were shifting from growth to value. The transition is becoming mainstream now.

Growth has sold off hard since new year, briefly declining to its April 25 lows. 

On the contrary, value has posted steady gains, leaving the performance gap far narrower.

The recent rotation has been unusually fast. 

Over a six-session window, value has outperformed growth by more than seven percentage points—an outcome that sits in the extreme tail of historical observations. 

These moves have previously occurred during the Financial Crisis and the late-90s unwind. 

This shows you investors are tired of paying up for high valuations when global growth is on.

We used to say ‘There is No Alternative’ (TINA). Now we have alternatives all over the place. We should call it TACO - There Are Countless Options.

The near-term preference is in assets with high cash flows, stronger balance sheets, and pricing power.

We wrote about CMCSA, Manulife Corp and other value stocks in our Lumida 2026 picks.

We expect more boring names to do well in 2026, especially international stocks.

Read here. 

THE BUBBLE IN VENTURE CAPITAL

What you are seeing unfold is a massive unwind in the venture capital market. 

Private market valuations bear no resemblance to reality. OpenAI is the poster chiled.

The Brex take under 50% of 2021 valuations is evidence of that. 

IPOs of dozens of VC darlings that tank is another. 

Companies staying private and avoiding the mark to market discipline of private markets is another example of that. 

75% of 2021 IPOs are below their IPO price (despite 20% inflation since!)

Over $1 trillion in private capital is now tied up in unicorns valued on internal marks, not clearing prices. 

Secondary markets trade at steep discounts.

It’s price discovery - delayed, painful, and unavoidable

DATABRICKS wants to IPO. 

They are valued at $134 Bn.

Meanwhile, SaaS multiples have crashed to earth. 

Do you think Databricks is worth $134 Bn in public markets?

No. 

Venture capital is making the world a better place for all of us. 

The tech, innovation, and moonshots… 

we need that. 

But, this is another example of the bubble in private venture markets. 

Maybe Lumida should open up an outsourced CIO service to help big money managers navigate. 

We see, study, and participate in ALL of the asset classes: public markets, pre-IPO, private credit, venture, commodities, etc. 

That gives us a unique perspective and ability to synthesize. 

(Helps that we are obsessed as well.)

GS: Software Is The New Newspaper

Goldman comparing software to newspaper stocks is about as bearish as it gets for IGV. 

It is a snapshot of where sentiment on SAAS has landed.

Analysts are flirting with “structural wipeout” language after a violent de-risking. 

Historically, that’s the type of framing you hear when expectations have already collapsed and positioning has already been washed out.

The data supports that read. 

Hedge funds have been dumping software and cutting risk aggressively, with net leverage compressing from roughly 16% to about 2%.

Prices and multiples have already adjusted. 

Software valuations have de-rated sharply on price-to-sales, falling from roughly 9x in September 2025 to about 6x today. This places the sector around the middle of its historical range since 2010. 

Importantly, the sector still trades at a premium to the equal-weight S&P 500 on P/S—about a 260% premium—roughly in line with the historical average. 

The real difference shows up in earnings valuations. P/E NTM have shrunk to 2014 lows, with the premium to S&P 500 nearing its all-time bottom.  

Software names have high margins and operating leverage to drive earnings higher even with lower sales growth.

Overall, software stocks aren’t dirt cheap. Many still are pricer than the S&P 500 (like Mongo DB). Others are much cheaper than the S&P 500 (like Salesforce and Adobe).

But, there is light at the end of the tunnel.

Software’s last 30-day rate of change is among the worst readings since 2008. 

In the few prior instances when downside momentum was comparable, the group was higher six and twelve months later each time. 

Overall, software has gone from crowded to avoid. 

Exposure has been cut. Leverage has been reduced. Multiples have compressed. Commentary has turned apocalyptic. 

Those are the ingredients that typically precede stabilization.

Lumida added to software exposure on Friday. We were early to purchase into the theme.

But we feel pretty good that our names - Salesforce, Adobe and Hubspot will do well in the next 6 months to a year.

Google Earnings

Google announced earnings this week with revenues surpassing $113B for Q4 (+19% YoY), with EPS coming in at $2.8, up 32% YoY. 

Cloud was the growth driver with revenues up 47% YoY with operating margins north of 50%.

But the real talking point of the quarter was capital intensity. Google guided $175–180B of capex for 2026.

Sundar Pichai notes Google has been “supply constrained even as we’ve been ramping up our capacity”. Demand across services and Cloud remains “exceptionally strong.” 

Anat Ashkenazi (CFO) noted that the “vast majority of our CapEx was invested in technical infrastructure, approximately 60% in servers and 40% in data centers and networking equipment.” 

The capex is driving returns. Gemini usage and engagement are scaling rapidly. 

Pichai highlighted that “our Gemini app now has over 750 million monthly active users,” with “significantly higher engagement per user, especially since the launch of Gemini 3.” 

He added that Gemini 3 Pro has “seen the fastest adoption of any model in our history,” processing “three times as many daily tokens” as the prior version. 

Across the ecosystem, “first-party models like Gemini now process over 10 billion tokens per minute via direct API,” underscoring both scale and throughput.

Importantly, this growth is coming alongside falling unit costs. 

Google disclosed it was able to “lower Gemini serving unit costs by 78% over 2025 through model optimizations, efficiency, and utilization improvements.” 

That combination—rising usage and falling cost to serve—is what enables capex to compound returns.

How does OpenAI compete with Google’s improving returns, and cash-flow funded capex?

Crucially, AI is not cannibalizing Google’s core businesses. 

Search revenues grew 17% YoY, with management describing the moment as “expansionary,” driven by longer, more complex queries and higher engagement. 

YouTube’s annual revenues surpassed $60B, supported by both advertising and subscriptions, while subscription, platforms, and devices revenue grew 17%, helped by demand for AI-enhanced plans like Google One.

Google was our pick back in July 25, when the fears of ‘AI-kills-search’ had led it to a ~20x P/E NTM valuation.

It has been re-rated significantly since then, and is now priced rather high amongst its comps.

We aren’t buyers here. A benchmark weight seems appropriate.

2026: The Trough For Managed Care

Molina Healthcare reported earnings this week, and the stock sold off sharply, down more than 30%. 

The headline numbers were weak - EPS was down 154%. 

The earnings commentary tells exactly why the sector is struggling.

Molina reports results deteriorated because medical costs rose faster than state rate updates. 

CEO Joe Zubretsky described 2025 as unusual: the medical cost trend accelerated from 4.5% to 7.5% in a single year. He remarks this as “an aberration, an anomaly by historical standards.” 

A key driver of that inflection was redeterminations. 

As pandemic-era eligibility rules rolled off, lower- and no-utilizers exited Medicaid first. That mechanically raised the average acuity of the remaining population. 

Zubretsky quantified it: “250 basis points of this 7.5% trend is attributable to the acuity shift from membership declines related to the final stages of redeterminations.” 

This is a critical context.

Utilization did not spike because behavior changed overnight; reported costs rose because the mix changed.

That mix shift exposed the lag in the rate-setting process. 

Molina’s CFO: “rates have not kept up with trend over the past six quarters.” 

Management believes “Medicaid plans are underfunded by 300 to 400 basis points.” 

The sector is under-earning because rates are stale

However, Zubretsky expects 2026 to be the “trough for managed Medicaid margins,” since the impact of redeterminations end by 2025.

We can see margins improving in 2027 and beyond.

Overall, US healthcare providers have struggled over the last few weeks- they have lost over 10% in the last 9 days. 

The decline began after lower than expected rates revision for Medicare Advantage.

We see the sector as an avoid.

AI Infrastructure Isn’t “Compute-Constrained”

Flex reported earnings with revenue of $7.1B (+8% YoY) and adjusted EPS of $0.87 (+13% YoY). 

More important than Flex-specifics, the call offered a clean sector read-through on where the demand is directed within AI picks and shovels.

AI capex is still accelerating, but the bottleneck has moved downstream into execution and deployment. 

Flex’s CEO Revathi Advaithi: “there’s tremendous complexity in the data center deployment”.

Customers increasingly need “an ecosystem of integrated products, capabilities, technologies, and services.” 

Flex’s read-through is straightforward:

The winners in the picks-and-shovels stack are the companies enabling faster deployment through integrated power + cooling + systems-level execution.

We also own Flex just in case we are wrong on our bearish assessment of OpenAI’s risk to the category. We can still participate in this way.

We believe we are hedging our views with a good investment in the space.

Latam Airlines (LTM): Our International Value Pick

LATAM is the dominant airline across South America, with scale advantages in Brazil, Chile, Peru, and Colombia. 

It has an expansive network that connects intra-regional travel with higher-yield international routes to North America and Europe. 

The tailwinds are structural. 

South America remains under-penetrated in air travel, and tourism flows and regional business travel are rising. 

LATAM transported over 87 million passengers in 2025, with full-year capacity up ~8% YoY, reflecting steady, demand-driven expansion.

International routes and premium travelers are enabling a better pricing mix, with leisure and premium demand improving.

What differentiates LATAM in this cycle is discipline. 

Coming out of restructuring, management has focused on route profitability, fleet efficiency, and customer experience, rather than chasing market share. 

That strategy is visible in the operating metrics: adjusted operating margin reached ~16% in 2025, up meaningfully from pre-restructuring levels. Adjusted EBITDA rose to ~30% YoY in the most recent quarter - highest in 10Y history.

Costs have remained controlled even as capacity expands, allowing yields to stay stable.

Operational execution supports the story. 

Customer satisfaction has improved materially, with record passenger NPS of 54, and premium traveler NPS at 58, signaling better service quality. 

Capacity growth has been selective.

The fleet has expanded efficiently, and cargo and international operations continue to diversify revenue away from pure domestic leisure demand.

The financial results are simply the consequence of that strategy. 

Q4 revenues grew ~16% YoY, net income rose sharply, and free cash generation exceeded $1 billion, enabling dividends and share repurchases. 

Despite this, LATAM trades at discounted valuations, with P/E NTM at 12.0x.

The bear case is familiar. 

Earnings remain exposed to South American macro volatility, fuel prices, FX, and regulatory or labor risks, particularly in Brazil and Chile. 

A sharp global travel slowdown or energy shock would pressure margins. 

But these risks are inherent to the airline sector and are more than reflected in the current valuation.

LATAM stands out because it combines regional market leadership, secular international travel tailwinds, and post-restructuring discipline, yet is still priced like a traditional cyclical airline. 

Lumida Curations

NVIDIA × Dassault: The Digital Twin Stack Goes Real-Time

NVIDIA and Dassault are collapsing decades of offline simulation into real-time, AI-accelerated digital twins—laying the infrastructure layer for the next generation of product design and engineering.

View Curation

Ken Griffin: Dollar Weakness Is Optics

Ken Griffin argues recent dollar softness reflects policy noise rather than structural decline, emphasizing that reserve-currency status—and America’s ability to accumulate and deploy capital—remains the core advantage.

View Curation 

Dry Sauna Is the Longevity Baseline

Short, high-heat dry sauna sessions (~20 minutes at ~200°F) reliably raise core body temperature and have the strongest evidence base, while infrared and wet saunas often fall short on heat stress or safety.

View Curation

Meme

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: Powell Agrees To An Expanding Economy; Weaker Dollar Is A Markets’ Tailwind

• Markets: Meta: Non-Consensus Alpha; Microsoft Was Misunderstood; Amazon: A White Knight For OpenAI?; Winner of CapEx Arms Race?; WSJ - The Contrarian Signal For CoreWeave

• Lumida Curations: Tariffs: Industrial Comeback or Hidden Consumer Tax?; Rebuilding the Arsenal-The $1.5T Push; The Simple Hack For Better Sleep 

Spotlight

This week, we had an exciting bit+bips podcast with Charles Edwards talking about Gold’s phenomenal rise in the last year, and where it goes from here.

Charles believes Gold has 5-10 year bull run every few decades, where it outperforms equities easily.

I had a contradicting opinion - $5K was a nice round number for Gold, and it’s usually around round numbers when markets re-assess the asset.

We saw this in Nvidia, where it sold off to $4.5T after becoming the first company to a $5T valuation.

The prediction on Gold turned out correct this week. It is now trading at approximately ~$4,900, down about 15% from its high.

We also talk about the geopolitical situation, bitcoin’s lag, and how rate cuts can impact markets.

Watch the full show here.

When To Have Paper Hands

Gold dipped 9.8% on Friday.

Recent buyers in Gold are “paper hands” or momentum chasers, and now are trapped buyers. 

They got a dose of reality and are rushing to exit.

If you are owning momentum, you want to have paper hands. The key is detecting and participating in early momentum. 

If you are a quality value investor, you need strong hands while your asset is mispriced or you will be shaken out. 

Paper hands are a form of mental flexibility. Diamond hands are a form of conviction. 

If you know the asset and have a concept of intrinsic value, choose diamond hands. 

Gold is a momentum idea. There is no valuation framework. The gold run has out paced ‘dollar debasement’ drastically. 

So, it is a momentum and crowd idea. 

You want to be paper hands there.

What we see is a classic intermediate top in Gold.

After that, we need to re-assess. Price targets don’t make sense in our view. It’s better to ‘re-assess’ with new information in the future.

Price targets create a false sense of precision. The world can unfold in many paths. The actions of real world actors matter - Warsh nomination, Trump’s action in Iran, dealings with Canada, China’s bid for Gold…

What we can say now is that the commodity rally needs to reset.

Note: The gold rally and international stock rally are both part of the same meta trade - fading U.S. assets. It’s reasonable to expect international stocks to pull back here as well (although we expect they will continue to outperform U.S. assets).

We also saw this week that Microsoft tanked 10% on news that 45% of its backlog is linked to OpenAI related agreements. The market is skeptical about Mr. Altman’s ability to honor $100 Bn+ in contracted agreements when the firm makes less than $30 Bn annually.

If you take a step back you have the following conditions:

1) Skepticism on OpenAI’s capex sustainability whose $100 Bn financing represents material earnings for semiconductor complex and large cap tech.

2) Commodities are having a blow off top.

Silver’s pullback should hurt Gold (also pulled back and we have a small short here initiated Friday) and even Copper. Copper has a great fundamental demand story… but in the near term all commodities should get hurt as Silver unwinds.

I shared my thoughts on how Silver would unfold several days before the commodity cracked [ here ], and suggested gold investors sell this past Monday [ here ].

3) Growth stocks, including software, remain under pressure. (It looks like we were early on our software call.) The category appears to be capitulating now.

4) “Old guard high beta” names like Robinhood and Palantir are in deep corrections.

5) Financials are weak due to credit card cap rate talk.

6) The United States has an Armada stationed outside Iran. Energy stocks are out-performing QQQ substantially. Energy leading in the S&P, and a rise in oil prices, is statically not good for equities.

7) We are entering a period of weak seasonality after we get past beginning of month inflows in February.

8)The Epstein files are back.

That will hurt Trump’s polling and dampen ‘Trump Bump’ related animal spirits including Crypto. Take a look at Coinbase stock decline...

9) We are in the first half of a mid-term cycle. Statistically, post GFC we see a 7% average drawdown.

10) Trump appears to back Kevin Warsh who historically is a hawk. Warsh wants to get the Fed Balance sheet back to normal. That’s not good for speculative assets, such as crypto (see Bitcoin breaking to new lows).

It’s also not good for the equilibrium PE multiple on stocks more generally.

It’s also a bit worse. Markets, including me, thought we might see BlackRock CIO Ric Reider as head of the Fed. The world’s largest buy of bonds (Reider) paired with the world’s largest seller of bonds (Bessent) could restore confidence and bring down long-term yields.

Markets rallied a bit on this news. Those investors have a rugpull here. More generally, markets have not had time to process a Warsh Fed much at all. Markets were trained on super-dove Hasset for many months, then Reider.

Markets don’t like surprises.

11) And this is a crucial point… as we have noted in recent weeks, the positioning in markets is high.

The AAII Bull Bear and multiple sentiment surveys show investors are ‘all in’.

We last saw these levels back in February 2021 which was a terrible time to own risk assets.

The good news?

Earnings growth is strong. In our view, however, this news will be outweighed by other risk factors.

There is a reasonable chance of a pull back here. The failure of Microsoft to rally on Friday after a 10% drop, despite a 24x forward PE and 20% EPS growth, suggests investors are skittish.

It’s a very complex picture. Day-to-day events matter. That’s also another way of saying that the pair-wise correlation across stocks (e.g., volatility) is going higher.

A de-escalation around Iran could help to spark a rally.

But, it’s hard to assume Trump re-positioned substantial military infrastructure and doesn’t want to take action.

We believe Trump sees himself as a historically significant figure in the style of Teddy Roosevelt whom he cites quite often.

Trump likes wins on the board. One of the few areas directly in Trump’s control is military action. The last military action on Venezuela was positively received by Venezuelans and Trump’s base.

We believe some type of strike on Iran is inevitable. (It would help shift the news cycle focus away from Epstein as a side consideration.)

That should help Oil Field services companies. Our name FTI, which we highlighted after the Venezuela Action, continues to do well.

Suffice to say, risks are unusually elevated. We took down our market beta this past week.

A Monday gap down is quite possible. Note: Buying Monday gap downs has historically been a winning strategy.

The news between now and Monday morning matters of course. If you’re looking for one of the best apps to monitor markets and real-time news, we suggest joining www.lumidainvest.com.

We use the app everyday, and it has a Bloomberg style real-time newsfeed. It even includes a Trump Watch feature - ideal for times like today.

Trumps VS Iran

On Friday, Trump said, “We have a large armada heading to Iran right now. Even larger than what we had in Venezuela.”

“If we don’t make a deal, we’ll see what happens.”

Will we see an escalation in Iran? It’s hard to say. My guess is ‘Yes and soon.’

This article forward to me by an ex-marine I trust suggests the same:

Here’s a podcast I did this week about the great power politics going on right now. We talk about John Mearsheimer, and his prophetic calls on Russia, China and the United States.

Watch it here.

Macro

Powell Agrees To An Expanding Economy

FOMC decided to keep rates unchanged this week as was expected. 

Powell’s commentary provides solid insights into the economy, and seconds our opinion on economic resilience. 

Powell: “The U.S. economy expanded at a solid pace last year and is coming into 2026 on a firm footing.”

“Available indicators suggest that economic activity has been expanding at a solid pace.”

Powell attributed economic strength to improved customer spending. 

"You've got strong consumption that's been happening before financial conditions have been supportive”

“Consumer spending has been resilient…” 

He also pointed to rising corporate investment, highlighting high business confidence: “...business fixed investment has continued to expand.” “...we're benefiting from the AI build-out of data centers.”

On Inflation, the tone was calmer than earlier, but he flagged elevated levels than the 2% target. 

“Inflation has eased significantly from its highs in mid-2022 but remains somewhat elevated relative to our 2 percent longer-run goal.”

Recent readings still reflect above-target levels:

“Total PCE prices rose 2.9 percent… and core PCE prices rose 3.0 percent.”

However, Powell highlighted that much of the remaining pressure is concentrated in goods rather than services:

“If you look away from goods and look at services, you do see ongoing disinflation in all the categories of services.”

Powell also had optimistic remarks on labor market performance, and says it is “stabilizing after a period of gradual softening.”

His comments suggest future rate cuts may be far off as inflation remains elevated from target and the economic activity has been resilient. 

Powell: “We want to see further progress on inflation before making adjustments to policy.” We want to “Finish the job of getting inflation back down to 2 percent.”

Overall, his comments second the economy is on a solid footing, and expanding.

Powell’s tenure as Fed chair ends in May '26.

Warsh is likely to be tested with market volatility as a new Fed Chair.

Markets don’t know if:

1) Warsh is a hawk as his history suggests

2) Warsh was a hawk but pivoted to dovish to get the job and is now a dove

3) Warsh was a hawk but pivoted to dovish to get the job but will now act like a hawk

We believe Warsh will gradually shrink the Fed Balance sheet and push to lower rates.

The weaker independence of the Fed will not help to lower long-term rates.

We recommend folks read our analysis of long-term rates here where we suggested long-term rates will face continued upward pressure.

Markets

Weaker Dollar Is A Markets’ Tailwind

This week, Trump commented the dollar is “seeking its own level” and is “doing great.” But, is it actually doing good? 

Dollar has fallen to its weakest level in roughly three years, dipping approximately 10% in the last 1 year. 

A weaker dollar is a risk-on signal for international markets, and also U.S. markets all things being equal. If you want to “re-balance” global trade, a weaker dollar (e.g., lowering the purchasing power of Americans) is necessary.

Since 2020, every meaningful dollar downtrend has lined up with double-digit EPS growth and strong equity returns.

When the dollar weakens, stocks tend to strengthen. 

Periods where the dollar hits lows have repeatedly coincided with equity markets pushing to highs. 

Forward returns after these setups have been consistently positive, often lasting months. 

With the recent weakening in dollar, markets have support for outperformance.

The Q4 earnings season also gives us a reason to be optimistic about the continuance of the bull market. 

So far, 72% companies have delivered a beat on their revenue estimates, while 78% have beaten on earnings. 

This outperformance is remarkable because revenues and earnings estimates were already revised higher after outperformance in Q3’25.

The beat rate has been the best since 2021 after Q3’25’s performance.

It’s difficult to argue about a bear market with solid fundamental performance.

That said, we believe headline risk will overshadow a strong fundamental earnings backdrop. The direction of the 10-year rate will be key to watch this week.

We continue to see the non-consensus opportunity in international stocks. 

The dollar weakness has supported a broadening out in international markets.

So far this year, global markets have broadly participated. 

Most country ETFs are up year-to-date, and have outperformed US markets. 

The US has been the 6th worst performing market, gaining only 1.4% in January.

Investors have been diversifying away from the US. 

You can see it in the outperformance of Brazil, South Korea, Turkey and many other markets. Such periods of outperformance tend to continue.

That said, we do expect international names to pullback this week. Markets never make it easy!

These are some of the markets we have been bullish on for a few months now. Over 35% of our exposure is now in international equities.

We discussed América Móvil in our last newsletter. You can read our thesis here. 

I also did a Lumida Non-Consensus Investing podcast explaining why I see this move continuing in the near term. View here. 

Meta: Non-Consensus Alpha

Meta reported earnings on Wednesday, delivering a double-beat and rising 10%. The stock was beaten down until a week earlier. 

We discussed Meta’s bull case in our last newsletter and how it had reliable growth levers and lower valuations compared to Mag-7.

Meta is also a 2026 Lumida stock pick.

The market agreed after earnings.

The biggest question around Meta was the returns on its AI capex, and the earnings commentary was more than comforting for investors.

Zuckerberg (CEO): “Our business performed very well, thanks to AI-driven performance gains.”

“We’re seeing very strong results from the ad performance investments we made throughout 2025 with year-over-year conversion growth accelerating through the fourth quarter.”

AI is increasing time spent and engagement, which converts into higher ad revenue. 

Zuckerberg noted “Instagram Reels had another strong quarter with watch time up more than 30% year-over-year in the U.S.”

“On Facebook, video time continued to grow double digits year-over-year in the U.S”

“Threads is also seeing strong momentum, benefiting from recommendation improvements. The optimizations we made in Q4 drove a 20% lift in threads time spent.”

Higher engagement has helped advertisers see better conversions, which is driving demand and higher prices. 

“The average price per ad increased 6% year-over-year, benefiting from increased advertiser demand, largely driven by improved ad performance.”

Meta also noted AI driving productivity gains across its business.

Susan Li (CFO): “Since the beginning of 2025, we’ve seen a 30% increase in output per engineer with the majority of that growth coming from the adoption of agenetic coding, which saw a big jump in Q4.”

“We’re seeing even stronger gains with power users of AI coding tools, whose output has increased 80% year-over-year.”

Overall, these comments accompanied with the double digit growth in revenues and profits helped the stock gain ~10% on its earnings day.

The world is going digital. Meta is positioned to monetize these consumer eyeballs for years to come.

Meta's ad revenue has grown 50% over the last two years at a base of ~$50B.

The margins, excluding R&D, come close to 80%.

Meta can choose to cut R&D and boost net income whenever they want. This is the  'real embedded option'

This is a high operating leverage business with rapid revenue growth.

The key question to ask for a business of Meta's size is can the revenue growth trend continue, or is this a mature business.

Ad revenue growth YOY is 18%, and it is accelerating!

We have already seen how AI-spending is driving better pricing and improved engagement, which will help produce reliable growth.

The best thing - Meta has plenty of fundamental growth to fund AI investment without diluting shareholders.

It’s the inverse of OpenAI, which needs to figure out how to monetize and fund its capex.

Meta is also exploring new sources of revenue growth, product innovation, and capex.

Zuck says "2026 is going to be the year that AI starts to dramatically change the way that we work.”

“We're starting to see projects that used to require big teams now be accomplished by a single, very talented person."

"Our vision is building personal super intelligence. We're starting to see the promise of AI that understands our personal context, including our history, our interests, our content and our relationships."

The early results are promising . 

“We're seeing good early traction with our AIs in Mexico and the Philippines, with over 1 million weekly conversations between people and business AI is now happening on our messaging platforms."

WhatsApp can also act as a growth lever. 

Zuck said “we expect to complete the rollout of ads in status throughout the year"

"Paid messaging within WhatsApp continues to scale as well, crossing a $2 billion annual run rate in Q4.” 

The outlook on Capex was also solid. 

"We will continue to invest very significantly in infrastructure to train leading models and deliver personal super intelligence to billions of people and businesses around the world."

How does OpenAI compete with this chart?

The main bear case for Meta is intense levels of capex spending and uncertain ROIC. 

But, at a valuation of low 20x NTM, I believe that is priced in. 

What's not priced in is Meta as a potentially dominant AI leader.

Microsoft Was Misunderstood

Microsoft also reported earnings on Wednesday, and it had an almost inverse reaction to Meta.

The stock dipped 12% after earnings despite producing a double-beat. 

Increased data center spending and slower than expected Azure growth (38% vs 40%) hurt investors’ sentiment. 

But, the market misunderstood its earnings. 

Although Microsoft's capex increased 63% from last year, most of this capex was for GPUs, which already have confirmed revenues for their life. 

Amy Hood (CFO): “Roughly 2/3 of our CapEx was on short-lived assets, primarily GPUs and CPUs”

The GPUs “are sold for the entire useful life”, meaning the capex will generate almost certain returns.

Amy Hood: “the remaining spend was for long-lived assets that will support monetization for the next 15 years and beyond.”

Moreover, Azure’s growth at 38% against expectation of 40% also had a solid underlying reason. 

Microsoft is using their compute capacities to meet their internal needs before they sell it ahead. 

Microsoft Copilot, where compute capacities were used, posted impressive gains - paid Copilot users increased by 75% YoY. 

Amy Hood acknowledges if compute was sold entirely to customers, “the KPI would have been over 40”.

Satya Nadella (CEO) notes it was a “record quarter for Microsoft 365 Copilot seat adds, up over 160% year-over-year.”

The bull case for MSFT looks interesting now.

Before earnings, Microsoft traded at ~28x P/E NTM. After the stock’s dip, and earnings beat, the P/E NTM valuation has declined to ~24x. 

The last time Microsoft traded at such valuations was back in Jan 2023. 

The business has grown about 50% in revenues and earnings since then. 

Microsoft’s stock generally bottoms reliabily at 22X forward PE. We aren’t there yet, and it’s not clear markets will go there either.

Cloud remains the Growth Lever. Intelligent cloud revenue was up 32% YoY, coming in at $56B.

The remaining performance obligations (RPO) grew by 110% YoY in Q2 '26, and is now at $625B. 

Management expects to realize 25% of it in the next 12 months, and an additional 39% in the next 24 months.

Satya Nadella says cloud “demand continues to exceed our supply.” This supports our thesis that capex spend is ROI positive.

Capital returns sweeten the bull case for Microsoft. Dividends and buybacks amounted to $32.7B in Q2’26, up 32% YoY. 

Capital returns show management’s confidence in the company's cash position, and their trust in the underlying fundamentals. 

However, the bear case of Microsoft is OpenAI.

OpenAI represents 45% of the $625B backlog.

This equals ~$280B in future revenue commitments. 

How does OpenAI fund it?

A difficult question.

OpenAI’s revenue performance obligations are like a metastasized Stage IV cancer. 

I think you could create a basket called ‘OpenAI Related’ and you would see it easily lags benchmarks. 

Names in there would be: Oracle, Microsoft, and others…

Fortunately, markets have priced in this risk sooner than otherwise. 

We won’t ever see a Dot Com level bubble ever again because markets are always learning and memory of that era is in practitioner’s minds. 

The pre-condition of the Dotcom bubble was the retirement of nearly every money manager in the Nifty Fifty bubble (except Warren Buffett).

We bought MSFT this week - it seems like an obvious trade with ~20% revenue and earnings growth on a base of ~$300B. 

The levers in cloud and AI monetization can help significant growth in future. 

Microsoft is non-consensus now, and that’s exactly where alpha lives. 

AMAZON: A WHITE KNIGHT FOR OPENAI?

Amazon is considering a $50 Bn investment in OpenAI. 

Since the disastrous Altman interview, markets have woken up to just how adjacent market cap rests on a pyramid of OpenAI counterparty risk. 

The valuation can’t be walked back, price is socialized and fixed with Softbank as the lead. 

So, non-priced terms, such as the spend OpenAI must commit towards AWS, must be in play. 

This is a ‘strategic’ not financial investment for Amazon. 

Meaning, if you are one of the Middle East SWFs throwing money at this, or Softbank, you are overpaying since you have no quid pro quo with a datacenter business. 

Pure financial investors see no commercial value exchange whatsoever. 

(That tells you this is a bad deal for non-strategics.)

Amazon is already an early and major investor in Anthropic. 

Amazon now has all sorts of confidential insight into OpenAI: including its product roadmap, and user engagement statistics, etc. 

Markets recovered from their lows immediately after this news was leaked. 

(That’s also why it was leaked - the market cap carnage would be greater than $50 Bn). 

Although I don’t think OpenAI is priced correctly, not by a mile, from the Trump admin point of view you could argue OpenAI is too big to fail. 

So long as markets can see a steady streak of financing into OpenAI, the spice will flow. 

There’s so much market cap resting on OpenAI - Microsoft, Oracle, Semi Complex, Datacenter names, Energy IPPs / Nuclear, etc that the Trump admin should want OpenAI financing to be successful. 

I do think an OpenAI round gets done and is successful. 

Note: Whenever markets feel stress, you see a leak like this to the Information or the WSJ. We saw this happen twice in the last 90 days…

But, I doubt we will see any smart money non-strategic investors involved. 

After this round of financing, OpenAI will need to prepare an OpenAI to fund its insatiable demand for capex - they have no choice. 

Wall Street will be the financier of last resorts, and you’re supposed to fade that. 

Winner of CapEx Arms Race?

The obvious winner for Meta, Microsoft and Amazon / OpenAI news?

Nvidia. And TSM. 

Nvidia’s EPS expectations are for 60% growth over the next 12 months. 

I believe those are low ball estimates. 

Meta ramped up capex. That means more for Nvidia in 2027. 

Nvidia is trading off of 2027 expectations now. 

We had trimmed our Nvidia exposure back when it hit a $5 Tn market cap, but added back in recent weeks with the Mag 7 correction.

It has worked well so far. 

Nvidia is not a lay up here. The market is still punishing companies with OpenAI exposure.

Still, we believe Nvidia can swap out OpenAI demand with other customer demand.

What’s truly in question is ‘When does Nvidia’s growth rate slow: 2027 or 2028?”

WSJ Is The Contrarian Signal For CoreWeave

WSJ published an article on Dec 16, titled "What's Ailing CoreWeave".

The stock had dipped 24% in 5 prior days. This article caused a low in the stock at around $66.

After the article, CRWV bounced back and rushed to $114 in less than 1.5 months, an almost 70% run. 

On Jan 26, WSJ published a positive headline of Nvidia investing in CoreWeave. 

This marked the top for the stock. Funny thing - we did trim our exposure to CoreWeave after this WSJ headline.

The stock has come down to $93 since then, erasing almost 18% of its gains. Now, we are looking to accumulate again.

That's classic non-consensus on both ends of a WSJ article.

We have been bullish on Coreweave - it is a Lumida 2026 stock pick. Read our thesis here.

The risk to CoreWeave is the OpenAI exposure. But, Nvidia stands ready to consume demand for any excess CoreWeave capacity. 

Lumida Curations

Tariffs: Industrial Comeback or Hidden Consumer Tax?

