If you purchased the S&P 500 40 years ago, reinvested dividends, and held through the myriad of crises over the intervening decades, you would have earned a 7,800% return, or about 11.2% per year.

87% of those returns (9.7% per year) came from the operating performance of the underlying businesses. Earnings per share grew 7.4% annually and the dividend yield averaged 2.3%. We call this the “business return” because it’s what a private buy-and-hold investor would earn.

Stock market investors in the S&P 500 earned an extra 1.5% per year on top of the business return because valuation multiples increased. Forty years ago, the S&P 500 traded at 14.5 times earnings. Today it trades for 26.0 times. The increase in valuation was lumpy and bumpy, and there were several years when valuations declined, but the net result was a 1.5% tailwind to annualized business returns.

The graph below shows the S&P 500’s business returns, in blue, and actual stock returns, in red.

The two lines follow the same long-term trajectory but occasionally diverge over a few years. The reason is simple: valuation multiples are mean reverting.

Business returns compound but changes in valuation are linear, one-off events. Over long periods of time, even a large increase in valuation multiple contributes relatively little to stock returns. For example, a multiple that expands from 10 to 20 over five years provides a 15% annual tailwind.

As long-term investors, we focus on the prospective business returns of stocks. This is because business returns:

1. Matter most for long-term investors;

2. Are easier to predict than short-term changes in valuation multiples;

3. Orient us towards owning good (high I-ROIC) businesses at low valuations; and

4. Naturally narrow our range of outcomes by helping us avoid risky investments.

Business returns matter most for long-term investment returns.

As the chart above showed, business returns account for the vast majority of long-term stock returns.

Over the short term (months or a year), changes in valuation multiples dominate stock returns. Flows, momentum, investors’ anxiety and euphoria, all cause stock prices to fluctuate more wildly than their intrinsic value and business returns. Look up any stock, even the bluest of blue chips, and you’ll see a 52-week high/low range of 50-100% of the stock’s price. Stock prices are significantly more volatile than their underlying businesses.

Over multi-year and certainly multi-decade time horizons, changes in valuation multiple usually net out to almost nothing, provided the starting or ending point isn’t extreme. Over the long term, earnings per share growth and dividends dominate investment results.

Business returns are easier to predict than short-term changes in valuation multiples.

The short-term forces that drive valuation change are difficult to reliably predict. Interest rates, prospective growth, taxes, and investor fear and enthusiasm generally drive changes in valuations. No single person affects a change in valuation, including management.

In 1936 economist John Maynard Keynes mused about two types of beauty contests. The first is the traditional contest where judges select the contestant they like best. In the second contest, now known as a “Keynesian Beauty Contest”, judges select the contestant they believe the other judges will like best. The first contest is straightforward. The second contest leads to round after round of circular reasoning.

Trying to predict short-term valuation changes is a Keynesian Beauty Contest. Success requires accurately predicting the future preferences of all market participants. We have no edge playing this game.

Our edge is in analyzing long-term business returns and buying when their stock prices are excessively discounted due to short-term fears or pessimism. When we succeed, we buy stocks with attractive business returns whose multiples are more likely to rise than fall.

If we’re wrong about the business returns, at least we did not pay a high valuation for them, limiting multiple compression. If we are right, we’re likely to experience multiple expansion, allowing our stock returns to exceed the business returns.

Business returns orient us towards owning good (high I-ROIC) businesses at low valuations.

Though we focus on business returns, we do not ignore valuation. Overpaying for a stock is the surest way to turn a good business into a bad investment. Valuation multiples are a key input into our calculation of business returns.

The diagram on the next page shows how valuation affects business and stock returns.

This diagram shows what we’ve already discussed: a stock’s return is the sum of its business return and change in valuation multiple. Business returns are the sum of growth and yield.

Growth is the product of a company’s reinvestment rate and its incremental return on invested capital (I-ROIC). Yield is the cash returned to shareholders as dividends and buybacks divided by the stock’s price.

The only equity value creation lever this equation doesn’t explicitly highlight is debt repayment. When debt is repaid, enterprise value transfers from creditors to equity holders, which can be a meaningful source of returns. It’s easiest to think of it as another form of yield.

Investors are often lumped into one of two camps: “value” or “growth.” This is silly. Our formulas show that growth is a component of value. A business that can reinvest the majority of its profits at a high I-ROIC for long periods of time is worth more than one that cannot. Even at an optically high valuation multiple, such a stock may be a bargain investment.

Our focus on business returns naturally captures this tension between value and growth. Valuation multiples are embedded in the yield component of our business returns analysis, alongside growth. Business returns, not an arbitrary multiple, govern the price we pay for a stock. The valuation multiple we are willing to pay is a by-product of our hurdle rate - the minimum business return we’re willing to accept.

Our emphasis on attractive business returns naturally disposes us to owning good businesses (high I-ROIC) at low valuations. As our stock return formula highlights, yield is a direct result of the price paid. As the price paid increases, yield and business return decrease, and vice versa.

Stocks with high valuation multiples and correspondingly low yield tend to produce low business returns. High-multiple stocks discount years of high growth, making them susceptible to the inevitable hiccups that every company eventually encounters. Missed growth expectations can lead to a rapid contraction in the stock’s valuation multiple and price. We’re leery of high-multiple businesses priced for perfection. History shows that very few businesses sustain meteoric growth rates for long.

Business returns naturally narrow our range of outcomes by helping us avoid risky investments.

Focusing on business returns keeps us inside our circle of competence. If we can’t understand, with a high degree of confidence, the growth and yield that a business will produce over the next five or ten years, we won’t invest in it. Underwriting a business’s growth requires understanding how management will reinvest capital and at what rate. Analyzing business returns naturally focuses us to think about bottoms-up unit economics, which requires a deep understanding of how the business actually operates.

We gravitate towards simple, predictable, and profitable businesses because they’re the easiest to analyze. Fewer moving parts means fewer places for flawed assumptions, which allows us to have greater confidence in our analysis. In investing, there are no points for difficulty.

Most businesses are hard to predict. Take AI-related companies, for example. Today they account for 40-53% of the S&P 500, depending on how you classify them. Though their stock prices have risen dramatically, it’s unclear what their business returns are. There is not enough data yet to determine what I-ROIC AI-companies are earning on their substantial investments.

Howard Marks said it best:

“I’ve never heard anybody tell me exactly what AI will be able to do, or when, or for whom, or how much profit it will produce, and for whom.”

Without a clear understanding of unit economics and I-ROIC, it’s impossible to estimate AI-related business returns with any degree of certainty. Nobody knows what the I-ROIC of the spending boom will be, including the executives overseeing it.

Stock prices imply that the I-ROIC of the AI buildout will be high. We are not so sure. The lack of conclusive data makes it speculative to bet on it in either direction. Should the data eventually show verifiably attractive business returns, we’ll happily change our minds.

AI may change the world and be widely adopted, but still fail to produce high I-ROICs. Cars and airplanes changed the world, but hundreds of manufacturers went out of business before the industries consolidated to a handful of manufacturers.

The history of every infrastructure and technological boom follows Stein’s Law, which says, “If something cannot go on forever, it will stop.” When it stops, slows, or pauses, investors may find that the business returns they implicitly underwrote were wildly optimistic. When growth slows, multiples contract and stock prices plunge. We seek to avoid this outcome.

We look for businesses in “replication mode” whose future is likely to resemble their past. This weeds out businesses subject to significant technological change or excessive reliance on factors outside of their control, like commodity prices.

It focuses us on companies whose future is reasonably knowable. The result is likely to be a high “hit rate” over time: hitting lots of singles and doubles, with few strike outs or home runs. It lops off the far-right tail (exceptionally large gains over short periods) of our returns distribution, as well as, more importantly, the far-left tail (large losses).

We think this is the best way to compound our capital over long periods of time while minimizing the chance that we fail to earn acceptable returns.

Conclusion

Over the last 40 years investors have enjoyed a mostly uninterrupted rise in valuation multiples alongside excellent business returns. This has produced almost magical results. Over the next 40 years perhaps we’ll enjoy the same tailwind in expanding valuations, though we have our doubts. A 1.5% annual tailwind could just as easily become a 1.5% headwind, or worse, as valuations mean revert.

We are not calling for a market crash, a bubble to pop, or any other doomsday scenario. We are simply calling attention to the fact that over the coming decades it may be as important as ever to rely on business returns to drive stock prices rather than hoping for valuations to continue rising.

Fortunately, the companies in which we’re invested have excellent prospects on both the business and the valuation side. We estimate that our aggregate portfolio’s business returns are compounding at a low-to-mid-teens annual rate and are valued around 13x earnings. This valuation is far below intrinsic value, given the predictable growth and yield of our businesses, which provides us a margin of safety.

Historically, our portfolio has experienced net multiple expansion. Our stock returns have exceeded our business returns. We don’t expect that to change, regardless of what happens with AI, the war in Iran, or whatever the next boom and bust cycle happens to be. When you buy good businesses at attractive valuations, good things tend to happen.

In the following sections we will provide an update on each holding and what we expect from its business returns, as well as comment on recent portfolio activity.

Subscribe now

EPC offers separately managed accounts, low minimums, and accepts most account types (including IRAs and 401k rollovers). Qualified Clients can elect a zero management fee structure inspired by the Buffett Partnership. EPC’s clients get complementary access to our premium Substack. To learn more please email info@eaglepointcap.com.

Disclosure: The author, Eagle Point Capital, or their affiliates may own the securities discussed. This blog is for informational purposes only. Nothing should be construed as investment advice. Please read our Terms and Conditions for further details.

Subscribe now

Do you have a “stranded” 401(k) from a past job that is neglected and unmanaged? Eagle Point offer separately managed accounts to retail investors, and 401(k) rollovers are often a good fit for our long-term approach. If you would like to invest with Eagle Point Capital or connect with us, please email info@eaglepointcap.com.

Disclosure: The author, Eagle Point Capital, or their affiliates may own the securities discussed. This blog is for informational purposes only. Nothing should be construed as investment advice. Please read our Terms and Conditions for further details.

• Simple and predictable business that we understand

• Replicable mode – the future will resemble the past

• Consistently free cash flow generative

• Dominant market position

• A demonstrated and enduring competitive advantage with high barriers to entry

• High returns on capital

• Limited exposure to extrinsic factors

• Resilient balance sheet

• Honest and able management who run the company for the benefit of shareholders

• A price which affords a margin of safety

Virtually no investment garners a perfect score on each criteria, but every stock we decide to buy meets this checklist to varying degrees (or so we think). Some businesses score through the roof on certain metrics - maybe it’s an extremely simple business trading for a rock-bottom valuation - while scoring slightly above average on others; maybe it’s only a 15% return on capital rather than 20%+ business.

We use this list as a filter and a guard rail to make sure we are staying within ourselves when making investments as well as while monitoring our existing portfolio.

Today’s investment scores abnormally highly across our checklist, and I’ll try to unpack the drivers behind the business quality as well as the current valuation disconnect.

In addition to being an excellent business, the opportunity follows some of Joel Greenblatt’s classic spinoff dynamics. The business is a virtual monopoly with tremendous pricing power, high recurring revenues, significant barriers to entry and a strong brand moat. These characteristics make the company a capital-light compounder. Further, because of some of the spinoff dynamics and underfollowed nature of the stock, it currently trades at a substantial discount to intrinsic value.

Do you have a “stranded” 401(k) from a past job that is neglected and unmanaged? Eagle Point offer separately managed accounts to retail investors, and 401(k) rollovers are often a good fit for our long-term approach. If you would like to invest with Eagle Point Capital or connect with us, please email info@eaglepointcap.com.

Read more

Recently We’ve Read:

• The Bottomless Cup: A Memoir of Secrets, Restaurants, and Forgiveness by Kevin Boehm

• The House of Morgan: An American Banking Dynasty and the Rise of Modern Finance by Ron Chernow

• We Are Bellingcat: Global Crime, Online Sleuths, and the Bold Future of News by Eliot Higgins

• 1873: The Rothschilds, the First Great Depression, and the Making of the Modern World by Liaquat Ahamed

The Bottomless Cup: A Memoir of Secrets, Restaurants, and Forgiveness by Kevin Boehm

If you’ve eaten out in Chicago, odds are you’ve been to one of Boka Group’s nearly 30 restaurants. My personal favorite is Zarella, which serves up one of the city’s best thin-crust tavern-style pizzas. As a Chicago native, I can attest that Chicagoans eat thin crust a lot more frequently than the deep dish we’re more closely associated with. Other Boka Group restaurants in Chicago include its namesake Boka, Momotaro, and Girl & The Goat (get the pig face).

Kevin Boehm co-founded Boka Group and tells his life’s story in this book. There are plenty of twists and turns that make this much more than a business success story. Boehm is a good story teller, and conveys a sense of the frantic adrenaline-fueled nights before a grand opening and non-stop chaos of a commercial kitchen (à la The Bear) that got him hooked on restaurants.

Unlike Anthony Bourdain’s restaurant memoir, Kitchen Confidential, Boehm’s comes from the front-of-house perspective. Boehm isn’t a cook, so he partners with chefs and handles the hospitality element.

Boehm’s passion is hospitality. He loves the art of hospitality like Buffett loves investment analysis. Most days, he was tap dancing to work. That’s the passion which has led to his success. I think that passion for your field is a requirement, especially in highly competitive arenas like restaurants and investing.

Restaurants are difficult businesses. Restaurants don’t have moats in the traditional sense. There are no switching costs, network effects, or barriers to entry. That Boka has 24 restaurants across the country suggests they’ve cracked the code and have some sort of moat or advantage.

Boka’s advantage must be its culture of hospitality. Boehm’s hard-won experience counts for a lot too, but opening so many restaurants suggests that Boehm and business partner Rob Katz have been able to codify and share their ideas across the organization to scale beyond their immediate reach.

Understanding the culture is one of the most important aspects of analyzing any business, but also the most difficult. It’s also one of the most important aspects to choosing a company to work for or work with as a vendor. It never shows up on financial statements and is rarely addressed directly by management. A company with a good culture can still have a bad year, like Boka in 2020.

Over long periods of time – decades – it becomes clear which organizations have a winning culture and which do not. It’s therefore a critical aspect to analyze for any long term investor. Since it’s rarely priced correctly, it can also help uncover bargains when macro factors like the pandemic temporarily overcome company-specific advantages.

I thoroughly enjoyed this book, and not just because I enjoy Boka’s restaurants. It’s not a business book, though Boehm’s presence on the business side of things means some of the mechanics of operating a profitable restaurant do get discussed. It’s more of a coming of age story, complete with plot twists and character arcs you won’t see coming.

Matt

The House of Morgan: An American Banking Dynasty and the Rise of Modern Finance by Ron Chernow

I’m slowly making my way through Ron Chernow’s books (they’re all like 1,000 pages so take me a long time to read) and The House of Morgan was his very first book written back in 1990. In true Chernow fashion the book covers every phase, major event, and important players from the founding of the original merchant bank in 1883 to the time of writing just after the 1987 crash.

It’s hard to boil down such a massive story into a few takeaways, but a handful of things that stand out are the shifting priorities of the various generations of leaders at J.P. Morgan, the consistency of regulatory cycles, and the theme of financial innovation ending in excess and destruction.

In the early years, Junius and Pierpont Morgan endured relentless market booms and busts, frequent bank runs, and bankruptcies of huge rail roads companies. Because of the nature of their environment they emphasized capital preservation, reputation, and risk management over trying to participate in speculative frenzies just to grow as fast as possible. This turned out to be brilliant as they only survived and were able to be alive 100+ years later because they weathered this period of turbulence.

