Meet like-minded investors at the Latticework 2026 summit in Chicago.

MOI Global members, scroll down for member news and updates.

11 Articles We Are Reading This Week

1. Fable - Shakeel Hashim

2. Cycles - Luke Bridgeman

3. AI theologies - Duncan Umphrey

4. AI exponential - Dario Amodei

5. Paying attention - David Epstein

6. Great switcheroo - Adam Mastroianni

7. Sample efficiency - Dwarkesh Patel

8. Principled thinking - Ray Dalio

9. Silent revolutionaries - Scott Sumner

10. Anti-monopoly fatigue - Noah Smith

11. Markets defy prediction - Devendra Bambardekar

Subscribe now

Letters, Presentations, Writeups…

Industry Insights: accounting | artificial intelligence | banking | software

Geographic Insights: Australia | Europe | Ireland | Japan | UK

Letters: GDS Investments | Merrill | RGA Investment Advisors | T. Rowe

Company Insights: Fiserv | Google | Hikari Tsushin | Intel | Paycom Software | Rentokil | SpaceX | Wise

Company Slides: Adobe | Campbell’s | Dollarama | EnerSys | Graham | Honeywell | Land’s End | Lufthansa | McGraw Hill | Oracle | Stitch Fix

5 Videos We Are Watching This Week

1. CNBC - Guy Spier talks about his cancer diagnosis and shares wisdom

2. Bloomberg - Inside Anthropic, the lab racing to create frontier models

3. David Senra - Gustav Söderström on the firm that Apple could not kill

4. iConnections - Short seller Jim Chanos on spotting frauds and bubbles

5. Brandes Center - Panel on value investing in today’s high-tech world

5 Podcasts We Are Listening to This Week

1. Latticework - Chris Bloomstran puts US equities in historical context

2. How I Built This - Tobias Lütke on how a snowboarder scaled Shopify

3. 100 Year Thinkers - Robert Hagstrom on the flaws of portfolio theory

4. Knowledge Project - Bill Gurley on systems and second-order impacts

5. Masters in Business - Chris Davis of Davis Advisors on market cycles

Member Updates & Events

The following is a private section for MOI Global members. Membership is by invitation only. If you would like to become a member, be sure you are on the annual plan. It is the first step toward full membership.

Read more

Chris is President and Chief Investment Officer of Semper Augustus Investments Group, the value-oriented firm he co-founded in 1998. His annual client letters, which routinely run past a hundred pages, contain what many regard as the most rigorous outside analysis of Berkshire Hathaway available anywhere. Few practitioners combine bottom-up valuation discipline with capital-cycle history the way Chris does. For a fuller portrait, we encourage you to listen to our exclusive audio program, Chris Bloomstran and the Semper Augustus story.

The most expensive cap-weighted market in US history

Chris opened with the table he has kept in his letters for over twenty years, comparing today against the secular peaks of 1929, 1937, 1966, 2000, 2007, and 2021. The S&P 500 ended 2025 at 26 times operating earnings, matching 1929 and approaching the 29 times of early 2000. The CAPE ratio stood at 40, a record. Adjusted for this year’s roughly 10% advance, price to sales is now 3.8 times and price to book 6.2 times. With about $70 of index dividends against an S&P at 7,600, the dividend yield sits near 1%, an all-time low. The yield reflects a record-low 27-28% payout ratio, down from roughly 60% a generation ago; a quarter-century of buybacks consuming half of operating cash flow never shrank the share count, because some 2% of shares are granted to insiders each year. His verdict: “You will not find a moment in the history of the United States cap-weighted stock market where valuations have been higher than today.”

The title is self-deprecating. Chris called a secular peak in his 2021 letter; 2022 obliged with an 18% S&P decline and 35% on the Nasdaq, but the recovery carried valuations above the 2021 highs. Borrowing Irving Fisher’s infamous “permanently high plateau,” he now calls it a secular plateau.

The arithmetic of buying high

The Ibbotson series puts long-run US equity returns at 10.1-10.2% annually from the end of 1925. Buy the exact 1929 top and you compounded at 9.7%; buy the June 1, 1932 low (the S&P at $4.40, down 86%) and you compounded at 12%. That spread compounds brutally: $100 invested at the 1932 low grew to $10,000 by the end of 1964, a hundredfold gain in a third of a century. Berkshire then turned each $100 of 1964 value into $6.1 million by 2025 (19.7% a year) versus $45,000 for the S&P 500. Berkshire’s stock could have fallen 99.26% and still matched the index. When Chris shared that statistic with Warren Buffett years ago, Buffett replied, “Ben would be proud, but let’s not test the math.” Charlie Munger’s in-person response at a Wesco meeting is worth the price of admission.

A T-bill over the index, but cash has a clock

Offered the choice between the cap-weighted S&P 500 and a Treasury bill today, Chris would own the bill. The excess sits not in every stock but in the Magnificent Seven and perhaps 50 or 60 other large companies at very full prices. Yet cash, he stressed, is a terrible long-term asset. His opportunity-cost table shows that with stocks compounding at 10% and cash at 3%, the market must fall 28% within five years, 48% within ten, for the cash holder to break even. Semper Augustus resolves the tension by owning roughly 25 businesses that look nothing like the index. “Time in the market is better than timing the market, and that is the absolute truth.”

Five factors, two case studies

Chris does not value companies using a DCF. His framework decomposes total return into five drivers: dollar sales growth, margin change, multiple change, share-count change, and dividend yield. Two Berkshire holdings illustrate it. Coca-Cola, bought for $1.3 billion at 15.7 times earnings, grew to $16 billion and 58.5 times by mid-1998, when it was half of Berkshire’s portfolio. Since then: sales up 3.4% a year, margins up to 26.7%, the multiple down 60% to 23, buybacks adding 0.5% and dividends 2.8% annually. Net result, 4.6% compounded for 27 years from a great business. Apple, bought at 13.9 times earnings (10 times net of cash), has compounded at 29.4%, but with the multiple now at 34.5 times and sales growth slowing, Chris’s scenarios produce zero to 7.5% forward returns. Berkshire has sold 80% of its shares, which he calls a very good sale. The 1998 General Re acquisition, paid for with 280,000 Berkshire shares at three times book, halved an overvalued equity portfolio and let book value compound at 11.2% since, ahead of the 9.2% the stock portfolio itself delivered. A masterstroke, in Chris’s words.

Margins, the Mag 7, and the road to Ibbotson

In his 1999 Fortune piece, Buffett observed that profit margins had ranged between 4% and 6.5% over his lifetime. The S&P 500 net margin ended 2025 at 12.8% and is headed for 14.5-15% this year, as a handful of companies, Nvidia foremost as its profit goes from $100 billion toward $200 billion, add some $300 billion to aggregate index earnings, taking them from $2.5 trillion toward $2.8 trillion. The Mag 7 earned their weighting, rising to 26.8% of index profits and accounting for 60% of all profit growth from 2021 through 2025. But the implied future is demanding. To deliver Ibbotson’s 10.5% over the next decade, margins must reach 20.7% at a constant 26 multiple, or the multiple must reach 43 times at constant margins. Hold both flat and the return is sales growth plus dividends, roughly 4.5%. Chris finds the 10.5% case implausible and expects the index to trade substantially below current levels at some point within ten years.

The AI capital cycle

The sharpest section of the talk applied capital-cycle history (canals, railroads, autos, electrification, fiber) to the AI buildout. The fiber precedent: roughly 1% of GDP spent, only 4% of fiber lit by 2021, most issuers bankrupt, yet the infrastructure eventually enabled YouTube and Netflix. Hyperscaler capex was $400 billion last year, is projected at $750 billion or more this year, and totals $1.3-1.4 trillion to date, with 2023-2030 estimates of $4 trillion and now as high as $7 trillion. The depreciation math is unforgiving: $400 billion of capex on a ten-year straight line creates $40 billion of year-one depreciation against AI revenues of $40-50 billion. A 20% return on $4 trillion requires $800 billion of profit, nearly a third of all S&P 500 earnings. Capex consumed 41% of the big four hyperscalers’ operating cash flow in 2023; in 2026 it will consume essentially all of it. Layer on roughly $650 billion of off-balance-sheet financing, such as Meta’s $30 billion, 5-gigawatt Hyperion data center, financed through a Blue Owl joint venture with just $500 million of Meta equity and a Meta debt guarantee, and Chris hears echoes of Enron and Lucent. He is waiting for the first hyperscaler to cite EBITDA on an earnings call.

Berkshire, the discipline case study

The counterexample is the company Chris knows best. National Indemnity shrank premiums from $366 million to $56 million between 1986 and 1998 while maintaining an underwriting profit every year until 9/11. Berkshire is again retreating as reinsurance pricing falls 20%, while holding $300 billion of the industry’s $600 billion of capital and writing premiums equal to just 7% of its own capital. A fixed GEICO runs combined ratios in the low-to-mid 80s and grew policies 5% last year. The insurance group dividended $100 billion to the parent over two years; about $120 billion of the $380 billion cash pile sits at the holding company. Then there is the new $10 billion Google purchase at 10 times sales, which lifts the position past $30 billion, now Berkshire’s third-largest equity holding. Chris’s read on what this signals about Greg Abel’s capital allocation, and the signposts he is watching, is alone worth hearing.

Hear the full session

This summary compresses a dense, data-rich session and a candid Q&A covering AI-driven job displacement, Google’s at-the-market equity offerings, housing exposure through Builders FirstSource, and the future of Berkshire’s culture. The audio replay captures the delivery and the room. The transcript preserves every number. The 54-page deck includes the secular peaks-and-troughs table, the five-factor decompositions of Coke, Apple, and the S&P 500, and the AI capex and depreciation schedules underlying the argument. All three are available now to Latticework members. If you are not yet a member, this session is a fine reason to join us.

Read more

We start with a story…

Gabriel-Julien Ouvrard was thirty-five years old in 1805, and he had borrowed an enormous amount of money against ships he could not see.

The ships in question were Spanish galleons carrying silver from the mines of Mexico and Peru. Ouvrard, the son of a provincial paper merchant who had become one of the richest men in France by quietly supplying the navy during the Revolution, had built a financial structure around them. He and two partners, Médard Desprez and Ignace-Joseph Vanlerberghe, ran a firm called the Compagnie des Négociants Réunis (CNR), which translates with a kind of corporate plainness that would not be out of place in a modern pitch deck as “The Merged Merchants Company.” The economists at the BIS who later reconstructed the firm’s balance sheet would call it “arguably even greater than that of Lehman Brothers in the 2008 financial crisis.” That comparison is not a metaphor. It is structural.