A heated Schiff–Pomp clash breaks down whether “Made in America” tariffs rebuild U.S. manufacturing—or simply push higher prices straight onto American households.

View Curation

The $1.5T Push to Fix America’s Supply Chain Weakness

Jamie Dimon warns that years of dependence on foreign producers left critical defense industries exposed—now massive capital is rushing in to rebuild the industrial backbone behind U.S. security.

View Curation

The Simple Hack For Better Sleep 

Small timing shifts—earlier dinners, screen wind-downs, and short pre-bed reading—can meaningfully improve sleep quality without medication.

View Curation

Meme

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: Accelerating Growth and Labor Resilience; The Reality Behind Slowing Wage Growth; Shifting Away From The US; DAVOS: REALPOLITIK IS BACK

• Markets: Software Is The New Restaurants; The Gains Were Due For Meta; The Bubble Around OpenAI

• Lumida Curations: AI-Powered Fighter Jets; The Future of Money Movement; Aging: A Solvable Problem

Spotlight

I did a Lumida Non-Consensus Investing podcast on Monday, talking about Economic populism, and the out-performance of international stocks vs the S&P 500.

I have never owned as many international themes as I do now. Watch the pod here.

We also discuss major themes, and sectors that might dominate markets in near term.

This Monday, I am speaking with Angelo Robles at the Uncorrelated Conference in Miami. 

We will be discussing real assets and what AI means for investing. 

Angelo is an amazing host.

If you’d like to attend it, the registrations are open. You can find details here.

WHY KARA SWISHER IS WRONG

[ I wrote this post on X, and it received some attention. Sharing here./ ]

Kara Swisher thinks entrepreneurs owe their gains to the state. 

She can’t be more wrong. 

California did not invent the product.

California did not write the code.

California did not take the risk.

California did not skip paychecks.

California did not sign personal guarantees.

California did not stare down payroll, lawsuits, or bankruptcy.

The founder did.

Entrepreneurs create value by solving problems people voluntarily pay to solve. That value did not exist before the founder acted. 

The state did not “provide” it. 

At best, it provided infrastructure—roads, courts, universities—things funded by prior taxpayers, including entrepreneurs themselves.

Infrastructure is not authorship.

A road doesn’t build a company.

A university doesn’t create a business model.

An ‘ecosystem' doesn’t ship a product.

People do.

Here’s the asymmetry Swisher ignores:

If you fail, the state still collects income tax, payroll tax, sales tax, penalties, fees.

If you succeed, the state claims moral ownership of the upside.

At no point does the state share downside risk.

That’s not a partnership. That’s rent-seeking.

You don’t “owe” someone who never co-signed the risk.

The real entitlement isn’t founders keeping what they built.

It’s the belief that the government deserves credit for success it did not create, risk it did not bear—while threatening those who can leave for daring to do so.

Entrepreneurs don’t owe California their wealth.

California owes its wealth to entrepreneurs.

Everything else is moralizing expropriation dressed up as gratitude.

I recommend watching "The Founder" this weekend. A great story about Ray Kroc and the origin of McDonalds.

Here is a special Non-consensus investing interview with the producer of that movie who himself is an entrepreneur.

The Opportunity In Geopolitical Tensions

Last weekend, Trump’s comments on Greenland made headlines, and this week, his focus has shifted back to Iran. 

On Friday, he said an American "armada" is heading toward the Middle East and that the U.S. is closely monitoring Iran’s activities.

Will we see an escalation? It’s hard to say. My guess is ‘Yes.’

But, one thing is clear. 

Geopolitical tensions are rising. And, countries are responding by ramping up defense spending. 

Global military expenditures surged by 9% last year—the steepest increase in over three decades. 

This rise in defense spending is directed toward military autonomy and AI. 

Nations are aiming to develop intelligent systems that operate independently.

Anthony Antognoli, executive officer of the U.S. Coast Guard, explained: “Our vision is to use robotics and autonomous systems as the foundation for Coast Guard missions.” 

He added, “It is impossible to carry out defense missions with humans and patrol cutters alone.”

These increased defense budgets are creating significant demand for defense tech companies. 

Venture capital has taken notice. In 2025, defense tech venture investments more than doubled compared to 2024, and this momentum is likely to carry forward as demand keeps increasing.

The bulk of this funding is concentrated in large, late-stage rounds focused on autonomy-driven defense firms. 

Companies like Anduril and its peer group are benefitting from these financing rounds. 

Those financings are only available to sophisticated private markets investors.

Lumida Ventures is also focused on investing in this category. We have had a string of investments across CoreWeave, Kraken, and Brad Jacobs QXO, and Canva.

Do reach out now if you want to be in the flow of our next deal.

If you are an accredited investor or qualified purchaser, sign up here to receive communications about our private deals.

Macro

Economic Growth Is Accelerating

This week's data releases highlight an economy running hot. 

Markets oscillate between fears of growth slow down and fear of running hot.

We’re in the latter camp. And, you can see it in commodity prices and a 10-year and 30-year yields that are creeping up.

The GDPNow model tracks real GDP growth at 5.4% for Q4-2025, a clear step up from an already solid 3.8% in Q2 and 4.4% in Q3. 

Corporate profits continue climbing to record levels in line with GDP, and cash flows remain at all-time highs.

The higher profitability and cash flows are translating into increased investments in tech and people. 

You can see this in the latest labor market data.

Unemployment insurance claims are continuing to trend lower.

• Initial claims came in below 200K for two weeks straight.  

• Continuing claims have also dipped to 1.85M compared to 1.88M last year.

Companies are firing less, and the unemployed are finding it easier to get in new roles.

This is a significant development from last year. 

There are two reasons behind this. 

Companies have gained more confidence in the economy, and the air of economic uncertainty is clear now. 

Corporates are more comfortable with new investments, and bringing on new employees. 

We saw this in our bank’s earnings analysis last week, where corporate deals are expected to be at record levels in 2026. You can read it here. 

Secondly, companies are also still working through AI's practical boundaries - they are starting to realize it's transformative for many processes but not yet a complete substitute in others.

As a result, they're gearing up to hire more workers to capture booming demand. 

It’s funny how we had recession fears until a few months ago.

The Reality Behind Slowing Wage Growth

Another talking point around this week's releases was the flattening disposable income graph. 

With spending picking up, a flat income chart meant consumers were using their savings to fund their expenses.

News headlines directed it to worsening affordability and weakening consumers, but with PCE generally in line with estimates, the argument didn’t have legs to stand on. 

But, here’s what the graph doesn’t tell you.   

A part of the decline in wage growth is driven by Baby Boomers retiring. 

They are often highly paid due to their experience, and are now exiting the workforce in waves.

This pulls down overall wage growth and disposable income figures. That keeps a lid on inflation, too.

However, this doesn't mean they aren’t spending. 

With over $85 trillion in net worth across the generation, they're spending it.

Retail sales have come in stronger consistently, with red book retail sales rising 5.5% YoY in Jan. 

You can also see it in rising demand for luxury and affluent services. 

We covered Carnival Corp's earnings in a recent newsletter, where Geoffrey T. Martin (CEO, CCL) said “Demand for cruise lines is proving far more resilient than traditional macro indicators would suggest.”

He also pointed to improved bookings at higher prices, suggesting more future demand. Read it here. 

Overall, the economy is running hot. 

Shifting Away From The US

On the global front, countries are stepping back from over-reliance on US dynamics amid ongoing tariff uncertainties. 

A standout example is Canada's recent deal with China, and Trump’s response.

The trade deal allows up to 49,000 Chinese vehicle imports annually at just a 6.1% tariff—a sharp cut from the 100% duties they aligned with the US in 2024. 

In exchange, China reduces barriers on Canadian goods like canola (tariffs dropping to about 15% from 84%), seafood, and peas, while adding visa-free travel for Canadians. 

This move isn't happening in a vacuum.

It's emblematic of a worldwide trade realignment. 

The EU is advancing pacts with South America, Mexico, Australia, and India.

Mexico is renewing ties with Canada. 

Regions like ASEAN and India are forging new agreements too. 

The common thread? 

Countries are now pivoting to reduce vulnerabilities, reorient supply chains and diversify partnerships away from US-centric models.

This shift also reflects in equity markets.

International markets like South Korea, Japan, Singapore, and Brazil have outperformed SPY in the last 1 year. 

Each week, we find ourselves increasing our equity allocations.

This shift started with the 10-year bond.

US10Y used to be investors’ preferred flight to safety, but in the last 1 year, yields haven’t moved much despite global uncertainties.

We saw the opposite recently, when TLT fall by 1% after Trump’s comments on Greenland.

Now, Gold has become the popular ‘safe haven’ - you can see it in its price moves around any geopolitical headline. 

We see this diversification from US markets and into international equity markets continuing in the near term. 

I discussed the reasons behind it in Lumida’s Non-consensus investing podcast. Watch here. 

Around 30% of Lumida’s exposure now is in international stocks. We discussed some of our picks a few weeks earlier. Read here. 

DAVOS: REALPOLITIK IS BACK

Few days back, I wrote Carney was the big winner from Davos. 

He positioned himself as a deal maker and the leader of the Free World (ex-US). 

His intellectual ability is ahead of Macron and other EU leaders. 

His storytelling skills and re-framing of the ‘Workers of the World Unite’ story displayed a skill we just have not seen in global leadership. 

And, interestingly, Trump would agree agree with most points in this speech:

1) Multilateral institutions such as the WTO have failed

2) Supply chains are points of vulnerability

3) There are risks to ‘extreme global integration’

4) You cannot rely on one party for defense or energy

However, it seems like Carney flew too close to the sun.  

Now, Canada is unfortunately staring down the barrel of 100% tariffs from its largest trading partner…  

…or the consolation prize: a China EV deal in a world where consumers still prefer internal combustion engines.  

Here’s the rub.  

Carney is a gifted orator, but he overplayed his hand.  

Tariffs are taxes on people. Full stop. They land on households and businesses—not on abstractions called “countries.”  

Carney didn’t have to tax his own people. That's the deal. A quiet deal.

There was no need to shout from the mountaintops a 'deal with China'.

Trade happens between individuals and firms, not governments. 

Governments don’t create trade—they constrain it, distort it, or weaponize it.

Carney poked his hegemon neighbor in the eye at Davos.  

Carney chose a highly visible move over a prudent one.  

Ultimately, he let his ego get in the way instead of advancing the interests of his people. 

And that brings us to history.  

In the late 1930s, prior to Pearl Harbor, the U.S. didn’t stop Japan with speeches or appeals to shared values.

FDR cut off oil, steel, and trade.  

The  best oratory in the world can't match military or economic coercion.  

This is the point of FDR's, 'Speak softly and carry a big stick?"  

In realpolitik, when push comes to shove, high-minded oratory doesn’t compete raw force —  whether through trade, capital flows, or threat of force.  

That’s not ideology. That’s reality.  

We are living in an age of realpolitik.   

(Was there any era in human history where we were not?)  

Canada’s neighbor is the world’s dominant hegemon.  

Did Carney seriously believe there would be no consequences?  

Now, If you are the hegemon—and you see the world forming trade alliances designed to exclude you —what do you do?

There is only one rational response:  

You respond swiftly.  

You move hard.  

You take the marbles while you still have advantage on the board.  

How does this play out?  

The United States is far more important in terms of trade, cultural ties, and capital flows than China.  

Trump wins this contest.

Carney overplayed his hand.

(Long-time readers now, we believe in the virtues of free trade and markets. I am not a fan of Americans paying taxes on Canadian imports. I’d like to see more trade between the US, Canada, and Mexico. I’m simply reading how I expect things to unfold. That’s our job.)

Markets

Software Is The New Restaurants

Last week, we wrote how software was approaching a major bottom and that we expected the turn to happen within about 2-5 business days.

The tape played out on Wednesday - IGV created a hammer, indicating buying pressure, and then, had two consecutive green days. 

Interestingly, IGV’s current chart coincides with prior instances of strong recoveries. 

IGV's bounce after a decline below 200-DMA with RSI under 30 has followed solid recovery in all instances since 2017. 

See this chart. H/T Seth Golden

The recovery in IGV is backed by solid fundamentals.

SAAS names, like Hubspot have double-digit revenue growth, expanding user base and consistent cash generation. 

The recent downturn has also bought them to better valuations.

Speaking of Hubspot:

(1) Lumida signed a contract with them last month

(2) I expect Google or Microsoft will seek to acquire them. Google tried before but failed under the Biden FTC. Google’s stock price is high, and SaaS valuations are low.

I could see Mag 7 names go on an M&A spree.

Moreover, the narrative around ‘AI is eating SAAS’ is fading. 

Large companies are not ripping out their incumbent SaaS vendors to rebuild everything internally. 

Instead, they’re leaning harder into existing platforms for AI add-ons and functionality. This shows up in increasing retention rates and improved revenue per user across major SAAS providers. 

This tweet from Obsidian Capital sums it up.

We increased our exposure in software on Wednesday by adding to Salesforce (CRM), Hubspot (HUBS), Adobe (ADBE), and Duolingo (DUOL).

We did sell Atlassian (TEAM). We noticed their Share Based Compensation was excessively high. That doesn’t mean the stock won’t rally anyway, it probably does with the broader peer group.

We talked about them in our last newsletter. View here. 

The Gains Were Due For Meta 

Another recovery we saw this week was Meta. 

The chart looks fairly similar to IGV - trading below its 200DMA and RSI touching 30. 

Meta’s recovery began on Wednesday after it announced the plan to roll out ads on Threads.

Threads has become the fastest Meta platform to gain 100 million active users - it has outperformed most social platforms. 

Published numbers show active users on threads are now outpacing X - Threads has around 141.5M active daily users compared to ~125M on X.

The monetization plan is well timed. This is likely to act as a significant tailwind to Meta’s revenue, and will also give the stock a breather before earnings. 

Overall, Meta is one of the most exciting businesses worldwide - it has a user base exceeding 3 billion, generates some of the strongest margins in large-cap tech, with operating margins north of 40%.

It has a history of producing consistent revenue and earnings growth with reliable cash flows.

The growth engine is advertising, and AI is making it more powerful. 

Management disclosed that the annualized run-rate of its AI-powered advertising stack has surpassed $60 billion (30% of 2025 revenue).

Recommendation systems across Facebook, Instagram, and Threads are improving engagement, driving more time spent and better ad performance. 

Meta is still significantly underpriced in the Mag-7 land with a P/E NTM of 22.5x and revenue growth north of 20%. It is also one of Lumida’s 2026 picks. You can read our thesis here. 

Meta is set to report on Wednesday - optimistic reports on capex returns can send the stock soaring, but any misses on revenue, earnings, or capex guidance will hurt the sentiment.

The Bubble Around OpenAI

OpenAI is chasing a $50B funding round from Middle East investors at a valuation of $830B. 

They have previously raised $40Bn from Softbank last year, and if this $50Bn raise completes, that’s 90% of what they were aiming for in this round. 

The capital infusion is good news for semis, and keeps SMH moving. It has had a good run since Nov lows.

However, the fundamentals supporting OpenAI's valuation remain questionable, particularly given its aggressive revenue projections. 

Sam Altman has suggested the company could reach $100 billion in annual revenue by 2027—an extraordinary leap from current levels of $13 billion annualized in late 2025.

This is only possible if they see lightning-fast monetization across early-stage products - not to forget they are competing in a market with Google and Meta. 

Google and Meta have durable cash flows, profits, and ecosystems that make adoption seamless – their AI models slot right into existing product suites.

For instance, Google users can use Gemini within Docs, Sheets, Gmail and other google products – users don’t have inertia in using already integrated AI.

Add to it, OpenAI’s weakening position in the competitive landscape. 

See this tweet from Jason. 

Note: Lumida recently switched the LLM that powers its Charting AI from OpenAI to Google Gemini’s Flash. We cut our costs by 50%. It took a few days to transition.

You can’t be sleeping well at night if you’re Sam Altman.

Simply put, OpenAI can’t command pricing power in an industry where competition is as good as them, or even better.

Moreover, OpenAI also sees headwinds in converting free users to premium models.

Users look at AI models as ‘web search’ - they don’t feel excited about paying for something as simple as a search. So, the pricing models used to project the $100M might not work. 

We are already seeing the signs. 

I predicted it back in Dec. 

With competition intensifying, and users’ preference for free AI models, OpenAI’s road to $100M revenue isn’t going to be easy. 

In my opinion, the bubble is in OpenAI’s $830Bn valuation. 

When will it burst?

Probably, when OpenAI launches its IPO.

Latin America's Connectivity King

We extended our international exposure by adding América Móvil (NYSE: AMX) to our portfolio on Friday.

América Móvil is a dominant wireless and broadband services provider across Mexico, Brazil, Colombia, and beyond.

It serves over three hundred millions subscribers in Latin American markets, and projects consistent subscriber growth due to growing digital demand.

The company has maintained its market leadership for years, and recent quarters show the business is accelerating.

Management is delivering solid returns with strong postpaid subscriber gains (especially in Brazil), improving revenue trends, and expanding EBITDA margins even as they invest heavily in 5G networks.

Mexico remains the crown jewel in AMX’s portfolio, with leadership in coverage and quality that keeps customers upgrading to higher-value plans.

The shift from prepaid to postpaid is bringing better economics and stickier relationships.

What stands out is the cash flow story.

Free cash flow has surged recently, giving the company room to reward shareholders through consistent dividends and buybacks while steadily reducing debt.

The balance sheet looks healthier with lower debt, and improved cashflows. Management stays disciplined—focusing on network upgrades and selective opportunities rather than overpaying for growth.

The stock started a downtrend after hitting its all-time high of $23.7 in late November (its rally had a gain of 72% post-April Lows).

But, the stock seems to have capitulated now, with recent price moves indicating the recovery is real.

AMX trades like a reliable compounder that's finally getting recognition as macro conditions stabilize in key markets.

The bear case is real but contained:

Currency swings in the region can create short-term noise on reported numbers, and any slowdown in economic activity could pressure the prepaid base or delay upgrades.

Debt levels, while improving, still require careful management, and occasional one-time items (like litigation) can pop up.

Yet the moat is wide—spectrum assets, scale, and market leadership make it hard for competitors to dislodge.

In a world hungry for steady cash generation and emerging-market exposure, AMX offers both without the hype.

Why chase crowded U.S. trades when you can own a proven cash machine riding structural tailwinds in accelerating geographies?

Lumida Curations

AI-Powered Fighter Jets are Revolutionizing Air Combat

AI-driven fighter jets eliminate the need for human pilots, enabling them to take on high-risk maneuvers to protect more valuable targets.

View Curation

The Future of Money Movement: Zero-Cost Transactions

Circle CEO explains how advancements in technology could reduce the cost of moving and storing money to zero, much like the evolution of data transfer and software publishing.

View Curation

Elon Musk on Aging: A Solvable Problem

Elon Musk believes aging is a problem that can be solved, suggesting that a synchronizing clock exists in our cells and that we could eventually discover ways to extend and even reverse aging.

View Curation

Meme

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: The Economy Is Resilient; The Consumer Is Strong; Corporate Confidence is Solid; AI Is Becoming A Real Efficiency Engine

• Markets: GPUs Aren’t iPhones; The Attention Asset With A Solid Story; Software: Dead Or Alive?; Constellation Software; International Value; Timing is Everything

• Lumida Curations: Fiscal Discipline Is the Market Risk; VO₂ Max Is the Longevity Signal

Software: Dead Or Alive?

Back in 2023, I made the case that AI is highly disruptive to firms like Adobe and SaaS more generally. I also made the case that Software as a category was over-priced.

What we see now is an inverse position.

We see quite a few software stocks whose share prices are battered, yet their sales and earnings continue to rise.

We’ll show you a few examples.

Take a look at Atlassian Corps’ sales growth vs. stock price.

Notice Sales (the blue line) is stair stepping higher year after year, but the stock price (the black line) has sold off.

Now let's look at Salesforce.

Same story.

Sales and earnings are headed up, but the stock price is dislocated.

Adobe has transitioned to a Cash cow - it's no longer a growth machine.

But, sales are still growing, and the installed base is sticky. The stock now has a 17X trailing and 12.6X forward PE with free cashflow yield of 6.6%.

HubSpot has the same story. Sales keep growing, but stock price is severely dislocated.

When an entire category sells off, but the fundamental sales are improving, then you have two situations:

(1) Narrative pressure. Here the narrative is ‘Death by AI’.

(2) Valuations are too high

For software, the valuations were indeed too high and some names are still too pricey. 

But, the EPS growth relative for at least some of the names, is looking attractive now.

(I say “some of the names”, because you still have firms like DataDog, MongoDB, and other software firms that trade at pricey valuations.)

Let’s look at narrative pressure.

Is AI truly disrupting SaaS?

Earnings calls shows management is incorporating AI.

And many of these SaaS firms have an installed customer base (e.g., data, integrations, hooks, etc.). It's not that easy to transition.

Timing is the hard part. 

Software reminds us of Restaurants. Both topped out around the same time and we were bearish on both back then.

Fast forward, restaurants appear to have exited their bear market, and are rallying.

We believe Software will do the same. 

The last phase of a bear market is the most difficult. 2/3 of losses happen in the last 1/3 of a bear market.

The best strategy here is to only be ‘feet’ in the water. You don’t want to be fully exposed to the price swings. You want to wait for ‘pivot points’ to emerge and demonstrate there is a change in trend. Then add to a position.

We believe software should bottom this coming week. We are within 5 business days of an intermediate low. Software probably bottoms around the same time as semiconductors top.

The two sectors continue to move in inverse directions.

IGV ETF

I

Semiconductor ETF

We purchased TEAM two weeks ago, but took an 18% tax loss harvest on the name and had a position size of about 2%.

At the same time, we purchased Hubspot (HUBS) and Intuit (INTU). TEAM is still a high-quality story.

Hubspot is Approaching 2022 Lows Despite Higher Sales

Intuit

But, not taking advantage of a tax loss harvest and rotating into another dislocated name that shares the same factor exposure would be a mistake.

If your marginal capital gains tax rate is, say, 42%, the IRS is effectively reimbursing ~42% of that loss in the form of lower taxes owed. That is a real, cash-equivalent value.

By selling our loss-making positions, we realize losses that can offset our gains - dollar for dollar.

It helps us mitigate taxes for our client, and achieve higher after-tax capital return. If you’d like more details about it, view our strategies here. 

The Rise Of Defense Tech

The USA’s attack on Venezuela led headlines last week. Trump has another long weekend, and Iran is on the radar. 

Polymarket also has non-zero odds for an attack on Mexico, Somalia, and Colombia.

Geopolitical tensions are intensifying, and it is not just the USA.

China is on the brink of a war against Taiwan.

Russia is already at war with Ukraine. Iran and Israel have been the headline for over a year now.

Countries are sensing the fear, and they are countering with increased defense spending.

World military expenditures rose 9% last year - the sharpest increase in more than three decades. 

Defense allocations are headed toward military autonomy and AI. Countries are aiming to have intelligent mechanisms that work on their own. 

Anthony Antognoli, USCG’s executive officer: “Our vision is to have robotics and autonomous systems as the foundation for the way the Coast Guard operates its missions.”

“It is impossible to do that work [defense missions] with humans and patrol cutters alone.”

These spending increases are creating massive demand for defense tech names.

Venture capital has realized this.

In 2025, defense tech venture investment was already more than double 2024’s total. We don’t see it stopping anytime soon.

The funding is concentrated in large, late-stage rounds of autonomy focused defense companies.

Companies like Anduril and its peer group are benefitting from these financing rounds.

Those financings are only available to sophisticated private markets investors.

Lumida Ventures is also focused on investing in this category. We have had a string of successful investments across CoreWeave, Kraken, and Brad Jacobs QXO, and Canva.

Past Results are not indicative of future returns.

Do reach out now if you want to be in the flow of our next deal.

If you are an accredited investor or qualified purchaser, sign up here to receive communications about our private deals.

Macro

Earnings Highlights: Large Banks

Earnings season started this week, with large banks –- Goldman, Morgan Stanley, JP Morgan, Wells Fargo, Bank of America, and Citi — reporting.

Bank earnings have been our gauge for the broader economy. 

These releases tell us exactly where the economy is headed, and unlike official government data, the comments don’t have revisions.

The Economy Is Resilient

The Q4 earnings commentary second what we have been saying about the economy. 

Economic activity levels are healthy: jobs are steady, business customers are operating with confidence, and growth is supported by a mix of consumer health and ongoing investment.

Bank of America’s Brian Moynihan: “It was a pretty good environment as we moved through the year 2025… Unemployment in labor market is stable.”

He described business conditions as constructive: “When you go to our corporate commercial customers… They had a pretty good year and good profits, including good credit quality and good money movement activity.”

Morgan Stanley’s Ted Pick framed the same story at a high level, calling out the durability of growth drivers: 

“In 2025, the U.S. economy proved resilient as ever. As predicted, the capital markets are kicking in with well-capitalized corporates and higher-end consumers driving the economy forward.”

Citi went even broader on the macro setup and why growth can keep carrying: 

“Turning to the macro. The global economy has powered through many shocks over the past few years, creating optimism and confidence that economic growth is poised to continue. With inflation now at normal levels globally, almost every central bank is becoming more accommodating.”

“Capital investment remains strong, especially in tech. And it's the combination of that CapEx, the health of the consumer, the tax bill benefits and anticipated rate cuts that should be enough to sustain growth.”

Wells Fargo also stayed in the same lane: “The economy and our customers remain resilient…”

The common message is that the economy is still producing the things that matter most for growth—jobs, corporate activity, and ongoing investment.

The Consumer Is Strong

Across the calls, banks are describing a consumer that is spending, confident, and not showing any kind of broad-based stress.

Bank of America’s Brian Moynihan (CEO) signaled towards stabilizing growth: 

“Consumer spending grew 5% over the 2024 levels. Account balances in the consumer business for that broad base of the U.S. consumer were stable through the year.” 

“Delinquencies and charge-offs improved in 2025 consumer credit. Unemployment in the market remains stable… This strong consumer health bodes well for the continued improvement in growth in 2026.” 

His CFO, Alastair Borthwick, reinforced that the trend is still moving in the right direction: 

“So we're still seeing the growth in each of the consumer categories. And that feels like it's in a position where it's likely to continue to grow from here.” 

Wells Fargo described the same dynamic from a fundamentals lens: “consumers continue to be resilient as income growth has generally kept pace with increases in inflation and debt levels.” 

Michael Santomassimo (CFO, WFC): “both credit card and auto losses were lower than a year ago.” 

Citi’s Jane Fraser pointed to engagement rather than retrenchment: “Branded Cards revenue grew 8% driven by robust engagement from customers in spend, borrowing and new account acquisitions…” 

And Citi CFO Mark Mason noted credit health was stable as well: “as we look at delinquencies, we're not seeing anything unexpected in either of the portfolios. And even when we cut it by different FICO scores and income brackets and the like.” 

The takeaway is optimistic: the spending engine is still on, and the credit data is not contradicting it.

Corporate Confidence is Solid

Bank earnings also signaled towards confident corporate boards. 

Deal Pipelines are rebuilding, and strategic activity is accelerating—usually the first sign that boards and sponsors are getting more comfortable putting capital to work.

Wells Fargo’s Charles Scharf (CEO) said, “We entered 2026 with our deal pipeline meaningfully greater than it has been at any point in the last 5 years…”

Morgan Stanley framed the same shift across sectors: “Investment Banking pipelines remain healthy, global and diversified across sectors. Strategic activity is accelerating.”

Goldman Sachs’ David Solomon (CEO) also echoed the same narrative: 

“Over the last year, we’ve seen high levels of client engagement across our investment banking franchise, and we expect the activity to accelerate in 2026.”

“I think the world is set up at the moment to be incredibly constructive in 2026 for M&A and capital markets activity.”

Citi’s earnings show they are already seeing the results of increasing corporate deals: “Investment banking fees increased 35%. M&A was up 84% reflecting a record quarter that closed a record year with momentum across several sectors…”

The read-through is bullish: strategic decision-making is restarting. 

When M&A, underwriting, and sponsor activity begin to accelerate, it typically means the corporate side of the economy is regaining confidence in the path ahead.

AI Is Becoming A Real Efficiency Engine

Banks are increasingly describing AI as a margin lever: less manual work, faster delivery, and more capacity without adding headcount.

At Bank of America, Brian Moynihan quantified the impact in software: “We have 18,000 people… who code. And by using AI techniques, we’ve taken 30% out of the coding… That saves us about 2,000 people… it’s several hundred million dollars.”

He also framed this as enterprise-scale enablement: “We have rolled out 365 Copilot across 200,000 teammates… we expect to get good leverage from that.”

BAC is also implementing AI in client service - Brian Moynihan flags how they have reduced service “touches” through proactive alerts: “you can set up alerts… the alerts are up to… billions a quarter…”

Brian Moynihan highlights AI in client service helps them in providing the “best customer experience at the lowest cost… which help [us] grow the business for 7 straight years in checking accounts”.

Citi’s Jane Fraser described AI as part of an operational transformation: 

“When combined with how we're deploying AI, this bank is being truly transformed in terms of its operational capabilities…”

She backed it with adoption: “Colleagues in 84 countries have now interacted with our proprietary tools over 21 million times… It’s now above 70%.”

The end goal is straightforward: “use AI tools and automation to… reengineer and simplify our processes… to improve client experience whilst reducing expenses.”

Goldman is still early in the journey, but is eyeing AI as an operating-model reset: 

“Goldman Sachs 3.0, our new operating model propelled by AI. We are excited to embark on this effort, starting with 6 work streams we identified as ripe for disruption. We will invest to reengineer these processes from the ground up.”

David Solomon tied it to multi-year productivity compounding: “GS 3.0 creates an ability for us in the next 5 years to accelerate the pace of Revenue per employee once again.”

AI is now being managed like a cost and productivity program. 

On a side note, the AI enterprise implementation is essential for the AI flywheel. 

Hyperscalers’ capex as % of OCF has reached its all time highs, and it doesn’t seem to slow any time soon.

The capex trickles into revenues and earnings across the flywheel from datacenters, chips, packaging, energy, infrastructure and more.

If this capex doesn’t drive returns, it will eventually slow down, hurting the flows for the entire chain. 

So far, banks’ commentaries show enterprises are actively considering and investing in AI implementation, which drives demand for the application layer. This gives us a reason to be confident.

Claims Say Labor Market Is Healthy

This week, we got initial jobless claims and continuing claims, both coming in better than expected. 

Initial claims are running below 2025 and 2023 levels and only modestly above the unusually low 2024 prints. 

This shows layoffs remain contained and the labor market is cooling through lower hiring.

Continuing claims tell the same story. 

They are sitting roughly where they finished 2025 and have performed better than expected.

Although the labor market looks stable holistically, we see increasing weakness in younger segments. 

The unemployment rate has ticked up to 15.7% for 15-19 age, and 8.2% for 20-23. It has been in an upward trend since 2023, and spurred higher in 2025. 

The problem isn’t really demand weakness - It’s skill mismatch.

See this tweet from Robbie Hendricks. 

Employers are “short labor” in specialized roles (engineering, technical manufacturing, defense/industrial work). 

Fewer students are completing rigorous STEM/technical tracks that companies actually need.

These jobs stay open for months, while more candidates compete for general roles, where competition is intense. 

This dynamic pushes up unemployment for workers who are newest and least credentialed, while the experienced labor pool stays employed and the layoff data stays quiet.

Women Moved Left

The partisan gap between young men and women has almost doubled in the last 25 years. 

And, it isn't the case for US alone. You can notice it in the UK, Germany, South Korea and most other nations. 

While the street argument suggests men have become conservative, it's the opposite. 

Ideology-wise, men have stayed with the same right-tilt, while women have drifted far more towards the left. 

The article by Vittorio discusses why this happened, and what it implies - a really interesting read. 

Markets

GPUs Aren’t iPhones

H100 rental pricing has been rebounding since the November lows. It’s now near an 8-month high.

Datacenter linked names including CoreWeave have rallied along these results.

This shows compute demand is running hot. 

But, it also pushes down on a more critical bear claim against AI infrastructure names that “GPUs become obsolete quickly once a new generation ships.”

Compute isn’t consumer hardware.

Each GPU generation gets slotted into a workload ladder. 

New chips handle the hardest training. Older chips migrate to inference, fine-tuning, and cheaper, performance-per-dollar work.

This is why we said Michael Burry was wrong in his October claims. 

His argument was that hyperscalers are stretching GPU life to understate depreciation and overstate profits.