As the Fed did not yet exist, J.P. Morgan was the lender of last resort in the early 1900s and repeatedly bailed out companies and the stock market as a whole because of its fortress financial position.

After the great depression that was partly brought on by banks’ risky behavior of comingling excessively risky investments with hum-drum lending, the government passed the Glass-Steagall act which dictated that banks that took customer deposits could no longer conduct investment banking activities. This caused the separation of J.P. Morgan, which retained the legacy banking business, and Morgan Stanley which became the investment banking arm.

For a long time, the two firms had a gentleman’s agreement, had mostly the same shareholders, and coexisted in a mutually beneficial way. For a few decades it looked as if J.P. Morgan would be left in the dust by the faster growing and far more lucrative Morgan Stanley.

In the 1970s and ‘80s, deregulation came back into flavor amidst the boom in hostile takeovers, and the “gentlemanly banker code” was shredded forever in favor of fast-paced casino capitalism (sound familiar?). Since then, J.P. Morgan and Morgan Stanley have been fierce rivals in almost every area of high finance other than traditional corporate lending and retail banking.

The most common theme throughout the book, regardless of era, is the constant flow between excess and bust that has been, and always will be, present in a capitalistic society. From the bank runs of the 1880s, to the general panics of the early 1900’s, to the great depression, to the leveraged buyout boom and bust of the ‘80s, to the stock market crash of 1987, like clockwork the next boom is always setting the stage for the next bust.

The House of Morgan has been around for, sometimes at the center of, and always survived on the other side, of every boom and bust of the past 150 years. My guess is they’ll be around for and survive the next.

Dan

We Are Bellingcat: Global Crime, Online Sleuths, and the Bold Future of News by Eliot Higgins

I read this book after hearing Chris Waller of Hidden Gems Investing discuss it on the Investing By The Books podcast. Chris has done some great scuttlebutt over the years, particularly on Watches Of Switzerland.

We Are Bellingcat is all about scuttlebut. It’s the story of how blogger Eliot Higgins built an open-source intelligence collective that has successfully exposed war crimes, assassinations, and international cover-ups strictly using publicly available sources. In most cases, the Bellingcat collective solves the mystery faster than the three-letter government agencies with proprietary information and much deeper pockets.

Bellingcat’s success is encouraging and inspiring for investors. Investors do not need proprietary data to gain an edge. The world today is drowning in publicly available data. What’s missing is the ability to synthesize it – what Bellingcat calls building the mosaic – and then analyze it with common sense, an eye for detail, and a few mental models.

The foundation of open-source intelligence (OSINT) is Mosaic Theory. No single tweet, Instagram post, public record, or shipping manifest has all of the data. The OSINT analyst’s job is to combine these disparate sources of information until they form a cohesive picture.

Bellingcat obsessed over the details of each piece of the mosaic they gathered. Shadows became clues about time of day and position of the photographer. Tattoos and insignias were clues which could identify a group or individual or track their movements.

Each piece of the mosaic had to be independently vetted and verified or else the conclusion would be untrustworthy. Bellingcat often reached conclusions faster than governments because Bellingcat was not hampered by the official story or assumptions. They followed the data and ruthlessly verified it, discarding what was unproven, even when it was the consensus.

Had Higgins chosen to pursue investing instead of OSINT, I have no doubt he would be extraordinarily successful. Successful investing is also about building a mosaic from a company’s filings, their competitors, and other open-source data. Investing requires being detail oriented, and understanding subtle shifts in depreciation schedules and option grants. Investors must follow the facts and change their mind when the facts change. Most importantly, investors must be willing to challenge the consensus narrative and think independently.

We Are Bellingcat opened my eyes to what it is possible to accomplish with these tools and shows just how powerful they are.

Matt

1873: The Rothschilds, the First Great Depression, and the Making of the Modern World by Liaquat Ahamed

Much like The House of Morgan, 1873 is another tail of market cycles, excess taken to ruin, and the ultimate triumph of liquidity over leverage. I think this book, while confusing and hard to follow at times, is one of the most important books for investors right now to read given the parallels then to the AI buildout now. In fact, Microsoft CEO Satya Nadella recently recommended shareholders read the book (that was kind of a head scratcher to me given Microsoft is sort of at the center of the current bubble, but whatever, I agree with him everyone should read it).

In the 1850s and 60s a transformative technology took hold; railroads. Massive amounts of capital poured into building new railroads across Europe and the U.S. The buildout was funded with cheap leverage and speculative investors eager to participate in the bonanza. The scale of spending almost exactly matched today’s global spending on the AI buildout when adjusted for 1860s dollars.

In 1873 the euphoria was brought crashing down when the Vienna stock market tanked. In April 1873 several massive new stock and bond issuances were floated on the Vienna market. Instead of investors gobbling these up hand over fist like in years past, there were suddenly a dearth of buyers; investors had had enough and speculators were short on cash.

The mania in Vienna was also partly tied to euphoria over the May Vienna World Exhibition (World's Fair). Vienna had been crushing it because of the speculative frenzy in its stock market, creating a wealth effect, which lead to immense hype over the fair. In the months leading up to the fair speculators drove up the stock prices of real estate, construction, and hotel stocks. The fair turned out to be lackluster due to poor weather and a cholera scare.

The illusion of profits from the World Fair and the failed capital raise a few weeks earlier caused a rapid panic and crash in the Vienna stock market which fell 45% in one day. The panic quickly spread overseas.

By September, a prominent U.S. investment bank called Jay Cooke & Co., which had its balance sheet tied to illiquid railroad debt, failed suddenly. The domino effect took hold and liquidity froze across the globe.

Making matters worse, several global powers decided to abandon bimetallism (gold and silver) in favor of a single gold standard. This caused a drain on the global money and credit supply and caused the global economy to go into a 20-year deflationary tailspin.

The global financial destruction was so great it sewed the seeds of disasters to come in the 1900s.

The ruin in the U.S. caused political support for Reconstruction to cease and ended federal support for civil rights in the South. Defaults by overly indebted nations triggered death spiral of the Ottoman Empire and led to Britain’s occupation of Egypt. The broad spread economic pain birthed the early deglobalization and populist swells and kicked off a run of antisemitism which wrongly scapegoated the Rothschilds as the root cause of the panic despite the fact that the Rothschild’s had been one of the few prudent and conservative lenders throughout the ordeal. None of these issues went away and the festered for the next 100 years through WWI, WWII, and the oppression of African Americans in the U.S.

Hopefully the current boom in AI spending turns out better!

Dan

The Best Of The Rest

• S&P500 Net Margins Soar +370bp; A.I. Productivity Gains, Or Something Else?

• The $1.5T Question Nvidia’s Earnings Can’t Answer

• Paul Graham: The Brand Age

• WSJ: Pizza Has Gone Cold. Dominos is Still Worth a Look.

• Reuters: Meta’s AI Agent Development is Going Slower Than Expected

• WSJ: Liquor Stocks Are Priced for a Big Tobacco Moment

• WSJ: Tobacco Stocks Are Coming in From the Cold

• Watch Pro: Value of American watch market could comfortably rise 50% to $30 billion

• Fortune Financial: Industrial Staples

• Legend: Bernard Arnault Interview

Subscribe now

Do you have a “stranded” 401(k) from a past job that is neglected and unmanaged? Eagle Point offer separately managed accounts to retail investors, and 401(k) rollovers are often a good fit for our long-term approach. If you would like to invest with Eagle Point Capital or connect with us, please email info@eaglepointcap.com.

Disclosure: The author, Eagle Point Capital, or their affiliates may own the securities discussed. This blog is for informational purposes only. Nothing should be construed as investment advice. Please read our Terms and Conditions for further details.

From 2010 to 2024 Churchill Downs’ stock generated an exceptional 23% annual return, or 2,200%+ total return, for shareholders. The last two years have been less kind to the stock, which is down 30% from the start of 2025, despite solid business performance. Since 2010 the company’s average PE (per ValueLine) is 27x earnings, but today the business trades for just 12x despite posting record earnings each of the past few years.

Management has become frustrated with what they perceive as a growing disconnect between the company’s share price and its underlying value, so they recently announced a major strategic pivot with the decision to sell the bulk of its regional casinos and use the proceeds to repay debt, return capital to shareholders, and focus on the faster growing and less capital intensive horse racing segment. Likely to the surprise of management, when the business announced this pivot the stock dropped a further 7%.

This situation rhymes with several we’ve studied or owned in the past, and the key question is, is this a knee jerk reaction to try to appease the market or a savvy capital allocation move that could enable the business to return to a long period of outperformance? Let’s get to it.

Overview

Churchill Downs operates three business segments, Live & Historical Racing, Wagering Services & Solutions, and Regional Gaming.

Live & Historical Racing

The Live & Historical Racing segment consists of the flagship Kentucky Derby and its associated race week along with venues that hold thousands of historical racing machines (HRMs). Ticket sales, sponsorships and licensing, and media broadcast rights drive the bulk of the revenue in this segment with Derby Week alone accounting for 25-30% of Churchill Downs’ full-year property-level profitability.

An HRM is basically a slot machine but instead of using a random number generator to determine payoffs it determines its winners based on the outcomes of past actual horse races. When a player starts a game the machine randomly selects a past anonymized race. The player is then presented with handicapping information and then can either manually select the winning order of the horses or let the machine automatically make a selection based on past odds. Once the bet is locked in the machine reveals the actual results of the race and the player wins or loses money.

HRMs enjoy regulatory moats by essentially exploiting loop holes in traditional state gambling laws. HRM operators can bypass laws that ban slot machines because they are not classified as gambling due to the pari-mutual wagering structure of the bets. Pari-mutual bets are wagers placed into a shared pool against other players which are then paid out to winners with the operator (in this case Churchill Downs) taking a percentage – usually around 20% - of the pot in order to administer the game.

HRMs also enjoy significant political protection. In staters where HRMs are popular, a mandated percentage of the wagers go to funding horse racing purses and local agricultural and breeding programs. Politicians are hesitant to create backlash from this community and have no incentive to upset the apple cart.

Only a handful of states allow HRMs, including Kentucky, Virginia, Wyoming, New Hampshire, Kansas, and Alabama.

Churchill Downs enjoys an incredibly strong position in the HRM market for a few reasons. In order to operate HRMs you must own a physical racetrack in the state and also have a racing license. States are not rushing to build new horse racing tracks and therefore new competitors are virtually non-existent.

Further, in 2023 Churchill Downs purchased Exacta Systems which is the dominant technology provider that powers HRMs, links the machines together, and handles the pari-mutual math. Competitors that offer HRMs often must pay Churchill Downs to use their software to administer the games.

By controlling the technology and owning physical racetracks in the states where HRMs are most popular, Churchill Downs essentially has the HRM market on lock with competitors effectively barred from entering.

The Live & Historical Racing segment sports 55% margins and is growing at a low double-digit clip.

Wagering Services & Solutions

Exacta Systems is part of the company’s second most important segment called Wagering Services & Solutions often referred to as the TwinSpires segment. In addition to housing the HRM software, TwinSpires operates an online horse betting platform that enables people to bet on live races (whether they are at the race or not). People often prefer to use the app to place bets instead of waiting in long lines while onsite, and users can just as easily place bets from their couch and watch live streams of the races on which they’re betting.

Churchill Downs makes money from this segment by keeping a percentage of wagers to administer the pari-mutual bets and it also sells data and tools (like horse racing statistical data) to third parties mainly consisting of other gaming operators.

Regional Gaming

Around 2010 traditional horse racing was flatlining and Churchill Downs’ prospects looked weak. HRMs were not yet a “thing” as the tech was primitive and not widely available and online horse betting was virtually nonexistent. The whole company depended on highly seasonal live horse racing which had a natural ceiling. Management needed something to kickstart the next leg of growth for the business and they found that by buying regional casinos.

From 2010 to 2022 Churchill Downs steadily acquired casinos in Mississippi, Maine, Illinois, and New York, among other places. These assets provided non-cyclical cash flow to fund the development of the HRM space, TwinSpires technology and transform the core racing and wagering segments into what they are today.

While on the surface one might observe that buying casinos was merely a distraction from Churchill Downs’ core business and a play to appease Wall Street, I think it’s more fair to say that without diversifying into the casino space it’s likely Churchill Downs would be a much smaller, seasonal, novelty business.

While diversifying into the casino space was likely the right strategic move, it’s not as if it was a financial was a bonanza. Buying and maintaining casinos required a lot of capital and earned ok returns. Over the past ten years the company has invested 140% of its operating cash flow at ~10% incremental returns. The upshot is Churchill Downs is a much larger business with new, promising, high-return segments that wouldn’t have existed without the cash flows from the casinos, but that strategy has likely run its course, and management has decided it’s time to divest. We’ll cover that more shortly.

Recent Stock Price Declines

Read more

“Every morning in Africa, a gazelle wakes up. It knows it must run faster than the fastest lion or it will be killed. Every morning a lion wakes up. It knows it must outrun the slowest gazelle or it will starve to death. It doesn’t matter whether you are a lion or a gazelle: when the sun comes up, you’d better be running.”

– African Proverb

Have you ever read a 10-K or Value Line report and wondered where all the money went? There are lots of companies reporting consistent profits, yet they are neither growing, paying down their debt, distributing cash to shareholder, or sitting on a growing pile of cash. The answer usually has to do with capex and working capital.

Financial statements simplify capex down to a single line item (“purchase of PP&E”) on the cash flow statement. This is an oversimplification because there are different types of capex. Some is good (growth), some is bad (maintenance), and some is plain ugly (defensive).

Maintenance Capex

Maintenance capex is the most important to understand. According to Berkshire Hathaway’s 1986 letter, it is the average annual capital expenditure required by a business to maintain its competitive position and unit volume. It’s how fast a company has to run just to stay where it is.

I called maintenance capex “bad,” but it is not bad to spend on maintenance. It’s absolutely critical. Maintenance can be deferred, but only to a point. Companies that defer maintenance too long pay the price via a compromised competitive position and/or lower productive capacity.

Maintenance capex is only “bad” to the extent that money spent on it is not available for shareholder distributions (dividends, buybacks) or growth capex. Companies with high maintenance requirements relative to their operating cash flow are capital intensive, earn a low return on invested capital, and, all else equal, are less valuable. We generally prefer to avoid investing in businesses which have to spend a lot merely to remain where they are. These are hamster wheel businesses: running fast but going nowhere.

Airlines are a perfect example of hamster wheel businesses. Commercial jets require continuous, legally mandated heavy maintenance. Parts are often sole-sourced because FAA approval is a long, costly process. This gives parts suppliers, like TransDigm, significant pricing power in excess of inflation.

Despite heavy maintenance, jets still wear out after 25-30 years and become nearly worthless. In the late 90s and early 2000s a Boeing 737-800 or Airbus A320 cost $30 million to $38 million (after standard discounts to list price). A modern replacement (Airbus A321neo or Boeing 737 MAX 8) costs $60-75 million. As a result, airline capex regularly runs 1.5-2.0x depreciation, and not because they’re increasing capacity significantly.

Auto OEMs are also hamster wheel businesses. Tooling, die replacements, plant re-fits for model year refreshes, and factory floor maintenance require non-stop reinvestment just to sell the same unit volume of cars as the prior year. Ford routinely spends $7 billion to $9 billion annually in capex, consuming 50% to 70%+ of its industrial operating cash flow, yet its unit sales aren’t growing.

Estimating Maintenance Capex

Most managers don’t know what their maintenance capex is, so it should be no surprise that it is rarely disclosed. Davita, Enterprise Products Partners, and Ecovyst are among the select few that disclose it.