Here is what the CNR did. Napoleon’s Treasury needed money to fight Britain. The Spanish Crown, which owed France a great deal of money, did not have any cash but did, theoretically, have access to enormous quantities of silver from its American colonies. Ouvrard negotiated to buy the future silver shipments at 3.75 francs per coin, knowing the same coins would be worth 5 francs once they reached Europe. To finance this arbitrage, he borrowed from everyone. He borrowed 10 million francs from Parisian banks. He borrowed 12 million florins from Hope & Co. in Amsterdam, the bluest of blue-chip lenders. He borrowed 100 million francs in credit from a Spanish institution called the Caja de la Consolidación, itself secured by discounted futures on the very same Spanish dollar coins. And, most consequentially, he borrowed from the Banque de France, where his partner Desprez sat as a director and owned 17.6 percent of the equity, making him by a wide margin the largest single shareholder. The second-largest held 8 percent. Most held less than 1 percent.

In modern terms, the partner of a leveraged investment fund was also the largest equity holder of the central bank, and the firm’s collateral consisted of physical commodities that had not yet been shipped from another hemisphere. By October 1805, Desprez’s exposure to the Banque de France had reached 105 million francs.

This worked exactly as long as the silver kept arriving.

The silver that ran the world

It is hard, from a vantage point in 2026, to grasp how thoroughly silver dominated the financial system of 1805. The Spanish dollar, formally the Real de a Ocho or Peso de a Ocho, was the bitcoin and the dollar of its century combined: minted in Mexico and Peru, shipped through Cádiz, accepted from Canton to Calcutta to Boston. Early central banks, including the Banque de France, the Bank of Amsterdam, and the Bank of England, held Spanish dollars as their primary reserves. Qing dynasty China absorbed staggering quantities of them in exchange for tea, silk, and porcelain. The entire pan-European credit system, denominated in bills of exchange that circulated like modern Eurodollars, assumed that more silver was coming.

What Ouvrard had built was, in essence, a trade financing the conversion of future silver into present credit. Every bill of exchange he discounted, every loan he took out, every advance to the French Treasury, was implicitly an option on the continuing flow of Spanish silver through the Atlantic. The Caja de la Consolidación in Madrid. The Hope partners in Amsterdam. The Baring brothers in London. The receveurs généraux who collected French taxes. The merchant banks of Marseille. All of them held paper that, traced back through enough layers, was an IOU on a galleon that had not yet sailed.

A modern reader will recognize the shape of what comes next.

The news that traveled by horse

On the morning of October 21, 1805, off Cape Trafalgar near the southern coast of Spain, the combined French and Spanish fleet under Vice-Admiral Pierre-Charles Villeneuve met the British Royal Navy under Vice-Admiral Horatio Nelson. By nightfall, eighteen French and Spanish ships had been captured or destroyed. Nelson was dead, shot by a sharpshooter from the rigging of the Redoutable. The British had paid a terrible price for it, but they now controlled, more thoroughly than at any previous moment in history, the Atlantic shipping lanes between Spain and Latin America.

In Paris, no one yet knew.

News from Cádiz to Paris in 1805 traveled at roughly the speed of a man on horseback, about 100 kilometers per day. The 1,800 kilometers between the two cities meant that the most consequential piece of financial information of the decade would not be read in Paris until November 6. The newspaper that carried it, Journal des Arts et des Spectacles, ran the report on November 9.

Two and a half weeks of latency between the event and the price action. There is something darkly amusing about this, in retrospect. Modern markets are obsessed with low-latency trading; firms spend hundreds of millions of dollars to shave microseconds off the trip between Chicago and New Jersey. In 1805, the news that the silver was not coming had to walk.

While it walked, the Banque de France’s discounting peaked. On November 6, the same day the news from Cádiz finally arrived in Paris, the Banque’s outstanding bills of exchange hit their crisis high. At precisely the moment confidence collapsed, exposure was at its maximum.

Three men in a room

The Conseil de Régence of the Banque de France met on October 2, 1805. The directors knew the Cádiz blockade was tightening. They knew the value of Spanish public debt, the vales reales, had collapsed. They knew Desprez was leveraged to the silver shipments. They did not yet know about Trafalgar. They had to decide what to do anyway.

Three positions emerged from that meeting, as documented in the Banque’s minutes and reconstructed by the BIS authors.

Director Benjamin Delessert, a textile manufacturer who would later found France’s first savings bank, argued that the Banque should halt discounts and withdraw support from Desprez. Continued lending, he said, endangered the Banque’s survival. He was, in modern parlance, the moral hazard hawk.

Director Rame argued the opposite. Hesitation had already triggered 25 million francs of deposit withdrawals at Desprez. Continued support, up to 30 million francs through November 11, would prevent panic and, in a phrase that would not feel out of place in the 2008 emergency meetings at Treasury, would “prevent defeating Napoleon in Paris.” He was the systemic-risk dove.

Director Sevene proposed the middle path. Partial support combined with shareholder solidarity to prevent the hoarding of specie. His appeal to moral restraint and the wisdom of the Paris merchant houses reads, two centuries later, like J.P. Morgan’s effort to lock the bankers of New York in a room during the Panic of 1907 until they recapitalized the system.

The Conseil temporized through October. Then, on November 6, the news arrived. On November 7, the conseil de régence explicitly linked excess note circulation to the crisis. On November 8, Delessert renewed his proposal to “suspend discounting until the banknotes were restored to par.” On November 11, the Banque intensified credit rationing, including toward the Treasury.

Between November 6, 1805 and February 11, 1806, the Banque de France cut its discounting of bills of exchange from roughly 120 million francs to 60 million francs. A 50 percent contraction of credit to the productive economy, in three months, achieved by a five-year-old central bank in order to protect itself.

What followed

The Banque de France saved itself. The metallic coverage of its banknotes rose from one quarter at the start of the crisis to 100 percent by late 1806. The Banque hoarded silver. It dispatched a representative to Madrid to extract more. It purchased silver piastres in December despite high prices. It rationed credit to private counterparties.

Approximately twenty major Parisian merchant houses failed. Eight commercial houses in Marseille filed for bankruptcy. The CNR was one of them. So was the house of Récamier, whose own director sat on the Conseil de Régence and watched his firm collapse from the inside. Aggregate annual discounts by the Banque de France collapsed from 847 million francs in 1804 to less than 400 million in 1806. The Banco de San Carlos in Madrid saw its silver reserves drained to near zero as the Paris hoarding pulled specie out of every peripheral market that would yield it.

Pierre Samuel du Pont de Nemours was commissioned to write a report on the crisis in 1806. His son had founded a chemical company in Wilmington, Delaware four years earlier. It is still operating.

The same story, told in oil

Read the previous five paragraphs again and ask yourself which words to change.

In February 2026, the United States and Israel went to war with Iran. The Strait of Hormuz, through which 20 percent of global crude oil and 12 to 14 percent of Europe’s liquefied natural gas had been moving as recently as January, was effectively closed to commercial shipping. Iran imposed a $2 million toll per tanker for transit. War-risk insurance premiums, which had been a baseline 0.125 percent of vessel value, spiked to between 0.2 and 0.4 percent.

West Texas Intermediate, which had traded at $67.02 a barrel on February 27, reached $102.68 by May 18. Diesel rose 49.8 percent. Jet fuel rose 56 percent. Urea fertilizer, the production of which depends on natural gas and the export of which is concentrated in the Persian Gulf, rose 25.9 percent.

The 10-year US Treasury yield spiked to 4.46 percent on March 27, the highest level since July 2025. Equities fell. Bonds fell. The dollar moved with crude oil rather than against it. The classic 60/40 portfolio diversification, which assumes that government bonds rally when stocks fall, did the opposite for six weeks running.

Meanwhile, the central banks of the world did what the Banque de France did in 1805. They hoarded reserves. The asset they hoarded was gold. Annual gold purchases by global central banks, which had averaged about 500 metric tonnes for years, doubled to over 1,000 tonnes in the run-up to the war. By January 1, 2026, Russia’s central bank held gold reserves worth $326.5 billion, or 43.3 percent of its total reserves, at a spot price of $4,400 per ounce. Turkey, facing a currency collapse with the Lira hitting 11 consecutive record lows in 16 trading days, began exploring whether its $135 billion of gold held at the Bank of England could be pledged as collateral for emergency dollar liquidity.

The precipitating event for all of this hoarding was not 2026. It was 2022, when the United States and Europe froze approximately $300 billion of Russian central bank assets following the invasion of Ukraine. From that moment, the world’s non-aligned central banks understood that dollar reserves carried a kind of political default risk that gold did not.

What is the silver of 2026?

The BIS paper makes a careful point. In 1805, the dominant safe asset was Spanish silver. In 2026, it is, in different ways, both gold and oil. Gold, because it sits outside the dollar system and cannot be sanctioned. Oil, because every modern fiat currency is, in the long run, a claim on the productive output of an economy that requires energy to function.

When the supply of either is severed, the consequences are not merely inflationary. They are systemic. The Banque de France in 1805 did not collapse the European economy because silver got expensive. It collapsed the economy because a credit system that had been built on the assumption of continuous silver supply suddenly required everyone to hold actual silver. The contraction of credit was the disaster, not the rise in the price of silver.

This is what investors in May 2026 must understand. The Iran war and the Hormuz closure are not, principally, an inflation event. They are a collateral event. The financial system has been operating on the assumption that energy would continue to flow through chokepoints that turn out to be more vulnerable than anyone had priced. When that assumption breaks, the credit system that depends on it has to recompose itself around different physical assets and different geographies. The companies that own those physical assets, in those geographies, are the silver of 2026.

This report is an attempt to identify them.

Key Findings

The macro thesis is structural, not cyclical.

The BIS framing and the 1805 analogy describe a regime where the central choke point of global trade and credit shifts to physical commodity routes and sovereign collateral. In 1805, the Royal Navy’s Trafalgar victory reorganized capital flows for a generation. In 2026, the Strait of Hormuz closure has done something similar to the assumed permanence of dollar liquidity and just-in-time energy. Investors are being repriced for a world where security of supply, not lowest cost, is the operative metric in energy valuation models, and where central banks treat gold as the politically neutral reserve asset of last resort.

The structural beneficiaries fall into eight buckets.

Energy sovereignty (LNG export, midstream, gas E&P), nuclear and AI power, defense and asymmetric warfare, agricultural inputs, LNG shipping, industrial gases and grid, reinsurance, and real assets (timber, farmland). The thread tying them together is that each owns or sells something that cannot be reshored quickly, manufactured on demand, or substituted away.