But his view assumed GPUs were short-lived and get replaced in a hurry. 

Here’s what we wrote back in Nov refuting his argument. 

The rental market offers the most compelling evidence on how he was wrong. 

If older supply were being stranded, pricing would be breaking down, not moving higher.

GPU Demand is also widening. They are no longer just required for frontier model training, but are also used for inference everywhere, across more companies, more products, and more use cases. 

That keeps older hardware economically relevant.

The rising rental prices for older GPUs is also a bullish evidence for CoreWeave. Longer useful life means the capex keeps generating returns for more years to come. This creates operating leverage, and better profitability.

We wrote how Coreweave is misunderstood in our previous newsletter. Read our thesis here. 

Micron - The Attention Asset With A Solid Story

Micron was up ~8% on Friday, after they initiated a ~$100B semiconductor manufacturing facility near Syracuse. 

Sanjay Mehrotra (CEO, MU) says this manufacturing facility will “power the AI revolution.” 

The project is also being supported by more than $6B from the CHIPS Act, plus over $8B in federal, state, and local tax incentives. 

Chuck Schumer, Minority Leader of the United States Senate, said it will give the U.S. “the lead in semiconductor manufacturing for generations.”

Micron had an investor group meeting this week, and the takeaway was consistent with what they said in Q1’26 call: customers want more bits than the industry can deliver. 

Mark Murphy (CFO,MU): “We are only able to meet about 50% to two-thirds of our demand from several key customers.” 

Mehrotra (CEO) echoed the same dynamic: “We see a very, very tight supply environment… there is a large gap between the demand and the supply.”

Mehtora mentioned MU is expanding capacity at a 20% YoY rate, while demand is expanding at 30%. He expects “Demand will likely continue to outstrip supply” till 2027.

When the gap is that wide, the market clears through price.

Murphy highlighted MU’s pricing power in Q1 '26 call: “We expect higher price, lower cost, and favorable mix to all contribute to gross margin expansion in Q2.” 

And, you can already see the initial signals in Q1’s numbers. 

In fiscal Q1’26, Micron did $13.6B (57%+ YoY) of revenue with 56.8% gross margin. DRAM revenue was $10.8B (69%+ YoY) and NAND was $2.7B. Free cash flow was $3.9B. Operating income was $6.4B (20% higher YoY with a 47% operating margin). 

HBM capacity is sold out through CY26, and management expects the demand growth to continue at the same rate till 2028.

MU has returned 80% in the last 3 months. This is one of Markets’ attention assets for 2026, alongside other memory stocks like Sandisk. 

Attention stocks are often driven by momentum and liquidity even when the valuations go bizarre, although not the case for MU with P/E NTM under 10x.

I did an X post on how the animal spirits basket for 2026 is different from 2025. Read it here. 

We bought MU on Dec 30th, when the stock traded at around $282 - today, it has risen to $362 - a 28% surge in 12 days.

Micron is an example of a lockout rally. The stock runs higher and doesn’t create an attractive entry for an investor. The main goal in these situations is to “get involved”.

When the trend breaks on a high momentum name like this, then we would sell.

Constellation Software 

I guess it’s fashionable to point out how ‘quality’ darling Constellation Software is down 40%. 

CSU was on Goldman’s conviction list last year if memory serves. 

I pointed out Constellation was insanely over-priced back in q4 ‘24 at 100x. 

You will have great grandkids before you see your principal. 

Interestingly, the name rallied 6 more months on momentum. 

Then it rolled over. 

A lot of other ‘quality’ names like FIGO and CAVA and others also cracked. 

A lot of ‘quality’ investors that went to school studying Buffett forgot that valuation matters. 

1) Price is what you pay, value is what you get. 

2) There is no single correct investment style for all times. 

If you identify as a growth investor or value investor or GARP investor or momentum investor, that’s picking ideology over science. 

3) The best investors in the world - Two Sigma and Citadel - have no such bias.

International Value

International markets are running circles around U.S. markets.

Greece, South Korea, Turkey, China, Brazil, Mexico, Canada.

Even France.

And Vietnam.

We continue to add international value names to our portfolio. 

We discussed our thesis about international outperformance in our last newsletter. Read here. 

I also did an FSD stream detailing how geopolitical uncertainty is also one of the reasons behind International value leading. 

Timing is Everything

Bitcoin purists see decentralized money as naturally having value.

I see it differently.

Bitcoin wouldn't have achieved adoption, nor value, without the platform formerly known as Twitter.

That's especially true for non-cashflowing assets that are stories about the future.

Value is social.

It's not a coincidence that Bitcoin took root just as X was taking off as a platform.

And, by the same logic, social media has transformed investing forever.

You really have to think differently about assets that build community around them...

This is the main lesson for the performance of Palantir, Tesla, and Robinhood stock prices.

But, like other attention assets, you need the concentric circles to keep growing to propel values to higher levels.

And when competing assets enter the picture - drone stocks, space stocks, <insert hot thing here> stock - those assets compete for attention and fragment liquidity.

Lumida Curations

Fiscal Discipline Is the Market Risk

Markets keep watching the Fed, but Larry Fink argues the bigger swing factor is Washington’s balance sheet. With US debt nearing $38T and deficits still rising, financing risk can push long rates higher even if inflation cools.

View Curation

Longevity: VO₂ Max Is the Longevity Signal

Cardiorespiratory fitness is the clearest single proxy for long-term healthspan—because it reflects how efficiently your heart, lungs, and muscles deliver and use oxygen under real-world stress.

View Curation

Meme

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: Fed Rate-Cut Remorse Is Creeping In; Did Rate Cuts Affect the Labor Market?; Housing Policy Whiplash

• Markets: TechnipFLP - The Upstream Play Around Venezuela; Atlassian Corp - Quality on Sale

• Lumida Curations: The Hidden Credit Crunch; CoreWeave's Secret Sauce; The Toxic Insider Wrecking Your Metabolism

Spotlight

This week, I had a podcast with Danny Goler. 

More and more studies from Johns Hopkins, NYU, and Imperial College are documenting the therapeutic benefits of psychedelics.

Danny is a ‘psychonaut’ and one of the few human beings to take a sustained DMT drip. He shares his experience and lessons. 

Here’s what we talk about:

• The “Code of Reality”: What Danny Goler Claims to Have Discovered

• Quantum Physics, Double-Slit Experiments, and Consciousness

• Are We Living in Parallel Worlds or a Holographic Universe?

• Humans as AGI: Intelligence, Alignment, and the Future of Civilization

Watch the podcast here. Danny is hosting a retreat in Costa Rica (details here.)

Run It Hot

GDP is running north of 5%.

The Fed cut rates into a blistering hot economy — rate cuts the economy does not need.

Stimulus checks are still to come..

Mortgage rates now have a five-handle.

That’s the setup.

Add to it Trump’s economic agenda, which taken together, is center-left and inflationary. 

Overall, we have a populist economic policy mix. It’s designed to win mid-term elections. It also has longer-run inflationary consequences.

Consider the mix:

1. Tariffs 

Milton Friedman would say no. AOC would say yes.

2. Ban on institutional ownership of homes 

“Corporations don’t live in homes.” Mitt Romney would disagree. 

Corporations rent out homes to people. They are pass-thru intermediaries that often can buy a home at a lower-cost of capital than renters locked out of a mortgage market,.

3. Cutting rates despite 4–5% GDP growth

Elizabeth Warren has always pushed for rate cuts at every Humphrey-Hawkins FOMC Chair congressional testimony.

4. Tax cuts framed as wage support, not for capital formation

Bernie Sanders would say yes. Glenn Hubbard would say no.

5. Opposition to entitlement reform

Deficits stay large.

6. Drug price controls

Both Elizabeth Warren and Bernie Sanders support this.

7. Buying mortgage-backed securities

An Obama–Bernanke era policy. It inflated housing prices and priced out the next generation.

8. Capping credit card rates at 10%

Sounds consumer-friendly. In practice, it restricts credit and pushes risk into unregulated markets.

You can debate each item individually.

But taken together, the direction is clear: more demand stimulus, more intervention, more inflation pressure.

What are the implications?

First, why would you own bonds here? Owning sovereign debt is a truly foolish proposition.

Especially when you consider there are dozens and dozens of public market securities with a 7% free cashflow yield and real earnings growth.

If you’ve followed us closely over the last two years, you know we have been consistent in the view that the U.S. consumer is strong,  and the real economy does not need rate cuts.

The 5%+ GDP print supports that view – as does inflation prints that are running at a 2.8% level three years after the Fed started raising rates.

What we have witnessed over the last several years is the game theory of rate cuts and inflation.

We are seeing how and why inflation cycles and rate cycles happen. All of this exacerbates the business cycle and leads to greater asset price volatility.

The simple fact is that the incentives of whomever is in power is to spend, cut rates, and win the next election.

The Fed Reaction Function

The Fed undoubtedly is having rate cut remorse. 

Note: The effects of the most recent rate ‘adjustment cuts’ have barely hit the real economy.

Monetary policy and fiscal policy should be counter-cyclical.

The logical response to all of this - after my hand-wringing - is to stay bullish on real assets.

The ‘run it hot’ policies mean you have an imperative to own assets that benefit during inflationary regimes.

Notice commodities of all stripes - Gold, Copper, Silver, Uranium - are rallying. Material stock indices and names like Freeport McMoran are breaking out.

Mr. Market sees inflation coming. I see it too.

The right way to approach this - stick to the trend. Avoid sovereign debt. Own assets that can pass thru inflation efficiently or re-price in nominal terms efficiently.

Here’s what FOMC officials are saying.

Minneapolis Fed President Neel Kashkari acknowledged the ‘higher for longer’ narrative on Friday. 

He noted policy “pretty close to neutral,” and the economy doesn’t appear to be under meaningful downward pressure at all. 

Philadelphia Fed President Anna Paulson reinforced that message, signaling that further cuts could be “some way off” and would be modest and conditional.

That’s exactly what rate-cut remorse sounds like.

In an environment with strong nominal growth, fiscal stimulus, and a policy mix that leans demand-side, why would you own bonds here?

This is a higher-for-longer setup. 

The third implication?

60/40 portfolios are not poised to do well here. 

Defense Spending Is Resetting Higher

Proposed defense spending is set to target a whopping $1.5 Tn.

That’s also inflationary. 

President Trump called this week for U.S. defense spending to rise to $1.5 trillion by 2027, more than 50% above the current $901 billion budget. 

This spending cycle is increasingly about autonomy, robotics, sensors, and AI-driven systems, not just traditional platforms.

Defense agencies are directing multi-year budgets toward technologies that reduce reliance on manpower and can operate at scale across large, complex environments.

Companies like Anduril, and its peers, are well positioned in that environment. 

They focus on autonomous systems, rapid deployment, and integrated AI platforms designed for modern defense missions. 

These are no longer experimental programs. They are becoming core infrastructure for how missions are executed.

The U.S. is refactoring its “defense tech stack”.

Trump has also been explicit that higher budgets will come with pressure: faster delivery, domestic manufacturing, and tighter control over cost overruns. 

That dynamic favors agile, tech-first defense firms over traditional primes.

Lumida Ventures is focused on investing in defense tech. 

We have had a string of successful investments across CoreWeave, Kraken, and Brad Jacobs QXO, and Canva. All of these but Canva (too new) have had markups or realizations..

Our strategy is simple. Front-run Pentagon spending. Find private companies priced attractively with product market fit that are poised to win government contracts.

Do reach out now if you want to be in the flow of our next deal.

If you are an accredited investor or qualified purchaser, sign up here to receive communications about our private deals.

Macro

Did Rate Cuts Affect the Labor Market?

The one thing the Fed wanted rate cuts to do was help the labor market, and the one thing that rate cuts have not changed is the labor market. 

The labor market was never cracking as the hyperbolic media would like you to believe, it was cooling and moving into balance.

New hires match fires and quit rates.

Even as hiring cools, economic growth remains intact thanks to the magic elixir of productivity growth. 

Hours worked are barely increasing, yet output continues to rise. 

In effect, the economy is growing with less labor input.

That shift carries important inflation implications. 

When productivity rises faster than wages, unit labor costs fall. 

Firms can pay workers more without passing those costs through to prices because each worker produces more. 

Automation and AI are the structural drivers behind this dynamic. 

Companies are not aggressively cutting headcount; they simply do not need to hire at the pace they once did. 

Software, automation, and AI are substituting for incremental labor, lifting efficiency and lowering marginal costs. 

The result is slower hiring, stable employment, and rising productivity—not labor market distress.

This is actually how AI flywheel strengthens. 

Companies are experiencing a positive return on AI and automation - This validates further investment in AI infrastructure, software, and compute. 

And, higher investments means the AI flywheel continues to expand.

One of the ways we are betting on AI adoption is thru an investment in Accenture.

The name entered a deep correction shortly after the Trump inauguration. It now has a 7% Free Cashflow Yield and the technicals have turned up.

We believe Fortune 500 companies need change management to re-factor their workflows around AI. Names like Accenture should do well.

Here is a chart of Accenture’s Free Cashflow Yield.

Here is a chart of Accenture’s stock price overlaid with key technical events.

The entry isn’t ‘perfect’  right now, but might be worth accumulating on the next down day.

I ask you again: why would you own bonds yielding 4% when you can own a business with a 7% yield and earnings growth?

Housing Policy Whiplash

Housing markets were hit with conflicting policy signals this week, and prices moved accordingly.

First, President Trump said he would push Congress to ban large institutional investors from buying single-family homes. 

Markets reacted immediately. 

Stocks tied to institutional housing ownership—Invitation Homes, Apollo-linked platforms, and related exposures—sold off sharply.

Institutional buyers act as a clearing bid for housing inventory. 

Whether a home is owned by a family or a corporation, it still gets occupied. 

What changes is ownership structure, not demand for shelter.

Removing institutional capital reduces one source of demand and slows inventory absorption—especially problematic when homebuilders are already sitting on rising inventories and weak earnings momentum. 

Less clearing demand means longer sell-through times and more pressure on prices. From that perspective, the market reaction made sense.

Banning institutions from owning homes doesn’t address home ownership affordability.

Affordability turns on mortgage finance costs.

Trump followed up with a call to purchase MBS.

Trump followed by calling on Fannie Mae and Freddie Mac to purchase $200 billion in mortgage-backed securities, explicitly targeting lower mortgage rates. 

Markets immediately understood the implication.30-year mortgage rates dropped 0.22% to 5.99%, first print (barely) below 6% since August, 2022.

The mere announcement of this intention caused benchmark mortgage rates to hit a 5% handle for the first time in many years.

This is bullish for U.S. consumers who can tap their home equity.

Lower rates improve affordability, and homebuilders rallied sharply on the news. XHB reflected that shift.

Between the two - this policy is the one that matters.

But lower rates come with trade-offs. 

Demand for housing will increase. That will spur home price asset inflation (not unlike Covid, but less pronounced).

In already tight markets, that dynamic risks pushing prices higher rather than improving affordability in the long-run.

Until policy makers address supply - via permitting, deregulation, NIMBY concerns, and allowing accelerated depreciation - it’s hard to see a sustained fix to housing supply..

Markets

SELL ME THIS PEN

OpenAI is building a pen.

This pen does not have product market fit.

Neither did the Apple Vision Pro.

(Is it made of silver, at least!?)

Question...

What financial projections attached to the 'pen' are baked into OpenAI's capital raise?

I've said it before, and I'll say it again.

The bubble is in OpenAI stock.

Google’s Gemini is growing more rapidly.

Do the math.

OpenAI wants to raise $100 Bn.

75% of that will go to Nvidia, who has a 75% margin.

Attach a 40x PE ratio to $50 Bn in earnings and you have about $2 Tn in market cap, give or take, linked solely to OpenAI capital raise.

Right now, Mr. Market will 'advance' OpenAI as it achieves financing milestones.

The semi market was set to crack a few weeks ago. 

But, the reports in the WSJ and Information saved semis.

The semi market was set to crack last week.

But, Softbank saying they are committing $40 Bn to OpenAI saved the day.

Each day, those news announcements arrested emerging downtrends.

The market is saying ‘So long as OpenAI can raise financing for capex, the show will go on.’

There’s a good logic to that. After all, earnings numbers for the semi ecosystem will hit if OpenAI gets financing.

We are now in the game of musical chairs. One of the most important variables to watch in markets is OpenAI’s ability to raise financing. When that dynamic stops - and we are in the later innings the IPO being inning #9 - you will want to be cautious around risk.

Each successive financing check is hard than the last.

OpenAI now needs to find another $60 Bn to close out the raise. That's bigger than most public offerings, and it's happening privately in non mark-to-market valuation land.

OpenAI must go public - by necessity. 

OpenAI is not going public as a "win" - but because they need capital to fund its capex obligations.

So, here we are.

Fanciful projections meet increasingly skeptical capital markets.

We are in the Sam Altman 'It's so over' -> 'We're so back' cycle.

If Brad Gerstner or 'smart money' doesn't invest in OpenAI's private round, markets will call bullshit on Sam Altman.

With Google Gemini gaining market share and growing even faster than OpenAI, it's clear to me OpenAI is not in a good place.

We’ll continue to keep an eye on this variable…

TechnipFLP - The Upstream Play Around Venezuela

The Venezuelan attacks led actions since last week - the obvious trade was Chevron. 

They stand to make real money from this—hundreds of millions in incremental annual free cash flow. 

But the better opportunity is not the operator. 

It’s the infrastructure behind it. That’s oilfield services. The subsea work. The execution layer.

If you look at that category—Halliburton, Schlumberger, Weatherford, and FTI—one name stands out. 

TechnipFMC (FTI)

TechnipFMC is the oil field service names most levered to Latin America—Brazil and Venezuela in particular—and it sits directly inside the execution layer of offshore projects. 

FTI was already moving before the news broke. 

It was up hard late last week, then gapped again on Monday, and it’s been up every day since. No pullback.

That usually tells you something. When a stock doesn’t give you an entry, it’s because investors are waiting for a good entry that never comes.

FTI spent years redesigning itself to make offshore oil projects attractive again. These are large, complex projects that take years to plan and billions of dollars to execute.

What FTI changed is how those projects get built.

Subsea 2.0 is their new operating model for building the equipment that sits on the ocean floor. Think pipelines, connections, and other systems that bring oil and gas up to the surface. 

About 80% of FTI’s work is now directly awarded, meaning clients don’t even put the job out to bid. 

They’re saying: “we trust you, we know you’ll deliver, and we’re not taking the risk of shopping around.”

FTI has a backlog—around $18 billion as of Q3—but the quality of that backlog matters more than the size.

The risks are straightforward. The Venezuela story is dynamic and evolving. It’s not clear how the regime will respond to the will of the United States. (We’re optimistic though.)

Atlassian Corp - Quality on Sale

We added to our Atlassian position on Friday.

This was already a Lumida 2026 pick, and after the recent drawdown, the setup improved enough for us to size it up.

The stock, along with software as a category,  had been out of favor for months.

Valuations compressed to levels we rarely see for software embedded in daily operations.

See TEAM against its peers. At current prices, Atlassian trades around 29x forward EBIT, with a free cash flow yield near 4%. 

Atlassian is the leading software engineering platform.

The market leader continues to grow at high-teens to 20% year-over-year revenue growth while free cash flow scales and margins expand. 

Atlassian’s products include: Jira, Confluence, and Trello. These tools run planning, execution, and collaboration across engineering-driven organizations. 

Once workflows are built on Atlassian, switching costs are real. 

That’s why retention remains strong and why the company continues to deepen penetration inside large enterprises.

The concern weighing on the stock is that generative AI undermines Atlassian’s relevance or compresses its economics. 

We think that framing is backwards. 

AI doesn’t replace coordination—it amplifies it.

Atlassian is moving customers from on-premise software to the cloud while investing aggressively in AI capabilities like Rovo. 

That transition pressures reported margins and earnings in the short run. 

Management has also shown confidence on their AI implementation. They note "AI is one of the best things that’s ever happened to Atlassian."

Over 300,000 customers including Ford and Wells Fargo are using Atlassian’s AI offerings. Cannon-Brooks(CEO) stated, “We are proud of our ability to deliver AI to customers, with over 3.5 million monthly active users leveraging our AI capabilities.”

He also highlighted, “Customers using AI code generation tools are managing over 20% more projects than their peers.”

Better customer experience means higher retention, which keeps revenues secured for longer.

The risk is if cloud migrations slow, AI monetization takes longer, or competition from Microsoft and ServiceNow intensifies, then the re-rating stretches out. 

But after the drawdown, the stock is priced like a growth company that stopped growing—which it hasn’t.

Atlassian is durable, mission-critical software trading at discounted valuations, with a growing cash-flow base and structural tailwinds still intact..

Bullish On International Value

I did a FSD livestream on how international value offers a compelling opportunity. Watch it here. 

These markets have outperformed the SPY, and by a significant margin.

Look at the charts.

Green (EWY) and blue (GREK) are market returns for South korea and Greece.

Over the past 12 months, Greece is up nearly 80%, more than 4x the return of the S&P 500. 

France is up over 30%. Japan, Brazil, Mexico, Canada, and large parts of Europe are all ahead of U.S. equities. 

These markets are experiencing non-consensus rallies ever since the March and April Trade war.

Sure, Europe is behind on technology and has all sorts of demographic issues. 

Even markets with well-known structural issues—regulation, bureaucracy, energy constraints—are delivering stronger returns.

The reason is simple: valuations, earnings, and momentum are all aligned.

International stocks are not unlike small cap stocks or value stocks. They lagged for several years… now we see a non-consensus rally in all of these areas.

Skepticism is yielding to momentum, and that momentum should continue.

We especially like the banks. You can buy banks at 10x earnings or less, often with improving balance sheets and positive operating leverage. 

Santander being up over 160% in 1Y is a good example— it’s cheap, profitable, and benefiting from improving fundamentals in Spain.

“Home-market bias” - the tendency for investors to invest in their local economy - has returned in international markets.

French capital flowed into France. The Brazilian capital stayed in Brazil. Japanese investors leaned into domestic banks and industrials. 

That shift coincided with international outperformance—and it hasn’t reversed.

There’s also been a deeper market change. 

In prior cycles, risk-off meant bonds rallied. 

This time, gold replaced bonds as the safety trade. 

This removed one of the structural supports for U.S. assets and made relative equity valuations more important.

From a portfolio perspective, this raises a simple question: do you own enough international stocks?

If you’re not forced to be U.S.-centric, that flexibility is an advantage. 

Momentum trends tend to last 1.5 to 2.5 years. We believe this trend continues until you see global central banks hike rates or dollar strengthening.

Greece, once written off, is one of the best-performing equity markets globally.

We had taken all opportunities to scoop international names in the last couple of months. 

If you want an international value idea - Coupang is a good one. You can read our thesis here. 

The name is completing its re-test now and we believe is putting in higher-lows.

Lumida Curations

The Hidden Credit Crunch Behind Tariffs, AI, and Onshoring

Governments are piling up trillion-dollar promises on AI, factories, and industrial policy—but with central banks sidelined and public balance sheets maxed out, the trillion-dollar question is: who is going to lend the private credit to actually pay for it?

View Curation

CoreWeave's Secret Sauce

Intrator says CoreWeave has quietly built the world's largest GPU fleet and uses its proprietary software stack to transform off-the-shelf hardware into a high-margin, premium AI cloud service that hyperscalers can’t replicate.

View Curation

The Toxic Insider Wrecking Your Metabolism

Visceral fat acts like a rogue endocrine organ, pumping out inflammatory toxins that rapidly trigger insulin resistance, hormonal havoc, soaring triglycerides, plummeting HDL, and serious metabolic damage—far more aggressively than harmless subcutaneous fat.

View Curation

Meme

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: The Misconception Around Labor Market

• Markets: Lumida’s 2026 Stock Picks; Is Goldilocks Priced In?

• Digital Assets: Stablecoins Aren’t Eating Banks—Yet

• Lumida Curations: The Mag 7’s Grip Is Loosening; Markets Are Defying the Fed; Hydration Signals Aren’t One-Size-Fits-All

HAPPY NEW YEAR

These is perhaps the best New Year resolutions I came across this week. 

‘Where there is hatred, let me sow love,

Where there is injury, pardon,

Where there is doubt, faith,

Where there is despair, hope,

Where there is darkness, light,

Where there is sadness, joy.’

Wishing everyone peace, light, love and laughter in 2026.

(h/t St. Francis of Assisi)

Venezuela News

Here’s a 9-minute livestream titled “The Donroe Doctrine” explaining my reaction.

From a markets perspective, I believe Oil Services will do well. (Look at names in the OIH ETF). Overall, we don’t view this as creating risk for markets.

We’ll share ideas next weekend.

New Years Gift: Lumida 2026 Stock Picks

We’ve curated an exciting stock stuffer featuring 20 of our top picks.

The common thread across these names is that they’re non-consensus—exactly where alpha is found. You can download the picks here. 

Let’s walk through the ideas.

EQT - Our Energy Pick for the AI Power Decade

Natural gas is becoming a critical input for U.S. computing power. 

AI, data centers, and electrification are driving demand that looks increasingly structural, not cyclical. EQT sits at the center of that shift.

EQT is one of the largest natural gas producers in North America, with both upstream scale and midstream infrastructure. 

EQT has signed long-duration data-center contracts that extend roughly 20 years, offering exposure to AI-driven power demand at a far more reasonable valuation than the data-center or semiconductor complex. 

Management highlights regional demand already exceeds current supply.

Export optionality is the second leg. 

LNG agreements with Sempra, NextDecade’s Rio Grande project, and Commonwealth position EQT to benefit if U.S. gas exports tighten domestic markets over time. 

This adds flexibility without requiring aggressive volume growth.

The execution backdrop has improved materially. 

Revenues and net income are up roughly 30% year over year, per-unit costs hit record lows in Q3, and EQT generated more than $2.3 billion in free cash flow over the past four quarters, translating to a ~7% free-cash-flow yield. 

The risk is straightforward: EQT is still exposed to natural gas pricing. 

A prolonged oversupply would pressure margins and slow capital returns, particularly given the company’s debt load. 

But if AI power demand and LNG exports reshape gas markets as expected, EQT offers a clean, cash-generative way to play the theme.

PagSeguro (PAGS) — Deep Value in Brazilian Fintech

PagSeguro trades like a broken fintech, but the underlying business looks far more durable than the valuation implies.

PAGS has evolved from a payments processor into a full-stack fintech platform serving Brazilian SMBs and consumers through PagBank. 

Payments remain the entry point, but the real value sits in the attached banking services: deposits, secured credit, insurance, and investments. 

That ecosystem has allowed PagSeguro to grow earnings while many global fintech peers have struggled with funding costs and margin compression.

PAGS trades at roughly 6–7x earnings, under 1x sales, and generates exceptionally high free cash flow, with FCF yields north of 20%. 

Margins remain strong—EBITDA margins in the mid-40s—and capital return is real. 

Dividends and buybacks together imply a double-digit shareholder yield, which is rare in fintech.

Credit is the swing factor. Management has been conservative, focusing on secured lending and prioritizing profitability over aggressive market share gains. 

That has kept losses contained even as Brazilian rates stayed high. If rates normalize, operating leverage could surprise to the upside.

The risk is clear: Brazil remains competitive and macro-sensitive. 

Nubank and StoneCo set a high bar, and if transaction growth stalls or credit costs rise, PAGS could remain optically cheap for longer. But at today’s valuation, the market is already pricing in a lot of bad news.

Meta Platforms (META) — The Value Stock in Mag-7 Land

Meta is the rare Magnificent Seven stock that still trades like it hasn’t fully earned its place. 

We talked about META’s mispricing in November - the stock is up 11% since then, but the potential is still ahead.

The market is treating META as a “show-me” AI story, but the fundamentals already say it has.

On valuation alone, Meta stands out. 

At roughly ~20–22x forward earnings, META is the cheapest name in Mag-7 land by a wide margin. 

Meta generates some of the strongest margins in large-cap tech, with operating margins north of 40% and returns on capital that rival — and in some cases exceed — its peers.

In other words, investors are paying less for a business that already prints cash.

The growth engine is advertising, and AI is making it more powerful. 

Management disclosed that the annualized run-rate of its AI-powered advertising stack has surpassed $60 billion (30% of 2025 revenue).

Recommendation systems across Facebook, Instagram, and Threads are improving engagement, driving more time spent and better ad performance. 

That virtuous cycle shows up directly in revenue growth and pricing power.

Capex is the concern, but it’s also the opportunity. 

Yes, Meta is spending aggressively on AI infrastructure, and that has created anxiety. 

But we’ve seen this movie before — Google faced the same skepticism before AI-driven efficiency showed up in results. 

Meta’s cash flow profile gives it the flexibility to invest without sacrificing profitability, and early returns suggest those investments are already paying off.

The risk is execution. 

If AI monetization lags expectations or capex continues to rise without incremental returns, the valuation discount could persist. 

But in a Mag-7 cohort trading at premium multiples, Meta remains the underpriced growth story hiding in plain sight.

Marqeta (MQ) — Payments Infrastructure at an Inflection Point.

Marqeta sits behind many of the most disruptive payment use cases—BNPL, on-demand delivery, expense management, and digital wallets. 

Its API-first platform allows customers to issue cards, control spend, and move money in real time. 

That makes Marqeta a critical layer for companies like Block, DoorDash, and Affirm. 

The opportunity is not flashy growth—it’s becoming embedded.

The near-term story has been painful. 

Revenue concentration, especially with Block, contract repricing, and delayed profitability have weighed on sentiment. 

But beneath that noise, the business is stabilizing. 

Total processing volume continues to grow, gross margins remain high (~70%), and operating losses are narrowing materially. 

Management is guiding toward GAAP profitability in 2026, and the balance sheet provides runway—Marqeta holds significant net cash and continues to buy back stock.

Valuation reflects deep skepticism. 

The stock trades near trough multiples on sales and gross profit, despite still growing and operating in structurally expanding digital payments markets. 

If execution improves even modestly—through customer diversification, cost control, or normalized pricing—the operating leverage is meaningful.

The risk is execution and concentration. If Block volumes weaken or new customer ramps disappoint, the turnaround takes longer. 

But at this price, the market is already assuming Marqeta never quite gets there. That’s what makes it interesting.

Qualcomm (QCOM) — The Underappreciated AI at the Edge Play

Qualcomm is not trying to be Nvidia, and that’s exactly why it works. 

While the market has focused on AI in the data center, Qualcomm is positioning itself where inference actually scales: on-device, at the edge, and inside cars.

At the core, QCOM still benefits from its dominant position in mobile silicon and licensing, but the business mix is changing. 

Non-Apple revenue is growing faster, and automotive and IoT are becoming real contributors.

Qualcomm’s AI strategy is centered on low-power, high-performance inference rather than brute-force training. 

That matters as AI workloads move closer to users and devices.

Valuation reflects skepticism. 

Qualcomm trades around ~14–15x forward earnings—well below most large-cap semis—despite strong free cash flow, a solid dividend, and consistent buybacks. 

The market is still discounting handset cyclicality and Apple dependence, even as management has been explicit about reducing that exposure over time. 

Meanwhile, margins remain healthy and returns on capital are strong, suggesting the core franchise is intact.

AI upside is optionality, not the base case. 

Partnerships in automotive AI, edge accelerators, and sovereign compute (including early-stage Middle East initiatives) offer long-term growth without needing immediate perfection. 

The risk is timing: handset demand could remain sluggish and AI revenues will ramp gradually, not overnight. 

But at this multiple, investors are being paid to wait for a business that is steadily becoming less cyclical and more strategic.

Manulife Financial (MFC) — A Cash Compounder with Asia Optionality

Manulife is not a growth stock in the Silicon Valley sense. It’s a balance-sheet business, and right now the balance sheet is doing most of the work.

MFC combines life insurance, wealth management, and asset management, with a footprint that matters most in Asia. That exposure is the differentiator. 

Sales momentum in Asia—particularly Hong Kong and broader APAC—has been strong, driven by rising savings rates, retirement demand, and underpenetrated insurance markets. 

This is long-duration, high-margin growth that does not depend on aggressive underwriting or leverage.

At the same time, Manulife is throwing off cash. Free cash flow yields are high, capital ratios are conservative, and shareholder returns are consistent. 

The stock offers a solid dividend, steady buybacks, and trades at a valuation that implies little growth—despite evidence that growth is happening where it matters most. 

Management has also been quietly repositioning toward higher-return businesses, including private credit and wealth solutions, which should lift ROE over time.

The risk is not existential, it’s cyclical. 