Most investors use depreciation as a proxy for capex, but this often underestimates it. In 2021 Venkat Peddireddy wrote his Columbia University PhD thesis about the difficulties of estimating maintenance capex. He concluded that maintenance capex exceeds reported depreciation by about 25%, on average, across industries.

There are two reasons: inflation and technological change. GAAP accounting uses depreciation charges to allocate the cost of an asset over its pre-determined useful life. The problem is, people rarely know ahead of time how long what they’re buying will last. I “invested” in CDs, thinking they’d last me a lifetime. Then music went digital, and I bought MP3s from iTunes, thinking the same. Now I subscribe to Spotify and my collection of music is obsolete.

The problem is particularly acute when there is rapid technological change. What happens when servers expected to last six years are obsolete after three? Earnings would be overstated because maintenance costs were understated.

Inflation is another problem, particularly for very long lived assets. Railroads depreciated track over 40 years, but inflation over the last forty years was a cumulative 204%. The CPI rose from 109.6 in 1986 to 333.0 in 2026. BNSF is still depreciating the track it laid in 1986 at cost. The true replacement cost today is multiples of the original cost in 1986.

Michael Mauboussin and Dan Callahan wrote a paper in 2022 called Underestimating The Red Queen which builds on Peddireddy’s thesis. They conclude that managers and investors perpetually underestimate maintenance capex, due to the effects of technological change and inflation, and therefore overestimate growth capex.

Growth Capex

Growth capex is money spent to expand the business’s competitive position or increase its productive capacity (unit volume). Examples of growth capex are a retailer building a new store or a manufacturer building a new factory.

I called growth capex “good” but the reality is a little more nuanced. Growth capex is only good when the business has a high incremental ROIC (IROIC). As Michael Mauboussin says, “growth amplifies returns.” A business with a high IROIC and high growth will produce a lot of value. A business with a low IROIC and high growth will destroy a lot of value. A business with a low IROIC should not spend money on growth capex; they should return profits to shareholders.

When the Santa Fe Railway (now BNSF) modified the height of bridges and tunnels along its LA-to-Chicago route to run double-stacked containers, it was high IROIC growth capex. It doubled the container capacity of a single train without adding much to the cost of running that train.

Chipotle’s drive-thru “Chipotlanes” are similar. They increase same-store sales 10-20% and leverage the existing kitchen, allowing for an attractive IROIC.

Growth capex doesn’t always mean increasing sales or unit volume. It could also be investing in more efficient machinery that will reduce operating costs and increase margins.

Cost-reduction capex is only growth capex if the company has a moat that allows it to retain the margin expansion. If the cost-reducing equipment is available off-the-shelf to every competitor, market forces pass those savings to buyers. What management pitched as growth was actually defensive capex.

Defensive Capex

The third type of capex is really a subset of maintenance capex, but its root cause is competition and industry structure, not entropy. I call it defensive capex, and it’s “ugly” because it looks like growth capex but functions like maintenance capex. It is easiest to explain with an example.

In Berkshire’s 1985 letter Buffett explains why he shut down the company’s namesake textile mills. The mills made commodity textiles. Price, not quality or brand, was their primary differentiator. Berkshire’s mills were disadvantaged because competitors in foreign countries had significantly lower labor costs. To combat this, Berkshire regularly spent considerable sums of money to buy more efficient looms to lower their costs and make them more profitable.

The problem was that all of Berkshire’s competitors were also making the same investments. As soon as everyone bought the new looms, the reduced operating costs became the new baseline for lower prices industrywide.

Each individual company thought they were making a rational investment to improve cost efficiency, but collectively, the decisions neutralized each other. Buffett likened this to standing on your tiptoes to see better at a parade. Competition dictated that the industry’s lower costs were passed onto the consumer, leaving the Berkshire mills with little to no profit on a higher base of invested capital.

Eventually Buffett opted out of the capex arms race and closed the mills. It was taking an ever growing amount of capex to continue earning paltry returns. Buffett realized that the capex was defensive; it had the promise of growth, but served merely to maintain the status quo.

A modern example is the auto OEMs’ investments in EVs. Ford and GM spent tens of billions of dollars retooling plants to make EVs to defend their market share from Tesla and Rivian. Ford and GM hoped they’d take share, but the investment was really about not losing share.

AT&T, Verizon, and T-Mobile’s foray into 5G is another example of defensive capex.

It started in 2020 when T-Mobile acquired Sprint for $26 billion. Sprint owned a ton of unused 2.5GHz mid-band spectrum.

5G signals rely on radio frequencies. Low-band spectrum travels far but is slow. High-band “millimeter wave” (mmWave) spectrum is fast but can’t penetrate a glass window or travel more than a couple of blocks. Mid-band spectrum (2.5 GHz to 4 GHz) is the “Goldilocks zone.” It offers multi-gigabit speeds and travels several miles per tower.

Back in 2020, Verizon was betting on fast high-band mmWave spectrum. They were trying to put antennas on every street corner. AT&T was preoccupied paying off its debt from its DirecTV and Time Warner acquisition.

When third-party network tests showed T-Mobile’s 5G speeds were significantly faster than Verizon and AT&T, subscriber churn began to tick up.

To avoid losing market share, Verizon and AT&T got into a bidding war for mid-band spectrum. Verizon spent $45.5 billion and AT&T spent $23.4 billion. They also paid a collective $13 billion in incentives and clearing costs to satellite companies (like Intelsat and SES) to vacate those airwaves early.

To deploy their new spectrum, they had to buy billions of dollars worth of new telecom equipment and upgrade the backhaul fiber to handle the surge in traffic.

Despite spending $100 billion, AT&T and Verizon did not acquire new customers. And they were not able to raise prices. They merely closed the competitive gap with T-Mobile and protected their competitive position and current subscriber base. Consumers were the real winners, getting faster 5G speed for no added cost.

The Capex Boom

A similar arms-race dynamic is playing out now, but in AI datacenters.

Buffett recently told CNBC’s Becky Quick:

“With Google and all its competitors now, they’re laying out hundreds of billions of dollars. That’s real money... That kind of money wasn’t even put in the railroad business. And that’s the game they’re playing now. They weren’t playing that game with computer software.”

The hyperscalers “don’t have any choice,” adding that they are “playing a game they don’t want to play.”

Buffett seems to think that the hyperscalers are engaged in a defensive capex arms race. Though, to be fair, Buffett did invest $23 billion into Alphabet this year: $13 billion in the open market plus a $10 billion private placement. Buffett must believe Alphabet is an AI winner, perhaps because of its full-stack capabilities (TPUs, cloud, compute, and frontier model).

Sundar Pichai, CEO of Alphabet, seems to agree that it is defensive capex, saying,

“When we go through a curve like this, the risk of under-investing is dramatically greater than the risk of over-investing for us.”

He elaborated that even if Alphabet turns out to have over-invested in GPUs and data centers, those assets have long useful lives and can be repurposed across Search, YouTube, and Cloud. Missing the shift entirely, however, would be catastrophic.

Mark Zuckerberg, CEO of Meta, said almost the same thing on The Circuit podcast:

“If we end up misspending a couple of hundred billion dollars, I think that that is going to be very unfortunate, obviously. But what I’d say is I actually think the risk is higher on the other side.”

Andy Jassy of Amazon wrote in his 2025 letter to shareholders:

“When you have shifts that are this momentous, you want to make sure that you invest... We’re not going to be conservative in how we play this.”

While any individual company’s capex may be rational and appear to earn a high IROIC, in totality they may only serve to reduce costs and maintain the relative competitive position.

As with the textiles arms race, consumers may be the real winners from all of the AI data center capex. After the railroad boom between 1870 and 1890 tens of thousands of miles of redundant track sat empty. Between 1865 and 1900, freight rates plummeted 50% to 70%. Farmers, merchants, and consumers reaped the benefits of cheap interstate transportation for decades, while equity investors faced recurring bankruptcies.

In the 1990s millions of miles of fiber optic cable was laid by Global Crossing, WorldCom, and 360networks, among others. In 2001, an estimated 95% of installed fiber-optic capacity was dark because supply was multiples of internet traffic demand. Bandwidth prices collapsed by over 90% in the early 2000s, and over $2 trillion in market value vanished.

The upside was that excess capacity built in the late 1990s kept internet transmission costs dirt-cheap for two decades. This helped make streaming media, like YouTube and Netflix, viable, as well as cloud computing.

The oil and gas industry runs into this problem all the time, most recently during the 2010s fracking boom. The drilling capex that each individual company could justify collectively overwhelmed the market with supply. As with the other booms, prices eventually crashed, stocks cratered, and consumers, not investors, benefitted.

There’s one last example that’s worth considering: the canal boom of the 1820s. After the successful completion of the Erie Canal, state governments and private investors borrowed heavily to dig thousands of miles of canals across Indiana, Ohio, Pennsylvania, and Maryland.

Unlike the other booms, excessive capacity didn’t do this one in. It was a technological change. Railroads emerged in the late 1830s as a better way to move freight, making many of the canals obsolete before they’d paid for themselves. Multiple states defaulted on their bonds in 1841–1842. Even though investors were wiped out, consumers benefitted from shipping rates 80-90% lower than before the canals and railroads were built.

In each of these situations the initial idea was correct. There was a credible reason to spend money on growth capex. The IROIC appeared to be there. “But as happens in Wall Street all too often,” wrote Buffett in his 1989 letter, “what the wise do in the beginning, fools do in the end.” Too much of a good thing turned out to be a bad thing for investors.

What’s an investor to do?

• Beware of investing in rapidly changing technologies. History shows that most technological innovations benefit consumers more than investors.

• Beware of the assumptions embedded in a capex decision. Capex may have an attractive IROIC over the assumed lifespan, but be obsolete long before then. Capex may have an attractive IROIC at current prices, but not if prices collapse. Prices in commodity industries (compute, NAND, oil/gas, textiles, freight, etc) are often significantly more volatile than management teams expect because managers fail to consider the industries’ collective capacity additions into their forecasts.

• Industry structure matters. When a competitor is determined to gain share, it can set off an arms race of defensive capex to maintain the status quo. We prefer to invest in stable oligopolies that change slowly.

• Capex to reduce costs has to be accompanied by a moat which prevents competitors from copying it and the gains accruing to customers, not investors. Otherwise it is defensive. Companies are perpetually launching cost savings initiatives and exclude the “one time” costs associated with those from their earnings, even though this is defensive/maintenance capex.

Figuring out what type of capex is happening is difficult, but crucial to estimating future returns. It is more profitable to avoid these thorny questions and focus our energy on the easier ones. As Ben Franklin wrote in 1736: “an ounce of prevention is worth a pound of cure.” Charlie Munger argued he was wrong. An ounce of prevention, Munger thought, was worth way more than a pound of cure.

We prefer to invest in capital light businesses with little exposure to technological change that are part of an oligopoly with stable market share. Royalty businesses like NRP have zero capex and can return 100% of profits to shareholders. There’s not where for it to slip through the cracks.

We filter out hamster-wheel businesses that have to run fast merely to stand still. We likewise filter out hyper-competitive businesses whose industries turn into arms races. We like stable, slow changing industries, like pharmaceutical distribution. Everyone in the oligopoly earns high returns on their capital employed, has roughly equal share, and benefits from the long term growth of American pharmaceutical consumption.

We like businesses in “replication mode” whose future will resemble their past (and whose past was pretty good). These businesses have a proven track record of high IROIC growth capex under their belt and a long runway ahead to continue doing more of the same thing. Watches Of Switzerland has a long runway to continue acquiring American Rolex authorized dealers. AutoZone has a long runway to continue building new stores, particularly in Mexico and Brazil, and mega hubs in the US. Constellation Software has relationships with 40,000 potential acquisitions.

These businesses are, by definition, not at the cutting edge and unlikely to be in the limelight. That’s an added bonus, because it is easier to think independently when the WSJ and Twitter are not full of smart-sounding commentary about them and their industries.

Subscribe now

Do you have a “stranded” 401(k) from a past job that is neglected and unmanaged? Eagle Point offer separately managed accounts to retail investors, and 401(k) rollovers are often a good fit for our long-term approach. If you would like to invest with Eagle Point Capital or connect with us, please email info@eaglepointcap.com.

Disclosure: The author, Eagle Point Capital, or their affiliates may own the securities discussed. This blog is for informational purposes only. Nothing should be construed as investment advice. Please read our Terms and Conditions for further details.

If you’d like to read more about these topics, I’ll link to our previous posts below.

Constellation Software (CSU)

• Constellation Software: AI’s Threats and Opportunities

• Elasticity, the value of data, and Constellation Software’s PEMS strategy.

• Constellation Software: Presentation At The Intellectual Investor Conference

Natural Resource Partners (NRP)

• Natural Resource Partners: High Uncertainty, Low Risk

• Updates on Natural Resource Partners, Cencora, and McKesson

Subscribe now

EPC offers separately managed accounts, low minimums, and accepts most account types (including IRAs and 401k rollovers). Qualified Clients can elect a zero management fee structure inspired by the Buffett Partnership. EPC’s clients get complementary access to our premium Substack. To learn more please email info@eaglepointcap.com.

Disclosure: The author, Eagle Point Capital, or their affiliates may own the securities discussed. This blog is for informational purposes only. Nothing should be construed as investment advice. Please read our Terms and Conditions for further details.

The attractive valuation exists for two reasons. First, a short report accused it of channel stuffing to make its numbers. This can be shown to be false. Second, it recently levered up to acquire an adjacent business. Net debt to EBITDA is 3.3x but will be below 2.5x by late 2026.

Read more

Each of these situations appear cheap with high prospective returns but with the real possibility of becoming a value trap. The common thread among each business is uncertainty around the direction of the moat. Matt wrote a great article about this concept and I’d recommend checking it out in conjunction with this post.

Each situation seems investible, but for various reasons I’m caught in a pattern of “thumb-sucking” as Munger would have said. Below are some of my real-time notes as I work through these, and I’d love to hear from any readers who might have any insights or differentiated views on these businesses.

Subscribe now

Do you have a “stranded” 401(k) from a past job that is neglected and unmanaged? Eagle Point offer separately managed accounts to retail investors, and 401(k) rollovers are often a good fit for our long-term approach. If you would like to invest with Eagle Point Capital or connect with us, please email info@eaglepointcap.com.

Molson Coors

We’ve passively followed Molson Coors for years and when skimming Value Line recently the valuation for the business really jumped out. After reaching an all-time high of $86/share in 2016 Coors has struggled to transition from a legacy beer company and adapt to shifting consumer tastes, volume challenges, and inflationary pressures. Ten years later the stock is down roughly 50% from it’s 2016 high.

As a quick overview, 85% of Coors revenue comes from beer but the business has been making a push into alternative beverages with it’s “beyond beer” category which includes hard seltzers, ready-to-drink cocktails and some non-alcoholic options. It’s an attempt to latch onto younger generations shift away from drinking beer to premium and better for you options.

On the surface there’s plenty to be interested in with Coors. The company currently trades for a rock-bottom 6.2x free cash flow and is returning significant capital to shareholders. Management repurchased 5% of its shares last year and is increasingly leaning into buybacks. The board recently authorized an incremental $2.5B repurchase which represents more than 35% of the company’s market cap. When combined with a juicy 5% dividend yield, Coors is returning around 15% of it’s market cap to investors this year (and presumably each year going forward if the stock doesn’t re-rate).

A 15% shareholder yield is rare and very enticing. It means the stock is priced to never grow and instead shrink in perpetuity. Management recently laid out it’s medium-term financial targets. These included low single digit sales growth, mid single digit pretax earnings growth, and high single digit earnings per share growth. 8-9% EPS growth combined with a 5% dividend gets you 13-14% business returns before any valuation re-rating.