The consensus winners are now too crowded for new capital, but quality alternatives are available.

Rheinmetall (up roughly 100 percent since end-2024), Frontline (up 162 percent), MP Materials (up 313 percent), Quanta Services (up 129 percent), Cameco (up 104 percent), Newmont (up 189 percent), and Hanwha Aerospace (up 291 percent) have done their work. Multiple compression is the dominant risk for these names.

The contrarian discipline is to find quality businesses that are exposed to the same structural drivers but have not yet been bid up. We have done that work. Each company profiled in this report was screened to exclude any name up more than 30 percent year-to-date or more than 60 percent since end-2024, then selected on the basis that the underlying business shares the same macro tailwinds as the consensus winners above. Several of the names we discuss are actually down on the year despite the structural backdrop. In our view, the set is available at attractive market quotations relative to underlying business value.

The gold equity opportunity has structurally closed under our criteria.

Every senior producer (Newmont, Agnico, Barrick), royalty company (Franco-Nevada, Wheaton, Royal Gold, Sandstorm, Triple Flag, Osisko), and mid-tier producer we screened is up more than 50 percent since December 2024. Investors who want gold exposure should pursue it via physical bullion rather than equity, since the equity now carries both the gold price beta and a sentiment premium.

The 60/40 portfolio remains structurally broken in the same way.

The 10-year Treasury spiked to 4.46 percent on March 27, 2026 as equities fell and the dollar tracked oil rather than acting as safe haven. The replacement diversifier is real assets with cash yield: midstream, royalty trusts, farmland, infrastructure, and reinsurance with hardening rates.

Key Themes and Ideas

• Theme 1: North American Energy Sovereignty

• Theme 2: Nuclear and AI Power Demand

• Theme 3: Defense and Asymmetric Warfare

• Theme 4: Agriculture and Food Security

• Theme 5: LNG Shipping

• Theme 6: Industrial Infrastructure and Grid

• Theme 7: Reinsurance and War-Risk Premium

• Theme 8: Inflation-Hedged Real Assets

• Things to Avoid or Approach with Caution

Theme 1: North American Energy Sovereignty

The Hormuz closure has revealed how dependent Europe and the Asian developed importers are on Atlantic-basin and Pacific-basin LNG and on non-Gulf crude. The structural winners are the US Gulf Coast LNG export complex, the pipelines that feed it, and the upstream gas producers that supply the molecule. The Permian crude royalty trade (Texas Pacific Land, Diamondback Energy) has already moved hard; we now express this theme through the gas-LNG export chain, where multiple names are still attractively priced.

Read more

SpaceX has now priced, and the largest IPO in history is no longer hypothetical. The offering has been set, unusually early, at $135 per share, which values the equity at roughly $1.8 trillion, with a final price expected Thursday and the listing scheduled for Friday, June 12. One thing should be clear at the outset: I have the highest respect for Elon Musk as an entrepreneur; on numerous occasions he has achieved the seemingly impossible. His companies have propelled humanity forward and promise to do so in the future.

I want to be careful about what I am and am not saying. I have no quarrel with what SpaceX might be worth a decade from now; for all I know the answer is enormous. The question that matters for anyone buying on the first day is what the company is worth today, on the strength of its actual financials, its actual markets, and what it actually does. Measured that way, $1.8 trillion reads less like a valuation than like a number the offering has manufactured, and a great deal of machinery has been assembled to bring retail investors into it at the open. Three pieces we are featuring in this week’s deck sharpened how I am thinking about the week ahead, and none of them requires a view on Mars or artificial general intelligence to be useful.

The first piece, a thought exercise entitled The Stock Market Crash of 2027 by James Emanuel, Mauricio Heck, and Hugo Navarro, presents a structural argument rather than a forecast. Three mega-IPOs, SpaceX, OpenAI, and Anthropic, are arriving inside roughly six months and together reaching toward $4 trillion or more, with Anthropic’s recent private round alone near $1 trillion. The dollars to absorb that supply have to come from somewhere, and the likeliest source is investors trimming the mega-cap technology positions they already hold, since these listings sit in the same bucket. In a market this concentrated in a handful of technology names, that rotation could pull the broad indices down on its own.

Two design features make the setup more fragile still. SpaceX is bringing only a small fraction of its implied value to market at the start, reportedly 3% to 5% of the equity, which manufactures the scarcity needed to support a price like this in the first place. And the lock-up is not a single 180-day cliff but a staggered series of releases, so across the back half of the year the float swells in tranches and the early scarcity reverses into a flood. The historical rhymes are not comforting. Facebook fell by roughly half after its 2012 lock-up rolled off, and the dot-com unwind was a rolling sequence of expirations rather than a single break. The investors most exposed to that sequence are the ones buying at the open, who pay the engineered price and then inherit the unlock calendar.

The second piece I include with real respect, and with a disagreement I want to state carefully. Aswath Damodaran revisited his SpaceX valuation once the prospectus was public and arrived at an equity value of about $1.3 trillion, roughly $500 billion, or close to 30%, below the $135 offering. Damodaran is the dean of valuation, and his practice of publishing his full work, spreadsheet included, for anyone to inspect and challenge is a genuine service to the investing community. On the headline point he and I do not actually disagree: even his estimate, which already leans on generous assumptions about the next decade, still lands well below the offering price, which is another way of saying there is no margin of safety for an IPO buyer.

Where I hesitate to follow is on the prior question, whether a business like this can be valued today with that degree of precision at all. The model is anchored on revenue figures for 2036 and on operating margins assumed for businesses, the AI line above all, that have very little public track record in the very markets said to make up most of the company’s opportunity. Damodaran himself flags the prospectus’s claim of a $28 trillion total addressable market, $26 trillion of it in AI, as closer to fantasy than forecast; he then doubles his own AI revenue target, to $160 billion, while trimming the assumed AI operating margin from 45% to 25%, and that single pair of choices moves the answer by a wide margin. When inputs this distant and this uncertain carry most of the result, my own instinct is that the honest output is a range too wide to anchor a decision rather than a single figure. None of this is a criticism of his rigor, which is beyond question; it is a difference about how much any method, however careful, can ask of facts this thin. SpaceX plainly contains valuable businesses. The question is the price today, and on that, his figure and my skepticism point in the same direction.

The third piece, from Phil Bak, turns to the index providers, and it is the one that genuinely encouraged me. Under heavy pressure to fast-track these mega-IPOs into their benchmarks from day one, several providers, including Nasdaq and FTSE Russell, moved to accommodate. S&P Dow Jones did not. It held to its twelve-month seasoning requirement and its GAAP profitability requirement and declined to make an exception for size, on the principle that financial viability, seasoning, and float standards should not be waived “solely based on market capitalization.” For a company that lost roughly $5 billion last year, that means S&P 500 funds will not be forced to buy SpaceX at any price for a long time. I do not think the stakes here are small.

Bending the rules to pull an unseasoned, loss-making company into the major indices at a nosebleed valuation would hand the bill to the passive investors least able to absorb it, retirees among them, who never chose the position. That is an abdication of the duty those benchmarks owe the people who entrust them with their savings. S&P is the adult in the room, and if it holds the line it is plainly acting in investors’ interest. One caveat, drawn from the first piece: the Nasdaq-100 has moved the other way, reweighting low-float names at three times their float and dropping the minimum-float rule that would have kept SpaceX out, so the relief is real for S&P 500 holders and only partial for those who own the Nasdaq-100.

None of this is a prediction that the week ends badly. It may be the start of real volatility, or it may not, and we will know more in a few days. What I do believe is that the moment calls for minding the downside with the same care most investors have lately reserved for the upside. The encouraging part is that this does not require sitting on the sidelines. There are still corners of the market that are reasonably priced, and a few that are outright cheap, so it remains possible to stay fully invested while stepping well clear of the excess on display in this offering. We will reconvene next Monday, when much of this should be clearer. This week’s deck follows.

Feedback on the Monday Morning Briefing

“Most of what I monitor, all in one place. Great value add.” —Brad Lummis

“Loving these Monday briefings!” —Jon Bartel

“Tightly presented and easy to digest. I just spent 20 minutes going through it, and it’s helped to level set me for the week ahead.” —Michael Loftis

“A great piece and thoughtfully assembled.” —Brian Wolf

“I don’t think I have ever seen more valuable content in one place.” —Bill Coleman

“Worth its weight in gold.” —Shree Viswanathan

A few words on the format

The Briefing is designed to answer a deceptively simple question. If you were sitting down before the weekly market open, as an investor rather than a trader, what would you want in front of you?

Each week, the Briefing walks through four parts.

• Weekly Review & Outlook covers equity performance, sector moves, the earnings just reported, and the earnings coming up, alongside curated editorial highlights from our Weekly Inspiration newsletter.

• Idea Generation surfaces candidates from screens we run: biggest decliners, names near 52-week lows, low multiples, high FCF yields, spinoffs, activist situations, buybacks, short interest, and more.

• Market Valuation & Positioning steps back to the index level: the Buffett Indicator, aggregate multiples versus history, S&P 500 concentration, equal-weight versus cap-weight, and long-run factor returns.

• Macro & Fixed Income rounds out the picture with rates, credit spreads, the Fed balance sheet, the dollar, labor, regional PMIs, and housing.

The new issue is attached below. We welcome your suggestions and ideas as we refine the format week to week.

Read more

Meet like-minded investors at the Latticework 2026 summit in Chicago.

MOI Global members, scroll down for member news and updates.