Earnings can wobble with interest rates, mortality assumptions, or regulatory shifts—particularly in Asia. 

Valuation is no longer distressed, and upside likely comes from compounding rather than multiple expansion.

MFC works in a portfolio because it pays you to wait. 

Comcast (CMCSA) — The Ignored Value Trade

It’s hard to justify owning sovereign bonds in a world where inflation stays higher for longer and real yields remain uncertain. 

Public markets offer alternatives that pay you to wait—Comcast is one of them.

We had talked about this before, and it paid back our trust - up 10% since the call.

The stock still has potential as markets recognize the ignored value in it.

CMCSA had been left for dead as cord-cutting dominates the narrative. Cable subscriber losses are real, but the stock price assumes far more damage than the business is actually experiencing. 

At roughly 7x forward earnings and near a 20% free cash flow yield, the market is effectively pricing in a permanent decline. 

Comcast has grown revenue and free cash flow in recent years, and expectations are for continued stability. 

Its businesses generate recurring cash flow, and management has consistently returned capital through buybacks, steadily shrinking the share count. 

This is disciplined capital allocation.

The overlooked upside sits in Content & Experiences—NBCUniversal, Peacock, Universal Pictures, DreamWorks, Sky, and theme parks—roughly 38% of revenue. 

Live sports anchor engagement, Peacock has scaled to 41 million paid subs with strong ad growth, and long-term NBA rights lock in premium content. 

The risk is broadband disruption and content missteps. 

But Comcast pairs connectivity with content optionality, trades cheaper than peers, and throws off cash.

Why own sovereign debt when you can own Comcast?

Atlassian (TEAM) — Mission-Critical Software, Temporarily Out of Favor

Atlassian has been our pick in software for some time. We flagged it in November, and the stock has risen 8% since then.

Jira, Confluence, and Trello sit at the center of how modern teams build software, manage projects, and collaborate. 

Once embedded, these tools are sticky—net retention remains strong, and Atlassian continues to deepen its footprint inside large enterprises.

The near-term problem is not demand, it’s digestion. 

TEAM is in the middle of a complex transition: moving customers to the cloud, investing heavily in AI features like Rovo, and absorbing higher R&D spend. 

That has pressured margins and kept reported profitability weak, even as revenue continues to grow at a healthy clip and gross margins remain elite. 

The market has focused on near-term earnings pain and valuation optics, not long-term operating leverage.

What gets overlooked is the durability of the model. 

Atlassian generates recurring subscription revenue, serves over 300,000 customers (including most of the Fortune 500), and benefits from secular trends toward distributed work, automation, and AI-assisted productivity. 

As cloud migration matures and R&D intensity normalizes, margins should expand meaningfully.

The risk is execution. 

If cloud migrations stall, AI monetization disappoints, or competition from Microsoft and ServiceNow intensifies, the re-rating takes longer. 

But after a sharp drawdown, TEAM is priced like a growth company that stopped growing—which it hasn’t.

Mercury General (MCY) — A Recovery Trade with Real Risk

Mercury General is a classic turnaround story, operating where few insurers want to be: California. 

Years of wildfire losses, frozen rate approvals, and underwriting pressure pushed MCY to the brink. The market still prices it that way. But the operating reality is changing.

Regulatory conditions have finally shifted. California has begun approving meaningful rate increases, allowing insurers to reprice risk closer to economic reality. 

Mercury has responded by tightening underwriting, raising premiums, and shrinking exposure where returns don’t make sense. The results are showing up. 

Underwriting profitability has improved materially, combined ratios have come down, and recent quarters reflect a business moving out of survival mode and into recovery.

At the same time, MCY trades like a distressed insurer despite normalized earnings power. 

The stock sits near low-teens earnings multiples with a strong free cash flow yield, assuming catastrophe losses remain within modeled ranges.

Investment income is also rising as higher rates flow through the portfolio, adding another earnings lever.

The bear case is obvious and real. 

Mercury remains highly concentrated in California, and another severe wildfire season could quickly erase progress. 

Capital flexibility is limited, and volatility will never disappear in this business.

But that risk is well understood—and heavily discounted. 

For investors willing to accept catastrophe risk, the payoff is asymmetric.

Progressive (PGR) — The Best Operator in a Tough Business

Auto insurance is cyclical, regulated, and capital-intensive.

Progressive has found the way to win - consistently.

The company’s edge is data. 

Progressive’s telematics, pricing models, and underwriting discipline allow it to price risk more accurately than peers, gain share when others pull back, and protect margins when loss costs rise. 

That advantage showed up again recently: premium growth remains strong, combined ratios are well below industry averages, and returns on equity remain exceptional. 

The recent pullback has more to do with normalization fears than fundamentals. 

After a period of unusually strong underwriting results, investors worry margins may compress as competition intensifies and advertising spend rises. 

That’s a fair concern—but it ignores Progressive’s history. 

When the cycle turns, Progressive tends to take share, not give it back.

Valuation now looks reasonable for a company with durable underwriting economics, strong free cash flow, and disciplined capital management. 

At low-teens earnings multiples, the market is assuming peak profitability is behind us.

The risk is that loss trends reaccelerate, regulators push back on pricing, or competitors close the technology gap. 

But even in tougher environments, Progressive has proven it can adapt faster than the industry.

Halozyme (HALO) — Cash-Flow Biotech with a Clock

Halozyme is not a typical biotech. It doesn’t live or die by binary trial outcomes. 

Its ENHANZE drug-delivery platform is embedded inside some of the world’s most successful biologics, quietly collecting royalties as partners scale blockbuster drugs like Darzalex and Phesgo. 

That model has turned Halozyme into a cash-flow machine, with exceptional margins, strong free cash flow, and a capital-light royalty stream that most biotechs can only dream of.

The numbers reflect that quality. 

Revenue growth remains strong, profitability is among the best in biotech, and capital returns are real. 

At a single-digit forward earnings multiple, the market is valuing Halozyme closer to a mature pharma than a high-margin platform business with multiple growth levers.

The hesitation is about time, not execution. 

ENHANZE’s core U.S. patent expires in 2027, and investors worry about what Halozyme looks like in the 2030s. 

Management is responding by extending the platform internationally, layering in new delivery technologies through acquisitions like Elektrofi, and reinvesting aggressively while today’s royalty cash flows are abundant.

Still, the risk is real. 

If next-generation delivery platforms fail to gain traction or partners renegotiate economics post-patent, growth could slow materially. Halozyme must prove it can reinvent before the clock runs out.

Today, though, the market is pricing the cliff, not the cash. 

Bread Financial (BFH): Mispriced Credit Optionality

Bread Financial is strategically positioned across the high-optionality sectors of consumer credit, Buy Now Pay Later (BNPL), and private-label cards.

It trades at ~8–9x earnings and roughly book value, despite generating strong cash returns. 

Recent quarters showed resilient profitability, with solid net income, rising tangible book value, and improved asset yields. 

Importantly, Bread benefits from deposit-based funding, which has helped offset higher interest expense and stabilize net interest margins in a volatile rate environment.

The company’s private-label and co-branded partnerships remain sticky, embedded directly at the point of sale for large retailers. 

Credit sales have continued to grow modestly, and are expected to pick up further as consumers stay resilient. 

Management has also been disciplined on capital allocation, returning cash via buybacks while maintaining ample liquidity.

The risk is consumer credit delinquencies increase.

Credit losses remain elevated, delinquencies could rise if the labor market softens, and regulatory scrutiny around fees and consumer lending is intensifying. 

Bread also runs with higher leverage than traditional banks, which amplifies downside in a sharper downturn.

Still, at today’s valuation, the market appears to be pricing in a severe and prolonged credit event. 

If losses stabilize and earnings normalize even modestly, BFH offers asymmetric upside with meaningful capital return along the way.

Norwegian Cruise Line (NCLH): Boomer Demand Meets Operating Leverage

If you’ve been following our newsletter, you’d know NCLH, and love it like we do.

Norwegian Cruise Line is a clean expression of a powerful, underappreciated demographic trend: aging but affluent consumers prioritizing experiences over goods. 

Baby boomers remain the core cruise customer, and this cohort has both the time and balance sheet to keep traveling. 

That demand has proven resilient even as inflation and higher rates pressured other discretionary categories.

Operationally, NCLH is past the hardest part of the post-pandemic recovery. 

Pricing has held up, load factors continue to improve, and forward bookings are running ahead of prior years with record advance ticket sales.

Management expects load factors to exceed 105% in 2026, reflecting strong mix, onboard spend, and disciplined revenue management. 

Capacity growth is modest and planned, with new ships supporting roughly mid-single-digit capacity CAGR through 2028 rather than flooding the market.

The leverage to earnings normalization is significant. 

NCLH trades around ~8–9x forward earnings, a discount to history and peers, despite improving margins and steady revenue growth. 

As fuel costs stabilize and operating efficiencies scale, incremental revenue should fall meaningfully to the bottom line.

The risk is leverage. Debt remains elevated, limiting flexibility if demand softens or refinancing costs rise. 

Competitive intensity from Carnival and Royal Caribbean, along with policy risks such as new cruise taxes, could also pressure margins.

Coupang (CPNG): Logistics Moat, Not a Data Breach Story

Coupang featured in our last two newsletters, and it has now made it to a consecutive third - Elite levels.

The customer data breach did not change its fundamentals - the over reaction made the business a bargain. 

Last week’s news also offered clarity around the possible repercussions of the breach, and they were better than most feared.

Coupang’s real advantage is not e-commerce. It is the logistics density. 

The company has already built the hardest asset: a nationwide, vertically integrated fulfillment network that enables same-day and next-day delivery at scale. 

That moat is extremely difficult to replicate and is now being leveraged across higher-margin adjacencies including Eats, advertising, fintech, and media.

The financials reflect this transition from growth to operating leverage. Revenue has grown at high double-digit rates over the last three quarters. 

Gross margins are near 30%, at the top of their historical range. Most importantly, gross profit growth (~30%) is now outpacing revenue growth (~18%), signaling improving unit economics. 

The key risk remains valuation: a high trailing P/E implies execution must remain strong. But structurally, Coupang’s logistics moat and improving economics remain intact.

CoreWeave (CRWV): Downside Priced, Upside Misunderstood

CoreWeave is yet another regular feature for Lumida ledgers - it had been volatile, gaining around 33% since our Nov 15 call, and then erasing all of it. Now, it sits around 10% higher, with Friday’s jump.

Market’s focus remains on Coreweave’s leverage, customer concentration, and execution risk, pushing valuation down sharply. 

As a result, much of the downside scenario is already reflected in the price.

What’s overlooked is the structure of the business. 

CoreWeave’s debt is largely collateralized by Nvidia GPUs—some of the most scarce and valuable assets in the AI ecosystem. 

These are not stranded assets; they are in constant demand, monetizable, and reusable across training and inference workloads. 

This materially limits true balance sheet risk compared to traditional leveraged growth companies.

Demand visibility is another underappreciated factor. CoreWeave carries roughly $55 billion in contracted backlog, reflecting persistent undersupply of AI compute. 

Enterprises and hyperscalers continue to seek dense, purpose-built GPU capacity, and CoreWeave has positioned itself as a critical supplier. 

That backlog provides unusually strong revenue visibility for a company still viewed as speculative.

CoreWeave has the Nvidia put - Nvidia has committed to utlizing any unallocated CoreWeave datacenter capacity. Nvidia is a quality counter-party. OpenAI is not.

The bear case is real: high interest expense reduces flexibility, customer concentration remains elevated, and delays in capacity buildouts could pressure near-term results. 

A slowdown in AI spending would amplify volatility.

But those risks are well known—and priced in. We believe the relevant metric to look at is P/S ratio - and CoreWeave is cheap relative to peer group.

What’s not is the upside if CoreWeave executes on its backlog and benefits from operating leverage as utilization scales. This is a high-risk stock, but increasingly an asymmetric one.

Kingstone Companies (KINS): Scarcity Creates the Opportunity

Kingstone is a small insurer operating in one of the most dislocated property markets in the U.S.: New York homeowners. 

As larger carriers retreat from the state due to regulatory friction and catastrophe exposure, Kingstone is doing the opposite—leaning into disciplined growth with tighter underwriting and selective risk taking. 

That supply vacuum matters. 

Insurance is a business where exits create pricing power, and Kingstone is a beneficiary.

The turnaround is already visible. 

The company has posted record profitability, driven by improved risk selection and the rollout of its “Select” homeowners product, which materially reduced claim frequency versus legacy books. 

With competitors pulling back, displaced policyholders are flowing to the remaining underwriters, allowing Kingstone to grow premiums without sacrificing underwriting discipline. 

Returns on equity and margins now screen at the high end of the industry despite the company’s micro-cap status.

Valuation remains reasonable at roughly 7x earnings, reflecting lingering skepticism from past missteps and the market’s general aversion to small insurers. 

That skepticism is not unfounded. Kingstone remains concentrated in New York, exposed to regulatory decisions and weather-related severity. 

A spike in catastrophe losses or missteps in geographic expansion could quickly pressure results.

If underwriting discipline holds, Kingstone’s recovery may be more durable than the market expects.

Molina Healthcare (MOH): Priced for Disaster, Positioned for Recovery

Molina has spent the past year being treated as the weak link in managed care. 

The stock was down sharply, margins were hit by a spike in medical loss ratios, and management cut full-year EPS guidance multiple times. 

We flagged the hidden value in MOH on 30th Nov, and the stock took off post call, gaining ~27%.

Molina is a pure-play Medicaid operator, serving roughly 5.7 million members across 19 states.

MOH is now entering a period where Medicaid rate increases are catching up to elevated medical costs, setting the stage for margin recovery in 2026.

Valuation reflects deep pessimism. 

The stock trades near ~8x FY2 forward earnings, among the lowest levels in its recent history, despite delivering ~20% ROE and aggressively returning capital. 

Shareholder yield is near 20%, implying the market expects buybacks to be cut materially. That’s a hard assumption to justify if earnings stabilize.

The bear case is clear: medical costs could remain elevated, Marketplace losses may persist, and further regulatory changes could delay recovery. Management execution has also been uneven.

But after a 50% drawdown and multiple guide-downs, much of that risk is already priced in. If margins normalize even modestly, Molina doesn’t need heroics—just mean reversion.

Labcorp (LH) — A Quality Compounder at an Inflection Point

Labcorp is a classic quality compounder and indexed to the Aging Demographics trend. Older boomers means more testing.

As one of the largest global diagnostics and life sciences platforms, LH benefits from long-term tailwinds—aging demographics, rising chronic disease, and growing demand for specialty and oncology testing. 

After several years of portfolio reshaping and post-COVID normalization, the business is quietly stabilizing.

The market’s skepticism is largely centered on margin pressure and regulatory overhangs, particularly around PAMA reimbursement cuts and cost inflation. 

Those risks are real, but increasingly visible and, in our view, largely reflected in the valuation. 

At ~15x forward earnings and a mid-single-digit free cash flow yield, LH trades closer to a utility multiple than a scaled healthcare services leader with strong barriers to entry.

Management’s focus on operational discipline is beginning to show. 

The Launchpad initiative targets $100–125 million in annual cost savings, while recent acquisitions in genetics and specialty diagnostics should support higher-quality revenue mix over time. 

Importantly, Labcorp generates consistent free cash flow and maintains an attractive shareholder yield, providing downside support.

The bear case hinges on slower margin recovery if reimbursement pressure persists or integration benefits disappoint. 

However, with leverage manageable, demand stable, and efficiency initiatives underway, LH offers a compelling risk-reward skew:  limited downside if conditions muddle through, and meaningful upside if margins normalize.

Shift4 Payments (FOUR) — the Payments Backbone of Hotels and Stadiums

Shift4 represents a scaled, profitable payments platform that is quietly consolidating share across some of the most complex and sticky verticals in commerce. 

Its core advantage lies in owning the full payments stack across hospitality, restaurants, gaming, and sports—markets where reliability, speed, and integration matter more than price. 

This has allowed Shift4 to build deep merchant relationships, high switching costs, and industry-leading margins.

The company is translating that positioning into results. Gross profit growth remains strong, free cash flow is robust, and EBITDA margins sit well above most fintech peers. 

The Global Blue acquisition meaningfully expands Shift4’s reach into luxury and international retail, strengthening its Unified Commerce strategy and broadening its addressable market. 

Over time, this positions Shift4 as a multi-vertical payments consolidator rather than a niche processor.

Valuation also tells part of the story. 

After a sharp sell-off, the stock trades at a far more reasonable forward multiple despite structurally higher margins and cash generation than many peers. 

The market appears to be pricing in a prolonged slowdown, even as management continues to execute and reinvest at attractive returns.

That said, the risks are real. Leverage increased post-acquisition, same-store sales have shown volatility, and competition remains intense. 

Integration execution will matter.

But if management delivers on deleveraging and sustains mid-teens growth, today’s price reflects far more downside fear than upside potential—setting up an asymmetric long-term opportunity.

Enova International (ENVA) — A Cash-Generating Fintech Levered to Underserved Credit

Enova was a 2025 stock pick, and we’re brining it back for 2025.

Enova is a non-prime lender that recently secured a bank charter. The bank charter will enable Enova to reduce and diversify its funding costs boosting earnings. 

Its core advantage is not balance sheet size, but underwriting precision. 

Using machine-learning models built over decades of credit data, Enova prices risk dynamically across consumer and small-business loans, allowing it to grow profitably in niche segments.

The results show up clearly in the numbers. 

Revenues and originations continue to grow at double-digit rates, driven by strong demand from small businesses and resilient consumer credit performance. 

Margins remain structurally high, with gross margins above 80% and ROE north of 20%, translating into substantial free cash flow. 

At roughly 11–12x forward earnings, the stock trades at a discount to its own growth profile, despite buybacks and shareholder yields that signal confidence in the durability of earnings.

The market’s concern is understandable. 

Enova operates with leverage, and any sharp economic slowdown or regulatory shift around high-APR lending could pressure credit quality or constrain growth. 

Charge-offs have ticked up modestly, and reliance on debt funding amplifies cyclical risk.

Still, much of that caution appears priced in. 

Enova has navigated multiple credit cycles, adjusted underwriting in real time, and emerged more profitable. 

If credit conditions remain merely “normal,” ENVA offers a rare combination of growth, cash generation, and valuation support.

The Rise Of Defense Tech

The USA launched an attack on Venezuela this week, taking Maduro in custody.

China is on the brink of a war against Taiwan.

Russia is already at war with Ukraine. Iran and Israel has been the headline for over an year now.

The Geopolitical situation is getting tense, and countries know this.

Global defense spending is accelerating.

World military expenditures rose 9% last year - the sharpest increase in more than three decades. 

Defense allocations are headed toward military autonomy and AI. Countries are aiming to have intelligent mechanisms that work on their own. 

Anthony Antognoli, USCG’s executive officer: “Our vision is to have robotics and autonomous systems as the foundation for the way the Coast Guard operates its missions.”

“It is impossible to do that work [defense missions] with humans and patrol cutters alone.”

These spending increases are creating massive demand for defense tech names.

Venture capital has realized this.

In 2025, defense tech venture investment was already more than double 2024’s total. We don’t see it stopping anytime soon.

The funding is concentrated in large, late-stage rounds of autonomy focused defense companies.

Companies like Anduril and its peer group are benefitting from these financing rounds.

Those financings are only available to sophisticated private markets investors.

Lumida Ventures is also focused on investing in this category. We have had a string of successful investments across CoreWeave, Kraken, and Brad Jacobs QXO, and Canva. All of these have had markups or realizations.

Do reach out now if you want to be in the flow of our next deal.

If you are an accredited investor or qualified purchaser, sign up here to receive communications about our private deals.

Macro

The Misconception Around Labor Market

The biggest surprise this year might be a rebound in payroll employment growth. 

You can see why in the claims data:

Initial jobless claims rolled over into year-end, and continuing claims have dropped at the same time. 

That combination matters—it suggests the duration of unemployment is shrinking. 

In other words, the labor market is healing at the margin even if the headlines still feel cautious.

It isn’t really a “no-hire, no-fire” framing. 

Hiring activity is not dead.

In October, the pace of hiring looked relatively normal—about 5.1 million workers.

The issue is that hiring is basically matching separations tied to quits and layoffs. 

In prior expansions, hiring typically ran ahead of separations.

Today, it’s closer to a “no-net-hiring” labor market: churn is happening, but the net adds are constrained. 

AI and automation are the differentiator in this regime- this reflects in higher productivity and improved margins. 

Markets

Is Goldilocks Priced In? Softbank financing OpenAI

I did a 20-minute explainer on how markets are already pricing in Goldilocks. Watch it here.

You can see it in positioning, sentiment, and what’s working in the tape.

The most bullish news this week was Softbank’s $40 Bn commitment to OpenAI.

Markets are trading around the sustainability of capex financing.

I was quite bearish on Tuesday evening…this news is positive and incremental and caused me to become constructive. So long as financing for capex is there, semiconductor markets will hang in there.

Positioning is the first tell.

Investor cash levels are at 3.3%. That’s very low. It means investors are already all in. 

If you’re looking for incremental demand to keep pushing the market higher, the question becomes simple: who’s not overweight?

Now layer expectations on top. 

We just finished ~4% GDP growth. It’s a Goldilocks economy. 

Macro conditions are sound. There’s no recession. But the economy is not the same thing as the stock market. 

The stock market is about expectations, and expectations right now are high. 

When expectations are low (think April trade-war stress), it’s easy to exceed expectations. That’s a great time to buy. 

When expectations are elevated, you look at history and you usually get weak performance.

Notice expectations were low in 2023... that was a 26% return year for the S&P.

The “smart money” is back in, too. 

Coming off the April lows through most of September, hedge funds, institutions, and retail were basically in a turtle shell—post-traumatic stress disorder, too much cash, not investing—while the market ripped. 

That rally was fueled by disbelief, even though the macro backdrop was Goldilocks and we just wrapped double-digit earnings growth (and the quarter before that, too).

Markets have figured it out now. People are back in. 

Fundamental long/short managers are at the ~80th percentile of net exposure over the past three years. In April they were closer to the 15th–30th percentile. They went from turtle shell to fully back in—fast.

Sentiment is the second tell. 

The Bank of America bull & bear indicator is at extreme levels—around 8.5. Everyone’s in the swimming pool. This is a contrarian indicator and it works when it’s at extremes. 

December 2021 is a good reference point: elevated sentiment, stimulus checks, record earnings growth—and it was a terrible time to buy stocks. 

Markets are forward-looking. Where we just left doesn’t matter. The setup ahead is what matters.

And it’s not hard to see why the “Fed rescue” narrative is shaky. 

We just had 4% GDP growth. You don’t cut rates in 4% GDP growth. 

Inflation is also still sticky enough that it’s hard to see how the Fed comes riding in to save risk assets if we get wobble. 

High beta is where you see the stress first.

The high beta vs low beta ratio is breaking trend—breaking down below the moving averages, with technical signals rolling over. 

High beta had an extraordinary rally off the April lows—massive short squeeze. 

But now you’re seeing the leaders lag. 

Robinhood is down ~26% while the S&P is just a few points from all-time highs. That’s an example of high beta fragility. 

Palantir is down less (~14%) but it’s flirting with a head-and-shoulders type setup—lower high, not clean. 

Tesla tried the cup-and-handle breakout and failed, and failure of a bullish pattern is quite bearish.

So what works in a mean reversion tape? 

Value over growth. 

Growth vs value is lower highs, lower lows—downtrend. Value is beating growth. If you’re going to be invested, you should probably be in value over growth. 

Value is near highs and still respecting its moving averages; growth hasn’t made new highs since October, which is classic weakness. The summary term is quality value.

And you can see it in individual names: Molina is straight up since late November—value + healthcare. 

ICE has been outperforming since early November.

We flagged both these names in our earlier newsletters right when they were beaten down.

Tactically, where we stand today, we believe high beta can do a counter-trend rally.

What makes markets tricky is the medium term and short-term views can conflict. This is one of those times.

It’s important here not to buy the dip in stocks that had incredible runs in 2025 like Robinhood and Palantir - or at least rotate after a bounce. Remember Cava or Chipotle? These also had incredible runs. After their euphoric peaks, they lag. And I expect we’ll see that here also.

Semis are the third tell—and it’s all tied to OpenAI. 

Semiconductors pulled back on news reports about concerns around OpenAI’s ability to fund its trillion dollars of obligations. 

Then OpenAI leaked to the Wall Street Journal they’re planning to raise $100 billion, and semis rallied. The market is basically running a “we’re so back / it’s over” cycle around OpenAI.

Why it matters: $100 billion from OpenAI gets spent substantially on Nvidia, which flows to TSMC, Applied Materials, KLA, Lam Research—the whole supply chain. 

$100 billion in revenue is substantial. Nvidia’s margin structure makes the market-cap math enormous. 

But markets are forward-looking—and they’ve already discounted a substantial amount of that future spend in semis and Nvidia. 

So now the debate is how much of that spend is real. That’s why you’re getting volatility.

The three things to watch in the first half of next year:

1. OpenAI financing—how does it go? It probably gets done, but does real smart money participate or not? If they don’t, it’s a very bad sign.

2. Midterm elections—generally you get a pullback in the first half. This year probably isn’t different. Healthcare outperformance could be discounting a change in Congress.

3. Growth-to-value and high beta—this rotation is already showing up, and it’s the cleanest expression of mean reversion.

One more thing: IPO supply.

There’s a lot of hope in markets—SpaceX and OpenAI at trillion-dollar valuations. That’s a lot of supply for markets to digest. 

To buy a share, you’ve got to sell another share. We’ve seen the IPO boom and IPO hangover cycle before. 

If the IPO ETF keeps bleeding out, it’s going to be hard to get an IPO market going.

That’s the setup.  

The Power of Compounding

I have a client - a divorced mother of two teenagers. 

She signed up in Spring 2024 with $3 MM in net worth. 

Her income requirements were $250K after tax. 

That means $300 K pre-tax. She also needs to combat inflation. 

The probability of her achieving this goal objectively according to various planning tools was challenging. 

She had a moderate risk tolerance. 

So, we put her in two equity strategies (one active), and two higher yielding strategies that generate predictable cashflow so she can sleep at night and fund living and travel expenses. 

We sprinkled some pre-IPO deals. 

Two years later, her net worth is $5 MM+. 

Now she could de-risk. 

What we are doing now is ‘defeasing’ the income requirement with part of her portfolio. 

We put the excess savings above her goal level into growth strategies. 

Lesson 1: Compounding works. 

Lesson 2: An ideal asset allocation must match the risk tolerance to ensure emotional adherence 

Lesson 3: Bull markets are amazing for wealth creation. 

There are many, many millionaire next door types that stuck to the process. 

Don’t fall for this Guaranteed Mediocrity nonsense. 

Lastly, 

Lesson 4: your best odds of making serious money is at the end of a bear market when psychology is frigid and you don’t want to buy.  

YOLOing when sentiment is flush is like buying high and selling lower. 

Out-performance comes from being Non-Consensus and Correct. 

Get to emerging trends early, and sell crowded and expensive themes.

Digital Assets

Stablecoins Aren’t Eating Banks—Yet

Despite persistent narratives around stablecoins “disintermediating” the banking system, the data tells a very different story. 

Bank deposits—across both large and small institutions—continue to grind higher. There is no evidence of meaningful capital flight from traditional deposits into stablecoins at the system level. 

In fact, deposits have steadily increased even as stablecoin market caps have grown, suggesting coexistence rather than substitution.

This matters because much of the bearish thesis around banks and bullish extrapolation for stablecoins rests on the assumption that digital dollars are hollowing out bank balance sheets. 

That simply isn’t happening today. 

Stablecoins currently function primarily as transactional instruments within crypto markets—used for trading, settlement, and on-chain liquidity—not as a wholesale replacement for savings vehicles, payroll accounts, or corporate cash management.

For now, stablecoins complement the financial system rather than disrupt it. 

They improve speed and programmability at the edges but do not yet replicate the full utility stack of insured deposits, credit creation, and regulatory trust that banks provide. 

The takeaway: digital assets may be reshaping how money moves, but not where it ultimately sits—at least not yet.

Lumida Curations

The Mag 7’s Grip Is Loosening

Markets are beginning to rotate beneath the surface, as leadership gradually broadens beyond the Mag 7 toward companies quietly benefiting from AI-driven productivity and margin expansion rather than headline growth.

View Curation

Markets Are Defying the Fed

Despite rate cuts, funding conditions remain tight as market-set rates stay elevated, signaling cash scarcity and forcing the Fed toward balance-sheet tools rather than conventional policy easing.

View Curation

Hydration Signals Aren’t One-Size-Fits-All

Common dehydration checks like skin turgor or nail color can offer quick clues, but age and body composition meaningfully affect how reliable those signs actually are.

View Curation

Meme

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: Economy Is Strong Heading Into 2026; Productivity Is The Missing Variable

• Markets: Why did Nvidia buy Groq?; Bullish On Markets; The Bubble In OpenAI’s 70% Gross Margins; Coupang Call Paid Dividends

• Digital Assets: Banks As The Next Mag 7? Where Are We In The Digital Asset Cycle?

• Lumida Curations: Structure Matters More Than Speed in AI Investing; Why the Fed Can’t Abandon the 2% Inflation Target?; Protein Isn’t the Problem

Lumida’s Twelve Days of Christmas :)

Look, we aren’t the best singers but we know how to make the tough calls!

Click the link above if you want to enjoy a chuckle.

Thank you for being a part of our journey.

Watch the video here.

Home Alone is a Christmas Classic.

But, here’s a thought. 

How much would the Home Alone pizza order cost today?

In the original Home Alone (1990), the McCallisters famously ordered 10 pizzas for a grand total of about $122.50. 

Fast forward to today…

If you placed the same order, you’d likely be looking at around $200 or more, depending on your pizzeria of choice.

In other words, over the roughly 35 years since Home Alone was released, the implied average annual inflation rate on that pizza order is about 1.6% to 1.7% per year. 

All things considered, that inflation rate is not bad as compared to healthcare, education, insurance and housing costs which have ballooned since then. 

Why?

1) Pizza prices are set in a hyper competitive unregulated market. 

2) The Federal government did not try to make Pizza a universal human right that is ‘inclusive and accessible’. 

3) Baumol’s cost disease - super under-rated theory

Overall view: Home Alone doesn’t really seem realistic in 2025. 

Why did Nvidia buy Groq?

I remember first interviewing the CEO, Jonathan Ross, on my podcast Lumida Non-Consensus Investing a few years ago.

I had seen a Youtube video of him demonstrating the LPU and the speed of inference at a tradeshow no one ever heard of.

Then when he joined the pod he explained ‘Groq is not Grok as in Twitter Grok, but Groq’.

I was listening to that show again this weekend. You might want to give it a listen. There’s a lot we were very early on.

The main point of the podcast is ‘learning in public’. Interviewing people who have unique vantage points into different sectors.

Dylan Patel, founder of SemiAnalysis, was another early podcast guess before his Substack blew up.

Kyle Samani was another one when Solana was exiting its bear market.

Subscribe to the podcast here on Youtube, Spotify or Apple.

Act on Early Trends Early

The main takeaway from all of this - act on trends early. Usually there is signal from a faint observation. By the time that faint insight becomes received fact, prices have gone up quite a bit.

One of my main lessons this past year is to act earlier and more decisively around those.

Trend shifts start as a faint glimmer - often more like a question than an answer like “Haven’t restaurants bottomed here? Isn’t Bitcoin going to benefit from the Federal Reserve Bank Term Loan facility? Isn’t TSM too cheap at 15x?”.

Recent examples of applying this include dialing up or exposure to Molina Health, Comcast, Chili’s (EAT), and others.

2026 Stock Picks

Incidentally, we did a bottoms up review of finding those mispriced assets in the ecosystem. There are easily dozens and dozens of mispriced securities out there — often in sectors such as software, recent IPOs, healthcare, consumer finance and so on. (Maybe even hombuilders or building supply materiasls given Trump’s shift to housing affordabiltiy, but I’m not sure there.)

This gives me confidence that the bull market has plenty of room to run. We’re going to see rotations and mean reversion from sold off sectors that have strong earnings growth.

Incidentally, I was looking at our 2024 stock picks. They are up 39% on an equal weight basis. There’s a lesson in there.

Avis was the worst performer in that list last year, now it’s up 60% YTD.

The main lessons for me is:

We tend to get in trends early. Stay in the trends longer. It takes about 1.5 to 2 years for a trend to play out.