This all sounds great on paper. I’m just not sure the business won’t shrink for the foreseeable future. This means it has all the makings of a potential value trap.

Beer market volumes have been declining for years, and Coors is facing intense competition in the beyond beer category. The company has also had a hard time offsetting inflationary cost pressures. When taken together earnings have gone nowhere for a decade, and I don’t have any reason to feel certain that will change soon.

My sense is the business doesn’t have much of a moat, and what moat it does have is likely shrinking. This makes it tough to pound the table on what the economics will look like for the business five and ten years from now. Management is currently leaning on what I consider mostly “empty growth” to hit revenue targets. Volume is falling at a low single digit percentage and the company is making up for it by raising prices to get to a net 0-2% revenue growth. I’m not sure this is sustainable especially given how weary consumers are of price increases across the economy.

That said, the Molson brand has been around since 1786 and this isn’t the first soft period they’ve experienced. Additionally, the price is low enough that buying Molson Coors stock may represent an intelligent speculation that management is able to stem volume losses and grow the business modestly. With a 15% shareholder yield, investors are being paid to wait, but don’t be surprised if the stock is mostly dead money if the beer market doesn’t perk up and the beyond beer initiatives fail to deliver.

Domino’s Pizza

We’ve followed Domino’s for years and if you had told me five years ago that, everything else equal, I could purchase the stock for ~15x earnings I would have told you with near certainty that I would have done so. Alas, everything is not equal, and things have changed for the business in recent years and here I sit, still not owning Domino’s despite the major valuation de-rating.

Read more

My slides and talking points are below.

Intelligent Investor Conference Csu 2026 6 24

672KB ∙ PDF file

Download

Download

Slide 1 — Title

Hi, my name is Matt Franz and today I’m going to discuss Constellation Software.

Slide 3 — Elevator Pitch

Constellation is a serial acquirer of VMS business. VMS stands for Vertical Market Software. I’ll explain what that means in a moment.

The stock has fallen 50% and now trades at 15 times free cash flow. This is a compelling valuation for a high quality business.

The stock fell over fears of AI disruption, the founder’s recent retirement, and concerns about the pace of capital deployment.

My view is that these are overblown and unlikely to impact the company meaningfully.

Slide 4 — What Is VMS?

VMS is hyper-specialized software for niche industries.

For example, Constellation owns businesses that

- run public transit for cities

- handle the billing for water utilities

- act as the operating system for country clubs, and

- manage operations at grain elevators.

They are not household names. They work behind the scenes and touch aspect of the global economy.

VMS are unusually good businesses. They are mission-critical, deeply embedded, have high switching costs, and low churn.

Customers cannot operate their business without Constellation’s software.

Despite this, Constellation prices for value. The average customer pays Constellation less than 1% of sales.

This is a strong value proposition and why Robert Smith called software “better than first-lien debt” back in 2018.

VMS produces predictable cash flow, driven by maintenance and SaaS fees, and benefit from fixed cost operating leverage as they grow.

These qualities make VMS good businesses. They also make VMS less vulnerable to AI disruption than your typical software business.

Slide 5 — Cumulative Acquisitions Graph

The problem with VMS is that they are small businesses with small addressable markets. Constellation typically buys them for a few million dollars, which is too small for private equity or venture capital.

Constellation was built to acquire lots of tiny VMS businesses, aggregate their cash flow, and use it to buy even more VMS businesses.

That’s exactly what they’ve done. They’ve made over 1,200 acquisitions since 1995, spread across 100 countries.

Slide 6 — Cloning High Performing Conglomerates

To make 1,200+ acquisitions over thirty years requires an unusual organization and culture.

When Mark Leonard founded Constellation, he studied conglomerates that performed well for several decades.

Two key insights came from this: the power of incentives and importance of decentralization.

Constellation pushes all business decisions to the lowest level, the business unit manager. Headquarter’s job is to distill beat practices and coach, not to manage.

Constellation even decentralized acquisitions. This is critical because it allows them to make hundred of acquisitions for just a few million dollar each every year. If M&A was centralized, like at Berkshire, Constellation would have to hunt for larger prey.

Even though there is extreme decentralization, there is a culture of high accountability.

Constellation rigorously measures its subsidiaries performance. Its favorite metric is the sum of ROIC plus Net Revenue Growth. That’s because improving one usually comes at the cost of the other.

At some Constellation events, managers have been required to post their numbers on their name tags.

Slide 7 — FCF & Stock Price

This has all worked out financially for them. Free cash flow has compounded at 26% since the company’s 2006 IPO. The stock has done even better, even in spite of its recent draw down.

Crucially, Constellation has not issued a single share since their IPO, which is highly unusual for software businesses. All of these economics have flowed to common shareholders.

Slide 8 — Source of Growth

Constellation’s growth has primarily come from acquisitions, not organic growth. M&A is the main driver of business results.

Organic growth has averaged just 1.5%, which is similar to the rate of inflation over this period. Being mission-critical, Constellation’s businesses has untapped pricing power. This is one of the most valuable qualities a business can have. If inflation were to run hotter, Constellation would likely be able to pass that through. This would materialize as higher organic growth.

Constellation intentionally does not push price in order to maintain a trusted relationship with its customers. It also does not want to create a pricing umbrella that invites competition.

Slide 9 — FCF Graph

Constellation used to trade for an extraordinarily high multiple, but recently its valuation has come crashing down to Earth, just like most software stocks.

Slide 10 —- Risk 1: AI

Now that we’ve talked about the good, lets talk about the bad.

Slide 11 — The Code Is Not The Moat

AI has reduced the cost to produce code, no question. But writing code is just one piece of a software company’s value, and it’s not Constellation’s moat.

Constellation’s moat comes from its close customer relationships. Its had customers for decades, who trust the operations of their business, and their livelihoods, to Constellation.

Constellation has spent those decades learning about their customer’s businesses and their workflows.

Constellation’s prices are already low relative to the value it provides, so there’s no room to undercut it. And Constellation’s software is mission-critical and deeply embedded, making the cost of switching high.

There are banks, even large money-center banks, which still run operations on mainframes running Cobol. Cobol hasn’t been high tech since the 1970s. That ought to tell you something about how switching costs and delay change, even in the face of technologic superiority.

Slide 12 — 1,500 Watertight Compartments

Constellation’s decentralized structure give an advantage when navigating change. Each business unit is left to respond in its own way. Headquarters looks at what’s working, distills best practices, and distributes them.

Constellation never has to make a top down, bet the company judgement call. The right answer arises organically.

Slide 13 — AI As A Tailwind

AI may even be beneficial to Constellation in a few ways.

First, it can roll out new products and features faster. As an incumbent, Constellation knows what its customers need and can bring those features to market fast.

VMS may embed AI models and act as the distribution and routing layer for them, which might be a pricing opportunity.

And a shift towards outcome based pricing could be disruptive, but also more lucrative, just as incentive fees tend to be larger than management fees, just less predictable.

Slide 14 — Risk 2: Capital Deployment

Slide 15 — M&A Is Systematic

Constellation does not wait around for the phone to ring. It treats M&A like a repeatable, data-driven industrial process.

They’ve built 40,000 relationships with VMS founders. These are not names purchased from a broker’s database. They are personal relationships. Constellation checks in with each one 3-4 times per year.

This helps Constellation learn about this business and build trust with the founder. It also keeps Constellation top of mind so that they are the first phone call wen the founder wants to sell.

Slide 16 — Lower Software Valuations Are A Tailwind

As Buffett says, if you are going to each cheeseburgers your whole life, you should root for the price of cheeseburgers to go down.

The price of VMS businesses have begin to decline, which benefits acquirers like Constellation.

They’ve ramped up capital deployment to the fastest in two years.

If AI leads to more software, there will be an ever-expanding pipeline of M&A targets.

Slide 17 — Beyond Niche VMS M&A

Beyond its core M&A, Constellation has begun investing in public equities as a minority, engaged, and permanent shareholder. Public valuations have declined faster than private valuations, and they’re leaning in. They’ve also done special dividends and spinoffs in the past. And they are studying other industries they could invest in.

Constellation’s management is thoughtful and aligned.

Slide 18 — Risk 3: Leadship Change

Slide 19 — Risk 3: Mark Leonard Steps down

Founder Mark Leonard recently announced that he was stepping down for heath reasons. I’m not worried because cultures tend to be sticky, for better or worse.

Mark Miller is taking over. He founded Constellation’s first acquisition, Trapeze Group, and joined Constellation in 1995, so he’s literally been there since day 1.

Slide 20 — Forward Returns

Here’s how I think about the stock’s forward returns. Constellation is dependent on its M&A engine to continue firing and for the inherent advantages of VMS to protect it from AI.

If it does I think shareholder do very well from here.

Slide 21 — Questions

Subscribe now

EPC offers separately managed accounts, low minimums, and accepts most account types (including IRAs and 401k rollovers). Qualified Clients can elect a zero management fee structure inspired by the Buffett Partnership. EPC’s clients get complementary access to our premium Substack. To learn more please email info@eaglepointcap.com.

• The Frackers: The Outrageous Inside Story of the New Billionaire Wildcatters by Gregory Zuckerman

• Churchill: Walking With Destiny by Andrew Roberts

• The Professor, the Banker, and the Suicide King By Michael Craig

• Build the Life You Want by Arthur Brooks

The Frackers: The Outrageous Inside Story of the New Billionaire Wildcatters by Gregory Zuckerman

The Frackers tells the story of three wildcatters—George Mitchell, Aubrey McClendon, and Harold Hamm—who pioneered hydraulic fracturing and horizontal drilling. Their approach was highly unorthodox at the time, but, once proven, changed the global energy landscape and geopolitical balance of power.

The book reads like a novel, leaning into the characters more than the science. For a generalist like me, it provided sufficient background knowledge.

There were two themes that stood out. The first was the importance of a can-do entrepreneurial spirit that does not take “no” for an answer. There is a reason Exxon and Chevron did not pioneer fracking—they’re too risk adverse. They already have fantastic businesses and do not want to risk their capital on lottery-ticket style payoffs. That left the opportunity to smaller, scrappier, headstrong entrepreneurs. Though these oil men were warned time and again that they would fail (and they did fail, sometimes for years), they kept at it and eventually succeeded.

The second theme is about capital efficiency. Oil and gas is capital intensive. Leasing land requires upfront commitments for uncertain payoffs. The more likely people believe that there is oil and gas below the surface, the more expensive leases are. The best deals therefore occur when the oil company has a correct and differentiated view, perhaps because of superior technology.

Harold Hamm, founder of Continental Resources, chose to finance his company conservatively, always paying cash for its land leases. Aubrey McClendon took the opposite approach at Chesapeake Energy, spending lavishly to amass as much land as possible. That Continental is still around and Chesapeake went bankrupt in 2020 tells you everything you need to know about flying too close to the sun.

The Frankers was published in 2013, before the 2015 oil bear market. That bear market was directly caused by excessive production from fracking companies like those profiled. It is a lesson that what may be smart and value-accretive in the beginning may be dumb and destructive in the end. The first companies to frack earned huge ROIs. Those that chased the boom at the end experienced massive losses.

Matt

Churchill: Walking With Destiny by Andrew Roberts

Reading biographies of those that lived in the Victorian era/Gilded Age through World War II is one of my favorite things to do. There was an incredible number of history-shaping events and trends from the late 1800s through the second world war. From robber barons building industrial monopolies to two world wars, a crippling depression, and countless panics, booms and busts, failing to closely study this time period almost feels criminal.

Winston Churchill was an important character through this entire time period and ranks up there as one of history’s most important figures. From a very early age Churchill sensed that his life would be extraordinary. As a teenager, Churchill foretold the challenges he would face later in life, saying;

London will be in danger... it will fall to me to save the Capital and save the Empire.

He was right. Churchill’s life was characterized by a series of setbacks and triumphs, all of which built into a crescendo of leading Great Britain through World War II. Churchill greatly admired his father, himself a prominent politician, but was never able to earn his approval before he passed away when Churchill was just 20 years old.

After graduating from the military academy he was determined to make a name for himself in the military. He quickly did just that by demonstrating tremendous bravery, bordering on stupidity. Churchill actively sought out danger to build his reputation. Always the first to volunteer to be deployed, he led daring frontline attacks in India in 1897. He partook in a reckless cavalry raid in Sudan in 1898 where he wound up surrounded and outnumbered and had to rapidly fire his pistol at point blank range to escape. In South Africa in 1898 Boer forces ambushed a train he was on. Instead of retreating he help free wounded men and ended up captured. He later made a daring escape across enemy territory.

Whether brave or plain stupid, acts like these quickly earned Churchill the reputation he sought. He also became a famous author after documenting his battlefield exploits which led to a prolific writing career for the rest of his life. The earnings from his books and articles was his financial lifeline during the depression when he nearly went broke.

One other aspect that Churchill developed early and stood out throughout his career was his masterful skills of oration. He was legendary for his recorded addresses to parliament and the public throughout his time in office, and he gave over 2,300 speeches in his career. His ability to rouse support were instrumental during WW II and I’d highly recommend listening to any number of his addresses during the war or his famous iron curtain speech.

In 1900 Churchill moved into government and held a number of posts over the next 30 years. This initial stint in government culminated when he served as First Lord of the Admiralty during WWI. He modernized the Royal Navy and displayed great leadership skills during the way but suffered a major setback when he orchestrated the Dardanelles and Gallipoli naval campaigns which resulted in significant defeats. He was effectively demoted after this and served in a few other positions before his party was defeated and he entered his “wilderness years”.

Churchill was out of the government for most of the 1930s, but remained one of the sole outspoken opponents of the British government’s appeasement of Hitler as he rose to power in Weimer Germany. Long in favor of rearmament and militarization, he was dismissed as a warmonger until it was almost too late. When England finally declared war on Germany in 1939 Churchill was summoned back into government which paved the way for his eventual election as prime minister 8 months later.

Throughout the war Churchill displayed his characteristic tenacity, doggedness, and eternal belief Britain wouldn’t suffer defeat. It didn’t look good for the Allies for many years, but Churchill knew if Britain could find a way to hold on until the U.S. entered the war there was no way Germany would come out victorious. He worked relentlessly, visited the frontlines, stood on his roof watching the battle of London, and made countless daring trips across Europe to meet with other leaders all while smoking 10+ cigars per day and consuming mass quantities of champagne and scotch with water - his “papa cocktail”.

Churchill was key in brokering alliances between Stalin and Roosevelt and instrumental in crafting the Allied path to victory, and there’s no telling what the western world would look like today if not for Churchill’s leadership during this formative age.

-Dan

The Professor, the Banker, and the Suicide King By Michael Craig

The books I’ve learned the most from have been about high-performing individuals, particularly businesses men with an “Outsider” approach. I wanted to study billionaire banker Andy Beal, but there is frustratingly little written about him. This book is one of the few that touch on him.

Born in Lansing, Michigan (not far from EPC’s office in Grand Rapids, MI), Beal was an entrepreneur from the get go. According to Wikipedia, “As a teenager, Beal began earning money by fixing and reselling used televisions with the help of his uncle. While attending high school he also installed apartment security systems. He also started a business moving houses and managed rental properties.”

The book explains how Beal got his start in real estate, buying, renting, and flipping apartments in Lansing. He dropped out of Michigan State in 1976 to pursue real estate full time. In search of a deal, he ended up at a federal foreclosure auction in Washington D.C. On a whim, he pivoted from bidding on apartments in Gulfport, MS to a distressed property in Waco, TX.

To his own shock, his own low-ball bid of $217,500 won. He had never seen the property. Beal put up just $17,500 of his own cash and secured a $200,000 loan to cover the rest. He moved to Waco and began fixing it up, often with his own two hands. Three years later in 1979 he sold it for $1 million.