11 Articles We Are Reading This Week

1. IPO wave - Mauricio Heck, James Emanuel, et al.

2. IPO info leakage - Francine McKenna

3. SpaceX valuation - Aswath Damodaran

4. S&P holds the line - Phil Bak

5. AI consumer surplus - Dan Rasmussen

6. Alphabet’s capital pivot - Ben Thompson

7. Evaluating an economy - Scott Sumner

8. World models taxonomy - Fei-Fei Li

9. Global macro long-short - Ray Dalio

10. LLMs and consciousness - Erik Hoel

11. Hedging the AI-pocalypse - Andrew Walker

Subscribe now

Letters, Presentations, Writeups…

Industry Insights: ai | art | banking | commercial real estate | defense | gaming | insurance | insurance brokers | crude oil | semiconductors | software

Geographic Insights: Europe | Japan | Taiwan | UK & Ireland

Letters: Aikya | Durkin | Long Cast | Merrill | Tweedy | Voss | Vanguard

Company Insights: Amadeus | Bristol Myers Squibb | EchoStar | Ferrari | Lennar | Limoneira | MasterBrand | Paragon Banking | Rollins

Company Slides: AEO | Apollo | Donaldson | Gap | GitLab | HPE | Lufthansa | Macy’s | Medtronic | SAIC | Samsara | Solaris | Ulta Beauty | VersaBank

5 Videos We Are Watching This Week

1. LSE - Ray Dalio on his investment principles, debt cycles, geopolitics

2. Re:signal - Rory Sutherland on AI, trust collapse and corporate growth

3. Ed Yardeni - Fear of missing out versus fabulous earnings momentum

4. Ronald Reagan - Jamie Dimon on the road ahead for the US economy

5. David Rubenstein - Jeff Aronson on private credit and macro outlook

5 Podcasts We Are Listening to This Week

1. All-In - Bill Ackman on his evolved investing philosophy and approach

2. Odd Lots - Lewis Hart on the hidden plumbing of commodity finance

3. Dwarkesh - Alex Imas and Phil Trammell on what is scarce after AGI

4. Knowledge Project - Mark Pincus on patterns from winning products

5. Invest Like the Best - Uber CEO Dara Khosrowshahi on AI and rides

Member Updates & Events

The following is a private section for MOI Global members. Membership is by invitation only. If you would like to become a member, be sure you are on the annual plan. It is the first step toward full membership.

Read more

Over the next couple of weeks the discipline of value investing runs headlong into the loudest event the public markets have seen in a very long time, if ever. SpaceX has filed for the largest IPO in history, with a listing reportedly targeted for June 12. OpenAI is expected to follow in September, and Anthropic is eyeing October. Together the three are seeking something on the order of $3.5+ trillion in combined market value. Three pieces shaped my thinking on what a long-term investor should make of this, and each comes at it from a different altitude.

Trung Phan reframes the SpaceX S-1 as an AI infrastructure story wearing a rocket costume. His word count of the filing tells the tale: “compute,” “xAI,” and “connectivity” now crowd the document alongside “launch.” The pitch is no longer rockets and Starlink alone. It is a $26.5 trillion AI total addressable market sitting beside a $1.6 trillion connectivity number and a $370 billion space number, with SpaceX positioning to own the compute. Phan is candid about the absurdity, invoking Matt Levine’s “Elon Markets Hypothesis,” the notion that some assets are priced not on their cash flows but on their proximity to Elon Musk. At $18.7 billion of 2025 revenue, which he pegs at roughly 93 times sales, that hypothesis is carrying an enormous load. What a value investor should notice is the structure of what is being sold: a narrative priced as though the option has already paid off.

Aurelion Research does the arithmetic the narrative would prefer you skip. They give the reframe its sharpest name, the AI Trojan Horse, a compute business carried to market inside a rocket company. Their sum-of-the-parts is the most useful work I read this week. Strip away the story, and Starlink, the AI segment, and the launch business support a valuation near $700 billion (still generous, in my opinion). To bridge from there to $1.75 or $2 trillion, you have to assign close to a trillion dollars to space data centers that do not yet exist and to a Mars optionality that may never produce revenue. Their cross-IPO comparison is equally clarifying. Of the three, OpenAI carries the worst loss profile and the deepest dependence on someone else’s compute, while Anthropic shows the fastest revenue growth, the earliest path to break-even, and the cheapest forward multiple. Aurelion states plainly that they will own none of the three, and that their own book offers better risk and reward. That is the right instinct. Relative value is real, but the least expensive of three expensive things is a ranking, not a margin of safety.

Ben Thompson lands on the most philosophically honest framing. He says outright that the filing cannot be justified by any financial model, calls the numbers absurd, and notes that growth slowed even as the xAI acquisition tipped the company into a $4.9 billion loss on $5.1 billion of AI research expense, money spent building a model that sits in fifth place. And yet he argues that orbital data centers are plausible, that SpaceX could become the dominant supplier of the world’s marginal compute, and that this alone might be enough. The rumored IPO valuation embeds extraordinary risk.

The three pieces agree on the one fact that matters and disagree only on whether it should trouble you. Phan, Aurelion, and Thompson all conclude that current financials cannot support the proposed valuations, with SpaceX being the most egregious and Anthropic being the most palatable. Phan and Thompson lean on the Musk reality-distortion field, the Elon Markets Hypothesis, and the Tesla precedent, to argue the price may hold regardless. Aurelion does the work to show exactly how much hope is embedded in the number, and then declines to participate.

This is the capital cycle in its most public form, and the frames from last week apply here without modification. The AI buildout is cresting, the spending curve is steep, and the most celebrated names are being floated at the very top of it. Paying up at that moment, on margins and multiples that are stretched at the same time, is the error the historical record warns against most consistently. None of this makes the businesses bad. Starlink is one of the finest assets to reach the public markets in a decade, and SpaceX’s launch economics are a genuine moat. It means the edge for a long-term investor lies in the willingness to decline mispriced hope, to read this IPO wave as a sentiment marker rather than a shopping list, and to hunt instead in the corners where fear, not hope, is the thing being priced in.

A bit of feedback

“Most of what I monitor, all in one place. Great value add.” —Brad Lummis

“Loving these Monday briefings!” —Jon Bartel

“Tightly presented and easy to digest. I just spent 20 minutes going through it, and it’s helped to level set me for the week ahead.” —Michael Loftis

“A great piece and thoughtfully assembled.” —Brian Wolf

“I don’t think I have ever seen more valuable content in one place.” —Bill Coleman

“Worth its weight in gold.” —Shree Viswanathan

A few words on the format

The Briefing is designed to answer a deceptively simple question. If you were sitting down before the weekly market open, as an investor rather than a trader, what would you want in front of you?

Each week, the Briefing walks through four parts.

• Weekly Review & Outlook covers equity performance, sector moves, the earnings just reported, and the earnings coming up, alongside curated editorial highlights from our Weekly Inspiration newsletter.

• Idea Generation surfaces candidates from screens we run: biggest decliners, names near 52-week lows, low multiples, high FCF yields, spinoffs, activist situations, buybacks, short interest, and more.

• Market Valuation & Positioning steps back to the index level: the Buffett Indicator, aggregate multiples versus history, S&P 500 concentration, equal-weight versus cap-weight, and long-run factor returns.

• Macro & Fixed Income rounds out the picture with rates, credit spreads, the Fed balance sheet, the dollar, labor, regional PMIs, and housing.

The new issue is attached below. We welcome your suggestions for improvement as we refine the format week to week.

Read more

Meet like-minded investors at the Latticework 2026 summit in Chicago.

MOI Global members, scroll down for member news and updates.

11 Articles We Are Reading This Week

1. IPOs - Aurelion Research

2. SpaceX - Ben Thompson

3. Bottlenecks - Anh Hoang

4. AI disruptees - Charles Carter

5. Electricity bills - Nic Carter

6. Dumbo could fly - Erik Hoel

7. After automation - Dan Shipper

8. Open source future - Bill Gurley

9. Geology to geography - Will Thomson

10. 250-year-old company - Eric Markowitz

11. AI and the material world - Alex Duffy

Subscribe now

Letters, Presentations, Writeups…

Industry Insights: artificial intelligence | banking | data centers in space | energy | oil and gas | semis | software | transport infrastructure

Geographic Insights: Australia | Europe | Germany | Japan | UK | Ukraine

Letters: 1949 Value Advisors | Crossroads Capital | Graham & Doddsville | Hayden Capital | J.P. Morgan | LVS Advisory | Merrill | SaltLight

Company Insights: Alico | Baker Hughes | Barratt Redrow | Bristol Myers | dLocal | DoorDash | Gibraltar Industries | Pure Cycle | Take-Two | Zoetis

Company Slides: Abercrombie | Agilent | CIBC | Costco | CSW Industrials | Dick’s | Dollar Tree | EasyJet | HP | Kohl’s | NetApp | Salesforce | Snowflake

5 Videos We Are Watching This Week

1. a16z - Apollo’s Marc Rowan on private markets and software repricing

2. Sohn - Bellas, Marques and Rosen on long-term fundamental investing

3. CNN - Rivian’s RJ Scaringe and OpenAI’s Bret Taylor on AI and jobs

4. Bill Gurley - McConaughey on careers, passion and long-term success

5. My First Million - Mohnish Pabrai on lessons that save investors time

5 Podcasts We Are Listening to This Week

1. How I AI - Felix Rieseberg of Anthropic on how he uses Cowork mode

2. David Senra - Rick Rubin on finding your life's work and creative path

3. In Good Company - Prosus CEO Bloisi on why AI is still under-hyped

4. Knowledge Project - Lessons from Hyundai’s founder Chung Ju-yung

5. Invest Like the Best - Dan Loeb on his 30-year evolution in investing

Member Updates & Events

The following is a private section for MOI Global members. Membership is by invitation only. If you would like to become a member, be sure you are on the annual plan. It is the first step toward full membership.

Read more

Join the Latticework AI Bootcamp and progress at your own pace. Participation is open to members and paid subscribers.

A note before we begin: This is the last lesson in a 16-lesson experiment. I have been doing every lesson alongside you, on the same tools, with the same constraints. Some lessons have landed cleanly. Some have lead to dead ends and needed rework. We’ve figured out what works, together.

If you are catching up, here’s what came before today’s lesson:

• Intro: Build Your Own Investment Idea Engine

• Lesson 1: How LLMs Work, and How to Defend Against Hallucinations

• Lesson 2: Prompt Patterns That Outperform Casual Prompting

• Lesson 3: Tools, Agents, and Structured Output

• Lesson 4: SEC EDGAR, the Primary Source

• Lesson 5: FRED Macro and Sector Data, the Free Read

• Lesson 6: FMP API Key and the First Checked Data Pull

• Lesson 7: Other Data Sources, and Idea Engine Formats

• Lesson 8: Spaces, Projects, and Connectors

• Lesson 9: Memory, Sub-Agents, and Parallel Research

• Lesson 10: Wide Research for Screening at Scale

• Lesson 11: Iterating Prompts and Structured Idea Write-Ups

• Lesson 12: Claude Code, and Building Your First Useful Script

• Lesson 13: Git, GitHub, and a Real Screener on FMP

• Lesson 14: Layer On EDGAR, FRED, and a Scoring Layer

• Lesson 15: Engine Spec, Narrative Layer, and End-to-End Run

Why This Matters for Investors

During most of our AI bootcamp, the engine has been a thing we ran by hand. Yesterday it became a pipeline: one driver, one dated CSV, one dated Markdown deliveries file, a YAML front matter block recording the spec version, the config version, the FRED observation dates, the EDGAR user-agent string, and the FMP endpoints used. The pipeline runs cleanly when we sit at the keyboard and type the command.