Speaking of, Restaurants have been in a bear market for about 1.5 years. It’s time to start looking for a bull market there and we have increased our exposure to names like Shake Shak (SHAK) and Chili’s (EAT).

Nvidia spent $20Bn on Groq’s acquisition - this is Nvidia’s largest purchase ever. 

I must say - all of the takes on X - including Gavin Baker’s seem wrong to me.

Note: a 1,000 word explanation for a large M&A transaction might be a hint that people are over-thinking things.

Here’s our thesis…

Google one month ago hit all-time highs and became a darling, in part on the success of the TPU. Analyst see Google as the first credible threat to Nvidia’s dominance.

(This is a complex topic in its own right which we won’t get into here, overall, we view Nvidia as maintaining strong position.)

The team that built Google’s original TPU founded Groq. It’s really that simple.

If you want a shot at Nvidia’s 75% margins, you buy Groq. The acquisition was Defensive.

Yes, there is an inference angle to it. Groq specializes in high-speed inference. And, Nvidia’s rival AMD is differentiating our inference.

But, M&A deals, often hinge on one primary consideration. Not a laundry list of pros and cons.

Keeping Groq’s TPU know how out of the hands of a competitor is the imperative. This also is consistent with Nvidia capturing the team and licensing the tech rather than doing a whole business acquisition.

Google’s fanfare 30 days ago is exactly what led to Groq’s acquisition today. You can imaging a conversation with Jonathan Ross and Jensen:

“We have some interest around the TPU IP we’ve built here given Google’s recent success. We are fielding interest from a number of parties. We think Nvidia is the leader. At the same time we need to do right by our shareholders. Does it make sense to have a conversation?”

Left independent, Groq represented a latent risk. If a rival bought it, the knowledge could have seeded a serious challenge to Nvidia’s moat. 

There’s also a strategic alignment angle that matters.

Nvidia is no longer selling chips. It’s selling entire data center solutions—full racks, integrated systems, turnkey AI infrastructure. Blackwell is the clearest signal for this. 

Groq was operating with the same philosophy: inference not as a component, but as a system-level product. Racks, software, deployment—end to end.

Groq strengthens Nvidia’s ability to think and sell at the systems level, while preventing an alternative architecture from scaling outside Nvidia’s control.

This acquisition solidifies Nvidia’s leadership position, and sets them up for the future.

We think AMD is going to struggle in the after effects of this purchase. AMD is also over-valued relative to Nvidia and not growing as fast.

AMD has a trailing PE of 112X and a forward PE of 38X.

Nvidia, by contrast, is priced at 47x PE and a forward PE of 27X.

AMD has lagged as of late, while Nvidia isn’t too far off from All Time Highs.

I can’t understand owning AMD here.

The Rise Of Defense Tech

The last twenty years were about software. 

The next 5 years are about hardware. 

Many of the most valuable companies in the world are hardware: Nvidia, TSM, Tesla, etc.

Where are we focused for our next theme?

Defense Tech. 

It is entering a new phase with Trump's latest focus on the US drug war. 

Agencies, like Southern Command, the Coast Guard, DHS, are now receiving multi-year budget increases to be spent on robotics and autonomous systems.

This is driving real demand for Defense AI startups. 

• Drone and imaging companies are helping track small narcotics vessels across millions of square miles. 

• AI platforms are mapping fentanyl networks the same way they once mapped insurgent cells. 

• Counter-drone systems from Ukraine are now deployed at the southern border to defeat cartel drones. 

Anthony Antognoli, Coast Guard’s program executive officer, notes “It is impossible to do that work with humans and patrol cutters alone.”

“Our vision is to have robotics and autonomous systems as the foundation for the way the Coast Guard operates its missions.”

Autonomy is becoming the center of military operations.

These counternarcotics and border security missions are becoming reliable growth engines for Defense technology companies focused on autonomy, robotics, and applied AI.

Lumida Ventures is also focused on investing in defense tech. 

We have had a string of successful investments across CoreWeave, Kraken, and Brad Jacobs QXO, and Canva. All of these have had markups or realizations.

Do reach out now if you want to be in the flow of our next deal.

If you are an accredited investor or qualified purchaser, sign up here to receive communications about our private deals.

Macro

Economy Is Strong Heading Into 2026

This week’s data releases were solid, and they reinforced a simple point: 

The economy is stronger and broader than most expected.

Q3 GDP grew 4.4%, well ahead of consensus and GDPNow - the underlying message was even better. 

Unlike the fears that followed Q1, growth is no longer narrowly concentrated in AI-related spending.

Consumption remained firm, exports rebounded, government spending contributed modestly, and—critically—investment stopped being a drag. 

Growth across Q2 and Q3 has meaningfully broadened, with non-AI categories accounting for roughly 85% of total growth over the past two quarters.

That distinction matters. 

Q1 briefly created the illusion that AI was the sole pillar holding the economy up, raising concerns that a slowdown in AI funding could undermine overall stability. 

The latest data points to the opposite conclusion. AI is additive, not singular.

Growth is being driven by services demand, recovering exports, stabilizing inventories, and renewed capex. 

That mix makes the expansion far more durable than the narrow-growth narrative implied.

Productivity Is The Missing Variable

Core PCE came in at 2.9%, in line with consensus expectation. It has now run above 2% for 19 consecutive quarters. 

Yet inflation has not reaccelerated in a way that would normally accompany this level of economic activity. 

The missing variable is productivity.

GDP per private hour worked rose over 4% annualized in the quarter, with the underlying trend now running closer to 2%+—a clear break from the post-GFC norm. 

It reflects a structural improvement in output per worker.  

Since COVID, firms have aggressively adopted AI, automation, and software-driven workflows to offset labor scarcity and rising costs. 

Processes that once required incremental hiring are now scaling digitally. Output is rising without a proportional increase in labor or input expenses. This reflects in rising corporate earnings and margins.

Improved productivity helps keep inflation contained despite sustained demand. They are absorbing growth rather than transmitting it into prices.

Markets

It’s not a Bubble

Every time, market closes in on its all-time highs, the “bubble” headlines start to surface. 

The data doesn't support it.

Start with cash flow. The median stock’s free cash flow yield is ~3.5% today, materially higher than it was at the peak of the Tech Bubble, when it bottomed near 1%. 

In 1999–2000, investors were paying extreme prices for companies with little or no cash generation. 

Today’s market is fundamentally different: profitability is widespread, and cash flows are real.

Valuations tell the same story. 

Even before adjustments, today’s multiples are nowhere near dot-com extremes. 

When the S&P 500’s forward P/E is normalized for profit margins, valuations sit close to their long-term average—far from bubble territory. 

Elevated margins are not a mirage; they reflect structural shifts toward software intensity, automation, and capital-light models. 

Those dynamics drive earnings—and earnings create value.

Investment breadth reinforces the point. 

Capex is accelerating across both the Mag 7 and the S&P 493. 

Bubbles are marked by capital scarcity outside a narrow leadership cohort. 

This cycle looks different: investment is broadening. 

When capex rises across the index, it signals confidence in future demand and profitability beyond a single theme.

Importantly, this is not speculative capex. 

Spending is flowing into AI infrastructure, automation, software, and productivity-enhancing assets—the same forces underpinning durable margins and cash generation. 

Companies are investing because returns justify it.

That is not what bubbles look like.

Here’s Jurrien Timmer’s tweet talking about returns since the beginning of the secular bull market. 

We believe equity markets will continue to outperform.

However, near term, we might see some fluctuations due to mid-term elections risk and excessive sentiment which we discussed last week.

We wrote about it a few weeks earlier. Read here.

Our inference from a number of factors is that investors should maintain a strong tilt towards value.

We had a nice high beta lock-out rally on the heels of OpenAI’s plans for fianncing. That sparked a small short squeeze.

As I noted earlier, there are plenty of mispriced assets out there - including those growing 20% revenue YOY with 5% free cashflow growth.

It’s observations like this that will cause the rotation to value… Investors will compare assets and arrive at these conclusions independently.

Our conviction comes from our process. We are looking at hundreds of charts and running stock screeners and we keep saying ‘Why would I own [ random over-valued stock ] when I can own this?”

The Bubble In OpenAI’s 70% Gross Margins

OpenAI is showing ‘improved margins’ on its API compute. 

I call bullshit. 

Gross margin excludes the dominant economic cost of the product.

(1) Model Training Costs are the most significant cost to OpenAI. 

Those are ‘below the line’ - it shows up as a Research Cost not Gross Margin. 

Gross Margin is based on Inference Costs, not Training Costs. 

We know LLM models go obsolete (GPT rendered 5.1 obsolete and so forth).

(2) OpenAI can make their margins look higher by using operating leases with datacenter companies like Coreweave 

This effectively converts large capital expenditures (GPUs, power, cooling, real estate) into a rental model.

The right way to assess OpenAI is Return on Invested Capital (ROIC) and WACC.

Coupang Call Paid Dividends

Last week, we talked about CPNG, and how it had a headline sell off after the data breach news. 

The stock sold off ~25% after Nov 04, but the fundamental business was intact. 

It has revenue growth faster than Amazon, better valuation metrics, and no looming hyperscaler’s Capex risk. 

This week, that call paid dividends. The stock had 3 green days, rising ~7% on Friday. 

The reason?

Clarity around the breach. 

The company disclosed that the breach was limited in scope, tied to a former employee, and did not involve sensitive financial data. 

To put it in numbers, management estimates the data leak involved a few thousand customers instead of the 33 million initially assumed. 

The news meant the worst-case scenarios being priced in were unlikely. Once that became clear, risk was repriced quickly.

The business was always solid, only the sentiment around it broke. That’s what we highlighted in our writeup. Read here.

Digital Assets

Banks As The Next Mag 7?

Tom Lee says JPMorgan and Goldman Sachs can become the next Mag 7 by rewriting their cost structures.

And, we agree.

AI and blockchain both attack the same problem: employee intensity. 

Financial services has historically scaled by adding people: more compliance, more operations, more middle-office headcount. 

Automation changes that. 

AI reduces labor per dollar of revenue. Blockchain reduces reconciliation, settlement friction, and back-office complexity. 

This results in structural margin expansion.

This leads to an important distinction:

• When margins expand because demand is strong, it’s temporary.

• When margins expand because workflows are digitized and automated, it’s durable. 

That’s why Lee argues these banks will eventually trade more like tech companies: 

Higher multiples, steadier margins, and operating leverage tied to software adoption rather than balance-sheet growth.

Overall, the structural benefit of AI and blockchain will further intensify the adoption of digital assets in mainstream finance.

We are already seeing the shift. Read our read throughs from bank earnings to see how banks are taking tokenization mainstream.

Where Are We In The Digital Asset Cycle?

At the passage of the Genius Act, digital assets reached the peak of inflated expectations. 

The narrative was clean. 

“Valhalla was coming. Progress felt inevitable. Upside felt frictionless.”

But cycles don’t move in straight lines. 

You don’t arrive at Valhalla without first climbing the Wall of Worry—and for digital assets, that wall was defined by the Gensler era. 

Regulation, uncertainty, and enforcement pressure slowed momentum and tested conviction. 

The inflection came after last year’s election. 

In a remarkably short period, the market pulled forward expectations that would normally play out over years.

Bitcoin didn’t grind higher through successive phases—it compressed an entire year’s worth of price discovery into a one- to two-month window. From the mid-60s to new highs, quickly. 

That compression was not random. It reflected a market where information moves faster, capital reallocates faster, and positioning resets more rapidly than in prior cycles.

That same structural shift showed up again when Bitcoin reached new all-time highs before the halving, something that had never occurred in earlier cycles. 

Then over the summer, a wave of DATs and IPOs came to market. 

New narratives. New tokens. New opportunities.

That was Valhalla - the peak of the cycle.

Retail investors chasing momentum got in with excitement.

These investors were not looking to compound patiently. They were chasing acceleration—lottery tickets, rocket ships, moon narratives. 

But, something unexpected happened.

Digital asset trade slowed down, while other assets kept working. 

Nvidia kept rising. Meta kept rising. Space stocks moved higher. Gold moved higher.

Predictably, many of those trades moved quickly from excitement into the trough of disillusionment.

“What did I buy?”

“I shouldn’t have owned this.”

“I just want out.”

That psychology is critical. 

Over time, the relative underperformance of digital assets became unbearable. The opportunity cost became too visible to ignore. 

Eventually, frustration forced capitulation.

And that moment—when investors sell not out of fear, but out of exhaustion—has historically marked something important. 

The Start Of The New Cycle.

My primary approach to gaining exposure now is through Crusoe-style off-grid Bitcoin mining.

We partner with operators that use excess natural gas to mine bitcoin at an effective cost of $40K. 

It doesn’t matter what your view on bitcoin is. 

If you can produce at a much lower cost than the market price, there is a quasi-arbitrage to take advantage of.

We have put together a comprehensive guide on how off-grid bitcoin mining works, and the opportunity for investors. Check it here. 

Lumida Curations

Structure Matters More Than Speed in AI Investing

As AI exposure becomes easier to access and more intermediated, investor outcomes increasingly depend not on picking the right theme, but on underwriting the structure, incentives, and discipline behind the investment.

View Curation

Why the Fed Can’t Abandon the 2% Inflation Target

Bessent believes Changing the target before inflation is fully defeated would undermine Fed credibility, un-anchor expectations, and teach markets that policy goals shift when conditions get uncomfortable.

View Curation

Protein Isn’t the Problem

When calories are controlled, higher protein—especially spread across meals—improves fat loss and muscle preservation. The real risk driver isn’t protein itself, but excess calories and obesity.

View Curation

Meme

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: US Economy Stays Steady; Labor Market Is Cooling; Is Automation Replacing Workers?

• Markets: Everyone Is In Markets; Exploiting the Low-Beta Anomaly; Cruise Line Names Shined This Week; Coupang- The Opportunity In Sell-Off; Ripple Effect Of OpenAI’s Raise

• Lumida Curations: Code Red or Code Slow?; The AI Bubble Isn’t the End of the AI Boom; Morning Cortisol Is the Longevity Lever

Are Boomers the Lenders Last Resort?

Overall, we are seeing a re-leveraging of corporate balance sheets. 

And, governments continue to spend and increase debt outsanding.

In equilibrium, it is impossible for all three groups to lever up at the same time.

That leaves consumers as the lenders of last resort. 

Consumers - specifically Baby Boomers - are sitting on record home equity and have strong balance sheets. 

Boomers are now the lender of last resort. 

And, they are doing this with $5 Tn in private credit outstanding and ownership of Treasuries.

This goes back to a theme we have written about before…

Lowering short-term rates is likely to have differing effects on household spend than the Fed anticipated previously.

Lowering short-rates may decrease boomer spending.

What is the direction of long-term rates?

A big question…

What is the direction of long term rates?

(1) On-shoring of supply chains adds to inflation pressure

(2) Massive capex build-out increases demand for borrowing in corporate debt markets

Most are high quality borrowers in the Mag 7 arena (see Amazon) but vendor financing will expand the high yield debt market

(3) Markets struggling to digest new offerings (see Oracle)

(4) Greater M&A increasing demand for debt (see Warner Brothers bid)

(5) Less foreign demand for US Treasuries and more demand for hard assets such as gold

Hard to see a world where long-term rates head lower. 

The counter to all of this: 

(1) A strong boost in productivity rates to combat inflation

(2) Labor market breakdown (unlikely) 

Why All This Matters?

(1) The rally from 1982 to 2021 was driven by a gradual lowering of long term rates. 

If rates are headed higher, valuations will face headwinds. 

(2) If you know the direction of long-term rates, you can position more accurately in rate sensitive names. 

(3) Tech capex bingers are becoming more rate sensitive arguably as they borrow tens of billions of dollars where they did not before. 

What do you think?

The Rise Of Defense Tech

Global defense spending is accelerating.

Europe is investing more in Defense as the U.S, pulls back.

World military expenditures rose 9% last year - the sharpest increase in more than three decades. 

Defense budgets are rising across all geographies after recent geopolitical tensions.

These budget increases are creating massive demand for defense tech names, and venture capital has realized this.

In 2025, defense tech venture investment has already more than double last year’s total. We don’t see it stopping anytime soon.

The funding is concentrated in large, late-stage rounds of autonomy focused defense companies. Companies like Anduril and its peer group are benefitting from these financing rounds.

Those financings are only available to sophisticated private markets investors.

Lumida Ventures is also focused on investing in this category. We have had a string of successful investments across CoreWeave, Kraken, and Brad Jacobs QXO, and Canva. All of these have had markups or realizations.

Do reach out now if you want to be in the flow of our next deal.

If you are an accredited investor or qualified purchaser, sign up here to receive communications about our private deals.

Macro

US Economy Stays Steady

The US economy continues to grow against market expectations.

Real growth is tracking near 3.5%.

Growth is being carried by business investment in equipment and software, which shows businesses are confident and willing to invest. 

Again, we have 1% of GDP growth coming from AI Investment - that’s not stopping anytime soon - especially if OpenAI raises $100 Bn!

Capex tied to productivity tends to persist across cycles, and acts as a solid indicator for future growth.

On the consumer side, we see a confident customer. 

Headline retail sales look flat… but I expect this will beat. I was at the American Dream Mall this weekend. Number 2 mall in square footage in the United States.

It was packed as if the Fed just cut rates and handed out mall gift cards.

Households are still allocating toward discretionary categories tied to lifestyle and convenience, while pulling back where higher rates directly bite, like autos. 

Another underestimated signal is on the supply side; Business formation is surging, and is reaching its 2021 high. 

Small businesses drive most employment growth - this is encouraging.

If new firms are being created at this pace, then economic strength is structural. 

New companies create new payrolls, new capacity, and new competition, which helps the labor market and economic growth.

Labor Market Is Cooling

This week’s labor market data agrees with the theme we have seen in the last few months. Labor market is stabilizing. 

After spiking over the summer (like last year, and the year before that) claims have rolled back over, and the 4 week average is now back to January level. 

Continuing claims reinforce the point. 

At 1.9 MM , they’re well below the 1.92 million threshold that would suggest compounding layoffs. 

If labor stress were building, displaced workers would be lingering longer. 

Instead, reabsorption remains fast enough to prevent duration pressure.

The Duke–Richmond Fed CFO Survey explains why the labor market is holding together. 

Revenue growth expectations are broadly stable, unit cost pressures are easing, and wage bill expectations are stable. 

CFOs showed confidence in the economy, and the future business environment.

Hiring behavior reflects the “low-hire-low-fire” environment. 

Most firms are replacing open roles, 40% are still hiring, and fewer than 10% report layoffs. 

Only 13.3% say staffing levels are ideal, implying labor demand hasn’t disappeared — it’s become more selective. 

Is Automation Replacing Workers?

Based on the CFO survey, When firms do slow hiring or reduce headcount, the dominant drivers are demand uncertainty and financial constraints. 

Automation is a secondary factor at best, cited by just 16.2% of respondents.

Firms are not cutting labor because of AI - they’re investing alongside it. 

84% of CFOs expect non-zero AI spending over the next 12 months, up from 56% last year, with adoption broad rather than concentrated.

CFOs also attribute AI to faster decision-making, higher efficiency, improved customer outcomes, and more time spent on higher-value work. 

We also see this trend in Vanguard’s study. 

Occupations most exposed to AI are outperforming. 

Job growth in high-AI-exposure roles is running 1.7%, more than double the 0.8% pace seen across all other occupations. 

Real wages tell the same story: +3.8% for AI-exposed roles versus +0.7% elsewhere.

Rather than displacing workers, AI is raising output per worker, increasing bargaining power and wage growth where it’s actually being deployed. 

This is an interesting post from a lawyer citing how AI crushed him at preparing a long legal brief.

But, here’s the thing. Only lawyers are licensed to carry out legal actions.

AI will cut the pricing for legal services and improve their margins.

That’s not mass job displacement.

Our legal frameworks - even beyond the law - require a concept of Accountability.

There’s always a “throat to choke”. So, while there will be shifts in hiring in certain markets (less demand for financial analysts and coders), labor markets will clear and aggreate earnings will increase.

Markets

Everyone Is In Markets

As a contrarian, I don’t like this chart. Not one bit.

Cash levels in the BofA Global Fund Manager Survey have fallen to a record low ~3.3%, which tells you something simple and uncomfortable: investors are already all-in. 

Historically, when cash drops below roughly 3.5%, forward returns compress sharply. 

One-month forward returns after similar readings have skewed negative (averaging -2%), even in otherwise healthy macro environments.

The BofA Bull & Bear Indicator confirms the same dynamic from a flow perspective. 

It jumped to 8.5, firmly into contrarian sell territory, driven by massive equity ETF inflows. 

I don’t like that either.

It reinforces my belief that you should position around Value rather than Growth.

When the 2021 market cycle topped, this Flow Show indicator also topped. Low cash levels also indicate there’s not much buying firepower left.

This shows investors positioning for what’s already happened.

We’re already seeing the early signs. 

The macro backdrop is supportive — inflation is cooling, growth is holding, labor is easing — yet markets are struggling to build momentum.

Growth has failed to keep up with value which recovered to all time highs first.

Add to the fact that we have mid-term elections coming and a typical drawdown during that period, and you should have more defensives.

We have over-weights in Healthcare and Utilities. We also like Financials - and especially Consumer Finance and Regional Banks - which are breaking out on strong fundamentals and monetary easing.

I think Utilities as an ETF will do quite well the next 3 to 6 months and that’s a cleanr expression. The 10-year is likely topping out around here, or at least capped, given the benign inflation print we saw.

Exploiting the Low-Beta Anomaly

Most people want to achieve outsized returns by owning high beta.

That can work in a bull market... unless you get caught up in a correction after which you can get wrecked.

It's a high risk strategy, and most people will not implement it correctly. 

There is a different way...

Oftentimes, there are higher risk-adjusted returns in low-beta stocks that are set to out-perform.

For example, in last week’s ledger, I had a write-up on Comcast.

Interestingly, Comcast is a "low beta" stock. It had a nice run up of 7% this week. We really nailed the bottom.

Now, if we had an ETF, I would have bought cheap call options on the name - because the beta of Comcast is low.

The main issue with a low beta asset portfolio is your total returns will lag the overall market in a bull market.

Utility stocks, insurance stocks, and staples - the kinds of businesses Warren Buffett likes - don't move higher for a given change of the S&P as high beta stocks.

But, their risk adjusted return is higher. 

Meaning, the return you generate per unit of volatility is higher. 

Mathematically, the right way to own this asset then is with appropriate leverage.

This is the same concept as owning the 'tangency' portfolio - something every finance student learns in school - and then using leverage to extend his returns and calibrate the volatility to his desired risk budget.

You own the asset mix that has the highest risk adjusted return. 

Then, if you are benchmarking to the S&P 500, you apply leverage such that the beta of this portfolio now matches the S&P 500.

If you've selected that portfolio correctly, you should now outperform the S&P 500.

Warren Buffett does exactly this.

He owns low-beta assets.

He uses non-callable permanent financing with AAA debt to bring his beta higher. 

The assets he owns have durable cash flows and wide moats - so there is no margin call risk.

Buffett's usage of leverage brings his portfolio beta higher. 

He needs to do that otherwise he would lag his benchmark.

Buffett never tells people explicitly that this is what he is doing - but rest assured it is one of several pillars to his strategy.

What's the point?

There are a lot of high-beta chasers on FinTwit.

That strategy works...provided you are on the right side of the beta factor.

When the tide goes out, it can be vicious. 

You're actually better off identifying low-beta assets and then calibrating to your desired level of risk.

I hesitate to write this, because you need sophisticated systems and factor models to implement this correctly.

But, this is the correct approach to modern investing.

And no one talks about it.

Here's one last way to make this point.

I have strong conviction Utilities will out-perform the S&P on a risk-adjusted basis with lower volatility over the next 3 to 6 months.

But, utilities have lower beta and volatility than the S&P.

So the S&P can still "beat" utilities... even though it is taking more risk to do so.

A more efficient strategy is gaining exposure to Utilities in a format that comps to the risk budget of the S&P 500.

Ripple Effect Of OpenAi’s Raise

Last week, we wrote how we weren’t convinced on semis going forward due to OpenAI’s financing risks, and exaggerated valuations in the sector.

The Semiconductor, SMH, had three down days on its correction, but this changed on Thursday. 

Reports leaked that OpenAI is raising roughly $100 billion at a valuation that could approach $800+ billion. 

The S&P 500 jumped ~1% that day. Semiconductors stabilized almost immediately. 

In prior weeks, the primary driver of market beta was Fed rate cut expectations for December, now it all hangs on the probability of OpenAI closing out a $100 Bn raise by end of Q1.

This is the most important news item to follow.

The central issue hanging over AI infrastructure has never been “Is demand real?”

It has been: Are the buyers solvent, committed, and capable of funding multi-year data center obligations?

That question got some answers after the news. 

OpenAI sits at the center of the data-center buildout. In particular, it is a key counterparty behind Oracle’s rapidly growing backlog. 

Those contracts only work if the customer can fund them. 

If the customer cannot put real equity behind those commitments, projects slow or get re-priced.

A raise of this size reduces that risk materially.

It improves visibility around Oracle’s ability to execute on its backlog. It lowers the probability of delays, renegotiations, or cancellations tied to customer funding constraints. 

That is why semiconductors stabilized almost immediately- we saw the effects across the flywheel.

The market was not repricing AI demand. It was repricing balance-sheet risk.

I am rooting for OpenAI to complete this raise - it’s crucial for the sector.

However, the $100Bn revenue projections used to justify OpenAI’s valuation remain unrealistic in my opinion. 

It assumes rapid monetization across products that are still early. They assume pricing power in a market where competition from Google and Meta is intensifying.

Google and Meta are players with durable cash flows, profits, and an already used ecosystem. 

The adoption of their AI models is significantly more easier because the models are seamlessly integrated within their product suite.

Also, I hardly see AI running its current pricing models in the long-run. Freemium ad models - those offered by Google and Meta - should rule the roost.

We will see more models, more free tiers, and more enterprise options, putting pressure on returns over time.

However, this Competition does not stop capex. The infrastructure still gets built, which is why semis saw recovery. 

Coreweave Join Genesis Mission

CoreWeave joined the Department of Energy's Genesis Mission.

DOE’s Genesis aims to speed up research by giving U.S. labs and institutions shared, scalable access to advanced computing for large AI models and datasets.

Coreweave provides the execution layer with on-demand GPU and AI-optimized cloud infrastructure. 

This deal diversifies CoreWeave's customer base toward institutional, public-sector demand, securing long-lived, less price-sensitive workloads that improve visibility and reduce reliance on volatile AI training demand.

This fits our long-term bullish view on CoreWeave.

We picked up 0.5% (on average depending on account) at the middle of this week before the 20% ish run-up. My only regret is not buying more. But, that’s how I always feel when these run-up. And, when it doesn’t work out, I’m glad I only bought 0.5%.

More generally, because we are constantly sourcing ideas, we never feel compelled to FOMO into anything or let positions size on any name get too large. We believe in diversifying our idiosyncratic risk. This is an important concept to understand - and it’s why Citadel, Millennium, and other quant shops do so well.

They have a small edge on many securities. They seek to capture that edge at scale. It takes a lot of investment infra to execute against this.

We have something like 50 positions and are up about 40%+ since the April lows. We are ‘diversified’ in several themes we like.

Did I tell you about the Wolfe Quant Factor holiday party we went to? Was funny to see the name tags of those in attendance - Point 72, Two Sigma… and Lumida.

It was a quiet party, not popular. Lot of nerdy quant types. I felt right at home. We are still very, very early.

Modern investing is not modern. Bill Sharpe and Harry Markowitz are still considered the leaders who ushered in ‘modern investing’ — that was 1950s technology… The investment philosophy has yet to evolve.

Incidentally, Coreweave was our very first private deals. We liked it because it has a good secular trend, a forward PE of 15x, and it was an ‘East Coast’ deal - not bid up to the moon by West Coast VCs.

We got in at a $7Bn valuation, and once it had an IPO, investors enjoyed a net realized gain of ~219%.

If you are an accredited investor or qualified purchaser, you can sign up with Lumida deals to get priority communication on our latest private deals.

Cruise Line Names Shined This Week

Cruise line names moved higher this week after Carnival Corp (CCL) reported. 

The reason: demand for experience-based spending remains strong.

Geoffrey T. Martin (CEO, CCL): “Demand for cruise lines is proving far more resilient than traditional macro indicators would suggest.”

Geoffrey pointed specifically to forward indicators:

“Customer deposits are up 7% year-over-year, hitting an all-time high.”

“Two-thirds of the business is on the books at higher prices.”

The improved bookings and prices highlight stronger demand going into the next year. 

This is why we saw rapid jumps in this sector.

NCLH was also up 15% this week- I flagged this name a few weeks ago, and that it was trading at a bargain then. The stock is up 30% since the post. Read our thesis on NCLH here.

However, our call on Resmed (RMD) was a miss. The stock dipped below its 200D this week. 

But, we are confident in its strong growth potential fueled by ongoing product innovation and a significant addressable market. 

Hence, we are holding it. 

Coupang- The Opportunity In Sell-Off

CPNG, the Amazon of South Korea, went on sale this week after a customer list data breach.

South Koreans love to trade. That also means they panic sell.

The stock sold off quickly, but the fundamentals didn’t change.

On valuation, Coupang is meaningfully cheaper than AMZN on price-to-sales and price-to-gross profit, despite faster revenue growth. 

Unlike U.S. hyperscalers, it also doesn’t carry a looming datacenter capex overhang that pressures near-term cash flow.

We scooped it at the lows, and then added to our position again on Friday after the 2.25% recovery. 

Coupang’s core advantage is not e-commerce. It is the logistics density.

The company has already built the hardest part of the business: a nationwide, vertically integrated fulfillment network that enables same-day and next-day delivery. 

That infrastructure is now being leveraged across higher-margin adjacencies like Eats, advertising, fintech, and media.

That shows up clearly in the numbers:

• Revenues has grown high double digits in last 3 quarters.

• Gross margins is near 30%, at the top of its historical range

• Gross profit growth (30%) is outpacing revenue growth (18%), signaling operating leverage

This is not a company burning cash to buy growth. It’s one converting scale into economics.

As for the breach headline, it didn’t cause user churn, engagement decline, or logistics disruption. But, it gave us an opportunity to buy the stock 10% cheaper. 

However, one can argue that high P/E TTM valuation (~110x), despite the dip, shows investors are already pricing in future growth, and any execution below perfect can hurt the stock.  

What Actually Creates 10x Stocks? 

This week, quant science published a quantitative study of 464 stocks that became 10x-baggers over 24 years.

They studied hundreds of companies.

Only a small fraction compounded 10x.

Let’s walkthrough the top 3 takeaways.

1. Size helps but only as runway

Small-cap stocks outperformed mid and large caps in 11 out of 12 cases.

But size itself isn’t the edge.

Smaller companies simply have more room to grow into their valuation.

That only works if expectations are modest at the start.

Small + expensive rarely compounds.

Small + reasonable can.

2. The “value effect” is really a warning: don’t overpay

Stocks with low book-to-market ratios — meaning investors paid far more than underlying assets justified — consistently underperformed.

This isn’t philosophical. Overpaying compresses future returns.

Multibaggers don’t start life as consensus favorites with heroic assumptions baked in.

They start with valuation slack.

3. Free cash flow yield is the most important factor

This was the key result of the study.

Across decades and hundreds of cases, free cash flow yield was the single strongest driver of 10x outcomes.

Why?

Cash does three things:

• Funds reinvestment

• Absorbs mistakes

• Compounds quietly while attention is elsewhere

Revenue growth tells a story. Free cash flow tells the truth.

Multibaggers are created by not overpaying, buying real cash generation, and letting time do the heavy lifting.

You will notice in nearly all (maybe all?) of our stock write-ups we are looking at Free Cashflow yield…

Lumida Curations

Code Red or Code Slow?

Sam Altman argues that simply bolting AI onto search won’t be enough for Google. Real disruption comes from rebuilding products AI-first, where agents don’t just assist, but decide.

View Curation

The AI Bubble Isn’t the End of the AI Boom

Demis Hassabis argues that froth and excess are inevitable in every major tech revolution, and while parts of AI will deflate, the underlying transformation is only accelerating.

View Curation

Morning Cortisol Is the Longevity Lever

The first 30 minutes of your day quietly shape stress, energy, metabolism, and aging. Small morning habits can either stabilize cortisol or push your body into survival mode before the day even begins.