Beal went on to perfect this playbook in the early 80s. In 1981 he bought two abandoned housing project buildings in Newark, New Jersey (the Brick Towers) for $25,000 and flipped them for $3.2 million two years later.

In the late 80s the Savings and Loan (S&L) Crisis unleashed a flood of foreclosed real estate. The government created the Resolution Trust Corporation (RTC) to liquidate it, and Beal was a frequent buyer.

Beal had more deals than he had capital, and access to attractive financing was the limiting factor. He realized that he could opened his own state-chartered, FDIC-insured bank and raise cheap, stable capital from the public via CDs.

Beal launched Beal Bank in 1988 with $3 million of his own money. Today it has $3.8 billion equity and Andy Beal remains the sole owner. He has never sold a share and does not intend to take it public.

This book focused on poker, one of Beal’s hobbies. Beal regularly flew to Vegas to play a consortium of pro poker players to test his skills. To skew the odds in his favor, he would insist the games start at 8 AM, ungodly early for the pros, who were often playing cash games until sunrise. He also insisted in playing heads up (one on one), and at extraordinarily high stakes. Millions would change hand every hand. This was change to Beal but meaningful even to Vegas’s wealthiest poker pros. Beal even developed methods to hide his emotions and randomize his raises to make himself unreadable.

These show how competitive Beal is and how good Beal is at thinking about risk and reward probabilistically, critical skills for any investor.

Matt

Build the Life You Want by Arthur Brooks

A few months ago I attended a lunch at which Arthur Brooks was the guest speaker. He teaches one of the most popular classes at Harvard called “Leadership and Happiness”. After his talk I immediately bought his recent book, Build the Life You Want.

Brooks explains that he is not innately a happy person. All his life he’s had to work to understand what makes people happy and try to apply the lessons to his own life. What he realized is happiness cannot be a goal, it’s an outcome. You can’t seek out happiness, but if you understand the ingredients, you can create circumstances where happiness and satisfaction are the result.

Brooks spends much of the book elaborating on what he refers to as the three “macronutrients” of happiness. After studying countless people and analyzing what underlies their happiness or lack thereof, he concluded that true well-being depends on having:

Enjoyment; which he defines as pleasure paired with meaning. Pleasure by itself is fleeting and ultimately unsatisfactory. Anything that brings pleasure with meaning or with some level of work or effort is not likely to aid in producing a happy life.

Brooks uses the example of eating. Gorging on junk food while scrolling through your phone may provide fleeting pleasure, but you’ll probably feel terrible later. Conversely, consider preparing a Thanksgiving meal with family and sitting around the table eating and making memories, this goes beyond the pleasure of the food and is instead an example of enjoyment.

Satisfaction; or the joy of working hard and achieving a goal. Whether you work hard to climb a corporate latter, work equally hard raising a family, are hammering away at a PhD thesis, or are training for an ultramarathon, everyone needs to be able to connect the work they do with an outcome they are trying to achieve. Meaningless work is a grind, and success without work feels empty. Satisfaction is the marriage of hard work and tangible progress.

A pertinent example of empty vs. true satisfaction is using AI to create something. Empty pleasure is asking AI to generate an image of something for you. It takes three seconds and takes almost no effort. Spending weeks sketching, painting, re-painting, and grappling with a portrait of your favorite summer vacation spot is a lot different. The real work it takes to create something meaningful is as much the point of the exercise as the end product. This is how satisfaction works.

Purpose; or the existential direction and reason for our life. Satisfaction is about what you achieve and purpose is about why you are here. Having a sense of purpose is the main reason suffering can be more easily endured. In fact, Brooks argues that finding meaning and direction in suffering often makes us happier compared to if the suffering never happened in the first place. Once again, you can find purpose in your work, in your family or friends, or in the community, and most of us have more than one purpose.

Brooks writes that finding a sense of purpose requires coherence (you need to step back and make sense of suffering and finding a coherent identify), direction (having goals in some area of your life), and significance (a belief that life matters).

The key is to find your purpose, work hard at bettering yourself towards the purpose, which will bring high levels of enjoyment to your life. Brooks’ books are easy to read and I always come away feeling more positive and energized.

-Dan

The Best Of The Rest

• Barron's: 5 Signs the Semiconductor Rally is Getting Silly

  • “In the last 25 trading sessions, the benchmark advanced nearly 54%, its best rolling 25-day performance since March 9, 2000, when it rose 56% at the peak of the dot-com bubble, according to Dow Jones Market Data. Between the end of March 2000 and the final trading day in September 2002, the index sank 80%.”

• Cal Newport: The Dangers of “Vibe Reporting” About AI

  • “The goal of this type of article is to create a pre-ordained vibe, not to get to the bottom of what’s really happening.”

• Construction Physics: Do Commodities Get Cheaper Over Time?

  • “Labor-intensive services (education, healthcare) get more expensive in inflation-adjusted terms over time, while manufactured goods (TVs, toys, clothing) get less expensive over time.”

• Bloomberg: The Utilities Analyst Who Says the Data Center Demand Story Doesn’t Add Up

  • “In a new report, CreditSights analysts led by Andy DeVries crunch the numbers to show that US utilities already appear to be in the process of building twice as much power capacity as is currently forecast to be needed by 2030. In other words, the data center buildout already looks to be outstripping demand for compute.”

  • See also: CreditSights: Utilities: The Great Data Center Overbuild

• The Coal Trader: Adding To My NRP Position

  • NRP is on track to be debt free in 2026 and substantially increase distributions. The stock should re-rate towards pipeline MLP multiples.

  • See also: Natural Resource Partners: High Uncertainty, Low Risk

• Fortune Financial: The Traits of an Ideal Industry

  • Industry structure can have a significant impact on a business’s long-term performance.

• Institutional Investor: Now Top Investors Are Doubting the AI Revolution

  • “Looking back at major technological booms, from railroads to the internet to shale, you see similar patterns emerge over time… Equity investors have not always seen returns that match early expectations.”

• Cal Newport: Why Hasn’t AI Made Work Easier?

• WSJ: AI Isn’t Lightening Workloads. It’s Making Them More Intense.

Subscribe now

Do you have a “stranded” 401(k) from a past job that is neglected and unmanaged? Eagle Point offer separately managed accounts to retail investors, and 401(k) rollovers are often a good fit for our long-term approach. If you would like to invest with Eagle Point Capital or connect with us, please email info@eaglepointcap.com.

Disclosure: The author, Eagle Point Capital, or their affiliates may own the securities discussed. This blog is for informational purposes only. Nothing should be construed as investment advice. Please read our Terms and Conditions for further details.

Brown & Brown started in 1939 as a two-man insurance agency in Florida. In the intervening decades it has grown into the preeminent middle market insurance broker defined by an entrepreneurial, decentralized, and M&A-driven culture. As a serial acquirer with a narrow focus, the company reminds me a lot of Alimentation Couche-Tard, Constellation Software, or HEICO, all of which we’ve written about (or owned) in the past.

Overview

The insurance brokerage market is extremely fragmented and Brown & Brown has been acquiring brokers of all sizes for decades. Revenue has grown organically at an average 3.7% clip since 1998 (the first year I can find easily disclosed organic growth rates). Overall revenue has grown at outstanding 14.4% per year over the same period thanks to the completion of over 650 acquisitions.

Historically Brown conducted primarily tuck-in or bolt-on acquisitions, but in 2025 the company acquired Accession for almost $10B, which was by far the largest deal in the company’s history. Accession is a massive player in specialty insurance and provided Brown & Brown with further scale in the highly profitable specialty distribution segment.

Of course, revenue means nothing if it’s not driving earnings and free cash flow growth. Fortunately, the story is even brighter there, as earnings per share have grown at 15.5% per year since 1998. The fact that Brown and Brown acquired so much revenue growth while generating even higher earnings growth points to management’s skill at identifying and integrating attractive acquisition targets over the last several decades. More on that later.

Business Segments

Brown & Brown operates two segments; retail and specialty distribution. Recall from last weeks’ post that the difference between retail and non-retail insurance brokerage is not what is being sold but how it is beings old. Both the retail and specialty distribution segment are primarily driven by specialty/niche insurance lines. In the retail segment Brown interfaces directly with the business buying the insurance. In the specialty distribution segment Brown interfaces with other insurance agents who are looking for insurance for their clients.

Here’s how I think of it:

Retail Segment: Brown & Brown Agent -> The Business Buying Insurance (Even if the business needs highly specialized coverage).

Specialty Distribution Segment: Brown & Brown Wholesaler -> An Outside Insurance Agent -> The Business Buying Insurance.

The retail segment makes up about 58% of revenue and has the below business mix.

The specialty distribution segment does business under the Arrowhead banner and offers wholesale brokerage as well as binding authority or “managing general agent” (MGA) services.

As a wholesale broker Brown is the “shopper”. They act as the matchmaker for retail agents when their clients have hard to place risks.

The MGA is more than a shopper, it is also the decision maker. MGAs are the entities/people that hold binding authority on behalf of carriers and do much more than “just” match retail brokers with specialty insurance carriers.

For example, a major insurance carrier might give Brown & Brown a bucket of money and say, "we want to write insurance for marinas, we don't have expertise in that area, but you do. Here are our parameters, go run the program for us."

Arrowhead MGAs do everything from marketing lines to underwriting and pricing risks and binding carriers to policies. They often manage claims as well, meaning an MGA often acts like an insurance company without taking the underwriting risk. The programs are then distributed through a network of thousands of independent agents.

Brown & Brown is largely U.S.-centric with 86% of revenue coming from the U.S. 10% of revenue is from the U.K. and 4% elsewhere.

Business Model Advantages

Insurance brokers are inherently good businesses. They are capital light and high margin, high ROIC, and provide an essential service with sticky client relationships.

Read more

Since the late ‘90s through early last year the publicly traded insurance brokers and risk managers had all compounded shareholder capital at attractive rates for several decades. Returns varied from pretty damn good at Aon (which topped the S&P 500 by about 1,000%) to other-worldly at Brown & Brown (which topped the S&P 500 by about 13,000%). Arthur Gallagher and Marsh also generated great returns.

The last year tells a bit of a different story with each stock down between 10% and 50% and the market up 25%+. Several concerns have mounted and caused a rare prolonged dip in share prices across the board.

Since early 2025 the insurance market has begun to turn from a hard market to a soft market. Further, the previously impregnable position of insurance brokers is suddenly being called into question. What is at the root of these fears? AI, of course.

To assess whether these fears are rational we need to examine past cycles in the insurance industry and whether the fears of AI disintermediation have any practical merit. Before digging into individual companies, I think it’s important to take a look at the insurance brokerage industry as a whole and the major players. That’s what we’ll do here, and next week I’ll publish some thoughts about an individual opportunity.

Industry Overview and Business Model

Insurance brokers sit in between insurance companies (carriers) and customers who need various types of insurance and who pay for insurance coverage. I like to think of insurance customers in a few segments; personal, commercial, and specialty.

Personal lines are largely commodity insurance products such as auto and home insurance. There isn’t a ton of differentiation in the types of coverage that are provided and it’s largely formula based. Large insurance companies compete mostly on price and there isn’t much value-add insurance brokers can provide in these areas. I don’t need to contact a broker to help me find auto insurance quotes as I can just do that online and compare the quotes and coverage. Commercial and specialty insurance is a different animal.

A business owner needs a plethora of insurance products and it’s a lot to sort through. Here is just a snippet of what types of coverage a business might need:

As a business leader, these are not insurance products I am just going to google and figure out on my own. The time required to understand which policies I need and which insurance carrier offers the best value for those policies is not something businesses want to spend time on, and the risk of picking the wrong insurance can be catastrophic. I need advice and tailored solutions, so I go to an insurance broker.

The insurance broker then takes a look at the insurance needs that are unique to my business and puts together a plan. They will reach out to a number of insurance carriers (like Travelers, State Farm, etc.) to find the right solution for the right price. Once the solution is in place the broker is paid a percentage of the premium by the insurance carrier at which the business was placed.

Brokers are extremely important to insurance carriers as they are basically an extension of their own sales force. Large brokers can negotiate favorable rates for customers because of how much business they bring to big insurance companies.

While standard business insurance is complicated and requires the advice of brokers, specialty insurance is yet another layer more involved.

Specialty Insurance

Specialty insurance products are lines that are non-standard and require flexible and tailored terms and pricing. The non-standard insurance market where niche insurance is purchased is often referred to as the Excess & Surplus (E&S) market. This is where non-standard policies live. Here are some examples of specialty lines from Brown & Brown’s 10K:

Professional programs. Professional liability and related package insurance products are tailored to the needs of professionals in the following areas: dentistry, legal, eyecare, insurance, financial, physicians and real estate title professionals. Professional liability programs also offer supplementary insurance-related products to include weddings, events, medical facilities and cyber liability.

Personal Lines: Personal lines programs offer a variety of insurance products to personal lines consumers including homeowners and personal property policies; residential earthquake; private passenger automobile and motorcycle coverage.

Commercial Lines: Specific industries and market niches are served by our commercial programs including automotive aftermarket, professional and amateur sports, special events and the entertainment industry; commercial transportation and trucking; forestry; manufactured housing; and workers’ compensation.

Public Entity: Public entity programs range from providing fully insured programs to establishing risk retention insurance pools, and excess and facultative specific coverages, including administration of various insurance trusts for cities, counties, municipalities, school boards, special taxing districts and quasi-governmental agencies.

Specialty: These programs include flood insurance, commercial difference-in-conditions (earthquake), all-risk commercial property, limited exposure captives, coastal property programs including wind, lender-placed solutions, sovereign native-American nations and parcel insurance.

As you can see, niche lines are not insurance needs that can be plugged into a standard formula and quoted within minutes. These often require judgement and underwriting on a case-by-case basis, deep sector expertise, and many insurers that provide standard (“admitted”) insurance do not provide these solutions off the shelf.

Standard insurance companies won’t touch high-hazard and complex risks, so business owners need access to the specialty market. The problem is, specialty companies that will write these policies don’t employ thousands of local street agents given the bespoke nature of the risks. Instead, the carriers rely on large wholesale brokers like Brown & Brown or Ryan Specialty to connect them with customers.

Because of these dynamics, brokers provide tremendous value in the specialty market. Let’s say I own a small winery in Napa, and I need a variety of non-standard insurance products. I need protection against things like “smoke taint” from wildfires whose smoke could ruin my grapes. I have several million dollars’ worth of wine aging in barrels. If a fire burns down my winery, how would an insurer value the wine in those barrels, at cost or at fair market value? How do you price a policy to account for that? I also have complex agricultural equipment that requires various types of insurance. I can’t just call up State Farm and get a suitable insurance solution, so I go to Brown & Brown or Ryan’s specialty unit and they help me piece it together.

Binding Authority

Within the E&S segment, specialty insurance brokers operate under what are referred to as binding authority agreements with large insurance carriers. This means they literally “hold the pen” and can bind carriers to insurance contracts in pre agreed upon areas without taking any of the underwriting risk. This segment is the secret sauce for many brokers, especially Brown & Brown and Ryan specialty.

Brown and Brown has a large binding authority business whereby they have expertise in various niche insurance applications and carriers have given them the authority to seek out, underwrite and price, and bind contracts on their behalf. These transactions are usually smaller on an individual basis but in total add up to a meaningful contributor to overall profitability as they are very high margin and sticky. In many cases the specialty broker will have exclusive or semi-exclusive arrangements with carriers whereby agents looking to place certain types of risk must go through certain brokers.

Often times there will be an outcomes-based component to the fees in a binding authority situation whereby the broker earns more when they successfully price contracts.