Today the engine stops needing us to start it. It runs on a clock. The output lands in the channel of our choice (e.g., our email inbox). The repository has a README that we can open in three months without re-deriving today’s reasoning. The first scheduled run produces an artifact we read critically, any issues get fixed in the My Analyst prompt file or the scoring config rather than in the script, and the retrospective writes down what surprised us while building and what we will change over the next three months or so. By the end of today the engine is durable enough for the future, not just usable in the present.

The reason this is the right place to end the bootcamp is the same reason we have been writing things down throughout. The engine that survives the test of time is not the engine with the cleverest scoring layer. It is the engine whose owner can still explain, three months in, why each box exists and how to change it. The durability is in the spec, in the diagram, in the My Analyst prompt file, in the YAML scoring config, in the .env file, and in the README we write today. The scoring weights will move. The exclusions will sharpen. The narrative questions will mature. The infrastructure underneath, version-controlled, source-grounded, parameter-driven, is what we keep. The discipline that keeps the engine trustworthy outlives the bootcamp.

There is one more reason today matters. The first lesson asked us to learn when an AI model is being merely fluent versus when it is being right. Every step since has been a small repetition of that question, applied to a different layer: defending against hallucinations, grounding every claim in a source, naming provenance, refusing to commit secrets, naming a user-agent string on every SEC.gov call, separating filter from score, requiring the narrative layer to earn the right to be investigated rather than recommend action. The retrospective at the end of today is where we read back the answer in our own voice. The verification habit is what made the engine worth running in the first place.

Let’s launch into today’s lesson.

Read more

Join the Latticework AI Bootcamp and progress at your own pace. Participation is open to members and paid subscribers.

A note before we begin: This is the 15th lesson in a 16-lesson experiment. I am doing every lesson alongside you, on the same tools, with the same constraints. Some lessons will land cleanly. Some will lead to dead ends and need rework. We’ll figure out what works, together.

If you are catching up, here’s what came before today’s lesson:

• Intro: Build Your Own Investment Idea Engine

• Lesson 1: How LLMs Work, and How to Defend Against Hallucinations

• Lesson 2: Prompt Patterns That Outperform Casual Prompting

• Lesson 3: Tools, Agents, and Structured Output

• Lesson 4: SEC EDGAR, the Primary Source

• Lesson 5: FRED Macro and Sector Data, the Free Read

• Lesson 6: FMP API Key and the First Checked Data Pull

• Lesson 7: Other Data Sources, and Idea Engine Formats

• Lesson 8: Spaces, Projects, and Connectors

• Lesson 9: Memory, Sub-Agents, and Parallel Research

• Lesson 10: Wide Research for Screening at Scale

• Lesson 11: Iterating Prompts and Structured Idea Write-Ups

• Lesson 12: Claude Code, and Building Your First Useful Script

• Lesson 13: Git, GitHub, and a Real Screener on FMP

• Lesson 14: Layer On EDGAR, FRED, and a Scoring Layer

Why This Matters for Investors

For the last three lessons the engine has been adding plumbing. We have a private repository, a .env file that holds the FMP key and the FRED key, a screener that hits FMP for fundamentals, an EDGAR layer that stamps every row with the most recent 10-K filing date, a FRED layer that stamps every row with five dated macro observations, a scoring config in YAML, and a composite 0-to-100 score column that ranks the names. The CSV is real and the columns audit cleanly. What we do not yet have is a written-down answer to the question, what is this thing actually trying to do for us, and how would future-us recognize when the engine is helping versus when it is just producing output. Today we write that down.

The one-page specification is the lesson’s first deliverable, and it is the part of the bootcamp where the personalization becomes binding. Every member of this cohort has the same scaffolding by now. After today, every member’s engine starts diverging from every other member’s engine, because the spec encodes universe, exclusions, source hierarchy, valuation lens, quality bar, schedule, delivery channel, and the maximum number of ideas we want to see per run. A generic spec produces generic ideas. A personal spec produces ideas we recognize as the kind of ideas we are willing to spend more time on.

The second deliverable is the narrative layer. The composite score sorts the CSV; the narrative layer earns the top three rows the right to be investigated. For each of the three names at the top of today’s run, we ask Claude, with the My Analyst system prompt from Lesson 2, to draft a one-paragraph qualitative take that answers four questions in our voice: why the name surfaced, what evidence in the columns supports the surfacing, what could be wrong with the surfacing, and what the next human check should be. The narrative layer does not recommend action. It earns the right to be investigated. That distinction is the entire reason we are writing it down rather than trusting our memory of the column values.

By the end of today the engine produces, in a single end-to-end run, a sorted CSV with provenance columns, a one-page spec saved alongside the screener, and three short qualitative paragraphs that turn the top three rows into a triage pack we can read over coffee. Tomorrow’s lesson schedules the engine so it runs without us, writes a README so future-us can change parameters without re-deriving today’s reasoning, and runs the first retrospective on what surprised us in the first week of output.

Let’s launch into today’s lesson.

Read more

It’s clear why this piece may be relevant to long/short investors, but why should long-only investors care? The answer is simple: When the unraveling comes, it may affect a broader swath of equities than many of us may like to admit. Companies such as Alphabet, Microsoft, TSMC, and numerous others that may appear reasonably valued at the moment will not be immune to the negativity that comes with post-bubble disillusionment.

This piece lays out a path for how the bubble may unravel. While focused on the high-flying data storage darlings, the potential sequence of events is instructive to investors in adjacent tech segments as well.

Table: The Current Mania, Quantified

The storage and memory complex is exhibiting mania characteristics that exceed the equivalent storage/memory cohort of 1999-2000 on most measurable axes. SanDisk (SNDK) trades at $1,478, up 3,866% over one year and 523% year-to-date. Western Digital (WDC) trades at $484, up 865% over one year and 181% YTD. Seagate (STX) trades at $812, up 621% over one year and 195% YTD. The trio carries a combined market cap of $568 billion. The peak-valuation comparables of the prior cycle, including EMC at 125-130x P/E, NetApp at ~700x P/E, Rambus at ~200x P/E, and Cisco at 201x P/E, eventually drew down 90 to 97 percent over the following 24 to 30 months.

Today’s stocks trade at trailing P/Es of 26 to 76x and 14 to 17x trailing sales on share prices that have advanced 9 to 40 times in 12 to 15 months. The mechanical analogue is uncomfortably tight; the differences are real but narrower than the bull narrative suggests.

Table: Peak Valuations and Drawdowns, Prior Cycle

This article quantifies the current setup, anchors it to the 1999-2000 mania with hard valuation numbers, builds a structured catalyst framework for the bear case, and sketches an unravel scenario. The bull case has genuine merit: hyperscaler capex visibility ($725 billion committed for 2026 by the Big Four alone, up 77% year-over-year), oligopolistic industry structure (HDD is now a three-player market), supply discipline in NAND, and real prepaid multi-year long-term agreements (LTAs) from creditworthy customers. But at current valuations the margin of safety is negative on any normalized earnings framework, and several of the leading indicators of cycle peak are already flickering.

Read more

Join the Latticework AI Bootcamp and progress at your own pace. Participation is open to members and paid subscribers.

A note before we begin: This is the 14th lesson in a 16-lesson experiment. I am doing every lesson alongside you, on the same tools, with the same constraints. Some lessons will land cleanly. Some will lead to dead ends and need rework. We’ll figure out what works, together.

If you are catching up, here’s what came before today’s lesson:

• Intro: Build Your Own Investment Idea Engine

• Lesson 1: How LLMs Work, and How to Defend Against Hallucinations

• Lesson 2: Prompt Patterns That Outperform Casual Prompting

• Lesson 3: Tools, Agents, and Structured Output

• Lesson 4: SEC EDGAR, the Primary Source

• Lesson 5: FRED Macro and Sector Data, the Free Read

• Lesson 6: FMP API Key and the First Checked Data Pull

• Lesson 7: Other Data Sources, and Idea Engine Formats

• Lesson 8: Spaces, Projects, and Connectors

• Lesson 9: Memory, Sub-Agents, and Parallel Research

• Lesson 10: Wide Research for Screening at Scale

• Lesson 11: Iterating Prompts and Structured Idea Write-Ups

• Lesson 12: Claude Code, and Building Your First Useful Script

• Lesson 13: Git, GitHub, and a Real Screener on FMP

Why This Matters for Investors

In the last lesson the engine started looking like a usable tool. We have a private repository, a dot env file that holds the FMP key, a dot gitignore file that protects it, and a working FMP screener that produces a CSV of names that pass our hard criteria.

Today we do three things on top of that scaffolding, and each one is a step from “screen” toward “engine we trust.”

• We add an EDGAR filing-freshness layer so the screen knows when each name last filed an annual report.

• We add a FRED macro-context layer so the screen carries the dated values of the rates, inflation, and labor series we actually want in the same row as the company data.

• We add a 0-to-100 scoring layer on top, with the weights living in a config file we can edit without rewriting the script.

The non-coder’s instinct on a day like today may be to flinch at the word scoring. We are not building a quant model. We are not estimating expected returns. We are making our preferences explicit, in numbers we can audit, on a screen we can defend. The default weights are a starting point, not a recipe. The point of putting the weights into a config file is the same point as putting the API key into the dot env file: the script becomes a thing we can change without fear, because the part that changes most often is the part that is easiest to find.

The filing-freshness layer is the most underrated of the three. A screen that does not know how old each name’s most recent 10-K is can quietly recommend a company whose disclosures stopped in 2024. That is not a value find. That is a name we should be running an extra audit on, not adding to the engine’s output without comment. The FRED layer is the smallest of the three but the one that earns its keep the fastest: the day the 10-year yield breaks out of the range we built our valuation assumptions around, the screen’s macro columns surface that fact in the same row as the company we are evaluating, with the observation date attached so we can audit the value against the FRED page directly.

By the end of today, the CSV we wrote in the last lesson has two new columns, one new ranking column, and a tiny config file alongside the screener that names what we think a good investment idea looks like.

Let’s launch into today’s lesson.

Read more

Join the Latticework AI Bootcamp and progress at your own pace. Participation is open to members and paid subscribers.

A note before we begin: This is the 13th lesson in a 16-lesson experiment. I am doing every lesson alongside you, on the same tools, with the same constraints. Some lessons will land cleanly. Some will lead to dead ends and need rework. We’ll figure out what works, together.