View Curation

Meme

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: Fed’s Bullish On The Economy; Watch the Long End

• Markets: Still a Bull Market; Time For Mainstream Ideas; The Ignored Value; Out Of Semis Into Software; Google Is Expanding; Disney And OpenAI Deal;

• Lumida Curations: Prediction Markets Are The New Information Layer; The DeFi Regulatory Mismatch; Magnesium Is the Migraine Hack

Lumida Wealth Is Hiring 

I am looking for a Director of Client Service (NY/NJ) to join our leadership team at Lumida Wealth.

The cheat code to create wealth is to join a rapidly growing company.

Must have deep experience in alternative investments, understand trusts and estate, and wealth management, and tech forward.

This is your chance to shoot your shot.

Target Total Comp: $300k

Apply here.

Unprecedented Investments in Defense Tech 

Pete Hegseth has previewed DOD spending plans. I’m surprised this hasn’t received much coverage. We believe it’s an opportune time to “Front Run the Pentagon”.

The next decade of defense spending won’t be about planes, ships, or missiles.

It will be about systems that can see, decide, and act without humans in the loop.

Trump’s $200Bn Golden Dome project is a step in that direction.

Golden Dome is a nationwide defense architecture that tracks missiles, drones, and hypersonic threats across space, air, and ground in real time.

You can’t build that with people and legacy platforms.

You need autonomy. 

You need AI.

A system like Golden Dome requires persistent sensing, automated threat detection, real-time data fusion, and autonomous command-and-control. 

This is where defense budgets are headed. We believe there are a number of private market opportunities that can become brands and successful investments like Anduril in the next few years.

Lumida Ventures is actively working on investment opportunities in defense technology.

If you’re an accredited investor or qualified purchaser, sign up here to receive priority communications.

We have previously done successful deals with Coreweave, Canva, QXO and Kraken - all of which have markups or realizations at gains.

You can read our success stories here.

Out Of Semis Into Software

I did a Tesla FSD livestream on why I am bearish on semiconductors.

Now that markets have recovered to the “Sam Altman Highs”, I believe we’re going to see rotation out of semiconductors and into software.

The issue is that hype exceeds reality on semiconductor spend.

Take a look at the valuation multiple for Applied Materials for example. We are now in the nosebleed section.

Investors are questioning how OpenAI will achieve its $1 Tn in committed spend with sub-$20 Bn in revenue.

That’s going to take a lot of debt financing…and debt providers are already holding back (see Oracle for example). It will also take a lot of equity financing.

But, OpenAI is already raising from Softbank for Project Stargate which is equivalent to ‘last call’.

OpenAI will need to cover approximately $207Bn in capital shortfall by 2030, as per HSBC, and it won’t be generating profit till then. 

We are also seeing intesifying competition from well-financed players like Google and Meta.

It’s hard to see a world where AI isn’t simply a free product (like other Google and Meta products) and how OpenAI fares in that world.

We wrote about Software a few weeks ago, and continue to see this theme do well.

We are positioned for this with exposure to ESTC and Workday, and a short position in SOX ETF. 

Software and semis are both subsectors in technology, and we have observed they tend to compete with one another for capital. When one is overbought, the other is oversold, and vice versa. 

Software has been beaten down.

But, these names have earnings growth, free cash flow and buybacks and distribution. 

The “AI kills software” narrative that dominated two years ago is largely priced in now.

Some software companies will struggle. But others are actually benefiting. 

They already have installed customer bases. They can layer in AI. 

They can become AI-relevant without blowing up their balance sheets.

Atlassian Corp is an example. Read our thesis on this name here.

The answer is not the analysis of whether it can or cannot. 

The approach to answer this is ‘what is priced into markets’.

If debt ratings start to come under pressure, then the cost of capital rises. When that happens, the perception around capex will be more restrained. 

And even a perception of slower spending is enough to move asset prices. This is being increasingly priced into the markets.

The bubble in AI is in Private Markets, but it will impact the semiconductor supply chain.

We are of the view that Meta is still cheap, and Nvidia at a sub-25x forward PE is attractive.

But the overall theme can be under pressure as we see a continued shift to value and “Main Street” type opportunities (look at retailers crank for example).

Macro

Fed’s Bullish On The Economy

The Fed cut 25 bps this week as expected. Since September 2024, the Fed has cut 175 bps, bringing the federal funds rate into the mid-3% range. 

Powell signals Fed is now “well positioned.”

His speech highlighted consumers are strong, business confidence is solid, and the economy is growing. 

Powell: “Available indicators suggest that economic activity has been expanding at a moderate pace. Consumer spending appears to have remained solid, and business fixed investment has continued to expand.”

“The consumer continues to spend. So it looks like the baseline will be solid growth next year.”

The main takeaway from the Fed is the dot plot and the Summary of Economic Projections.

The FOMC sees a pickup in growth next year, and a decline in inflation, and growth in productivity. We agree with that outlook - and it’s a constructive tonic for risk assets.

Don’t forget the stimulus checks too, which will serve as a boost to “Main Street”.

Look at the Retail ETF index and you can see markets are already discounting spend in the quarters ahead:

This indicates the Fed is underwriting productivity growth- something I’ve been emphasizing for a while.

Powell highlighted this in his speech. “Productivity has just been almost structurally higher for several years now."

We are already observing glimpses of the future.

This week, McDonalds launched their fully automated restaurant chain in Texas, which does not involve any human staff at all.

This is a concept store. But it’s a preview of what’s to come.

Watch the Long End

The key risk to monitor is the long end of the curve.

We saw in September 2024 what happens when the Fed cuts but long rates rise driven by bond vigilantes. You have to watch that carefully.

I think this time is different. 

Markets perceive a cooling labor market, not because labor is weak, but because growth in labor supply and demand is lower.

That’s just the physics of the system.

Importantly, corporate income and personal income tax receipts remain strong. And, consumer spending is strong.

These are reliable indicators for the real progress of the economy. 

Our thinking is that consumer discretionary linked categories will do well next year, and healthcare, and leaders in technology (esp software).

We’re still sorting out our views on homebuilders - we believe this category would rally if the 10-year breaks below 4%. It’s hard to get a view on that.

Add in tax cuts and rebates, and you have a setup for improving consumer confidence and spending next year.

Markets

Still a Bull Market 

The Dow Jones and S&P are cranking higher, while tech index Nasdaq lags.

The category broke out to all time highs.

Small caps are outperforming other categories.

Value is also outperforming growth. (Take a look at our newsletter called the “Revenge of Warren Buffett” which anticipated this shift.)

We expect that to continue next year - benefitting from lower interest rates, and consumer stimulus.

We continue to believe high beta categories of animal spirits have put in a local top. Robinhood, Palantir, Bitcoin, Uranium stocks are lagging.

Meanwhile, quality names like ICE and S&P Global are quietly rallying. We suggested buying those on this day a few weeks back. Read it here.  

Markets look constructive into the year-end.

We’re near all-time highs, yet consumer confidence is near the lows.

That disconnect matters.

Confidence is low not because the economy is collapsing, but because people are anxious. 

Why are they anxious? Doom scrolling.

Geopolitics. Headlines. Televised news. Social media. Constant doom-scrolling.

That’s driving sentiment, not fundamentals.

Market tops don’t form when people are anxious.

Historically, major tops come after long periods of euphoria. 

You see it in positioning. You see it in sentiment. You see it in behavior.

That’s not what a market top looks like.

This is a classic “wall of worry” market.

I was speaking with an RIA recently who joked that his office feels like a therapy couch. 

Clients come in trying to reconcile how bad the world feels with the fact that markets are still holding near highs.

That tension tells you a lot.

People don’t trust the rally. They’re cautious. They’re hedged. They’re worried they’re missing something.

That’s not how bubbles end.

Wary bulls are the best kind of bulls. 

When sentiment stays skeptical and positioning isn’t euphoric, rallies tend to persist longer than people expect.

This doesn’t mean there won’t be pullbacks. In fact, the 1H of a midterm election often does see a 7% on average pull back.

But, those pullbacks are buyable.

The Ignored Value  

It’s hard to make the case to own bonds in a “higher for longer” inflation environment.

There are plenty of public markets securities that offer dividend yield and earnings growth.

Why own Sovereign debt? I wanted to deep dive into one example of that.

This is not a bad idea for the conservative part of your portfolio - and value is working now.

A lot of attention now is on Warner Brother. We see ignored value in Comcast Corp which also has unique IP and content.

Comcast (CMCSA) has value in premium content libraries, supported by strong free cash flow. 

CMCSA’s stock has languished as cable cord cutting continues and its overall membership declines. We think it’s overdone with the name at 7x forward PE and a 20% free cashflow yield.

You have to believe that free cashflow yield is going to get cut in half. It’s hard to see that…

In fact, Comcast has grown their revenues and free cashflow in recent years, and analysts expect that to continue.

Comcast’s businesses generate steady, recurring cash flow from its business lines.

Comcast has been doing buybacks for years. They are prudent allocators of capital. Take a look at its outstanding sharecount.

Take a look at Comcasts PE ratio. We’ve never seen it this low…and revenue and cashflow is growing.

How about the technicals?

Comcast is sitting at multi-year support…despite the business growing during this time period.

The distance of price from the 200 day moving average is also severe. This is a “mean reversion” opportunity.

Comcast also has valuable IP content that could make it an attractive asset for failed suitors of Warner Brothers. IP content has a lot of value in the age of AI. (See Disney’s deal with OpenAI for example.)

Comcast owns NBC Universal. They explored combining these assets along with the streaming business - but ultimately lost out to Netflix’s top bid for Warner Brothers.

This is where Comcast’s story gets interesting. 

Its Content & Experiences segment—NBCUniversal, Peacock, Universal Pictures, DreamWorks, and Sky Studios—accounts for roughly 38% of revenue and delivers the upside. 

Live sports, including Sunday Night Football, keep viewers glued to NBCUniversal.

Streaming platform Peacock now has 41 million paid subscribers, benefiting from higher subscription prices and over 20% growth in advertising revenue. 

Comcast secured exclusive NBA broadcast and streaming rights in an 11-year, $76B deal through 2035–36.

Comcast is also simplifying its portfolio with divestitures.

Comcast is focus on higher-growth assets—NBC, Peacock, Universal Studios, theme parks, Xfinity, and Sky.

The bear case?

Starlink disrupts cable broadband, and the last of the boomer generation cuts their cord. Netflix builds a behemoth and content for NBA slides and there’s no real M&A interest in the asset.

But unlike peers like Charter (CHTR), Comcast pairs connectivity with content optionality, trading at a lower PE (~4.5x vs 5.7x) and generating a 21% free cash flow yield—roughly five times what it spends on buybacks.

Why own sovereign debt when you can own Comcast?

Disclosure: We own Comcast

Google Is Expanding

Gemini is being rolled out as the first enterprise AI on the Pentagon’s GenAI.mil platform, designed to support millions of users.

Gemini is already embedded across Google’s core products - Docs, Gmail, Sheets, Search, and Cloud. 

This is driving rapid adoption. 

If AI shows up where people already work, they use it by default. There’s no behavior change required.

Google has three things most AI-native players don’t:

• A massive installed user base

• Deep enterprise and government relationships

• Infrastructure that already runs mission-critical workloads

That combination lowers friction. It also lowers switching costs for users and buyers.

We wrote last week how Google is catching up against OpenAI. Read it here. 

We won’t be the buyers of Google at current valuations - it was a solid purchase in July when AI-kills-search narrative led to massive drawdowns. It was priced at 20x P/E NTM then-  today, it’s at 30x.

Why is Amazon lagging so much?

This post was at the bottom of the 2022 bear market. 

Back then, Amazon’s PE was 100x. 

After 3 years of growing earnings and going nowhere, the multiple is around 30X. 

I’d still rather own META which is growing earnings faster and is priced at 21x. 

The capex rising is the bear case for Meta. 

But, their CFO and Mark can essentially ‘paint by numbers’ their earnings trajectory by controlling these levers. 

And, I expect LLM competition to diminish. 

Have a look at Jason’s take here.

DISNEY AND OPENAI DEAL

Disney announced that it is investing $1 Bn in OpenAI, and will use OpenAI's APIs.

A few observations:

1) Disney could have adopted the APIs without spending $1 bn and being a de facto committed customer

2) Does Disney have any competitive advantage in late stage ventures? Is Disney's responsibility capital allocation?

(No.)

3) Did Disney derive any competitive advantage from the investment? 

(No.)

4) Disney did license its characters to OpenAI.

This part actually makes sense... 

The mistake here - Disney should be receiving major revenue from OpenAI from licensing content. 

Reddit is making money licensing its content, why isn't Disney.

5) Bog Iger over-paid for the Star Wars franchise. 

Now he has done the deal with Open AI in reverse. 

Bob is desperate to get some AI multiple in the stock. 

Bob Iger is a highly over-rated CEO and is driven by ego not shareholder returns.

Investor Tip:

How is it possible for the S&P 500 to out-run GDP growth year after year? Is that sustainable?

Take a step back.

2023: S&P up 20%+

2024: S&P up 20%

2025: S&P up 16%

How is this possible? 

GDP hasn't kept pace... We have deficits, wars, etc.

This is your self-test to understand markets.

Bill Gross famously got this wrong - the link between GDP and market returns - a decade ago.

Think about it for a minute before reading on.

The S&P can and will out-run GDP growth due to Earnings Growth.

Where does it come from?

1. Operating Leverage:

A company can grow revenue 5% but earnings will go up 12%.

GDP captures the top-line ('revenue').

Mr. Market focuses on the bottom line ('earnings')

The company gets disproportionate earnings due to fixed costs that don't scale with revenue.

2. Productivity Growth

The one free lunch. Companies become more efficient with technology, cost cutting, outsourcing, and lower input or energy costs.

Lower oil and nat gas prices directly boost corporate margins and reduce the cost of energy for consumers.

3. Interest expense

Lower interest rates enable firms to refinance debt. 

This reduces left-tail insolvency risk, and increases corporate profits.

4. The Market Prices the Future, Not the Present

GDP is backward-looking.

Markets discount expected earnings 6–18 months ahead.

A weak GDP print today may coincide with falling inflation, lower rates, rising productivity, and future earnings growth—all bullish for equities.

Lumida Curations

Prediction Markets Are Becoming the New Information Layer

Once dismissed as gambling, prediction markets are rapidly emerging as real-time signals for public sentiment, economic outcomes, and major events - often faster and more accurate than traditional media.

View Curation

The DeFi Regulatory Mismatch

DeFi exposes a growing gap between legacy financial regulation and internet-native markets, where trust shifts from intermediaries to code and responsibility increasingly sits with the user.

View Curation

Magnesium Is the Migraine Hack

One of the most evidence-backed supplements for migraine, magnesium works by calming neuronal excitability and improving cerebral blood flow, targeting the biology of migraines at the root.

View Curation

Meme

My thoughts on OpenAI..

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: PCE Adds Support For December Cut; How’s the Labor Market?; AI Is A Productivity Engine

• Markets: Growth Vs. Value; The EU’s War On Meta; How Far Can Netflix Drop?; Google Is Catching Up Fast; Defense Tech: The Next Structural Growth Lane; The Midterm Election Tailwind; Trump Bump?; International Equities Are Leading

• Digital Assets: All Markets On Chain; The Founder Behind Polymarket

• Lumida Curations: Bitcoin’s Four-Year Cycle Is Breaking Down; AI Is On Track to Run at Near-Zero Energy Cost; Alcohol Truly Delivers “Empty Calories”

Spotlight

I was in Miami this week, attending the Art Basel 2025 - funny thing: I don’t invest in art. I invest in businesses - investments that generate cash flows. 

I did two livestreams with exciting investors from the event. You can watch them here:

1. Non FSD: Art Basel with James K.

2. Art Basel: Indian Creek Island with Patrick

I also visited San Francisco on the way back for a housewarming party hosted by Vance Spencer- Framework Ventures. 

Got a chance to use Waymo during this trip - I have been a big fan of Google, and Waymo is one of the reasons why. I did two livestreams during my rides. 

1. Thoughts on Waymo vs Tesla FSD, Defense Tech, China vs Taiwan

2. Adderall & Sleep medication

The Rise of Defense Tech

The last twenty years were about software. 

The next 5 years are about hardware. 

Many of the most valuable companies in the world are hardware: Nvidia, TSM, Tesla, etc.

Where are we focused for our next theme?

Defense Tech. 

It is entering a new phase with Trump's latest focus on the US drug war. 

Agencies, like Southern Command, the Coast Guard, DHS, are now receiving multi-year budget increases to be spent on robotics and autonomous systems.

This is driving real demand for Defense AI startups. 

• Drone and imaging companies are helping track small narcotics vessels across millions of square miles. 

• AI platforms are mapping fentanyl networks the same way they once mapped insurgent cells. 

• Counter-drone systems from Ukraine are now deployed at the southern border to defeat cartel drones. 

Anthony Antognoli, Coast Guard’s program executive officer, notes “It is impossible to do that work with humans and patrol cutters alone.”

“Our vision is to have robotics and autonomous systems as the foundation for the way the Coast Guard operates its missions.”

Autonomy is becoming the center of military operations.

These counternarcotics and border security missions are becoming reliable growth engines for Defense technology companies focused on autonomy, robotics, and applied AI. 

Lumida Ventures is working on an exciting deal with a Pre-IPO company in Defense technology.  Sign up here to be on the priority communications list if are an accredited investor or qualified purchaser.

Wither Trump Bump?

Back in 2024, I shared a thesis called the ‘Trump Bump’

(See this thread for the evidence of a Trump Bump in equity markets leading up to the election.)

I am seeing evidence after the blue sweep in October that markets are starting to discount a DNC win in Congress. 

I expect a continued back and forth on this theme in markets.

1) What’s the biggest sector winner since those elections?

Healthcare stocks. 

2) Solar stocks stand to benefit in a DNC regime.

First Solar (FSLR) is cranking. 

3) How about Day care service providers, and other family benefits?

Bright Horizons (BFAM) is cranking off the lows post election.

4) META brought on UFC chief and Trump friend Dana White on its board and terminated a DNC aligned advisor after the election. 

Meta had an unusual sell off post election. 

5) Crypto assets have also sold off hard post election results. 

What other examples do you see?

(Below is my thread showing the Trump Bump hypothesis before it was Consensus.)

It’s Time To Take Action!

Rooftop and solar tax credits are set to wipe down at the end of December, and investors are rushing to cash in this opportunity.

And, why shouldn’t they?

Solar Investment Tax Credit (ITC) allows you to offset federal taxes by financing qualified solar projects. 

In practical terms, an investor with a $100,000 tax bill could put roughly $60,000 into a qualified solar project and reduce their tax liability by nearly 40%. 

Corporates have been doing this for years; in 2024 alone, more than $20 billion in solar credits were traded, up four-fold from the year before.

Lumida can help you avail these solar tax credits, and reduce your tax liability substantially.

We have a limited capacity on this opportunity; so, now is the time to act! 

Click here to know more about this and other tax mitigation opportunities. You can also book a free consultation call here, and learn how Lumida can help you save taxes.

Macro

AI Is The Productivity Engine

The benefits of AI are appearing in a number of industries.

Slower price growth is increasingly being driven by productivity gains, especially in service-heavy sectors where AI and workflow automation are starting to reduce unit labor costs. 

We’re already seeing the results:

• Goldman Sachs estimates that generative AI could raise labor productivity by ~1.5% per year over a decade and add roughly 7% to global GDP.

• PwC’s 2025 AI Jobs Barometer notes that industries “most exposed” to AI saw 3x higher growth in revenue per employee than the least exposed sectors.

When firms can produce more with the same headcount, or hold staffing flat while output rises, the pressure to pass through higher prices diminishes. 

Lower PCE also means Fed will have one less thing to worry about for the December meeting.

We think the Dec rate cut is almost certain at this point - the economy doesn’t need it, but markets are pricing it. 

Inflation Is Behaving Well

The inflation print for September came in exactly where pro-rate cut policymakers hoped:

Moderate, stable, and below 3% on both headline and core. 

How’s the Labor Market?

This week, we got two divergent data points on the labor market. 

Initial jobless claims print delivered one of the sharpest surprises of the year. 

Initial claims cratered to their lowest level since 2022, and excluding that period, the lowest since 1969. 

Remember how everyone thought the labor market was cracking over the summer? Next time you see those headlines, just take a deep breadth and remember how bad the data quality is.

The economy is an organism, fuelled by earnings and population growth, and it exhales. The data that we use to measure it (outside of corporate earnings) are skewed, noisy, and have massive standard errors.

Continuing claims, which had been drifting 5% higher YoY through the fall, abruptly rolled over as well. 

All of this is quite bullish - especially for breadth and consumer discretionary stocks.

Take a look at how Dollar Tree, Dollar General, American Eagle and other retailers are doing.

Costco (over-priced bubble stock) meanwhile is giving back ground to names investors ignored.

The ADP Report is Garbage

The ADP report showed a contrasting view with payrolls declining by 32000 in Nov. 

What do we make of it? Nothing!

ADP reports have structural deficiencies. 

Simply put, ADP clients aren’t representative of the U.S. labor market.

Trends in Offshoring, contract work, gig economy, (illegal) immigration change the data.

ADP spikes/dumps in yields tend to mean-revert.

It’s noise. Look out for any over-reactions. 

Overall, the underlying message around labor markets is unchanged: 

It is cooling - we are in a “low-hire-low-fire” environment.

Businesses are leveraging AI and automation to increase output without increasing workforce, which is reflected in higher productivity, and improved earnings. 

The December rate cut won’t fix the labor market, but it will most likely create asset-based inflation.

Markets

Growth vs. Value

Every few months I like to trot out this growth to value chart.

Growth remains over-priced vs. value, and investors are beginning to realize it. 

The S&P 500 Value ETF (IVE) ended this week near its All time highs, showing a strong recovery post its low in November. 

Growth isn’t breaking down either. For the first time in months, it’s failing to lead. 

It’s more pronounced when you look under the hood. Fintwit darling Axon is in a bear market, like Cava before it, like Chipotle before it, like Fico before it… It has a 176x trailing PE ratio.

Small caps tell the same “tilt to value story” more clearly.

Over the past six months, small-cap value has steadily outperformed small-cap growth - a sign that breadth is widening beneath the index and that investors are reallocating toward businesses with tangible cash flow, pricing power, and lower duration sensitivity.

We wrote how small cap values are poised to benefit in this shift to quality two weeks back, and markets are playing out the tape. (I have a hazy crystal ball) 

We flagged Ziff Davis as a beneficiary. The stock has risen 21% since then. Read the full thesis here. 

(Note: We sold Ziff Davis this Friday, it was extended, might get it later.)

Historically, when IVE hits a 52-week high at the same time IVW fails to do so, value tends to outperform over the next 3, 6, and 12 months, with the spread widening along the horizon.

The EU’s War on Meta

The EU is going after Meta for integrating AI into WhatsApp.

This is really absurd.

Meta is in near last place among Super Apps for AI dominance.

AI is a product feature for WhatsApp.

All apps must be reimagined as AI-native.

Any app that is not AI-native will disappear into the wastebin of history.

The EU wants Meta to offer competing chatbots.

That’s also absurd.

That’s like demanding Apple ship the iPhone with Samsung’s camera app.

Or requiring Tesla to install BYD’s drivetrain.

Or telling OpenAI to bundle Anthropic by default.

No regulator in their right mind forces a company to embed a rival’s core product into its own interface.

European regulators are treating AI as some separate market rather than the foundational layer of all software.

AI isn’t a standalone product category — it’s the new UX, the new distribution layer, the new operating system.

It’s anti-innovation, anti-consumer, and ultimately anti-competitiveness.

This is what regulatory over-reach looks like. 

No wonder entrepreneurs continue to flee the EU for the United States.

Code Red For OpenAI

This week, Sam Altman shared a “Code Red” memo with OpenAI employees, and it tells exactly what we signaled a few weeks back.

People compare OpenAI to Amazon…but they forget Amazon made money for the vast majority of its history and could self-finance growth.

And the valuations were never as outrageous as OpenAI.

You can be an AI bull, like me, but at least differentiate winners and losers.

The competitive tides are turning against OpenAi, and the landscape is changing faster than expected. 

Google’s Gemini 3.0 didn’t just post better benchmark scores; It demonstrated that an incumbent with global distribution, integrated products, and near-infinite compute can shrink technical gaps almost overnight. 

OpenAI senses the fear. 

The Code red memo signals OpenAI will reallocate resources away from longer-dated initiatives and back toward shoring up the core ChatGPT experience. 

The focus will now be on enhancing speed, personalization, and reliability. 

In the last few weeks, OpenAI has gone out of favor - if it was a listed stock, the drawdown would have been Figma-esque.

Look at the difference between the losses OpenAI and Anthropic will incur before becoming profitable. 

Anthropic is also front running OpenAI for an IPO in 2026. This week, they signed law firm Wilson Sonsini to launch the IPO process and are seeking a valuation of $300 billion. 

Overall, AI leadership is transitioning from model breakthroughs to the harder, more durable moats of deployment, infrastructure, and ecosystems. 

Firms with cash flow and distribution scale -  Google, not OpenAI - are structurally advantaged. 

I did a Waymo video during this week’s SF trip, comparing Google and OpenAi. Watch it here. I discuss my views on Defense Tech there as well.

How Far Can Netflix Drop?

Netflix has lost 19% in the last 3 months. But, it's still more expensive than NVIDIA.

They recently announced the acquisition of Warner Bros at $82.7Bn, a ~30% premium to WBD’s current market cap. 

We have been bullish on WBD- here’s our thesis from July- it pays to be contrarian. 

The funny thing about that post… The one and only one reply asked sincerely “What is the bull case?”

Bro, the valuation and FCF yield was the bull case!

The best ideas are those that people struggle to recognize. Cognitive dissonance is a contrarian tell.

It goes in reverse too. Look at SoFi. How many people will hold on to this after the $1.5 Bn secondary sale this Friday?

Twitter tweet

The Midterm Election Headwind and then Tailwind

Markets are going to gyrate in 1H 2026 as they assess the path of policy. A 7% drawdown would be normal and expected in the post-GFC era.

(We remain bullish looking three months out and expect all time highs. As for next week, we believe markets need some time to breathe and give back after an incredible rally.)  

The SPY-election study shows markets almost always rally after midterm elections.

Across the last century, the S&P 500 has averaged:

• +5.8% in the 3 months after midterms

• +10.5% in 6 months

• +14.8% in 12 months

These gains occurred even when the 12 months preceding the election saw large drawdowns. 

With an expanding economy, higher corporate earnings, and rate cuts on radar, this seasonal pattern adds another structural tailwind. 

The early market behavior is already tracking that script.

Digital Assets

All Markets On Chain

SEC Chair Paul Atkins forecasts “All U.S. markets will be on chain within two years.” 

Tokenization isn't just a futuristic idea anymore. It's becoming a rapidly accelerating reality, with enormous implications for market efficiency and transparency.

Paul Atkins believes placing securities on the blockchain will help remove opacity from the system. "There's much more transparency as to where it [a security] is." 

Atkins also highlighted blockchain’s potential for T+0 settlement, which can help reduce the risk associated with trade settlement - "The idea of on-chain delivery versus payment is a prospect of de-risking the markets.”

Atkins points out that in the past, “The SEC was always a little bit behind the market,” but this is changing. 

In fact, "We need to be embracing [new technology] to keep the United States at the forefront of cryptocurrencies and whatnot." 

Tokenization will drive the next phase of market evolution, and the SEC’s support will ensure that the U.S. remains at the cutting edge. 

The future of financial markets is on-chain, and it's coming fast.

The Founder Behind Polymarket

Shayne Coplan has had quite the founder’s trajectory: a college dropout building the platform in three months during COVID, and now leading a $9 billion company.

Lumida Curations

Bitcoin’s Four-Year Cycle Is Breaking Down

Michael Saylor argues the halving no longer drives Bitcoin’s behavior: structural credit, derivatives growth, and TradFi adoption now set the real market cycle.

We believe the top is in for Bitcoin and we won’t cross $125K this year. It’s competing with two many narratives, and the DATs are a terrible investment category for the forseeable future.

Rallies will be sold.

View Curation

AI Is On Track to Run at Near-Zero Energy Cost

Jensen Huang breaks down why accelerating compute efficiency is driving AI toward a future where it runs everywhere, all the time, on almost no power.

We are still many, many years away from Small Modular Reactors.

The fact that we have 80 nuclear submarines but can’t build an SMR tells you a lot about regulatory dysfunction and the challenges ahead.

View Curation

Alcohol Truly Delivers “Empty Calories”

Andrew Huberman explains why alcohol is metabolized as a toxin, offering no usable nutrients and imposing a heavy processing cost on the body.

View Curation

Meme

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: No Support For Doomers; Consumers Are Confident; Amazon’s Nightmare 

• Markets: The Bullish Trigger; Every Day Up…; Our Energy Pick for the AI Power Decade; Top Trade in Managed Care; Our Bet in Biotech; OpenAI Needs To Raise At $1Tr; Investment Tip: When Bubbles Burst

• Lumida Curations: AI Will Define This Generation; The Market Just Hit Reset; Protein is the Breakfast Metabolic Hack

Happy Thanksgiving.

I’m deeply grateful for the privilege to serve as a steward of your capital. 

Investing is an act of Trust. 

That trust is earned slowly and tested often, and we never take it for granted.

To invest on someone’s behalf is to shoulder the volatility, and carry the emotional load of markets so you don’t have to. 

It’s a duty we embrace. 

Our job is to stay steady when the world feels unsteady, to make disciplined decisions when the noise is loudest, and to protect your ability to pursue what matters most: 

family, purpose, and the freedom to choose your path.

This year reminded me that challenges arrive with a kind of quiet purpose. 

There’s a wisdom that only reveals itself through tension.

I’m grateful for those experiences because they carve out the resilience and perspective that ease alone could never teach.

I’m thankful for clients and followers who challenge our thinking. 

Curiosity keeps us honest. 

It reminds us that markets don’t reward ego — they reward learning. 

Every cycle gives us new data, new puzzles, and new opportunities to refine our process. 

Wealth is a tool for possibility. 

Our duty is to protect and expand those possibilities so you can stay focused on building a meaningful life. 

That responsibility is a privilege — and we treat it as such.

Wishing you and your families clarity, rest, and a moment to appreciate the progress — visible and invisible — that brought you here. 

Thank you for trusting us with work that truly matters.

I did a stream from my FSD on Thanksgiving night. Watch it here.

Keep More Of Your Capital Gains

Every spring, investors write one of their biggest checks of the year to the IRS. 

However, you can save your taxes and contribute to economic and social welfare instead.

Actions taken today can mitigate income tax owed to the IRS in the spring.

The solution is Solar Investment Tax Credit (ITC). It allows investors to offset federal taxes by financing qualified solar projects. 

In practical terms, an investor with a $100,000 tax bill could put roughly $60,000 into a qualified solar project and reduce their income tax liability by 40% to 45% if you live in a high-tax state. 

Corporates have been doing this for years; in 2024 alone, more than $20 billion in solar credits were traded, up four-fold from the year before.

We have a limited capacity on this opportunity; so if you have a substantial tax bill, now is the time to act.

Click here to know more about this and other tax mitigation opportunities. You can also book a free consultation call here, and learn how Lumida can help you save taxes. 

Macro

No Support For Doomers

This past Sunday, I did a podcast highlighting how there is significant value in stocks due to a pullback driven by the Fed hawkish cut and our “no recession” call. Watch the Non-consensus investing podcast on Youtube here.

(The fact that one has to make such a call and that it is amazingly non-consensus tells you we are nowhere near a market top.)

The economy is growing, and this week’s data releases continue to prove it.

• Black Friday sales hit a record $11.8Bn, up ~10% annually. 

• Q2 GDP growth was revised higher to 3.8%, with Q3 now tracking around 3.9%. 

• Initial jobless claims also came in lower than expectations, consistent with the “low-fire-low-hire” environment. 

To sum it up, we have a growing economy with spending consumers and confident businesses.

Take a step back. We have a boomer led economy. Boomers have record household balance sheets. Corporate and individual real incomes are up. We have transfer payments and robust federal spending. We have a services economy driven by healthcare. We have a massive capex boom adding 1% to GDP. (Swings in manufacturing just don’t matter).

It is very difficult to derail an economic boom given the above conditions.

Why are people still fearing a recession?

Some other data points…

Commercial real estate is showing signs of bottoming in various hard hit markets like Austin or San Francisco.

I will do my best to tell you when you should be concerned. Now is not that time.