Ryan Specialty provides a helpful graphic in their 10K of how the E&S market works:

In the graphic above Ryan Specialty shows how regular insurance agents will come to them with E&S lines and Ryan can either route these lines on an unbound basis to a specialty insurer (like Berkshire Hathaway or Lloyds of London) who is then in charge of setting pricing and terms, or they can act as the underwriter and binding agent and submit the lines to the specialty insurer on a bound basis (the second two boxes in the middle of the graphic, “underwriting manager” and “binding authority”).

In most cases brokers are much more than just insurance matchmakers. In addition to finding the right insurance products they service the accounts and provide analytics and risk management consulting services. When a claim arises, often times the insured will call the broker and they will help facilitate the claims process from claim through resolution. Because the broker is so ingrained into the insurance workflow they become an extension of the customer and the insurer and therefore enjoy high customer retention.

Revenue Model

In addition to the types of insurance brokers place, it’s helpful to delineate how they interface with customers. Brokers generally refer to operating either as “retail” brokers or “wholesale” brokers. The difference in these models is who the end customer of the broker is.

Retail brokers serve the businesses who are buying the insurance. Wholesale brokers serve other insurance agents or other brokers who are trying to place insurance for a customer but lacks the expertise or access to the E&S market so they must use an E&S wholesale broker like Brown & Brown or Ryan Specialty.

Many brokers operate both retail and wholesale segments and the retail segment can utilize the wholesale segment when non-standard risks need to be underwritten. More on that below.

Key Industry Players

There are a number of publicly traded insurance brokers, and they can be grouped into a few buckets.

At the top end we have Aon, Marsh & McLennan, and to a lesser degree, Willis Towers Watson. These three are less pure-play brokers and more enterprise risk and benefits consultants. They focus on very large corporate clients, such as Fortune 500 companies, and have tailored their business around advising multinational businesses and the risks associated with them.

They serve a smaller list of customers who generate huge premium volumes and need more risk consulting and analytics as well as benefits consulting services and even things like wealth management. Aon and Marsh can bring E&S lines directly to specialty insurers without using wholesale brokers because of the size of the contracts. About 60-65% of the large player’s business is from insurance brokerage with the balance coming from consulting and advisory services. These businesses are Wall Street’s brokers.

The large players don’t really play in the wholesale brokerage market as they made the decision to divest those divisions years ago as a result of Elliot Spitzer bringing charges against Marsh, Aon, and WTW for “double dipping” on commissions. A retail broker is obligated to act on the best interest of its client, and if a broker owns a retail broker and a wholesale broker they could favor the in-house wholesale brokerage solution (even if it’s not in the best interest of the customer) and charge commissions at the retail and wholesale level. Rather than deal with the red tape and bad press, the brokerage behemoths divested their wholesale segments around 2005.

Moving down the corporate value chain, Brown & Brown dominates the middle market. Instead of targeting large corporate accounts, Brown & Brown serves businesses like regional construction companies, auto dealerships, and mid-size manufacturers. If Aon is Wall Street’s broker, Brown & Brown is America’s main street broker.

Brown’s relationships involve a higher volume of transactions and do not require the consulting services that Aon and Marsh offer. Accordingly, Brown & Brown is basically a pure-play broker with nearly 100% of its revenue coming from commissions from placing either standard (or “admitted”) or E&S lines of insurance.

Ryan Specialty is a pure-play E&S broker founded in 2010 by Patrick Ryan, who also founded Aon and led that business for over 40 years. The key difference between Brown & Brown and Ryan is that Ryan does not operate a retail arm, they act purely as a wholesale broker, or a “brokers broker”. This avoids potential conflicts of interests and allows Ryan to build deep specialty in specific niche insurance lines.

Now that we have the industry sorted out, next week I’ll zoom in on Brown & Brown and look at some of the current fears, risks and opportunities I see at the former compounding machine.

Brown & Brown is down about 50% over the last year and trading at a valuation not seen in over a decade. I’ll try to wrestle with the question of whether something has fundamentally changed about the business or the recent price declines are overdone. We’ll explore what past insurance cycles have looked like and what the current soft market might bring to the business and to the stock.

EPC offers separately managed accounts, low minimums, and accepts most account types (including IRAs and 401k rollovers). Qualified Clients can elect a zero management fee structure inspired by the Buffett Partnership. EPC’s clients get complementary access to our premium Substack. To learn more please email info@eaglepointcap.com.

Disclosure: The author, Eagle Point Capital, or their affiliates may own the securities discussed. This blog is for informational purposes only. Nothing should be construed as investment advice. Please read our Terms and Conditions for further details.

In an effort to encourage longer-term thinking, we consciously avoid quarterly letters and Wall Street’s typical reporting cadence. For the same reason, our letters focus on bottom-up business fundamentals rather than macroeconomics and stock price volatility. Our goal is to think and act like long-term business owners.

Clients receive our letters on or around April 1st and October 1st each year. Premium subscribers receive full access to our letters on a 45-day lag.

Above all, we aim to provide our clients and our premium subscribers the information we would want if our positions were reversed.

Thank you for reading and for your support. Please don’t hesitate to reach out with any questions or comments.

Read more

Soda Ash

NRP had an unusually tough quarter because of its soda ash business. Prices are below the cost of production for most producers, driven by a supply glut from China and weak demand for flat glass.

Last quarter, NRP said that they would inject $39.2 million equity into its Sisecam Wyoming JV. The managing partner wanted to reduce debt in light of the cyclical downturn in soda ash pricing. This quarter that transaction materialized in the financial statements.

Collectively, NRP and its partner (Sisecam Chemicals Wyoming LLC) invested $80.0 million into Sisecam Wyoming. The full $80 million was used to pay down the JV’s bank credit facility. $60 million of debt remains; $29.4 million is NRP’s pro rata share.

NRP’s free cash flow was -$5.4 million and gross debt increased from $33 to $60 million. Since the quarter ended, net debt has been reduced to $45 million.

Read more

In the early 1800s, agriculture was the engine of the American economy, employing over 80% of workers and accounting for nearly 20% of GDP. In 1900, when automation began its march, it took 56 man-hours to grow an acre of wheat. By the late 20th century, that dropped to 3 hours. Today, agriculture employs less than 2% of the workforce and contributes only 1% to GDP.

Automation—tractors, threshers, combines, and the cotton gin—increased productivity massively. People, however, can only eat so much. Beyond a certain point, lower prices do not drive incremental demand. Economists call this an inelastic response. It means demand is not sensitive to changes in price. Since higher agricultural productivity was not met with a higher demand, employment fell.

Air travel is the opposite. Deregulation in 1978 and technologic advances drastically lowered the cost per mile flown. In 1970, a round-trip ticket from New York to London cost approximately $550, or 140 hours of labor for the average worker. Today, the same flight costs about 13 hours of labor. As the real price of airfare fell, demand exploded. Global Revenue Passenger Miles (RPMs) have grown from roughly 350 billion in 1970 to over 8 trillion in 2025.

Demand for air travel was highly sensitive to a change in price. People did not fly the same amount for less money. They flew more: low-cost fares created demand for weekend trips, international vacations, and in-person business meetings.

Moore’s Law describes one of the highest price elasticities in human history. Since the mid-1960s, the semiconductor industry has delivered an average 20% to 30% annual decline in the cost of manufacturing a transistor. In 1987 the global sales of semiconductors was $33 billion. In 2024 it was $775 billion.

As the price of a transistor fell, we didn’t just use the same amount of compute for less money. We put computers into everything from greeting cards to cars.

Historically, demand for software has looked more like airfare and semiconductors than wheat. It is highly elastic.

As programming languages became more efficient (moving from machine code to high-level languages like Python and JavaScript), the price of building an app dropped. Instead of hiring fewer programmers, companies built exponentially more complex software for every aspect of life, leading to a massive hiring boom in the sector. Between 2019 and 2025, the number of professional software developers worldwide grew from 14 million to nearly 21 million.

We didn't use the efficiency gains of high-level languages to hire fewer programmers; we used them to migrate the entire human experience—banking, social life, and infrastructure—into the cloud.

Software development jobs continue to trend higher and are meaningfully higher year over year. This runs counter to the narrative that AI is destroying software development jobs. As W. Edwards Deming, the father of the lean manufacturing movement, said: “In God we trust. All others bring data.”

Read more

When I first became interested in investing I thought that valuation was the key way to make money. I’ve now come to appreciate that valuation is actually the primary driver of how to avoid losing money. Further, we used to be more rigid when it came to valuation and deciding when to sell. We sold too early and were too quick to exit winning positions just because the valuation had risen a bit.

Valuation is a complex subject and is easiest for me to think about in three different “buckets”; the buying, the holding, and the selling. Valuation can help and hurt you in different ways in these three buckets, and what follows is how we think about valuation in relation to the various stages of stock ownership.

EPC offers separately managed accounts, low minimums, and accepts most account types (including IRAs and 401k rollovers). Qualified Clients can elect a zero management fee structure inspired by the Buffett Partnership. EPC’s clients get complementary access to our premium Substack. To learn more please email info@eaglepointcap.com.

Valuation and Buying

I’ll start with the most important way to think about valuation – as a margin of safety. We’ve written extensively about the benefits of buying low absolute valuations, particularly when combined with a capital return program, so I’ll briefly rehash some of those ideas.

It’s not rocket science to understand that if you pay a low price for a business that is returning cash to shareholders it makes it harder to lose money. For example, if a company is returning 8-13% of your purchase price (meaning you paid ~8-12x for the stock) every year, and you’re right about the cash flows, it makes it very hard to lose money over an extended period of time. It’s unlikely that a company that trades for 10x earnings and is generating a 10% shareholder yield will de-rate to 5x earnings for any meaningful period of time. If this occurred, shareholders would experience a 50% decline in market price but would then be earning a 25% shareholder yield. If things continued this way, you’d be right back to breakeven in two years after collecting annual 25% yields.

Public markets are generally too efficient to allow for investors to collect extraordinarily high yields without the stock price adjusting (i.e. going up). In this sense, when you pay a low price you don’t need a valuation re-rating to make good returns, you simply let the business kick off cash and you earn a good base return via dividends, share repurchases, and debt paydown, which transfers equity value from debt to equity holders.

When you pay a low valuation for a stock it doesn’t need to grow or re-rate in order to earn satisfactory returns. Expectations are usually low and positive surprises are easier to come by. An important caveat is that this is generally true for businesses that are able to grow, even if just by a little. When you pay a price that implies a terminally shrinking business (~10x or less) but the business is likely to grow, you’ve found an asymmetric opportunity that tilts the odds in your favor.

This means that you can be wrong and still likely not lose money (or at least not too much).

Let’s say you pay 10x for a business that you believe will grow. After holding it for two years you realize you were wrong. It hasn’t grown, it’s prospects for growth are not good, and the stock de-rates further to 8x earnings. You lose 20% of your purchase price due to a valuation de-rating, but because the business was returning 10% per year via dividends and share repurchases you essentially breakeven. You were wrong but you didn’t lose money and your investment process worked.

Plan A is to not be wrong. However, it is a dangerous game to assume you won’t be wrong a material amount of the time when investing, and paying a low price is the best way we know to guard against the times we will inevitably be wrong. It allows us to weather our mistakes and keep plowing ahead. This brings us to our valuation mantra when buying:

Buy at a valuation cheap enough where we should make good money by the stock’s business returns (growth and yield) alone, and at a price cheap enough whereby we’re unlikely to lose much if we’re wrong.

By maintaining strong discipline on the buy side to protect us when we’re wrong, it allows us to be more flexible on valuation for times when we’re right. Which brings us to the next step in valuation discipline; holding.

Valuation and Holding

We used to sell stocks too early. This left money on the table and created meaningful opportunity costs. The main culprit for selling too early was that we put too much focus on the valuation multiples of stocks we had purchased for low prices.

The most common scenario of selling too early would occur when we purchased a stock and it in short order went up significantly (say 25-50% in the first year). We don’t usually invest in companies whose earnings go up 25-50% in a year, which means when a stock we own rises quickly it’s due to a rising valuation multiple. This naturally makes us uncomfortable given we prefer cheap stocks. We’ve learned it’s critical to fight this discomfort. If you buy a stock thinking you can make 3 or 4x your money over 5 years, well, you can’t make a 4x if you sell after a 25% pop. It requires patience and fortitude to hold even in the face of a potential near term “retracement”.

Usually when a stock you own goes up a lot it’s because something good has happened. You were right about something. When you sell too early you don’t allow yourself the time to be proven really right and earn the fruits of significant compounding. Selling at the first sign of being right is like cutting a flower as soon as it begins to bloom. We’ve come up with a new framework to help maintain patience and let our positions that we were right about continue to reward us;

Don’t feel an urge to sell stocks just because the valuation multiple has risen. Provided the business is still performing, don’t feel the need to sell an above average business for a below average, or even average valuation, just because it’s not as cheap as when you bought it.

If the forward returns of a stock (growth + yield + change in valuation multiple) in aggregate still comfortably exceed our targeted return hurdle, we don’t feel pressure to sell. The market can often reward stocks with higher valuation multiples than we would assign to them when things are going well, and it’s important to harness this fact to not unnecessarily truncate your returns from positions about which you were right.

We’ve also frequently been asked by new investors if we allocate new capital to existing positions that have risen dramatically. The answer is yes for all of the reasons above. Oftentimes stock’s whose valuation has risen since we purchased them could actually be viewed as less risky than when we bought them at a lower valuation. All else equal lower prices equal lower risk but usually when a stock has started to re-rate whatever uncertainty that existed when we purchased (and that was causing a low valuation) has begun to clear. When this happens we are just beginning to enjoy the fruits of being correct. Of course, we won’t hold, let alone buy for an incoming investor, stocks whose forward returns don’t meet our internal hurdle rate expectations; more on that below.

As is always the case in investing, there are no hard and fast numerical valuation rule I can provide as to when you should be patient in holding a stock that is working. It depends on the business. A business that we expect to grow 15-20% is easy to hold at 15-20x earnings whereas a business that we expect to grow 5%-7% we are unlikely to hold at 20x earnings. In either case the guard rail is the same; hold stocks when forward returns exceed our hurdle rate and the business is performing in-line or better than expectations over the long run.

A particularly important caveat is necessary for all of this. The above discussion on valuation only holds true when we believe the core thesis of why we bought a stock is intact. If something fundamentally changes about the business and we realize our thesis was wrong, we sell. Valuation is not an important part of the equation when we realize we made an error in judging the business. When you’re wrong it is not time to be patient or hold a stock just because it’s cheap. In that instance, almost always the right thing to do is move on.

Valuation and Selling

Deciding when to sell is directly related to holding, so the two concepts are tied at the hip. We try to make things easy on ourselves when it comes to using valuation as the only reason to sell; don’t do it except in the extremes or unless you have a far superior opportunity.

When you’ve bought a good business for a good price Matt likes to say we should make the market pry it out of our hands. This means we generally need to be offered a rich price to sell a great business that is performing well. This rule helps us maintain patience and default to inactivity. But what does a “rich” price mean?

As with buying and holding, we again look to our forward returns framework to help decide what price is too high for an existing position.

When we believe a stock’s valuation is high enough that it will provide a meaningful headwind to returns such that our forward returns drop well below our hurdle rate, we usually sell.

For example, if we own a stock that’s growing at 5% and returning all excess free cash flow to shareholders, when it trades at 10x earnings our forward business returns are 15% (5% growth + 10% yield). We’re also likely to get a tailwind from a valuation re-rating, taking forward returns to 20%+. Because yield is a direct function of price, if the same stock rises to 25x earnings but nothing has changed about the business, the forward business returns have dropped to 9% (5% growth and 4% yield). Further, a 5% growing stock is unlikely to remain valued at 25x earnings, and if the stock drops back to 15x, a fairer multiple, that will provide a 40% headwind to forward returns, likely wiping out our 9% annual business returns for several years. In this case, we’ll most likely be looking for a new place to park our capital.