If you are catching up, here’s what came before today’s lesson:

• Intro: Build Your Own Investment Idea Engine

• Lesson 1: How LLMs Work, and How to Defend Against Hallucinations

• Lesson 2: Prompt Patterns That Outperform Casual Prompting

• Lesson 3: Tools, Agents, and Structured Output

• Lesson 4: SEC EDGAR, the Primary Source

• Lesson 5: FRED Macro and Sector Data, the Free Read

• Lesson 6: FMP API Key and the First Checked Data Pull

• Lesson 7: Other Data Sources, and Idea Engine Formats

• Lesson 8: Spaces, Projects, and Connectors

• Lesson 9: Memory, Sub-Agents, and Parallel Research

• Lesson 10: Wide Research for Screening at Scale

• Lesson 11: Iterating Prompts and Structured Idea Write-Ups

• Lesson 12: Claude Code, and Building Your First Useful Script

Why This Matters for Investors

In the last lesson we put a working script into a folder. Today we do two things that turn that folder into the start of an actual idea engine. We put it under version control with Git, push it to a private GitHub repository, and we add the first real screener on top of the FMP API. The two changes are not as different as they look. Version control is what lets us change the screener tomorrow without losing today’s. A private repository is what lets us keep credentials out of our screens, our screens out of our chats, and our chats out of anyone else’s hands. The screener is what we will be editing tomorrow and the day after that.

The non-coder’s instinct here is to flinch at the word Git. We are not going to learn Git. That is not what today is about. The goal is narrow: keep a private versioned record of the engine so we can change it without fear, and keep our API keys out of the repository by default. Those two outcomes are worth the thirty minutes even if we never type a Git command again on our own. Claude Code does the typing. We read the plan, we read the diff, and we approve.

The screener side of today is where the engine starts to feel like a tool we will actually use. FMP gives us fundamentals, ratios, ownership, and the screener endpoint in one place, on a free tier that is enough to complete this lesson. We are going to ask Claude Code to write a small Python script that hits the FMP stock-screener endpoint with our criteria (in the default scaffold: US-listed, market cap between five hundred million and five billion dollars, return on invested capital above fifteen percent over the trailing five years, and insider ownership above ten percent), and saves the output as a CSV in the project folder. The criteria are ours to change. The script is the part we keep.

By the end of today we will have a private repository, a clean separation between code and credentials, the EDGAR risk-factors script from yesterday committed with a plain-English message, and a first screener output we can sanity-check against the names we already know.

Let’s launch into today’s lesson.

Read more

This week’s briefing wrestles with a question that refuses to be settled: are we in an AI bubble, are we priced for perfection, or has the narrative already quietly inverted? Three pieces shaped my thinking, and each takes the question from a different angle.

Cullen Roche argues there is a “bubble in bubble calls.” His claim is empirical, not rhetorical. Corporate profits as a share of GDP sit near record highs, the AI trade is the driver, and in his framing high valuations are warranted as long as that chart stays elevated. To his credit, he acknowledges that when expectations surge the margin for error thins and sequence risk rises. Where I part company is on the conditional itself. Mean reversion in profit margins is one of the most reliable patterns in financial history. Treating the current margin level as the sustainable baseline, and capitalizing it at premium multiples, is precisely the move that has historically marked the most punishing entry points. The time to pay up on multiples is when margins are depressed and earnings are understated, not when both are stretched.

Dan Rasmussen and Chris Satterthwaite sharpen that concern from a different angle. Invoking Mordecai Kurz’s “correlated beliefs” framework, they note that the semiconductor complex now trades at roughly 55x earnings, which implies 16.5% EPS growth for a full decade. About 75% of the present value of the global semi industry is derived from cash flows beyond year ten, in territory clouded by deep uncertainty. Microsoft itself disclosed on its Q1 2026 call that two-thirds of recent capex went to CPUs and GPUs with three to five year useful lives, so the hamster wheel is spinning fast. Shiller’s CAPE sits at 42x, within a whisker of the 44x peak in late 1999. The conclusion, that those who underwrite the first $5 to $10 trillion of AI capex may not capture the hoped-for returns, lines up with everything we know about prior capital cycles, and it is the right caution to hold in mind.

Logan Shearer offers what I found perhaps the most actionable frame. He calls it Bizarro World. Software names that were left for dead two or three years ago, with Adobe collapsing from 60x EV/EBIT to roughly 11x, now sit at depressed multiples with stable growth and stable margins, while the picks-and-shovels beneficiaries of the AI buildout have enjoyed simultaneous growth, margin expansion, and multiple expansion. Shearer’s instinct is classically value: buy mispriced fear, avoid mispriced hope. He warns against the capital cycle error of paying up at the top of the spending curve, and reminds us that tacit knowledge, the kind that separates a hobbyist from a professional, remains a moat the LLMs have not yet breached. That is exactly the orientation I would commend to long-term value investors today.

Rasmussen and Shearer point in compatible directions. Verdad sets the macro guardrail by quantifying just how much future growth is already in the price of the AI complex. Shearer turns that observation into a hunting ground by identifying corners of the market where fear has been mispriced and the capital cycle is working in the patient investor’s favor. Roche is correct that the AI trade has real fundamentals underneath it, and he is right to push back on lazy 1999 analogies, but the conditional at the heart of his argument, that today’s elevated margin level will persist, is precisely the assumption the historical record warns against.

The rest of this week’s briefing brings the data: activist 13D filings, fresh spin-offs, micro-cap Tiny Titans, names trading near 52-week lows, and the cheapest broad-universe stocks on P/TBV and three-year EV/EBITDA, alongside the highest three-year FCF and earnings yields. Read it with these three frames in mind.

A few words on the format

The Briefing is designed to answer a deceptively simple question. If you were sitting down before the weekly market open, as an investor rather than a trader, what would you want in front of you?

Each week, the Briefing walks through four parts.

• Weekly Review & Outlook covers equity performance, sector moves, the earnings just reported, and the earnings coming up, alongside curated editorial highlights from our Weekly Inspiration newsletter.

• Idea Generation surfaces candidates from screens we run: biggest decliners, names near 52-week lows, low multiples, high FCF yields, spinoffs, activist situations, buybacks, short interest, and more.

• Market Valuation & Positioning steps back to the index level: the Buffett Indicator, aggregate multiples versus history, S&P 500 concentration, equal-weight versus cap-weight, and long-run factor returns.

• Macro & Fixed Income rounds out the picture with rates, credit spreads, the Fed balance sheet, the dollar, labor, regional PMIs, and housing.

A bit of feedback

“Loving these Monday briefings!” —Jon Bartel

“Most of what I monitor, all in one place. Great value add.” —Brad Lummis

“A great piece and thoughtfully assembled.” —Brian Wolf

“Tightly presented and easy to digest. I just spent 20 minutes going through it, and it’s helped to level set me for the week ahead.” —Michael Loftis

“Worth its weight in gold.” —Shree Viswanathan

The new issue is attached below. We welcome your suggestions for improvement as we refine the format week to week.

Read more

Meet like-minded investors at the Latticework 2026 summit in Chicago.

MOI Global members, scroll down for member news and updates.

11 Articles We Are Reading This Week

1. Bubble - Cullen Roche

2. Perfection - Dan Rasmussen

3. Bizarro world - Logan Shearer

4. Cheap evidence - Mauricio Heck

5. Homo agenticus - Rohit Krishnan

6. Global fertility crisis - Derek Thompson

7. Moats and value traps - Kai Wu

8. Commencement speech - Ray Dalio

9. China’s optimism and fear - Zilan Qian

10. Counterpoint to zero-sum thinking - Eric Markowitz

11. Swiss watches and the dream equation - Trung Phan

Subscribe now

Letters, Presentations, Writeups…

Industry Insights: banking | consumer health | gaming | meatpacking | neo-clouds | platinum group metals | semiconductors | software | technology

Geographic Insights: China | Europe | Ukraine | US

Letters: First Eagle | Longleaf | Merrill | Robotti | Silver Beech | Tactile

Company Insights: Brambles | Enogia | Establishment Labs | GrafTech | MercadoLibre | Monzo | Nvidia | Sato Foods | Sea Ltd | ServiceNow

Company Slides: Borr Drilling | Deere | Eagle Materials | Enersys | Golar LNG | Keysight | Nvidia | Ryanair | Star Bulk | Take-Two | Walmart | Workday

5 Videos We Are Watching This Week

1. Norges - Rubenstein, Griffin, O’Leary, and Grew discuss global markets

2. Founders - Strauss Zelnick, the man who built Grand Theft Auto 6 hit

3. Bloomberg - Jamie Dimon on hiring more for AI, fewer bankers at JPM

4. Prof G Markets - Aswath Damodaran on the AI rally and IPO outlook

5. In Good Company - Pfizer CEO Bourla on AI in medicine and China

5 Podcasts We Are Listening to This Week

1. Acquired - Ben Gilbert and David Rosenthal trace Vanguard’s history

2. Squawk Pod - Jeff Bezos with Andrew Ross Sorkin at Blue Origin HQ

3. How Leaders Lead - Mark Smucker on how iconic brands stay relevant

4. Invest Like the Best - Gavin Baker on watts, wafers, future of AI infra

5. Behind the Balance Sheet - Jeremy Grantham on bubbles and crashes

Member Updates & Events

The following is a private section for MOI Global members. Membership is by invitation only. If you would like to become a member, be sure you are on the annual plan. It is the first step toward full membership.

Read more

Thesis summary:

Oro is a Tokyo-listed software company whose high-quality Cloud Solutions ERP franchise is obscured by a struggling Marketing Solutions segment, a cash-heavy balance sheet, and a resulting conglomerate discount.

Jiro sees an opportunity to own a founder-led Japanese compounder at a bargain price, with catalysts tied to a potential Marketing Solutions divestiture and further capital-allocation improvement. Founders Atsushi Kawata (CEO, 37.67% ownership) and Yasuhisa Hino (17.02% ownership) started the company in 1999 and, per Jiro, engage openly on shareholder-return and portfolio-optimization topics.

• Cloud Solutions, which serves project-based businesses such as IT services, system development, advertising, and consulting, contributes 68% of revenue and 94% of operating profit at segment OPM above 40%. FY2025 segment revenue was 5.6B yen and operating profit 2.5B yen, with 10-year revenue and profit CAGRs of 14.9% and 20.8%. Growth is being driven by new customer additions of 80–90 per year against a base of 1,100 clients, rising licenses per client, NRR of 116%, churn of 0.3%, and the January 2023 shift to SaaS-only contracts, which Jiro expects to support further margin expansion. Management targets 4,000 clients against an estimated 44,000 domestic targets.