Consumers Are Confident

Like Druckenmiller, our primary datasource for understanding the health of the real economy is studying the US consumer thru earnings transcripts.

We discussed banks in October, now we have data on retailers.

Headline: The US consumer is strong and spending.

Management teams across retailers describe shoppers as “choiceful,” “value-seeking,” and “engaged”. 

Target’s CCO Richard Gomez captures the mood: “Guests are choiceful, stretching budgets and prioritizing value. They're spending where it matters most.” (Note: Target’s numbers look dreary and we think the stock should be avoided.)

Costco’s CFO Gary Millerchip echoes that “members are spending towards necessities and remain very choiceful in their spending on discretionary items.” (Costco remains over-valued and is now going thru multiple compression).

Dollar Tree notes that “customers are seeking value and convenience more than ever.” (Dollar Tree’s numbers are impressive, we want to buy on the next pull back.)

The spending pattern beneath that language is a reordered wallet, not a shrinking one. 

Retailers have also shown optimism around consumers’ health, and guided towards a strong holiday season.

Walmart, Ross Stores, and Dollar tree also raised their guidance encouraged by the strong business momentum.

Walmart’s numbers and product innovation - including e-commerce - is very impressive. The multiple is now higher than Amazon and by more than a few turns (37x vs. 30x).

Amazon’s Nightmare: Walmart’s E-Commerce 

Walmart is scaling e-commerce as a profit engine, and markets are beginning to feel the difference. 

John Rainey (CFO, WMT): “our business model continues to evolve with operating income increasingly influenced by improved e-commerce economics.” 

Walmart had a remarkable 28% e-commerce growth, its seventh straight quarter above 20%, driven by pickup, delivery and ad monetization. 

Fulfillment speed is closing what used to be an Amazon-only moat:

“Approximately 35% of store-fulfilled orders were expedited or delivered in under three hours,” with sales through fast channels up nearly 70%. 

Combine that with market-share gains in grocery and general merchandise, and the readthrough is clear: 

Walmart is building Amazon’s flywheel on top of a value-anchored consumer base and a physical network Amazon doesn’t have.

It is hurting Amazon’s share, and is one of the reasons why Amazon’s lagging.

Amazon is also ceding capture of cloud market share growth to Google and Microsoft on the margin.

We aren’t in the ‘buy Amazon’ club because of this two prong flank attack given Amazon’s current valuation vs its peer group. Especially when you can buy Microsoft just off the 200 Day Moving Average.

Markets

The Bullish Trigger

Last week, we saw SPY drop below its 50D, but it recovered with 4 consecutive green days and 70% of NYSE issues rising. 

This is one of the strongest Thanksgivings in the market history books.

A lot of capital exited the market these last few weeks and are now falling over themselves to get back in. 

That’s why we have this string of gains. 

This triggered a bullish signal. 

Over the past 50 years, this pattern has coincided with powerful forward returns:

+12.8% at 6 months, +23.2% at 12 months, with a 10/10 positive hit rate over one-year periods.

With the economy expanding, businesses growing, and consumers confident, we might see markets play out similar to what the indicator implies. 

EVERY DAY UP…

We see many names and themes doing well now thru year end. 

A short and shallow pullback early this week given tactical overbought conditions would be expected. After that, best to deploy if you are not fully invested.

The best businesses in the world went on sale, and we were quick to scoop them. 

We see this across a variety of names. That last time we saw the best businesses in the world on sale was November 2023. They’ve now returned to reasonable valuations.

On the broad indices, I expect we’ll get to All Time Highs on or before year end. 

Rate cut probabilities are close to 90%

However, If the Fed does not cut, then the rally might stumble. 

It’s rare to find a variety of quality businesses on sale at the same time. 

When that happens, it’s best to scoop them up.

Our Energy Pick for the AI Power Decade

Natural gas is becoming the most important feedstock for U.S. computing power - the sector is experiencing demand like never before.

This week, we initiated a position in EQT Corporation (ticker: EQT).

(The stock is up 7% since we wrote about it on X last week, so maybe you wait for a pullback.)

EQT is the second largest natural gas producer in North America and has midstream pipelines. EQT has 20 year datacenter contracts - this is a neat way to bet on datacenters at a much more reasonable valuation.

Tn EQT’s own words, demand from the region is already “greater than what we can currently provide.”

At the same time, EQT has also signed LNG contracts with Sempra, NextDecade Rio Grande, and Commonwealth to build export optionality.

From a numbers perspective, the engine is humming:

EQT’s revenues and Net income increased by ~30% YoY - it had record low per-unit costs in Q3, and generated over ~$2.3B in free cash flow over the last four quarters, taking FCF yield to 7%. 

Olympus Energy, acquired mid-year, was integrated in 34 days, the fastest operational onboarding in company history. This shows the execution culture.

The most important risk is commodity price sensitivity: 

Even with long-term contracts, EQT’s earnings still move with Henry Hub natural gas pricing, and a prolonged gas oversupply could compress margins and delay capital returns.

EQT also has over $8Bn in debt, so a decline in cash flows could hurt their ability to cover repayments. 

Top Trade in Managed Care

Healthcare stocks were back in business this week thanks to reports around ACA subsidies extension. 

We bought Molina Healthcare (MOH) this week because the stars are aligning: 

The political risk around subsidies is fading, valuation is depressed, and the long-term structural runway remains intact.

Politicians will continue to make the irresponsible decision and spend, spend, spend. And Molina Health is already priced for spending cuts, despite a mismatch in political rhetoric.

Molina is a leading managed care organization that provides healthcare services primarily through government-sponsored programs. It operates in 19 states and serves approximately 5.7 million members.

Molina had been priced for disaster, losing over 50% in the last 1 year.

It now trades at roughly 8× 2Y Fwd P/E, which is amongst the lowest levels in its 3Y History.

It’s hard for us to imagine this business trading at 8x PE in two years. Shouldn’t it price at least at 12x PE? That would imply a 50% return over the next two years.

And healthcare stocks have rallied sharply, the backdrop is solid.

The company delivers a high return on equity (~20%) and has leaned into aggressive buybacks. 

Shareholder yield is now at 20% - among the highest in the sector. The market is acting as if this is going to get cut in half. That’s hard for us to imagine.

The core Medicaid book remains disciplined; the company is continuing to secure state wins, including a new Medicaid contract in Florida that is expected to bring substantial premium volume over time. 

The bear case is that Molina saw a significant spike in Medical loss ratios, jumping to ~93% from 89% an year earlier. Now, this is true for plenty of other insurers that have subsequently rallied off of the May lows.

Management has also cut full-year EPS guidance for the third time, from an initial ~$22 to about $14 now. We think the worst is priced in.

We see a number of hedge funds scooping up the name in the recent 13Fs.

They all bought way too early. The stock continued to drop. We think now is a reasonable time to buy. (It’s satisfying when the other hedge funds did the work and you can get in at a better price. :)

We are getting in at about a 25% lesser price than the hedge funds.

We like being on the same side of the table, just at a meaningfully better entry point.

Incyte (INCY): Our Bet in Biotech 

INCY presented a good opportunity this week, and we took it.

Incyte sits at the intersection of proven drug economics and pipeline-driven upside. 

The business started its story around Jakafi - the dominant JAK inhibitor for myelofibrosis and polycythemia vera. It continues to be Incyte’s cash-engine at scale, and has funded the company’s pipeline for a decade.

But Incyte isn’t a one-drug story anymore. 

Over the last two years, INCY has built a second growth pillar in immunology and dermatology.

Opzelura, a topical formulation with expanding label reach, is emerging as the next flagship asset, with revenue up ~35% YoY last quarter. 

Beyond it, the company is stacking multiple new oncology and rare-disease products: Minjuvi/Monjuvi +33.6% YoY, Iclusig +26.4% YoY, Pemazyre +10.1% YoY, and even smaller assets like Zynyz are inflecting off a tiny base (+3167% YoY).

Moreover, you rarely find a biotech with this level of balance-sheet quality: 

~56%+ gross margins, ~25% EBIT margin, ~30% ROE, strong free cash flow, and effectively no debt. 

The business funds R&D internally, returns capital through buybacks, and isn’t dependent on capital markets to survive. 

Valuation is not stretched either: ~13.8x forward P/E and ~6–7% FCF yield, putting it mid-pack for healthcare but with better profitability than most peers.

The overhang, and the reason this isn’t priced like a growth compounder, is clear: Jakafi loses exclusivity around 2028. 

Everything between now and then is a race to diversify revenue and prove that Opzelura + the oncology pipeline can absorb the earnings gap. 

Management has shown strong execution, and we believe markets aren’t pricing INCY for its potential (YET!).

Note: Biotech investing is extremely difficult. We would not allocate more than 1 to 2% to the name.

AI Adoption is Driving Mainstream Productivity Growth 

AI implementation is now the norm across industries. Companies are reporting efficiency gains, cost take-out, and dollar-linked ROI from AI rollouts. 

Banks and insurers are the furthest along. 

JP Morgan CFO Jeremy Barnum said AI is already improving “fraud resolution times and analyst throughput”. This represents a “structural efficiency unlock rather than a cyclical benefit.”

Insurance is echoing similar efficiencies. 

Progressive CEO Tricia Griffith noted that AI-supported claims handling is “materially improving adjudication speed while reducing manual intervention,” enabling volume growth without equal headcount growth.

Retail and logistics are embracing AI far faster than consensus assumes.

Walmart CEO Doug McMillon said AI-driven inventory and replenishment tools are helping reduce stockouts, improve accuracy, and enhance e-commerce flow.

Target CIO Brett Craig highlighted that AI-led routing is reducing logistics miles and increasing routing efficiency, translating directly into cost-per-unit gains.

Healthcare, historically slow to adopt technology, is now discovering throughput leverage.

UnitedHealth COO Dirk McMahon stated that automation is “removing manual claims friction and speeding intake workflows,” allowing more throughput without additional labor. 

Healthcare is 18% of the U.S. GDP, even marginal productivity gains here ripple through inflation, wages, and national cost curves.

Overall, this enterprise adoption of AI will only drive the demand higher, and help the AI productivity growth story continue.

We have previously written how AI is driving returns for hyperscalers - Read it here.

OpenAI Doom Loop? OpenAI Needs To Raise At $1Tr

HSBC estimates that OpenAI could need $207 billion in financing by 2030 to meet all of its compute capacity commitments.

If OpenAI needs $200 Bn+ in funding, here are their choices:

1) Cash from Operations (CFO): $12 Bn going to [ $50 to $100Bn if you are bullish ]

Not enough, and that’s assuming pure margin. 

2) Cash from Debt Financing:

They would need a debt to equity structure like CoreWeave, except CRWV can defend its liabilities with term revenue contracts but OpenAI cannot.

3) Cash from Equity Financing aka Shareholder Dilution. 

OpenAI needs to raise at a $1 Tn valuation. 

The stock would be down 30% to 40% easy in the last two weeks if it was publicly traded.

FOMO is the enemy of returns. Size is the enemy of returns. Capex is the enemy of returns.

Lumida Curations

AI Will Define This Generation

Sundar Pichai says Those who learn to harness AI will shape the future, unlocking new creative and economic opportunity.

View Curation

The Market Just Hit Reset

After fear-driven volatility and a VIX spike, conditions now resemble the start of a rally not the end of one.

View Curation

Protein is the Breakfast Metabolic Hack

A high-protein morning switches on satiety and calorie control, while sugar sends appetite in the opposite direction.

View Curation

Meme

Not subscribed yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: The U.S. Economy Keeps Defying the Bears; Rate Cut Odds Soar Higher; How was the Labor Market Data?

• Markets: The Dot Com Flashback; Time To Buy Quality; Fear Is Maxed Out; Software Vs AI Update; Opportunity In Panic; Coreweave; Positioning For An Oil Rebound?; The Return Of Animal Spirits?

• Digital Assets: Bitcoin’s Free Fall

• Lumida Curations: Future of Work; Silicon Valley’s Blind Spot on Defense; The 2–4 AM Wake-Up Call

Lumida hosted a holiday party this Tue evening in NYC.

Fun fact: the event was hosted by a member of our telegram group; we had not met before. He found us through the DCG Genesis debacle, and invested in Coreweave Pre-IPO and made money on some of our stock picks.

Thanks David!

I did a podcast from my FSD on our way back from the event. View it here.

This is the Season to Mitigate Tax

Every spring, investors write one of their biggest checks of the year to the IRS. 

However, you can save your taxes and contribute to economic and social welfare instead.

Actions taken today can mitigate income tax owed to the IRS in the spring.

The solution is Solar Investment Tax Credit (ITC). It allows investors to offset federal taxes by financing qualified solar projects. 

In practical terms, an investor with a $100,000 tax bill could put roughly $60,000 into a qualified solar project and reduce their tax liability by nearly 40%. 

Corporates have been doing this for years; in 2024 alone, more than $20 billion in solar credits were traded, up four-fold from the year before.

We have a limited capacity on this opportunity; so if you have a substantial tax bill, now is the time to act! 

Go to our website at www.lumidawealth.com and click ‘Learn More’ to learn about this and other tax mitigation opportunities.

We also offer tax mitigation strategies for our wealth management clients. 

Macro

The U.S. Economy Keeps Defying the Bears

Earnings season just confirmed how resilient the U.S. economy is. 

Over the last three months, more than 70% of companies beat expectations on both earnings and revenues, versus long-term averages closer to the low 60s. 

The share of companies raising guidance is sitting near the top of its historical range. 

Real GDP has been growing around 4% annualized in Q2 and Q3.

Productivity has also ticked upwards thanks to AI and automation.

Overall, these statistics suggest the economy isn’t on the brink of recession as some headlines might suggest; it is the backdrop of an economy that can keep surprising to the upside.

How was the Labor Market Data?

Overall, this week’s labor market releases painted a mixed picture.

September payrolls rose by about 120,000 versus expectations of 50,000. 

However, the unemployment rate has ticked up to the mid-4s due to increasing labor force participation, which rose to 62.4%. 

There is undoubtedly anxiety in labor markets. That’s not enough to cause a recession.

Take a step back: we have a baby boomer economy, led by services, and funded by transfer payments, and significant capex and federal spending.

There’s no recession coming folks.

Interestingly, this uptick in participation has coincided with a decline in crypto markets. 

Under the surface, the story is “low fires, low hires.” Initial jobless claims remain low and within the post-covid range.

However, continuing claims rose by about 5% YoY, which means it is taking longer to land a new job, especially for younger workers and recent college grads. 

We have observed a similar pattern across the last few months. 

Companies are retaining workers, and implementing AI, which reflects in higher output without increasing hours worked.

This indicates a structural change in the labor market, which rate cuts aren’t likely to fix.

Rate Cut Odds Soar Higher

NY Fed John Williams’ comments on Friday raised the rate cut odds back to 70% from 33% a day earlier. 

The S&P surged on Friday after these comments.

The Dec Fed rate cut probability is the most crucial to watch.

Governor Williams: “My assessment is that the downside risks to employment have increased as the labor market has cooled, while the upside risks to inflation have lessened.” 

“Therefore, I still see room for a further adjustment in the near term to the target range.”

Our view remains that the economy is growing and doesn’t need rate cuts. But, rate cuts are coming so we will position accordingly.

Markets

The Dot Com Flashback

Small-cap value led the tape on Friday, while large cap growth names were on the losing end. 

We can draw a parallel here from the post-dot-com bubble era. 

In the 5 years following the 2000s bubble burst, tech indices lost almost 30% of their market cap, while small-cap value names compounded close to 90%. 

In 2025, markets had again been discounting Ziff Davis - it lost 44% in the last 1 year, while SPY gained 13%. 

However, in the rotational move on Friday, Ziff Davis received the love and rose 7.7%. 

Ziff Davis is a $1.3B digital media and tech company that now trades around 4.3x P/E NTM with a FCF yield at 21%. 

The latest quarter was solid.

Revenue grew nearly 3% YoY to $364M, marking the fifth consecutive quarter of growth, with an adjusted EBITDA margin a little above 34%. 

Adjusted EPS rose 7% to $1.76, helped by buybacks that retired roughly 3.3M shares. 

Free cash flow was $71M, up 18% YoY, and trailing-twelve-month FCF is about $261M, which management is recycling into what they believe is ROIC-productive M&A and repurchases.

The bear case? 

Roughly 60% of revenue is still tied to advertising, gaming & entertainment, which declined by about 4% YoY. This revenue stream is highly cyclical. 

Nonetheless, the mix is improving: three of the five segments are growing.

Health & wellness is up 13% YoY. Cybersecurity, and performance marketing are all positive as well. 

We own some of this in small cap value.

The Revenge of Warren Buffett

In February, I wrote a newsletter called ‘The Bubble in Quality’. We noted how stocks like Costco and other high quality businesses (FICO, etc.) were sporting bubble valuations.

Costco’s PE went from 50x to 45x since then. I still wouldn’t own it.

But, other quality names have sold off quite a bit and may present attractive entries. They aren’t bargain basement prices. But, are cheap historically. 

Buffett is back in the market and has picked up quite a bit of Google - one of our top 3 positions.

As Animal Spirits fade, we believe a rotation into its opposite - quality and value may be in order.

We got two “quality” Buffett type stocks this week: Intercontinental Exchange (ICE), and S&P Global (SPGI).

ICE is a dominant market leader in derivatives and clearing. They own the NYSE and operate across asset classes.

The stock is down 17% from its high in July. P/E NTM has also compressed from 26.0x to 21.3x today. 

In Q3, revenue was about $2.4B (+3% YoY), adjusted operating income grew 3% to $1.6B, and adjusted EPS hit $1.71 (+11% YoY). 

ICE also bought Black Knight - a mortgage analytics software firm. The acquisition enhances ICE's competitive edge in Mortgage Technology, giving it over 50% market share in mortgage origination and servicing software.

We also like ICE’s CEO - Jeff Sprecher. It’s a Founder led business.

ICE has paid roughly 2Bn back to its shareholders in the last twelve months, with Shareholder Yield at 5%.

The bear case: you are paying a 28x P/E TTM multiple for a financial stock that grew revenues by 3% YoY in Q3. 

However, management has guided towards a 7% increase in Q4 thanks to the strong performance within the exchange-aided business.

This business I put in the category of ‘quality compounder’. If you hold for 5 years, I expect it would compound well.

S&P Global

SPGI is another of our high quality picks this week.  The firm makes money from licensing its indices to asset managers like BlackRock. They also run a ratings business.

These businesses will be around in 50 years - that’s a hallmark of “quality”.

It trades at roughly 26x forward earnings.

In Q3 2025, revenue grew 9% YoY to $2.9B, operating margin expanded 330 bps to 52.1%, and adjusted EPS jumped 22% to $4.57. 

Management has raised guidance on all metrics for Q4, and EPS is now expected to grow by ~14%. 

Management has returned nearly $1.5B to shareholders year-to-date with Shareholder Yield of ~4%.

SPGI has a pristine business model, but the valuation is pricier than others. But, take a look at the trailing valuation multiple - you can see we are at levels that are typically met with institutions buying:

The bear case is that a prolonged slump in primary markets would hit Ratings and asset-linked fees.

Nonetheless, in Q3 earnings, management remarked 8% higher rating volumes expectations in 2026, and the same trend to continue till 2028.

Fear Is Maxed Out

Inversions in the VIX term structure have pushed the 3-month/1-month VIX ratio into the “fear zone.” 

It is the level that has historically lined up with short-term market lows and subsequent bounces. 

A recovery might see capital rotating into oversold themes. 

We had been running our net exposures at around 70 to 80% the last few weeks. We are close to 100% net exposure as of this Friday.

One theme we like around here is consumer discretionary.

Take a look at the chart. Looks poised to bounce around the 200 DMA.

Airline stocks, cruise lines and other travel & leisure that baby boomers spend on are the right focus in our view.

If you believe our ‘no recession’ thesis then these will do well.

The sector dipped significantly over the last few weeks, and is now trading at oversold levels. 

We added to our Norwegian Cruise Lines position this Friday. We think it has bottomed.

Have a look at our thesis on NCLH from last week. 

SOFTWARE VS AI UPDATE

I’ve been bearish and correct on software stocks since 2024. 

The bear case is that AI is displacing software form factors. 

Do we really need Salesforce? Adobe?

Corporates are allocating capex to AI POCs instead. 

But, I am seeing at least two reasons indicating why the selling in software may be overdone. 

This may be early, but it’s worth ‘stalking’. 

Here are ideas.

1. Some SaaS stocks are actually indexed to AI growth in a “picks-and-shovels” way. They sold off with the broader theme. 

2. People that say ‘valuations don’t matter’ are wrong. Both the software and restaurants call - two themes in a bear market in an otherwise strong bull market - were valuation driven. 

Now is the time to look for themes with solid fundamentals that investors have ignored.

We added exposure to Atlassian Corp (TEAM) this week.

TEAM hasn’t really caught a break since its high in February, and is now available at a deep discount.

The P/E NTM has dropped to 29.0x from 84x in Feb - the lowest in its 10Y history. The FCF yield is at approximately 4%.

Revenue is growing ~20% with >80% gross margins. 

Management is expanding the enterprise portfolio with Rovo AI while keeping a clean balance sheet and healthy FCF.

The stock has been de-rated on the “AI code-gen kills Jira” narrative, but management answers the risk in Q3 earnings. 

CEO Mike Cannon-Brooks noted they are “not seeing any impact from this [AI coding] on any numbers” and that customers integrating code-generation tools show no change in usage and “really healthy user growth.” 

This supports the view that Jira and the broader Atlassian suite remain embedded in developer workflows even as AI adoption rises.

Atlassian corp has a 3.2Bn Backlog in Q3, up 42% YoY. 

The “>10K“ customer base has increased 13% to 53,000+ customers. 

Over 85% of Fortune 500 companies use products from the Atlassian suite. 

Overall, it’s a quality business, but there are some visible risks. 

The EBIT margins have depressed as TEAM invests further into AI.

TEAM also failed to produce a GAAP positive bottom line in Q1’26 due to high stock based compensation (349M- 1% market cap).

Nvidia Earnings and the Datacenter Theme

I have a fully video dedicated to breaking down Nvidia’s earnings and the datacenter theme here if you want a full treatment.

The data-center trade has been under pressure for the last few weeks. 

It started with Coreweave missing out on delivery schedule and Micheal Burry’s comments around GPU deprecation. 

Michael Burry argues that hyperscalers are inflating profits by stretching GPU lives from roughly 3 years to 5–6 years. 

If his argument is correct, It understates hyperscalers’ depreciation expense by ~$170B+ across 2026–28, inflating their capex returns. 

However, Burry is wrong here. 

Jensen Huang answered Burry’s argument in Nvidia’s latest earnings: “The A100 GPUs we shipped six years ago are still running at full utilization today, powered by a much stronger software stack.” 

Similarly, Google has also been running datacenters on CPUs for over 10 years now, and they are still operational.

This is hard evidence on the actual useful life of a GPU. 

The demand side for datacenters is also durable. 

AI Capex has been increasing rapidly, and the long-awaited Nvidia earnings show it. 

Nvidia’s data-center revenue grew at 60% YoY off a base that already exceeds $50 billion.

In this week’s earnings, Jensen Huang said “Demand for AI infrastructure continues to exceed our expectations."

He estimates “$3 trillion to $4 trillion in annual AI infrastructure build by the end of the decade.”

This capex is also translating into higher revenues and profitability for hyperscalers, which justifies continued investment in AI. 

Jensen Huang highlighted Meta’s revenue conversion from AI in Nvidia’s earnings release.

“Meta reported over a 5% increase in ad conversions on Instagram and 3% gain on Facebook feed driven by generative AI-based GEM.”

Nonetheless, we might see datacenter theme lag. Technically the picture around semiconductors shows evidence of over-positioning.

COREWEAVE

AI Hedge Fund, Situational Awareness, loaded up on CoreWeave.

So did Paul Tudor Jones, Citadel and DE Shaw.

CoreWeave had a final unlock last week.

The stock hits a local low after every unlock

The same happened this time around. 

The oversold RSI levels suggest an intermediate low has already occurred.

At a P/S of 3.6x with strong revenue growth, it's hard to see how this isn't up significantly in 12 months.

We had a deep dive on CRWV in our last newsletter. Read here.

Positioning For An Oil Rebound?

Oil is trading near one of its cheapest points in history when measured against money supply. 

Each time we’ve been here, investors have argued “this time is different” and that prices can stay depressed.

Historically, that narrative has not held.

One of our bets in the oil industry is Riley Exploration Permian (REPX). It trades on a P/E NTM of 8x, with a FCF yield of 17%. 

Q3 oil production grew more than 30% and upstream free cash flow was roughly $39 million on $82 million of revenue. 

Management is integrating the Silverback acquisition, drilling more wells, and upgrading midstream capacity. It would support production volume growth and a 5–6% dividend even if oil stays range-bound.

The risk? 

REPX has high sensitivity to oil price volatility, which has led to muted revenue growth and earnings de-growth over the last few quarters. 

If the downturn in crude prices prolongs, it would pressure earnings and cash flows, hurting the stock’s long term returns.

The Return Of Animal Spirits?

The 'Liquidity Factor' measures the performance of stocks with the highest turnover.

Notice how this factor continues to dip lower.

There has been a forced deleveraging amongst traders that own this basket.

And, the factor is now oversold.

Some of you may be thinking that this or that stock is down because of some news (e.g., ROE on GPUs, GLP1 innovation, credit risk, etc.)

No, it's not that.

It's just deleveraging across a correlated basket of stocks that people on Twitter own in unison.

eth

IonQ is a good example: the stock rose 300% between March to Oct, but has declined about 50% since its Oct high, returning to its June Levels.

The stock appears to have topped after JP Morgan helped them sell $1 Bn of stock to foolish faddish trend buyers.

We think the top is in for quantum stocks and any rallies will be sold. Better to focus elsewhere.

SoFi shows a similar pattern- it had a big run earlier in the year, but has declined by 21% since Nov high.

QBTS is no different - it has dipped by over 55% since its high in Oct. 

We believe this rotation will continue from high beta names to quality compounders and value in near-term. 

Get Lumida CoPilot on Your Apple Phone

Lumida is in the final stages of launching our app - our app is your copilot in the investment world. It uses advanced AI models to help you pick top stocks, themes and sectors.

If you’d like to be the amongst first few external testers., write to us on our X account or go to our website and click ‘Learn More’.

Digital Assets

Bitcoin’s Free Fall

Bitcoin has now dropped to 83K - its lowest since April.

Here’s what we wrote about it last week, and it has turned out accurate so far. 

High-momentum “attention assets” have all been hit at once: Bitcoin, gold, small-caps, quantum names, and other story stocks. 

These trades went up together as investors crowded into the same high-beta narratives; they are now coming down together as leverage is forced out of the system.

Under the surface, this unwind is being driven by:

• Fed’s pushback on a December cut.

• Hangover from the DAT IPO frenzy, most of which are now underwater or back to fair value.

• OG crypto holders rotating out of long-held positions after a strong run.

Is Bitcoin oversold? Yes. Can it rally? Yes. Can it drop? Yes.

Why not focus on easier ponds to fish in?

You should read our view on attention assets from early November- it’ll help you interpret the broader market context.

Lumida Curations

Future of Work: Musk’s Vision vs. Bessent’s Reality

Elon Musk says AI could make work optional and money less relevant, but Scott Bessent argues history shows productivity booms create new roles rather than erase them.

View Curation

Silicon Valley’s Blind Spot on Defense

For decades, the world’s top tech hub treated defense as a moral hazard, ignoring the one technology mission that ultimately determines national power.

View Curation

The 2–4 AM Wake-Up Call

If you regularly wake between 2 and 4 AM feeling wired, it may be your liver flagging hidden stress, poor nutrition, and an overloaded metabolism.

View Curation

Meme

Not subscribed yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: Consumer Credit Is Stable and Improving; Loan Demand Is Strong; AI-Driven Underwriting and Calibration Are Material Levers; Cruise demand is resilient and experience-led; Challenger’s Report Was Dramatic

• Markets: Restaurants Remain Bearish; The Crypto Pullback Is an Attention Reset; Hyperscalers Ignite a Memory Supercycle; Samsung Eyes U.S. Card Market

• Digital Assets: The Fed: Stablecoins Move Markets; Tokenization Is the Bridge Between Worlds; Banks Vs. Non-Banks

• Lumida Curations: Banks Quietly Reenter the Crypto Arena; AI Enters Its Goldilocks Phase; Olive Oil Supercharges Your Fast

Spotlight

This week, I attended the Asset on Blockchain (AOB) conference at the Harvard club. It was exciting to meet SEC chair Paul Atkins - a fan of his leadership.

We need interpretive guidance (not rule by enforcement), an industry SRO, and Congressional action.

I wrote an op-ed in the WSJ with Arthur Levitt - former SEC chairman- last fall laying out a framework for responsible crypto regulation. I read it again today, and it still makes sense. You can read it here.

I did a livestream on the conference’s highlights - you can watch it here. 

DNC Sweep - What to Make of It?

Markets sold off this week as the same time as a DNC sweep took place in various gubernatorial and city elections this week.

These counter-trend elections following a Presidential election are typical. Still, the margins combined with high turnout contain signal.

We believe the read-thrus are:

1) Consumers want ‘stability’ not fast-action cockpit fighter televised diplomacy from week to week. The electorate just can’t process that much information and it’s overwhelming.

2) Inflation remains an election issue. Trump won on bringing down grocery prices and immigration. The year-to-date focus has shown a shift on international politics and tariffs.

Democrats are starting to own the inflation message now.

Going into next year, we’d expect both parties to continue to err on the side of stimulative monetary and fiscal policy as we start look towards the midterms

That’s bullish.

Also, stock prices rise with earnings. Markets, we believe, have now priced in the DNC sweep and we suggest being fully invested as we approach year-end.

We’d expect record retail sales ahead to close out the year.

How To Reduce Your Tax Liability

Every spring, investors write one of their biggest checks of the year to the IRS. 

However, you can save your taxes and contribute to social welfare instead.

The solution is Solar Investment Tax Credit (ITC). It allows investors to offset federal taxes by financing qualified solar projects. 

The credit covers 30% of a project’s cost, with potential “bonus” credits for systems built with U.S. materials or located in low-income or energy-transition communities.

In practical terms, an investor with a $100,000 tax bill could put roughly $60,000 into a qualified solar project and reduce their tax liability by nearly 40%, while also earning passive income from power generation. 

Corporates have been doing this for years; in 2024 alone, more than $20 billion in solar credits were traded, up four-fold from the year before.

Lumida has been actively exploring tax mitigation strategies to help our investors reduce their tax liabilities.

If you’d like to be on our priority list, sign up using the form here. If you’d like more details on how Lumida can help you reduce your tax liability, write to raymond@lumida.com.

We are hosting Lumida Holiday 2025 on 3rd Dec - the event brings together founders, family offices, and investors to share ideas, discuss opportunities, and have fun. We would love to host you. Please sign up here.

Earnings Highlight: Consumer Finance

We turn our focus to consumer finance companies this week. Their earnings offer great read throughs on consumers’ health, lending patterns, and their confidence in the economy.

Consumer Credit Is Stable and Improving

The U.S. consumer remains far healthier than the headline narrative suggests. 

Each week, our team analyzes the earnings calls across a variety of sectors to glean insight into the economy.

The performance data from LendingClub, SoFi, and Upstart show a picture of consumers meeting their commitments and even declining delinquencies.

Credit risk is normalizing, and the economy remains stable.

Loan Demand Is Strong

Discover’s CEO Michael Shepherd noted: “Card spending growth remained positive and personal loan demand was strong as consumers continue to manage household balance sheets proactively.”

These signals show consumers are confident in the economy, and they are willing to take on more loans, helping demand growth.

AI-Driven Underwriting and Calibration Are Material Levers

AI is quietly reshaping consumer finance from the inside out. 

Operators, from SoFi to Upstart, emphasized how data science, model calibration, and real-time cash-flow analytics are becoming the new edge in lending.

If you are still not convinced with AI’s potential, you should also read our last newsletter, where we discuss how hyperscalers are using AI to boost their returns.

Baby Boomers: Cruise Demand Is Resilient And Experience-Led

Cruise line earnings confirm the pattern visible across the economy: consumers are confident, and they keep spending - No recession in sight!

Consumers are spending on travel, and their preferences continue to focus on toward experiences. 

Jason Liberty, CEO of Royal Caribbean, noted: “Consumers continue to prioritize experiences and make room in their budgets for meaningful vacations… Roughly three-fourths of consumers intend to spend the same or more on vacations over the next 12 months.” 