We also generally require a new idea whose forward returns far exceed the forward returns of the position we are selling in order to exit a position. We almost never sell a position and move the allocation to cash just because the valuation has gotten a little ahead of itself. Fortunately, we generally aren’t short on ideas.

Our framework towards valuation is not the only or even the most correct one, it’s just what works for us. Plenty of investors have different definition of what a richly valued stock means, and loads of investors make more money than we do by buying stocks at valuations we would be happily selling. This framework is what works for us and it suits our style of focusing on durably good returns.

As a general rule, when a stock we own is valued at a multiple that starts with a 2 or higher (i.e. 20x+ earnings) we start to get uncomfortable. The S&P 500 trades at 26x earnings, extremely high in our view by almost any measure, and we find it hard to hold onto stocks that we consider clearly above average businesses at a sizeable discount to even the S&P 500 index. We’ve gotten better about patiently holding onto positions, but we still have work to do, and our attitude towards valuation is nuanced and evolving.

To sum it up, we like to think of valuation as:

• our safety net when we’re buying;

• a tailwind and guardrail to keep us patient when we’re holding, and;

• a tool to assess opportunity cost when deciding what and when to sell.

We try to remember it’s ok to be greedy in both directions. It’s easy to be greedy when buying a stock and demand a low valuation, but it’s often equally important to be greedy when selling and demand a high valuation, just as a private business owner would do.

Subscribe now

EPC offers separately managed accounts, low minimums, and accepts most account types (including IRAs and 401k rollovers). Qualified Clients can elect a zero management fee structure inspired by the Buffett Partnership. EPC’s clients get complementary access to our premium Substack. To learn more please email info@eaglepointcap.com.

Disclosure: The author, Eagle Point Capital, or their affiliates may own the securities discussed. This blog is for informational purposes only. Nothing should be construed as investment advice. Please read our Terms and Conditions for further details.

Sirius XM operates a satellite radio network and the Pandora music streaming service. It is a predictable and profitable business with recurring revenue, low churn, and significant free cash flow.

In late 2024 Sirius separated from Liberty Media, becoming an independent public company. 2025 was a year of stabilization. Key metrics were largely flat, but management progressed several initiatives which will serve Sirius over the long term.

Our thesis is that Sirius is a stable business whose capital expenditure cycle is peaking. We expect flat revenue but increasing free cash flow as satellite spending slows. As debt gets repaid, we expect management to use free cash flow for repurchases, adding to the stock’s 4.7% yield.

Sirius’s 2025 performance met these expectations. Revenue fell 2% while free cash flow increased 24% to $1.26 billion. Self-pay subscribers ended the year at 32.9 million, a decrease of 301,000. The quarterly results show an encouraging trend. A loss of 303,000 subscribers in Q1 narrowed to a loss of 68,800 in Q2 and a loss of 40,000 in Q3. In Q4 Sirius added a net 110,000 subscribers.

The loss of 303,000 in Q1 was the result of a December 2024 decision to focus on the in-vehicle experience and deprioritize high-cost and high-churn streaming-only customers. Sirius has an advantage in-car, where it has its proprietary hardware and fewer services competing for its customers’ attention. Q1’s high churn was a deliberate cleansing of the subscriber base.

In July the company launched SiriusXM Play, a low-cost, ad-supported option that costs less than $7 per month. It is designed to widen the top of the funnel by targeting price-sensitive drivers.

In December Sirius introduced companion subscriptions. These allow customers to add a second vehicle or family member to their plan at a reduced rate. Companion subscriptions are like the family plans Netflix and Spotify offer. They have proven to reduce churn because they lock-in an entire household, not just one driver.

Companion subscriptions drove the surge in Q4 subscribers. Of the 110,000 net new subscribers, 80,000 were companions. Management expects this to be a one-time effect and cautioned that the 2025 results will make for difficult comparisons in 2026. Stripping out the one-time 80,000 companion additions leaves 30,000 net adds, which is still a strengthening trend.

Churn remained a best-in-class 1.5%, an improvement from 1.6% in 2024. This indicates that while it is difficult to acquire new subscribers, existing subscribers are sticky.

Average Revenue Per User (ARPU) was $15.11, a decrease of $0.10 or 0.7% for the year. Sirius raises prices every two years and did so on schedule in March 2025. SiriusXM Play and companion subscriptions pressured ARPU and offset the price hike. This was intentional. In exchange for a few cents of ARPU, management built a larger funnel of subscribers on lower priced tiers they can upsell. The ad-supported tier is dilutive to ARPU but accretive to gross profit as it brings in more ad dollars not measured by ARPU.

Advertising revenue was up 1%. Despite the topline stability, ad revenue shifted from music to podcasts. Podcasting revenue grew 41% in 2025, lapping 12% growth in 2024. The SiriusXM Podcast Network reached the #1 spot in the nation for weekly listener reach (according to Edison Research) and now represents more of the Top 20 podcasts than any other network. In February 2026 Sirius re-signed Howard Stern to a new three-year contract.

Pandora’s monthly active users fell 5%, and ad revenue fell 13%. Listening hours decreased 3% to 9.5 billion as users shift towards on-demand music streaming services like Spotify and Apple Music.

Management expects total ad revenue to remain stable as podcasting growth continues to offset the managed decline of Pandora’s music streaming.

Sirius returned $501 million to shareholders in 2025. $365 million via dividends and $136 million via opportunistic repurchases. The stock currently yields 4.7% and repurchases added 1.2%. Total debt fell by $669 million, reducing leverage from 3.9x to 3.6x. Management’s target is the low-to-mid 3s, which it should achieve in Q4 2026.

In February 2026 Sirius refinanced $1.25 billion of 3.125% senior notes due in 2026 with $1.25 billion of 5.875% notes due in 2032. Interest costs will increase $34 million per year but are locked for six more years.

Sirius trades at a rock-bottom 5.6x free cash flow, an 18% yield. Management expects free cash flow to increase 7% to $1.35 billion in 2026 and 11% in 2027 to $1.5 billion. Deleveraging should finish in Q4 2026, making an additional $600+ million available for repurchases. We expect Sirius to return nearly $1.2 billion to shareholders in 2027, a 15.6% yield.

Forward returns are compelling. A combination of an 18% free cash yield plus 10%+ annual free cash flow growth puts business returns squarely above 20% per year. If free cash flow continues to grow and gets returned to shareholders, we think a valuation re-rating is likely. That would push the stock’s returns even higher.

Subscribe now

EPC offers separately managed accounts, low minimums, and accepts most account types (including IRAs and 401k rollovers). Qualified Clients can elect a zero management fee structure inspired by the Buffett Partnership. EPC’s clients get complementary access to our premium Substack. To learn more please email info@eaglepointcap.com.

Disclosure: The author, Eagle Point Capital, or their affiliates may own the securities discussed. This blog is for informational purposes only. Nothing should be construed as investment advice. Please read our Terms and Conditions for further details.

If you feel like the world is changing at an ever-faster pace, you are not alone. Since we started Eagle Point Capital in 2017, we’ve witnessed the rise of cryptocurrencies, EVs, autonomous vehicles, drone warfare, GLP-1s, generative AI, and large language models. We’ve watched fads in ESG, FAANG, SPACs, memestocks, memecoins, and the metaverse come and go, and watched the Magnificent 7 compound into the most concentrated market since the 1930s.

We survived a global pandemic that severed supply chains and ushered in an era of protectionism and tariffs. Inflation soared to forty-year highs, ending the zero-interest rate era and causing a wave of bank failures.

A hot war has raged in Europe for four years, along with conflicts in Israel, Palestine, Venezuela, and Iran, reshaping the geopolitical order and causing defense budgets to bulge.

We could never have predicted these events. Even if we had, we wouldn’t have predicted the market’s reaction to them. If you knew what was going to happen, would you have predicted that the S&P 500 would return 15% per year from 2017 through 2025?

Our investment strategy is grounded in our business-owner mindset. We own meaningful stakes, at attractive valuations, in a selection of the world’s best businesses. This is exactly what the world’s wealthiest individuals have always done.

Though we own publicly-traded securities, our mindset is that of a private, long-term owner. Instead of trying to anticipate change, we invest in areas that are unlikely to change.

Technology changes but human nature doesn’t. Humans have been consuming nicotine for over 10,000 years and gambling for at least 3,000. That bodes well for British American Tobacco, Philip Morris, and Brightstar Lottery. Humans have been wearing jewelry as a status symbol for at least as long, which benefits Watches of Switzerland. AutoZone and Dollar General deliver convenience, which never goes out of style.

McKesson, the oldest company in our portfolio, is a great illustration. McKesson began delivering pharmaceuticals by horse and buggy in 1833, almost 200 years ago. The world has changed a lot since then, but what hasn’t is the need to get pharmaceuticals from point A to point B safely, securely, and efficiently. That is a societal need that will never change.

Today McKesson uses trucks and vans, but someday it may use drones or something else entirely. The world is certain to change in ways we cannot fathom. Those changes will inevitably have a trickle-down effect on McKesson. We’re confident McKesson can weather the change for a few reasons.

First, as discussed, McKesson’s core service of delivering pharmaceuticals will never be obsolete. The specific means of delivering them may change, but the core service won’t. McKesson is a technology consumer, not a technology supplier, and will adopt beneficial technologies as they become available.

Second, McKesson operates atop a stable and consolidated industry with a fragmented customer base. Growth plods along at a slow but steady pace. Scale advantages protect the industry from new competitors. The total addressable market is not large enough for a new competitor to justify the tens of billions of dollars of investment it would take to replicate McKesson’s physical infrastructure. It cannot be disrupted overnight.

We favor businesses with physical infrastructure advantages because the physical world changes slower and is harder to disrupt than the digital world. We particularly like physical networks, like McKesson, AutoZone, and Dollar General, which are hard to replicate. Unique and irreplaceable assets, like Brookfield’s 235 hydroelectric facilities, 30 port terminals, and 14,300 miles of natural gas pipelines, are great too.

Digital businesses like Zoom, Dropbox, and Evernote were once thought impregnable. Low switching costs, a rapid pace of change, and fierce competition proved otherwise. Now they are considered commoditized products that can be swallowed by larger platforms or replaced by a better user interface. They are neither enduring nor robust.

AI coding tools will accelerate the pace of competition and change in the digital world. The physical world will be affected, but slower. Physical businesses will have a better opportunity to adapt. It would take a lot of time and money to replicate McKesson’s physical infrastructure, with no promise of any customers when it was done.

Third, McKesson provides a mission-critical service. The American medical system rests on its shoulders. Yet it earns net margins of 1-2%. It provides significant value relative to its cost, which gives it bargaining power.

To generalize, Eagle Point looks for:

1. Products and services that satisfy timeless human desires;

2. A stable and consolidated industry with predictable and moderate growth;

3. A fragmented customer base;

4. A competitive advantage or “moat” which creates a barrier to entry;

5. Hard to replicate physical infrastructure;

6. A mission-critical product or service;

7. Products with a low price relative to their value; and

8. A business that is not dependent on rapidly changing technology.

Few businesses have all of these qualities. McKesson is one of the rare ones. The more qualities a business has, the better they’re positioned to confront change from a position of strength.

Businesses that hinge on technology are too fragile to interest us as investments. We prefer cockroach-like businesses: very hardy and almost impossible to kill. Cockroaches have existed for 320 million years, outlasting the dinosaurs and surviving every mass extinction in Earth’s history. They’re not pretty, but they are robust, just like our investments.

The Giant Panda is the opposite of the cockroach, and an example of what we avoid. While cute, cuddly, and beloved worldwide, they are evolutionary specialists hyper-optimized to live in a narrow band of bamboo forests in central China. They thrive in this environment, but struggle to adapt to even minor changes. As climate change and deforestation shrink their natural habitat, they teeter on the precipice of extinction.

If McKesson is a cockroach, adapting to the tumult and technological change of the last two hundred years, then Nvidia and TSMC may be pandas. They’re beloved by investors because they dominate the cutting edge of their industries. But the technology in their fields is changing rapidly. It is unclear if they will be able to change with it. They must continuously disrupt and reinvent themselves before their competitors do. The odds are that eventually they’ll stumble. No one bats one thousand forever. History is littered with companies, like Intel, that looked dominant but could not adapt as their environment changed.

Chegg is a panda with a cautionary tale. The company spent a decade building a database of 120 million answers to textbook questions and sold access to students for $20 per month.

ChatGPT’s release in late 2022 disrupted Chegg’s business model overnight. It gave students instant, unlimited, and customized homework help for free. Students, notoriously tech forward and price sensitive, immediately cancelled their Chegg subscriptions en masse. Three years later Chegg’s stock has lost 99% of its value.

McKesson provides a physical service that technology improves. Chegg provided a digital service predicated on information scarcity that technology erased. Chegg’s customers did not have any switching costs, so the pace of change was rapid and severe.

If the last nine years have taught us anything, it is that the future will be volatile, uncertain and unpredictable. Fortunately, we don’t need to predict it to thrive. By focusing on timeless human desires and mission-critical physical infrastructure, we’ve anchored our portfolio to the areas of the world least likely to change quickly. By stocking our portfolio with cockroaches instead of pandas, we gain resilience to the technological changes that inevitably occur.

We may not make trendy investments, but we are confident that by prioritizing durability over novelty, Eagle Point will continue to survive and compound through whatever environment the next nine years may bring.

Subscribe now

EPC offers separately managed accounts, low minimums, and accepts most account types (including IRAs and 401k rollovers). Qualified Clients can elect a zero management fee structure inspired by the Buffett Partnership. EPC’s clients get complementary access to our premium Substack. To learn more please email info@eaglepointcap.com.

Disclosure: The author, Eagle Point Capital, or their affiliates may own the securities discussed. This blog is for informational purposes only. Nothing should be construed as investment advice. Please read our Terms and Conditions for further details.

Whether it was experimenting with unique SPAC structures in 2020, becoming deeply involved in geopolitics (particularly on twitter) in 2024, taking over Howard Hughes to turn into his version of Berkshire Hathaway, or attempting to raise a new large closed end fund, it’s proven hard for Bill Ackman to sit still and just focus on his portfolio. This week he announced a very unique idea to revive a previously failed U.S.-listed closed end fund.

On March 10th Ackman outlined a plan to complete a dual IPO for Pershing Square U.S. (his long-sought closed end fund) and Pershing Square Inc. (the management company for all of his funds). Ackman has had perhaps the most influence of any investor on the characteristics we look for in businesses - though how we apply them is often vastly different than how Pershing Square applies them - so I can’t help but dig in whenever he comes out with a new offering.

Ackman has long dreamed of raising a huge sum of permanent capital in one shot; hence his plans to start a closed-end fund listed in the U.S. To say it’s hard to raise a publicly traded closed end fund nowadays is an understatement. Publicly listed closed end funds nowadays almost always trade at a discount to their underlying NAV. This is a big problem when trying to seed raise capital. With the rise of ETFs and open end fund structures, no one wants to lock up capital at one price only to see it trade at a discount days later and have no recourse to redeem their shares at par. For those that aren’t up to speed on the background, I can’t summarize the lead-in to this situation better than Matt Levine’s Bill Ackman Wants Your Money Again article from earlier this week:

“In 2024, Bill Ackman came to the stock market with a proposition: You could give him $50, and he would invest about $49 of it in some stocks for you. (The other $1 would go to underwriting fees.[1]) This is called a “closed-end fund,” and Ackman’s is called Pershing Square USA Ltd., or “PSUS.”