• Marketing Solutions, concentrated on Nissan and Aeon, generated 2.6B yen of revenue but only 0.1B yen of operating profit in FY2025, with a 10-year profit CAGR of negative 5.6%. Jiro believes a sale for 2–3B yen is plausible, removing a visible drag and allowing the market to reprice the Cloud franchise on its own.

Capital allocation has inflected. Net cash has grown from 1B to 10B yen, or 70% of total assets. Buybacks have stepped up from 500M yen in 2024 to 1B yen in 2025 and another 1B yen announced for 2026, with payout exceeding 40% and a 2.5% dividend yield. Jiro believes Oro should distribute more than 100% of profit going forward given the modest capital needs of the business.

Oro recently traded at 6.7x EV/EBITDA and 15.7x P/E, or roughly 10x ex-cash, a discount to Japanese software peers trading at a median 13.5x EV/EBITDA and 23.4x P/E. VARECS’ three-year base case assumes a 7.6x EV/EBIT multiple on FY2028E EBIT of 4.6B yen and a 10% share-count reduction, yielding ~66% upside from a recent 1,900 yen. A spin-off plus multiple expansion to 10x implies 104% upside; combined with a 25% buyback, upside reaches 145%.

Disclaimer

Asian Investing Summit 2026 was held from April 14-21, 2026. The content of this website is not an offer to sell or the solicitation of an offer to buy any security. The content is distributed for informational purposes only and should not be construed as investment advice or a recommendation to sell or buy any security or other investment, or undertake any investment strategy. There are no warranties, expressed or implied, as to the accuracy, completeness, or results obtained from any information set forth on this website. BeyondProxy’s officers, directors, employees, and/or contributing authors may have positions in and may, from time to time, make purchases or sales of the securities or other investments discussed or evaluated herein.

Slides

Oro Presentation

2.04MB ∙ PDF file

Download

Download

Let’s take a closer look.

Read more

Join the Latticework AI Bootcamp and progress at your own pace. Participation is open to members and paid subscribers.

A note before we begin: This is the twelfth lesson in a 16-lesson experiment. I am doing every lesson alongside you, on the same tools, with the same constraints. Some lessons will land cleanly. Some will lead to dead ends and need rework. We’ll figure out what works, together.

If you are catching up, here’s what came before today’s lesson:

• Intro: Build Your Own Investment Idea Engine

• Lesson 1: How LLMs Work, and How to Defend Against Hallucinations

• Lesson 2: Prompt Patterns That Outperform Casual Prompting

• Lesson 3: Tools, Agents, and Structured Output

• Lesson 4: SEC EDGAR, the Primary Source

• Lesson 5: FRED Macro and Sector Data, the Free Read

• Lesson 6: FMP API Key and the First Checked Data Pull

• Lesson 7: Other Data Sources, and Idea Engine Formats

• Lesson 8: Spaces, Projects, and Connectors

• Lesson 9: Memory, Sub-Agents, and Parallel Research

• Lesson 10: Wide Research for Screening at Scale

• Lesson 11: Iterating Prompts and Structured Idea Write-Ups

Why This Matters for Investors

So far we have asked assistants to do things for us. Today we start asking an assistant to build small tools that do those things on our behalf, on our schedule, without us in the room. This matters because everything we have built to date (the Spaces, the Projects, etc.) was research scaffolding. It is excellent at one-off questions and it improves our judgment in real time. What it cannot do is run while we sleep, screen overnight, hand us a triage pack at 7 a.m., or remember last week’s pull date when it builds this morning’s. That is what the next five lessons are for, and Claude Code is the workshop where we build it.

The temptation, when a non-coder opens a terminal, is to either freeze or pretend. Instead, we will walk the middle path: ask Claude Code to plan before it edits, ask it to explain every line in plain English, and refuse to keep any script we could not change later. The non-coder’s superpower is not that we suddenly write Python. It is that we know how to interrogate a thing we did not write until we trust it the same way we trust a sell-side model only after we have verified the inputs.

There is one more reason to take Claude Code seriously today rather than treating it as a developer’s toy. The investment-research artifacts we care about (a screener that runs daily, a watchlist monitor that pings us only when something material changes, a wide-research pass that produces a CSV every Friday morning) are exactly the kind of small, durable tools that hand-coding makes expensive and pair-programming with an assistant makes nearly free. The cost of building the first one used to be a developer relationship. The cost now is thirty minutes, a folder, and the willingness to read every line back.

By the end of today we will have Claude Code installed, a small Python script that pulls a recent 10-K from SEC EDGAR and prints the first 500 words of the risk-factors section for any ticker we choose, a line-by-line plain-English explanation of that script, and the start of a habit we will use for the rest of the bootcamp: plan first, smallest working version next, explain in English, keep only what we understand.

Let’s launch into today’s lesson.

Read more

Join the Latticework AI Bootcamp and progress at your own pace. Participation is open to members and paid subscribers.

A note before we begin: This is the eleventh lesson in a 16-lesson experiment. I am doing every lesson alongside you, on the same tools, with the same constraints. Some lessons will land cleanly. Some will lead to dead ends and need rework. We’ll figure out what works, together.

If you are catching up, here’s what came before today’s lesson:

• Intro: Build Your Own Investment Idea Engine

• Lesson 1: How LLMs Work, and How to Defend Against Hallucinations

• Lesson 2: Prompt Patterns That Outperform Casual Prompting

• Lesson 3: Tools, Agents, and Structured Output

• Lesson 4: SEC EDGAR, the Primary Source

• Lesson 5: FRED Macro and Sector Data, the Free Read

• Lesson 6: FMP API Key and the First Checked Data Pull

• Lesson 7: Other Data Sources, and Idea Engine Formats

• Lesson 8: Spaces, Projects, and Connectors

• Lesson 9: Memory, Sub-Agents, and Parallel Research

• Lesson 10: Wide Research for Screening at Scale

Why This Matters for Investors

In the last lesson we ran a twenty-name wide-research triage pass under one strict schema; and the schema, not the universe, was the thing that made the table honest. Today we take that lesson one layer deeper. A prompt is not a clever sentence. A prompt is a small operating procedure that has to behave the same way on the fifth ticker as it did on the first, on a sector we know cold and on one we do not, on a name with a clean filing history and one with a recent restatement. If we do not test it deliberately, we will keep noticing failures only when they show up inside an investment decision.

There is a second reason to slow down here. Most non-technical members get into trouble in the same way: they write a prompt that produces one good output on a ticker they already understand, conclude that the prompt works, and then are surprised when the same prompt produces a polished but unsourced write-up on a name they do not know. That second output is the dangerous one. It will be fluent, it will be plausible, and it will smuggle in assumptions about discount rates, terminal margins, share counts, or filing dates that the model invented because the prompt did not forbid it. Prompt iteration is how we close that gap.

The other half of today is the structured idea write-up itself. Research that does not land in a comparable shape is hard to compound. If every company note has a different structure, we cannot scan across them, cannot revisit our reasoning later, and cannot tell whether our process is improving. A five-bullet house-style write-up, generated from a tested prompt, gives us that comparability without flattening the thinking. And once the write-up exists, we have earned the right to run a simple reverse DCF on one name we know well, where the discipline is to separate market-implied assumptions from our own and to refuse to let the model invent inputs.

By the end of today we will have a small prompt-evals sheet with five rows, one documented failure pattern, one targeted prompt fix, and one reverse-DCF output whose inputs we have personally checked.

Let’s launch into today’s lesson.

Read more

Thesis summary:

Lycopodium is a Perth-based engineering and project delivery services firm serving the resources sector, founded in 1992 and listed on the ASX in 2004. Sidd and Kimi describe a 30-plus-year-old business that derives roughly 94% of revenue from miners in gold, copper, lithium, rare earths, uranium, titanium and mineral sands, with no equity raises since listing. The three founders — Rodney Leonard, Michael Caratti and CEO Peter De Leo — remain involved and collectively own about 30% of the equity, aligning management with outside shareholders. Market cap recently stood at about A$530 million (about US$365 million) at a share price of A$13.30.

The core of the thesis rests on Lycopodium’s predominantly EPCM (engineering, procurement and construction management) model, which Sidd and Kimi contrast with the lump-sum EPC model that dominates listed peers. In EPCM, the owner bears cost-overrun risk while Lycopodium earns a cost-plus professional services fee, making the business capital-light, free of balance-sheet exposure, and capable of generating 30%-plus ROE (FY25). Today roughly 80%-plus of revenue sits in EPCM, with the EPC-heavier exposure largely ring-fenced via a 40/60 JV with Monadelphous. Kimi emphasizes a “study-to-EPCM flywheel”: the firm runs 40-plus scoping, PFS, FS and DFS studies at any point, and miners who use Lycopodium for feasibility almost always award it the EPCM, which in turn feeds recurring optimization work. Committed contracts stood at A$415 million with a A$1.3 billion opportunity pipeline as of December 2025.

Competitive strengths include on-time, at-budget delivery track record, two-thirds of revenue from repeat clients, a client list that includes Newmont, Rio Tinto and Anglo American Platinum, and 30-plus years of metallurgical and process data that is not replicable by new entrants. About 57% of FY25 revenue came from Africa, an area of relative strength given few established peers. In 2024, Lycopodium entered the Americas through its acquisition of Argentine firm SAXUM, which management estimates expands its TAM by about 40%. Tailwinds include high gold prices (six-plus active gold EPCM projects), a looming copper deficit driven by the energy transition, and battery-mineral demand (lithium, nickel, graphite, rare earths). The principal risk, as highlighted by the 2013-2017 commodity downturn and reinforced by Kimi, is a mining capex cycle that compresses miners’ access to financing, which historically forced Lycopodium to take on EPC risk and produced its only loss year (FY15).

Revenue grew from A$162 million in FY21 to A$340 million in FY25 (about 20% CAGR), while NPAT grew from A$14 million to A$42 million (about 31% CAGR), with net income margins expanding from 8.8% to peaks of 14.5% before softening to 12.4% in FY25. The balance sheet carries A$79 million of net cash and zero debt. The payout ratio averaged in the high 60s until FY25, when management cut payout to 33% to fund SAXUM — an unusual attribute of a business that has compounded earnings while distributing two-thirds of profits. Over the 20 years since listing, revenue compounded at 8.2%, net income at 10.8%, dividends at 11.3%, and total shareholder return (including dividends reinvested) at 14.9% CAGR.