Demand remains historically strong. 

Harry Sommer, CEO of Norwegian Cruise Line, said: “Bookings… marked the strongest third quarter in company history, with bookings up over 20% from last year,” adding that intent is “at or near record levels.” 

This strength is pulling bookings further forward.

Josh Weinstein, CEO Carnival Corporation, highlighted: “Booking trends have continued to improve, nicely outpacing capacity growth at higher prices… with nearly half of 2026 already on the books at higher prices.”

Restaurants Remain Bearish

CAVA reported Q3 earnings this week. Revenue rose ~20% YoY to ~$289.8M. EPS was $0.12, about 1c below consensus and slightly down vs last year’s $0.13.

Management trimmed FY25 revenue guidance to 3–4% increase, causing a 4% stock decline. 

CAVA’s transcripts offer insights on what’s working for restaurants (and what is not!). 

Brett Schulman (Cava CEO): “Consumers today face a challenging environment; we’re not seeing anything geographically to call out… it’s more around the macro environment and the pressure on the consumer.”

Overall, Full-service restaurants remain in a bearish market. The dining basket (EAT, CAKE, TXRH, DRI, etc.) slid to new lows during Q3 earnings.

Challenger’s Report Was Dramatic

October saw 153,074 announced job cuts, concentrated in technology and warehousing.

These are industries, where automation and artificial intelligence are boosting output per worker and reducing labor demand. 

We expect a number of those impacted in the tech industry will go on to start or join AI companies to drive a deepening of AI adoption across the economy.

The private payrolls this week showed 42,000 new jobs in October, following small declines in the prior two months. 

While this number looks small, it is exactly where payrolls should be to keep the unemployment rate stable.

St. Louis Fed notes slower immigration and lower labor-force participation mean “the economy only needs to add ~50,000 jobs per month to keep unemployment stable… 75–100k would be meaningful positive news.”

Similarly, the Dallas Fed estimates the economy needs only 34,000 new jobs to maintain stability in the unemployment rate.

We have been talking about this for several months now; labor supply constraints are structural. Immigration has slowed, and labor force participation remains below pre-pandemic levels, especially among younger cohorts and working-age men. 

This dynamic is reshaping the labor market: AI is replacing some forms of work, while boosting overall output. 

Productivity is rising, and hiring needs are flattening. Meanwhile, earnings continue to march higher.

The labor market feels “soft” on headline job numbers but remains tight in supply - a mix that supports disinflation without signaling recession.

Markets

The Crypto Pullback Is an Attention Reset

The recent crypto drawdown is part of a broader unwind in high-momentum “attention assets.”

Bitcoin’s recent dip has moved in lockstep with weakness across gold, small-cap equities, uranium, and quantum-computing names like Rigetti. 

These trades have risen and fallen together this year as investors crowd into the same high-beta exposures that dominate narrative cycles.

The root of the selloff traces back to the following factors:

• Fed’s October meeting where Powell walked back prospects for a December cut.

  • Some concerns on the liquidity in the Treasury General Account (TGA).

• Government shutdown that continues to extend and is a form of fiscal tightening

• Hangover effect from the DAT IPO frenzy — many of which are now underwater or close to fair value (the exception is Canton Network…)

• OGs rotating from long-held positions (see chart)

Our view is that it will be difficult for Bitcoin to break $125K this year.

We would need a surprise shock that the market is not anticipating, and most certainly would need December rate cuts or shadow appointment of a Dovish Fed leader.

This isn’t a structural crash as some articles might suggest. It's an attention rotation. Capital is simply cycling out of over-owned trades after a strong run. 

Macro conditions remain constructive, regulatory momentum is improving, and institutional interest continues to deepen beneath the surface.

The shakeout looks less like the end of a cycle and more like a pause for breath in a crowded trade.

Datacenter and Chips Demand Remain Strong

Sandisk (SNKD) delivered a triple play for Q3; their results show the industry is seeing a sharp surge in demand from hyperscalers.

CEO David Goeckeler: “Demand for our NAND products continued to outpace our supply… customers are proactively seeking long-term commitments given the critical nature of our technology.” 

Data center revenue surged 26% YoY. “We’re now working with five major hyperscale customers through active sales and strategic engagements.”

CFO Luis Visoso added that “our products are currently on allocation across all end markets.”

Notably, Micron and SK Hynix and other memory stocks were in the dumps for the better part of 2024 and earlier this year.

We believe investors should wait for a pullback here, not chase memory stocks.

A more interesting area to focus on would be Meta which we believe presents an attractive valuation entry. Meta is gradually becoming our largest holding. 

Samsung Eyes U.S. Card Market

Samsung is making a serious push into consumer finance. The company is in advanced talks with Barclays to launch a U.S. credit card, with Visa likely to serve as the payment network. 

The move signals Samsung’s ambition to go beyond devices, and build an ecosystem that ties together payments, savings, and loyalty much like their archrival Apple has.

Samsung’s internal documents describe a vision where a Samsung credit card earns cash back that flows into a Samsung Cash account, then a Samsung savings account, creating a seamless loop between spending and saving. 

The company is also exploring high-yield savings, digital prepaid accounts, and buy-now-pay-later products - all designed to make its digital wallet indispensable.

For Barclays, the partnership represents a chance to rebuild its co-branded credit-card portfolio after losing American Airlines. The bank gains access to Samsung’s vast customer base - over 71 million active mobile users and 97 million customers across its U.S. businesses.

The race between Apple and Samsung now extends from smartphones to financial ecosystems.

Digital Assets

The Fed: Stablecoins Move Markets

Fed Governor Stephen Miran called stablecoins “a force to be reckoned with”, and agreed their rise could push interest rates lower. 

“This innovation [stablecoins] has been unfairly treated as a pariah by some, but stablecoins are now an established and fast-growing part of the financial landscape.”

Miran said stablecoins are expanding the “net supply of loanable funds”. 

When global investors hold dollar-backed tokens, they’re effectively financing the U.S. Treasuries and bank reserves. 

He argues extra liquidity “puts downward pressure on r*, the neutral rate of interest,”.

If r* falls, the Fed’s baseline policy rate will follow. 

We believe it’s way to early to draw these conclusions.

Further, issuance has shifted to the front-end of the curve where the Fed sets rates. And, stablecoins are funding on the front-end of the curve.

So, how exactly are stablecoins pushing down R*?

Further, we believe R* is a flawed economic concept that assumes a fairy tale equlibrium rate. Reality is complicated by concepts such as Animal Spirits as we’ve written about before.

Tokenization Is the Bridge Between Worlds

Tokenization is emerging as the connective tissue between the physical and digital economies.

AI and robotics digitize how the world operates. Tokenization digitizes how the world owns. 

Together, they create a feedback loop: AI builds smarter real assets; tokenization lets those assets trade, fractionalize, and move at internet speed.

The implications are massive. 

Real-world assets, from energy infrastructure to real estate and data centers, are being pulled onto blockchain rails.

Tokenization doesn’t just make them tradable; it makes them programmable, composable, and eventually integrated with AI-driven financial systems.

Ribbit Capital explored this concept in their 2025 token letter; you can read it here.

BANKS VS. NON-BANKS

Since the GFC, banks continue to retreat from "owning the customer" and directly servicing them.

On the payments side, firms like Stripe are acquiring merchants and then working with banks on the back-end.

On the lending side, private credit firms acquire and service customers. Then a bank on the back-end provides the "lending to the lender" back leverage.

Now, SuperApps and neo-banks (also built on top of infra banks) are increasingly owning the customer.

Hard to foresee how anything will change this trend; and it takes risk out of the banking system.

Tokenization should accelerate this and a handful of banks - those that can gather deposits and source lending - should do quite well.

Lumida Curations

Banks Quietly Reenter the Crypto Arena

Michael Saylor says new regulatory clarity from the Fed, OCC, and FDIC is ushering traditional banks back into digital assets, signaling a fresh wave of institutional adoption.

View Curation

AI Enters Its Goldilocks Phase

The AGI hype is cooling as AI settles into a balanced phase: steady, polytheistic progress driving real productivity rather than existential panic.

View Curation

Olive Oil Supercharges Your Fast

Fasting isn’t about eating less. It’s about sending the right metabolic signals, and olive oil flips your body from storing energy to repairing itself.

View Curation

Meme

Not subscribed yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Markets: Big Tech Is Spending More Than Ever on AI; AI Capex Is Paying Back; Bubble Talk Is Lazy Analysis; Open AI To Go Public; AI & Productivity: Carmax (?!); Power Is The New Bottleneck

• Macro: Automobile Sales Show Consumer Strength; The Hawkish Rate Cut

• Digital Assets: The Future Is Hybrid Finance; Visa’s Stablecoin Superhighway; Flows From Gold To BTC?

• Lumida Curations: America’s Supply Chain Reset; How To Fix Your Sleep Cycle?

Spotlight

This week on Bits + Bips, I joined Austin Campbell, Chris Perkins, and Joseph Chalom (Ex-BlackRock and Co-CEO of SharpLink) for an insightful conversation on:

• Why the Fed cut rates despite a ‘Goldilocks’ economy

• How AI is causing labor market shocks

• Stablecoins going mainstream in FX and settlement

• How JPMorgan, Citi, and others are quietly racing into blockchain

• Japan’s first yen-backed stablecoin and the future of cross-border money

• Why CZ’s pardon and Mike Selig’s CFTC role matter for crypto’s next chapter

You can watch the Bits and Bips podcast here.

Lumida Wealth Is Hiring

I am looking for a VP of Sales to join our leadership team at Lumida Wealth.

The cheat code to create wealth is to join a rapidly growing company. 

This is your chance to shoot your shot.

Requirement:

Must have seen the epic movie: Glengarry Glen Ross.

Target comp: $250 K to $350K and equity.

Apply here to see the full job description.

This week, I was in Vegas to attend Money 2020. 

There are a lot of financial institutions that want to do tokenization. 

They have no idea what they're doing.

They're doing POCs that don't make any sense. 

I mean, there are FIs doing tokenization of assets in their data warehouses.

I talked to the sales guy three times. He explained it three times, but it still didn’t make any sense.

I told him: “it's just a database with tokens”

He's like: “yes, but they're automated actions”.

But, those are called workflows. Zapier can do an automated workflow for you. 

You don't need tokenization for that.

The clear takeaway: Everything is getting tokenized.

Watch the livestream here.

Earnings Season Highlights: Big Tech

This week was an important one for the markets with hyperscalers’ earnings and the FOMC meeting. 

Sam Altman is saying he wants to spend $1 Tn+ on capex… and they are doing $13 bn in revenue. Even if revenue 10xs — the spend does not make any sense.

Rather than buy OpenAI, you should front-run Sam Altman. Take a look at this post more for the analysis.

Concertely, what that means is owning the supply chain — and the most mispriced part of the supply chain is the datacenter energy space. Liquid natural gas will play an important role in enabling turbines and generators to power datacenters.

Elon Musk’s Memphis datacenter is nat gas powered. We’ll see a lot more of this in the future.

Twitter tweet

Big Tech Is Spending More Than Ever on AI

Capital spending at Big Tech has entered a new phase: coordinated acceleration.

Meta, Google, Microsoft and Amazon are scaling compute to stay ahead of AI demand and defend moats.

Alphabet’s CapEx hit $24 billion this quarter, almost double from last year, with 60 percent going to servers and 40 percent to data centers and networking. 

CFO Ruth Porat said the company expects $91–93 billion in capex in 2025 and a “significant increase in 2026.” 

Microsoft framed its expansion in similar terms.

Satya Nadella said the company will “increase total AI capacity by over 80 percent this year and roughly double our data-center footprint within two years.” 

He also highlighted investments in the 2 GW Fairwater campus in Wisconsin as a model for next-generation AI plants. 

Amazon’s CFO Brian Olsavsky echoed the pattern.

“Cash CapEx was $34.2 billion in Q3, roughly $125 billion for 2025, and we expect that amount to increase in 2026. Most [capex] is [for] expanding our data-center footprint for generative AI.” 

Meta is not far behind. 

CFO Susan Li told investors, “We currently expect 2025 CapEx at $70–72 billion. Compute needs have expanded meaningfully, and we expect to invest aggressively.” 

Across the four companies, the capex is going in the same direction: GPUs, custom silicon, and data-center power. 

Amazon and Microsoft both cited NVIDIA directly; Amazon added AMD and Intel; Alphabet continues to scale its in-house TPUs.

The ripple effects of the increasing AI Capex reach every layer of the supply chain: semiconductors, power utilities, fiber networks, cooling systems, and construction firms tied to hyperscale builds.

Nvidia, Coreweave and Nebius seem poised to win from the increasing demand. You should read our capex receiver thesis to help decide where to bet in the AI flywheel.

Meta Earnings

Meta’s stock dropped about 12% after one-time tax related earnings charges.

In April of 2024, Meta dropped 20% after earnings. That was a great time to buy the dip and we noted.

Take a look at Meta’s trailing valuation history. It’s not as oversold as Google was. But it does offer compelling value and warrants an overweight in our view. The risk is that Meta’s spend on capex and AI engineers causes earnings to slow.

Still, Meta’s earnings growth is strong. They have distribution. They have a right to win and a Founder led CEO that has a proven track record of pivoting from Desktop to Mobile, and now AI.

AI Capex Is Paying Back

The defining feature of this earnings season wasn’t the size of AI investment; it was the visibility of payback.

Big Tech is reporting customer adoption, backlog growth, and early monetization of their AI services.

Alphabet reported the sharpest acceleration in enterprise demand.

Sundar Pichai said the company has entered “the generative AI era.” 

“AI Mode now has over 75 million daily active users… Most importantly, AI Mode is already driving incremental total query growth for Search.”

Ruth Porat noted: “Over 70% of existing Google Cloud customers use our AI products.” 

Google Cloud’s backlog had climbed to $155 billion, up 82 percent year-over-year, with “more billion-dollar deals in nine months than the past two years combined.” 

Sundar Pichai added adoption for AI products “is rapidly accelerating. In Q3, revenue from products built on our generative AI models grew more than 200% year-over-year.”

In May we noted Google was the only Mag 7 stock not priced for AI. We made it our largest position. It’s up 65% since this post.

Microsoft showed a similar curve. 

Nadella noted Microsoft is seeing “increasing demand and diffusion of our AI platform and family of Copilots.” 

The Copilot rollout is their first software product to monetise AI at scale: “More than 90 percent of the Fortune 500 now use Microsoft 365 Copilot… PwC added 155 thousand seats and credits it with saving millions of hours.”

Bookings strengthen the ROI story: commercial RPO reached $392 billion, up 51 percent Y/Y, while Azure AI usage rose more than 80 percent.

Amazon’s demand pipeline is just as aggressive.

Olsavsky said, “As fast as we’re adding capacity, we’re monetising it.”

At the application layer, Amazon’s retail AI assistant Rufus now touches 250 million active customers, who are “60 percent more likely to complete a purchase,” driving “over $10 billion in annualised incremental sales.” AI is moving directly into revenue conversion.

Meta’s version of AI demand is rooted in its ad and commerce ecosystem.

Mark Zuckerberg said the company’s “AI-powered ad tools have passed a $60 billion annual run rate,” and that the same models, driving engagement, also power business automation. 

CFO Susan Li added, “Investments we’re making within our ads and organic engagement will enable strong revenue growth in 2026.”

The pattern is clear: AI spending is being validated by usage and backlog - it isn’t speculative anymore.

AI & Productivity: Carmax (?!)

CarMax cut 300 call center jobs due to the benefits of AI. 

CarMax is a sub $10 Bn company. We don't own it.

But, the point is that productivity gains are spreading to broader parts of the economy, including small to mid-sized companies.

This is bullish for the real earnings growth of the economy and disinflation.

You are not bullish enough on the promise of AI.

Automobile Sales Show Consumer Strength

Auto makers closed Q3 with a tone of cautious confidence. Consumer demand remains resilient and tariff headwinds are easing faster than expected. 

Consumer appetite for high-end and core models remains strong.

Ford’s CEO Jim Farley said they have added “up to 1,000 new jobs to increase F-Series production to recover lost volume and fulfill strong customer demand.”

President Andrew Frick highlighted that U.S. market share rose to 12.8 percent, led by flagship models: “The all-new Expedition is very strong, gaining over 3 points of segment share, with 75 percent of customers choosing high-end trims like Tremor.”

General Motors echoed that consumer health remains intact despite macro uncertainty.

CFO Paul Jacobson said, “History shows that rising unemployment usually has an impact, but right now we’re in good shape and seeing customers stay strong.” 

Moreover, tariff impacts look more controlled now.

GM reported that its $3.4 billion EBIT included a $1.1 billion tariff impact: “a little less than we had expected.” 

Ford expects similar relief. 

CFO Sherry House cut the company’s 2025 tariff forecast in half: “We now expect tariffs will be a $1 billion net headwind for 2025, down from $2 billion.” 

Taken together, the auto sector looks stable. 

The Hawkish Rate Cut

The Fed delivered a 25 bps rate cut this week - its second in the last two meetings, but Jerome Powell’s future indications were hawkish. 

“There were strongly differing views about how to proceed in December,” he said. “A further reduction in the policy rate at the December meeting is not a foregone conclusion. Far from it.” 

After Powell’s comments, equities sold off and the 10-year yield climbed back above 4 percent. 

Polymarket odds for a December cut dropped to 66% from 90%.

What’s funny is the economy doesn’t need rate cuts at all. 

The labor market’s cooling is structural, not cyclical, and Powell acknowledged that reality. 

“Conditions in the labor market appear to be gradually cooling,” he said. 

The slowing supply “likely reflects a decline in the growth of the labor force due to lower immigration and labor force participation.” 

He added, “Both layoffs and hiring remain low… we do not see the weakness in the job market accelerating.”

Overall, the moderation in job growth is due to slowing of both labor demand and supply. Businesses are relying on higher efficiency to cover the gap. 

Productivity growth is picking up, unit labor costs are contained, and the economy is generating more output per worker. 

This setup means lower rates aren’t likely to “fix” the labor market, but they are also not likely to reignite consumer inflation. 

Instead, easier policy is more likely to fuel asset-price inflation, particularly in equities, as liquidity flows into productive and financial assets.

Markets

Zoom out. 

Big banks are growing earnings. 

Big tech are growing earnings. And briskly!

GM and consumer names are seeing growth. 

Even names that sold off, like cruise lines, are seeing strong growth in their core business.

I first ask myself: ‘How is the underlying business doing? and, what about the core drivers?’ before looking at how it did versus expectations. 

Take Virtu for example. 

Strong double digit growth YOY and benefits from trends in ETF and retail trader growth. 

Dropped due to a ‘miss’.

The core business is great. 

We scooped it up on the gap down on Wednesday and are sitting on gains now. 

This is a bread and butter tactic.

Bubble Talk Is Lazy Analysis

‘Bubble’ - every cycle gets the same label before it’s understood. 

The 1999 analogs miss the point. This isn’t a liquidity-driven bubble. It’s a cash flow–backed investment cycle.

The top of the Nasdaq in 2000 was funded by debt, story stocks, and vaporware. 

Today’s AI leaders are cash-rich, profitable, and reinvesting out of free cash flow. 

NVIDIA, Microsoft, Alphabet, and Meta are financing record CapEx internally. The balance sheets are pristine, and the ROI math already works.

AI isn’t speculative; it’s productive. 

Cloud margins are rising, GPU utilization is climbing, and enterprise adoption is accelerating. These are capital projects with yield.

As Jurrien Timmer notes, NVIDIA’s valuation may be high, but it is nowhere near dot-com extremes.

Cisco traded at 214× earnings in 2000; NVIDIA’s forward multiple is around 32×. The difference is that this time, earnings are compounding with each compute cycle.

Open AI To Go Public

OpenAI is planning a $1 Tn IPO in 2027.

That could potentially be the time to be cautious on AI stocks. (We'll have to assess then.)

2000s: Goldman Sachs IPO was a top signal

2007: Apollo and Blackstone

2021: SPAC and IPO boom ends

That large new issue would need to be absorbed.

We're going to learn a ton from their S-1 filing in 2026.

CMG needs a reality check

Chipotle is down 20.2% since it reported earnings.

1) Shareholders are running for the border

2) Investors are losing their appetite after Ackman’s 13F

3) Chipotle’s dwindling portion size is matched by dwindling returns.

Notably, it still remains pricier than META despite this pullback.

Management cites consumers in the 25 to 35 age cohort are pulling back.

CEO Boatwright cited a "broad-based pullback in frequency across all income cohorts" and pointed out, "the gap has widened, with low to middle-income guests further reducing frequency". 

This group, which represents about 40% of total sales, is "dining out less often due to concerns about the economy, and inflation."

Interestingly, if you look at EAT's latest results, you'd see the opposite comments. 

Kevin Hochman(CEO): "Under $60,000 household is actually growing faster than other cohorts right now." 

The reason: " I think we've got the sweet spot of being positioned as a great value and then delivering a consistent experience when guests come in"- Kevin Hockman

Consumers switch to substitutes if you are falling behind. 

This is business 101.

Here’s a hilarious video to understand why CMG is lagging.

Overall, restaurants remain in a bear market.

Power is the new bottleneck

Energy demand is quietly becoming the binding constraint on AI. 

Recent projections suggest data-center electricity use will reach about 20 % of U.S. consumption by 2030, with AI alone accounting for roughly 15 %. 

The winners will be natural-gas midstream players, private-market plays in modular generation, and turbine/generator firms servicing new builds.

Their earnings already reflect the demand. 

Nat gas midstream backlogs are up ~60 % with ~90% of it being gas-related. 

In latest reports, KMI’s leadership projects a 28 Bcf per day increase in natural gas demand by 2030, nearly double today’s level, driven by AI power loads.

Halliburton’s CEO Jeffrey Miller says: “The demand for power and for AI is like nothing I’ve ever seen in terms of demand growth.”

However, listed nat gas midstream players look inflated. 

KMI, for instance, has a P/E NTM of 19.3 compared to P/E NTM of 16.0 of S&P 400 Mid caps. 

Nonetheless, the private market offers some exciting opportunities.

In Alberta, innovative O&G companies are turning stringent flaring rules into an advantage. 

Flaring regulations prohibit oil and gas producers from venting natural gas during oil production.

The alternative, capturing and transmitting the gas via pipeline, can be expensive, sometimes forcing operators to curtail mining.

This creates opportunities for firms that understand both oil operations and datacenters. 

They capture excess gas, combust it to generate power, and use that electricity to mine crypto or run datacenters, earning revenue from both oil sales and mined assets.

Crusoe Energy pioneered a similar model and is now a $10B private company, though inaccessible to retail investors. New players are emerging with significant upside potential.

At Lumida, our team seeks undervalued assets often overlooked by the broader market, offering qualified investors opportunities typically reserved for institutions. 

If you’d like to be considered for upcoming deals, please sign up using the form, or email ray@lumida.com. We have several ideas in the industry with proven execution.

Moreover, Lumida ventures is actively exploring opportunities around tax mitigation to ensure you can optimize your after-tax returns. You can sign up with Lumida's Tax Shield, or write to ray@lumida.com understand how our existing strategies work.

Digital Assets

The Future Is Hybrid Finance

Legacy finance is crossing the bridge into crypto.

This week, Western Union partnered with Anchorage Digital and Solana to launch USDPt, a stablecoin built for cross-border payments across 150 countries. 

Anchorage Digital Bank, the only federally regulated stablecoin issuer, will mint and redeem USDPt under U.S. oversight. Solana provides the rails, and Western Union supplies the global distribution.

Traditional firms are no longer fighting crypto, they’re absorbing it. Instead of “disruption,” we’re seeing integration. 

Legacy financial networks bring compliance, licensing, and consumer reach; digital-asset infrastructure adds speed, finality, and programmability. 

The result is a new settlement architecture that feels inevitable in hindsight.

Anchorage’s co-founder Nathan McCauley called it a “landmark partnership,” and he’s right. 

For decades, Western Union was synonymous with paper slips and remittance counters. Now it’s experimenting with blockchain rails to modernize dollar transfers in real time. 

Solana’s low-cost throughput gives it the performance layer needed to make that work at scale.

The next decade of payments will be built on hybrid finance, not pure crypto or pure banking. 

The winners will be firms that can merge regulatory credibility with blockchain speed, and those that move first are already defining the template.

Visa’s Stablecoin Superhighway

Visa just made its biggest on-chain move yet.

CEO Ryan McInerney announced the company will add support for four stablecoins across four blockchains, spanning two fiat currencies and convertible into 25 traditional currencies. 

The goal: make digital dollars spendable anywhere Visa runs, which is practically everywhere.

The data tell the story. 

In the fourth quarter, stablecoin-linked Visa card spend quadrupled from a year earlier. Since 2020, the company has processed $140 billion in crypto and stablecoin flows, including $100 billion in purchases using Visa credentials. 

It now has 130 stablecoin-linked card programs in over 40 countries.

McInerney said Visa is also enabling banks to mint and burn their own stablecoins. 

This move paints a broader picture: instead of crypto firms trying to replace the networks, the networks are absorbing the tech.

Stablecoins are becoming settlement assets on global payment rails.

This is what convergence looks like.

Flows From Gold To BTC? 

Gold peaked in mid-2020 just before Bitcoin began its parabolic run. 

The chart suggests we may be watching the sequel play out.

Liquidity cycles tend to move from hard collateral to high-beta assets once rate expectations flip and real yields stabilize. That transition favors crypto.

Lumida Curations

America’s Supply Chain Reset

With China controlling 90% of rare-earth refining, the U.S. is moving to restore domestic processing capacity and reduce dependency through reshoring and new investment initiatives.

View Curation

How To Fix Your Sleep Cycle?

Late-night eating spikes glucose and wrecks deep sleep. Finishing dinner early and walking after meals improves recovery and next-morning energy.

View Curation

Meme

Not subscribed yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | Youtube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax

As Featured In

Disclaimer: Lumida Wealth Management LLC (‘Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transact business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions,forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.

Here’s a preview of what we’ll cover this week: 

• Macro: Main Street Still Strong; Credit Normalization; Private Credit Risk Is Controlled

• Markets: AI & Productivity Growth; The AI Buildout Isn’t a Bubble; Energy Companies are Major AI beneficiaries; America’s Defense Machine Is Running

• Digital Assets: BlackRock Is Tokenizing the Financial System; JPMorgan Brings Bitcoin Into the Banking System; Fed Bear hugs Crypto

• Lumida Curations: Meme Coins Are Changing Finance; Omega-3 Deficiency: The Hidden Brain Drain; Gold’s Rally Nears Exhaustion?

Spotlight

This week, I was invited to a podcast with Pio Vincenzo for a wide ranging discussion on investor psychology, fears around the US economy and opportunities in equities and crypto.

Here’s what we discuss:

• Fear vs fundamentals + Political perception

• Goldilocks economy & global peace narrative

• Fear gauge & political polarization

• Banks, Insurance and Mortgage plays

• Options & covered-call strategies

• Bitcoin 4-year cycle & market attention dynamics

Watch the podcast here.

On Thursday, Lumida hosted a Best Idea Dinner in NYC with an incredible group of family offices. If you want to join, shoot us a note at info@lumida.com and tell us a bit about yourself.

Each guest shared a two-minute idea from market views to emerging trends, companies, and opportunities they’re most excited about.

We got some thoughtful insights - overall, great energy around the table.

The S&P started the week with gloom & doom all over the headlines.

I posted this snapshot from the WSJ on X Monday morning. The pessimism is thick. Notice the headlines have nothing to do with what matters — earnings.

Then the several indices finished the week at all-time highs.

I did a solo-podcast on Friday about how we are in Goldilocks, precisely because no one realizes it’s Goldilocks. Watch Lumida Non-Consensus Investing here or on Youtube or Spotify.

Earnings Season Highlights: Regional Banks

Over the last few weeks we have been writing about how bank earnings are set to do well.

This past week we saw regional bank earnings, and they were better than most expected. 

The KRE rose ~5% as a result, recovering after last week’s sell-off which was muchado about nothing. 

Here was a link to my FSD livestream the week before when KRE index fell to multi-month lows. The main message here is that this is not a credit risk story.

Main Street Still Strong

The broader themes in regional banks were exactly the same as major banks, and they second: economy is strong, the consumer prevails, and credit risk is near lows. 

Ken Vecchione (CEO, Western Alliance Bancorporation):  "Right now, I think the overall backdrop to the economy is pretty good.”

“You’ve got GDP growing at 3.8% to 3.9%. You’ve got the 10-year rate coming down to under 4%. Employment, for as much as people are talking about and nervous about it, is still in a rather low 4% area.” 

“You have greater investments being made into the country from foreign countries. That should help continue with economic growth. You’ve got a pro-business president. That’s the backdrop to a lot of things that we’re seeing.”

Richard Fairbank (CEO, Capital One Financial) shared a similar view:

“The US consumer and the overall macro economy have been quite resilient so far in 2025.” 

“Unemployment has ticked up a bit recently but remains quite low by historical standards. Wages are growing in real terms and debt servicing burdens remain stable and close to pre-pandemic levels.”

Bruce Van Saun (CEO, Citizens Financial) reinforced that the consumer remains intact:

“Credit continues to perform better than expected. Consumers are adapting well to higher rates and maintaining strong balance sheets.”

The slowdown doomers have been expecting hasn’t arrived (and isn't likely to arrive). 

Credit Normalization

Credit metrics are normalizing. Charge-offs and delinquencies are ticking up from record lows, but remain well below historical averages - a sign of a healthy credit cycle.

Stacy Kymes (CEO, BOKF) described the trend:

“These are abnormally strong credit numbers.” “There will be a time where both charge-offs and criticized levels kind of revert to the mean. That doesn’t mean credit’s deteriorated. It just means there’s been a reversion back to normal.”

Harris Simmons (CEO, ZION) echoed that same macro framing:

“We’ve continued to see solid credit quality across the portfolio. Nonperforming assets remain low, and reserve levels are very healthy relative to risk. What we’re seeing now looks like a normalization to pre-pandemic levels, not deterioration.”

We saw similar sentiment in large banks’ earnings as well. Consumers are strong - they are taking new loans, and paying back on time. Read the last newsletter here.

Private Credit Risk Is Controlled

Last week, ZION’s $50MM write off and WAL’s $100MM fraud case led to a sharp sell-off in regional banks. Markets had fears of GFC crawling all over again.

Nonetheless, banks’ leadership muted fears of any systemic risk, and showed confidence in their credit portfolio. They highlight private credit as well-capitalized, collateralized, and diversified. 

Harris Simmons (CEO, ZION) addressed the issue directly:

“We’ve reviewed our portfolios and believe this was an isolated incident [Canton Group]… We continue reviewing with an external party to make sure that we’re learning from experience and seeing what we can improve upon.”

Other than the $50MM related to Cantor, ZION reported only an additional $6MM in write-offs in Q3, highlighting the one-off nature of the default.  

Kenneth Vecchione (CEO, WAL) also showed confidence in their private credit quality:

“We have operated in the private credit business for over 15 years.”

“We view our underwriting expertise, ability to evaluate structured credit and sophisticated approach to minimizing uncovered risks through strong collateral with low advance rates as core competencies of the bank that prevent and mitigate losses.”

“Over the past 5 and 10 years, our net annual charge-offs averaged just 10 and 8 basis points, respectively, placing us among the top 5 U.S. banks with assets greater than $50 billion.”

Derek Steward (Chief Credit Officer, Zions) detailed how their nonbank exposure is structured:

“NDFI exposure is actually about 3% of our total loans… The majority of it would be equipment leasing type transactions; yellow iron, trucks, things like that… It’s very well diversified across a lot of lending segments.”

Stacy Kymes (CEO, BOK Financial) noted that even within specialized finance, risk remains tightly contained:

“Our exposure to NDFIs is approximately 2% of total loans, with the vast majority in the two highest credit quality subcategories: subscription lines and residential mortgage warehouse lines. ”

Bank private credit exposure is minimal and controlled. Credit risk is idiosyncratic - financials look in good shape.

Market

Friday’s cooler-than-expected CPI report spurred equity markets to new highs after odds of rate cuts increased. CME Fedwatch shows a 90% probability of two rate cuts in 2025.

What’s funny is the economy doesn’t need rate cuts at all, any more than it needed them last year.

The non-farm payrolls reports which is notoriously noisy and public psychology is ushering in rate cuts.

That’s fine - we are well positioned with exposure to small caps that will benefit.

AI Names Have More Fuel