Kimi and Sidd frame valuation by projecting the past 20-year track record forward 20 years. Starting from a current market cap of A$531 million, they estimate cumulative dividends reinvested at 3% of A$1,795 million, terminal NPAT of A$262 million, and apply a deliberately conservative terminal P/E of 7x (compared with a trailing P/E of about 15x today and peer multiples of 17-40x trailing) for a terminal value of A$1,837 million. Total expected value of A$3,632 million implies an expected return of about 10.1% CAGR over two decades. The shares recently traded at a trailing P/E of 15x against a depressed FY25 earnings base; Kimi notes that low near-term earnings — with FY26 expected roughly flat to FY25 — are precisely what create the current opportunity as feasibility projects convert into EPCM delivery and SAXUM contributes to the Americas pipeline.

Disclaimer

Asian Investing Summit 2026 was held from April 14-21, 2026. The content of this website is not an offer to sell or the solicitation of an offer to buy any security. The content is distributed for informational purposes only and should not be construed as investment advice or a recommendation to sell or buy any security or other investment, or undertake any investment strategy. There are no warranties, expressed or implied, as to the accuracy, completeness, or results obtained from any information set forth on this website. BeyondProxy’s officers, directors, employees, and/or contributing authors may have positions in and may, from time to time, make purchases or sales of the securities or other investments discussed or evaluated herein.

Slides

Lycopodium Presentation

489KB ∙ PDF file

Download

Download

Let’s take a closer look.

Read more

Thesis summary:

Motilal Oswal is an India-based, vertically integrated capital markets firm with a $4.5 billion market cap and franchises across asset management, wealth management, broking, investment banking, alternates, and financial distribution. Gokul Raj presented the company at the Asian Investing Summit 2026 as a way to participate in the financialization of Indian household savings. Insiders own more than 75% of the equity, and the business has compounded book value per share at 21%+ in USD terms (25%+ in INR) over the past decade, with revenues, AUM and profits all up roughly 10x over that period. The model is capital-light: the firm IPOed once, has not raised additional equity, has executed two buybacks, and has generated FCF since inception, which has been recycled into a treasury book that earns a low-20s IRR and provides funding advantages for the operating businesses.

The structural thesis rests on India’s underpenetrated equity culture and a rare cluster of growth drivers: 7%+ GDP growth, demographics, formalization, an under-levered household and corporate balance sheet, and a regulatory regime that has built a fully digital, transparent capital markets infrastructure. Gokul Raj estimates cumulative Indian gross savings of roughly $47 trillion over 15 years, with $3-4 trillion potentially flowing into capital markets at current allocation rates. Domestic investors now own ~85% of the market and SIP flows have been a relentless monthly bid, even as foreign investors have sold roughly $50 billion. Indian equities have underperformed EM by 25%+ over two years, the median stock is down 50% from its highs, and Motilal has corrected with the market – which Gokul Raj views as the entry point.

The mix shift in earnings is at the core of the re-rating case. Asset management has scaled 3x in five years, with mutual fund SIP AUM up ~6x and incremental flow market share of 8-10% versus a 3% stock share. Alternates – PE, real estate and a newly launched private credit fund – has delivered 20%+ IRRs across vintages, allowing each successive fund to be 2-3x the size of the prior one, with ~60% of carry accruing to shareholders given heavy insider ownership of the listed entity. Private wealth, entered in 2016, has grown net revenues and AUM 4-5x in five years, runs at 50%+ margins at scale, and benefits from RMs that are still on average only ~3 years vintage. The legacy broking franchise, now positioned as a full-service wealth distribution channel, has absorbed share losses to discount brokers by consolidating the tail and has built a high-rated lending book (margin trading and LAS) that has delivered near-zero credit costs across cycles.

Quality of earnings has shifted from broking-led to wealth- and asset-management-led, which now contribute over 50% of operating PAT and are tracking toward 70-80% within two to three years. Annual recurring revenue is ~60% of consolidated revenues; firms that have crossed the 65-70% ARR threshold (e.g., 360 ONE) trade at 30-40x PAT. Blended ROE will keep rising as the housing finance unit – the only capital-heavy and historically problematic operation, now cleaned up at <1% GNPA and 12-14% ROE – is monetized via IPO or sale within two to three years. Founders have pledged ~$500 million (10% of holdings) to education-related philanthropy over the next 5-10 years; the resulting promoter dilution would lift the regulatory cap on buybacks and likely accelerate repurchases when shares are cheap. Succession is in place, with both founders’ sons in operating roles and ~$300 million of equity held by non-family insiders.

The shares recently traded at less than 14x trailing operating profit after tax and 3x book value, with the treasury (~20% of firm value, ~70% public equity / 30% alternates) carried at what Gokul Raj considers a conservative 20% holdco discount. Stripping out MTM noise that has cluttered reported P&L over the past six months, Gokul Raj argues the operating business can compound earnings at 15-20% over the next decade, with optionality from margin expansion, ARR mix shift, buybacks, and a multiple re-rating toward the 30-40x PAT range that the market awards to pure wealth and asset management franchises. The downside is largely tied to a prolonged Indian equity drawdown – a scenario in which earnings might be hit by ~10% on a worst-case basis, while the high-beta franchise would offer leverage to any subsequent recovery.

Disclaimer

Asian Investing Summit 2026 was held from April 14-21, 2026. The content of this website is not an offer to sell or the solicitation of an offer to buy any security. The content is distributed for informational purposes only and should not be construed as investment advice or a recommendation to sell or buy any security or other investment, or undertake any investment strategy. There are no warranties, expressed or implied, as to the accuracy, completeness, or results obtained from any information set forth on this website. BeyondProxy’s officers, directors, employees, and/or contributing authors may have positions in and may, from time to time, make purchases or sales of the securities or other investments discussed or evaluated herein.

Slides

Motilal Oswal Presentation

1.76MB ∙ PDF file

Download

Download

Let’s take a closer look.

Read more

Session summary:

Hoosik Min expects improving ROE across Korean companies as industrial structure and capital-market reforms take hold. The KOSPI rose ~110% from early 2025 to March 2026 (2,400 to 5,052 points). Pine targets businesses with strong cash generation and reinvestment-led growth, run by managements with sound capital allocation, acquired at undervalued or fair prices. Roughly 20-30% of Korean listings recently traded below 1.0x PBR. KOSPI’s ten-year ROE has ranged 6-12%; Hoosik estimates that if ROE reaches 10-15% and OP margin expands to 14-16%, PBR could re-rate from 0.9-1.5x to 1.5-2.5x.

Korea carries real cost-of-equity risks: the North Korean border sits ~50 km from Seoul, chaebol governance opacity in a market just ~2% of global capitalization, and a super-aged demographic with a 0.8 fertility rate. Offsetting factors are accumulating: May 2024 Value-Up guidelines, a July 2025 amendment expanding directors’ duty of loyalty to shareholders, January 2026 treasury-stock disclosure tightening, and pending rules requiring cancellation of treasury stock within one year. Net cash is ~25% of KRX market cap; Samsung Electronics, SK Hynix, and SK announced Q1 2026 buybacks of KRW 15.7, 13.5, and 5.0 trillion, respectively.

Pharma Research (214450), a medical-devices and cosmetics business, is anchored by the Rejuran brand (PN technology IP) in skin beauty. From 2020 to 2025, sales grew ~36% per year, net profit ~46%, and OP margin ran 30-40%; the stock rose 12x to KRW 403,000 at end-2025. After a recent correction, Hoosik views the shares as priced in a low range versus intrinsic value. Management emphasizes cash profits, has trimmed low-margin channels, and pursued small-scale M&A (Healer, Botox) for diversification. Risks include new entrants, compression of the 40% OP margin, and global-expansion costs.

Samyang Foods (003230), a Korean consumer-goods company, has ridden Buldak Ramen through five years of reinvestment-led growth. From 2021 to 2025, revenues grew ~29% per year, net profit ~42%, and OP margin expanded from 10% to 22%; the stock rose 13x to KRW 1,231,000 at end-2025. Exports moved from 10% of sales in 2015 to 80% in 2025. The Miryang factory (2020-2022) delivered operating leverage; a China facility starts in 2027. Global share is 3-4%. Risks include new competitors, trend durability, and a thin second growth engine.

Leeno Industrials (058470) supplies Pin & Socket components for semiconductor test equipment. Over 2011-2025, sales grew ~13% per year and net profit ~16%; the stock rose 35.8x over 15 years to KRW 60,500 at end-2025. By company estimate, Pin share runs 60-70% globally, with end clients including Apple, Qualcomm, TSMC, and Samsung Electronics. A new factory twice the size of the current one begins operations in 2026, extending capacity from customer-specific R&D to mass production. The principal risk is valuation.

Seers Technology (458870) is Hoosik’s newest case — an early-stage medical-equipment company focused on hospital operational efficiency. Consensus expects sales to grow ~78% per year and net profit ~99% over the next two years, with OP margin near 45% in 2026-2027E. Mobiecare (2020) is an AI arrhythmia diagnostic; ThynC (2024), ~90% of revenue, is a Bluetooth bedside monitoring gateway now expanding into the Middle East. Of Korea’s ~700,000 hospital beds, ~300,000 are addressable; ThynC sits near 10% penetration. Risks include low entry barriers, limited domestic TAM, and overseas software-security validation.

Hoosik frames the thesis as “South Korea Investment Season 2” — Season 1 being Warren Buffett’s 2002-2003 Korean purchases at 3-4x P/E after the IMF bailout. Season 2 centers on creative innovation and differentiated business. Capital-heavy industries are restructuring; shipbuilding, defense, and energy infrastructure have shifted from general-purpose to custom-order output, as HBM and server memory have in semiconductors. Korean brands are converting 20-30 years of cultural accumulation — PSY, BTS, Parasite, Squid Game, Han Kang’s 2024 Nobel — into price premium, with cosmetics exports compounding at a 16% CAGR. Hoosik likens the arc to Japan’s Value-Up program 14 years prior.

Disclaimer

Asian Investing Summit 2026 was held from April 14-21, 2026. The content of this website is not an offer to sell or the solicitation of an offer to buy any security. The content is distributed for informational purposes only and should not be construed as investment advice or a recommendation to sell or buy any security or other investment, or undertake any investment strategy. There are no warranties, expressed or implied, as to the accuracy, completeness, or results obtained from any information set forth on this website. BeyondProxy’s officers, directors, employees, and/or contributing authors may have positions in and may, from time to time, make purchases or sales of the securities or other investments discussed or evaluated herein.

Slides

Korea's Second Act in Investing Presentation

1.23MB ∙ PDF file

Download

Download

Let’s take a closer look.

Read more