How much is that proposition worth? How much should you pay for $49 worth of stocks selected for you by Bill Ackman? Some possible answers include:

• $49, because $49 worth of stocks is worth $49;

• less than $49, because Ackman will charge you fees for picking the stocks, which will reduce the value of your stake over time[2]; or

• more than $49, because Bill Ackman is good at picking stocks and will pick ones that go up more than the average stock.[3]

Ackman obviously argued for “more than $49,” because he is quite confident in his ability to pick the stocks that go up. Others disagreed. Closed-end funds often trade at a discount to net asset value (i.e., less than $49), so it would not be unreasonable to worry that Ackman’s would also trade at a discount. In fact Ackman already has a closed-end fund in Europe, called Pershing Square Holdings Ltd., and it trades at a discount, so.

This is a big problem when it comes to raising money. If Bill Ackman comes to you and says “give me $50 to buy $49 worth of stocks,” and you think that proposition is worth $52, you give him the $50 and everything is great. But if you think that proposition is worth $47, there’s a problem. You can’t give him $47 to buy $49 worth of stocks: Where would he get the other $2? (He could just buy $46 of stocks for you, but that’s not worth $47 — that’s worth like $43. Etc.) Ackman can only raise new money from people who think that his fund is worth more than net asset value; he can only get this started by selling shares at a premium.

Anyway this didn’t work. Ackman set out to raise something like $25 billion for Pershing Square USA, cut that back to $2.5 billion, and ultimately raised $0. Oops.”

The Pershing Square team went back to the drawing after 2024’s flopped fundraise and came up with a new idea. Instead of asking investors to invest in a closed end fund at NAV and explaining all the reasons why their fund won’t trade at a large discount to NAV like the other closed end funds, Ackman is going to give investors who invest in Pershing Square U.S. (PSUS, the closed end fund) shares in his management company, Pershing Square Inc (PSI). I have to admit, this is pretty creative and I can’t wait to see how it unfolds.

Here’s how it will work; for every 100 shares that IPO investors invest in PSUS they will receive 20 shares in PSI. The larger the fundraise for PSUS, the more management fees that will go to PSI and the more benefit the IPO investors will receive, making the deal somewhat reflexive.

Pershing Square Inc. is a fantastic business for its owners. Look no further than Ackman’s comp from last year; he made $142.8 million and his lieutenant, 40-year-old Ryan Israel, made $44 million. The management company is an asset-lite business that has rapidly compounded it’s earnings power over the last several years thanks to solid returns from the flagship fund.

Receiving shares in a fantastic business is always worth considering, the question is at what price are investors receiving these “bonus” shares in PSI, and will that make up the difference in value if and when PSUS trades at a discount to NAV? As usual, the intricacies of this Ackman deal are numerous and it’s not light on complexity. Before exploring the math behind the deal, let’s look at the rationale and structure.

IPO Rationale

In a letter to investors, Ackman explained that the best investments:

“are simple, predictable, and free-cash-flow-generative. They have impenetrable ‘moats’ or large barriers to entry protecting them from disruption, a growing risk in a world in the midst of massive technological change. They are often asset-light, and generate recurring and rapidly growing, inflation-protected, free cash flows that do not need to be reinvested in the business for the company to grow. They do not need large amounts of leverage to generate high returns on capital. And they have limited exposure to extrinsic threats outside of the control of the company… In this combined offering, along with your investment in Pershing Square USA, Ltd. (PSUS), we are providing you with an interest in Pershing Square Inc. (PSI), an alternative asset management firm that we believe meets all of the above criteria.”

He goes on to outline why it makes sense to offer shares in his business:

“PSI will become a publicly listed company concurrently with the PSUS IPO, and will trade on the New York Stock Exchange along with PSUS. By providing an extra incentive to invest in the PSUS IPO, the distribution of PSI shares to PSUS IPO investors should help us to execute a large-scale, successful PSUS IPO. A highly successful, large-scale PSUS IPO will make PSI more valuable by materially increasing PSI’s fee-paying assets under management (AUM), revenues, and cash flows. The larger the PSUS IPO, the more valuable the bonus shares of PSI you will receive.”

Basically, Pershing Square is offering shares in its lucrative management business in hopes that the shares of PSI more than offset the discount that PSUS might trade at after it IPOs and to induce investors to put as much capital as possible into PSUS because their PSI shares will benefit from the increased AUM at PSUS. Later we’ll look at what valuation PSI needs to achieve vs. what discount PSUS could trade at for this deal to make sense.

Pershing Square Inc.

Ackman believes PSI should achieve a premium valuation in the public markets. Pershing Square does enjoy a durable competitive advantage thanks to its largely permanent capital structure. Roughly 96% of Pershing’s capital is non-redeemable, the majority of which is housed in Pershing Square Holdings, a European listed closed end funds (that happens to trade at a large discount to NAV).

Competing against investment firms that offer daily, monthly, or quarterly liquidity is a major advantage for long-term investors as the investment managers do not need to acquiesce to short term performance mandates and can instead invest with an eye on optimizing long-term returns. I fully agree with Ackman that this represents a real advantage over time.

In addition, the permanent capital nature of Pershing Square in theory means they do not need to dedicate undue resources to fundraising – except apparently this one time when raising PSUS. Most alternative asset managers like KKR, Apollo, Brookfield, etc. manage long-term capital but it is usually held in fund structures that have finite life cycles and the capital ultimately needs to be returned to investors. This means that growth for these businesses depends in large part on the managers ability to continually raise capital in order to grow their fee-bearing AUM. They have, however, had no problems to date raising capital at an ever-increasing rate.

In his investor letter Ackman writes:

“If PSI never launches a new fund nor raises a dollar of new capital but continues to generate strong investment returns, PSI’s revenues, cash flows, and intrinsic value will grow at a rapid rate. The substantial majority of the cash flow that PSI generates can be returned to shareholders as our business model requires limited capital for reinvestment. PSI is one of the rare businesses that can grow at a high rate while returning the substantial majority of its capital to shareholders.”

While I agree that PSIs permanent capital is highly desirable and enables to them grow without raising new capital, there are puts and takes to PSI’s model.

On the positive side, PSI charges a management fee on an asset base that generally goes up as they earn investment returns (assuming they do a good job, which they largely have). This is great. They also earn performance fees on mark-to-market performance, which is common for public investors and is also great. Alternative asset managers usually don’t earn performance fees until assets are actually sold, making carried interest inherently lumpy and difficult to value for public markets investors.

On the flip side, other alternative managers mostly play in the private markets and charge management fees on a more stable equity value. PSI’s management fees can easily decline 20-30% in a year (right now Ackmans flagship fund is down ~15% this year, and management fees are down commensurately). Alternative asset managers don’t have this problem; it’s almost inconceivable to imagine KKR or Brookfield’s management fees compressing at all, let alone dramatically.

The last point Ackman highlights is PSI’s operating efficiency. Because the firm manages a concentrated public portfolio they do not need scores of employees. Unlike private alternative managers who wholly own and operate businesses – which takes tremendous manpower – PSI enjoys one of the highest ratios of fee paying assets to investment professionals in the industry.

This high assets-to-employees ratio allows PSI to recruit and retain very high quality talent because they can pay a boatload. This is undoubtedly another advantage PSI should enjoy for years to come.

There are puts and takes to both models, and at the end of the day they are both exceptional business models, so it probably isn’t worth splitting hairs. PSI is a great business as are the other alternative asset managers.

Structure

Pershing Square Inc. earns fees a few different ways, all of which will impact shareholders once the stock IPOs. In 2025 about 30% of the revenue was from fixed management fees. Most of this is from the 1.5% fixed fee they charge on Pershing Square Holdings (PSH). PSH is a closed end fund that trades in Europe and manages about $17 billion.

The fund has generated solid returns with a 16.9% annual NAV return over the past 10 years and 12.4% since inception (12/31/2012). Ackman went through a brutal stretch from 2015 – 2018 when his large investment in Valeant tanked at the same time his short position in Herbalife went way against him. He righted the ship and has done well since then and has an excellent long-term track record, though the fund is down substantially this year.

Two other small private funds make up another couple billion of AUM and earn fixed fees.

The real upside to Ackman and team comes through performance fees. They earn a 16% performance allocation from their existing funds. The kicker is there is no hurdle rate, just a high-water mark, so 16% of all gains above the fund’s previous high watermark go to PSI. Performance fees made up 55% and 70% of PSI’s revenue in 2024 and 2025, respectively. PSUS will not charge a performance fee and will only have a 2% fixed management fee due to regulations in the U.S.

There are a few weird aspects of how these fees will be split with shareholders post-IPO.

Once PSI goes public, the performance fees from PSH will be split between shareholders and employees. The performance fees from the first 5% net returns will belong to shareholders – these are referred to as “Preferred Performance Fees”. Let’s say the fund returns 10%; common equity holders will receive 16% of the first 5% of returns. If the starting asset base for performance fees is $15B then the Preferred Performance Fee would be $15B x 5% x 16% = $120 million.

Anything above this threshold will go to employees and be called “Subordinated Performance Fees”. This is pretty weird. Ackman is doing this for a few reasons.

First, the public markets tend to discount performance fees when valuing investment managers. As we’ve written about, Brookfield’s carried interest is likely to be sizeable and very important to shareholders, but the market hardly incorporates any carried interest into BN’s valuation because it’s lumpy and hard to predict. Ackman’s goal is that analysts will more easily be able to value the preferred performance fee stream because most years it’s likely to be earned given the fund just needs to return 5%. So, instead of discounting this fee stream Ackman is hoping the market ascribes a higher multiple to these fees.

Next, he explains that PSI needs to keep plenty of upside to entice his team to stay at the firm. He argues that by keeping the upside beyond a 5% return he won’t have to issue equity to employees and public shareholders won’t be diluted. I mean, I get it, but it sure seems like employees are keeping a lot of the upside for a public company.

The final weird element to PSI is the relationship to Howard Hughes (HHH). To make a long story short, Pershing Square took majority control of HHH last year and has turned it into another perpetual capital fee-paying vehicle. Ackman intends to try to turn HHH into his mini Berkshire and has already agreed to acquire an insurance business. Howard Hughes pays PSI an annual $15 million management fee along with an annual percentage of the market cap compared to a baseline; another highly favorable fee arrangement for PSI.

Once PSUS goes public PSI will have another long-term $100 million+ high margin fee stream without adding any material incremental expenses.

Valuation / Prospective Returns

There are two questions that really matter for IPO investors; what is the “bonus” stake in PSI worth and at how big of a discount to NAV will PSUS trade? Obviously I don’t know the answers to these questions but we can make some educated guesses.

As to the first question; Pershing Square sold 10% of PSI in 2024 in a private transaction that valued the firm at about $10B. Is this realistic?

Asset managers are usually valued either on a percentage of AUM or, more traditionally, as a multiple of distributable earnings (I prefer this method). Either way you look at it, $10B was a handsome valuation.

The firm had about $17B of AUM (“core AUM” including Pershing Square Holdings the two private funds), so at the time PSI was valued at roughly 60% of AUM. At first glance this looks stratospheric, but it’s not as egregious when you look under the hood. Alternative asset managers like Blackstone, KKR, Apollo, and BN are valued at 10-15% of AUM. This isn’t the most useful metric because of PSI’s insanely high margins and high AUM-to-employees. Price to distributable earnings is a far more relevant metric.

On a distributable earnings basis, Brookfield and KKR trade around 15x and Blackstone is slightly more expensive at 19x distributable earnings. A $10B 2024 valuation for PSI equates to about 32x trailing distributable earnings for PSI, about double most comps. But this isn’t the number we really care about because it doesn’t incorporate the new fee stream from the launch of PSUS (the whole reason PSI is IPO’ing in the first place!).

On a pro-forma basis, the valuation is more reasonable. PSUS is targeting a $5-$10B initial fund raise at the IPO (a few billion has already been committed via a private placement), which they refer to as “scenario 1” and “scenario 2” in the S-1 filing. If we add in incremental fees from PSUS and account for other weird things like an offset agreement with Pershing Square Holdings, a $10B valuation for PSI would equate to a 23-27x multiple of DE; still high but maybe not absurd? A premium to the other public alternative asset managers is perhaps justifiable because of the permanent capital nature, long growth runway of PSI, and all the other reasons discussed above.

That’s all well and good, but what we care about even more is what could PSI be worth a few years from now? Here is how DE has compounded over the past five years (adjusted for the new comp structure with the Preferred/Subordinated performance fee split):

Distributable earnings have grown at a 19% compound annual rate since 2020 which closely tracks the firms CAGR in AUM which has been a whopping 24% per year since 2020. Core AUM has compounded at about 14% per year with the balance of the increase coming from consolidating HHH which now counts as AUM for Pershing Square.

If PSI can keep growing distributable earnings at a mid-to-high teens rate for the foreseeable future, a 20x+ multiple for the stock will prove reasonable. Put another way; business returns for PSI at 20x distributable earnings and 15% growth are 20% (5% yield + 15% growth).

Finally let’s look at the combined offering; investing in the PSUS IPO and getting shares in PSI. If the $10B PSI valuation sticks, that puts the share price at $25 (they have 400 million shares outstanding). An investor that invests $10K into the IPO will receive 200 shares in the fund (PSUS) and 40 shares in PSI. Total NAV of the two combined holdings will be $11,000, meaning as long as PSUS trades at less than a 10% discount to NAV, you’re in the money.

What about the forward returns? That requires assessing how you think Ackman’s portfolio will do. If you take his super long-term returns as a proxy for forward returns, which I think is reasonable, the portfolio, and therefore AUM and fees, could compound at 15-16% per year.

If he achieves similar returns over the next decade, PSI’s intrinsic value will compound at 15%+ per year as will the investment in PSUS.

The bad news is that earnings are likely going to be depressed this year because PSH is down quite a bit to start the year. However, given the concentrated nature of the portfolio, things can turn around quickly. On the other hand, it also means that growth could be super-charged for those that buy into PSUS at potentially depressed prices (I have some doubts about how his recent investments will work out, but we’ll see).

I’m fascinated to find out a) what valuation the market assigns to PSI and b) what discount PSUS trades for.

I think the bonus shares in PSI, the reduced fee burden, Ackmans celebrity status, and the U.S.-based listing make it unlikely that PSUS trades at more than a 10% discount to NAV, other than maybe temporarily. If he comes out of the gates hot, there is an outside shot that PSUS could even temporarily trade for a premium to NAV, though I think it’s unlikely.

The dual listing IPO is a unique structure and makes for a fascinating study. The combined offering doesn’t strike me as an obvious fat pitch, though it seems there’s a reasonable chance IPO investors will do well. I’ll be most interested to follow PSI over time. During a tough market or an occasional blunder from the Pershing Square investment team the management company’s valuation could temporarily become dislocated from reality and prevent an excellent investment opportunity in a capital-light compounder.

Further reading

Bill Ackman Details Plan to Take Hedge-Fund Firm, New Fund Public Simultaneously

Ackman’s Compensation and Other News We Learned From Pershing Square’s Filings

Bill Ackman Has a Stroke of Genius. Buy His New Fund and Get Shares of His Pershing Square IPO

Bill Ackman Wants Your Money Again

Do you have a “stranded” 401(k) from a past job that is neglected and unmanaged? Eagle Point offer separately managed accounts to retail investors, and 401(k) rollovers are often a good fit for our long-term approach. If you would like to invest with Eagle Point Capital or connect with us, please email info@eaglepointcap.com.

Disclosure: The author, Eagle Point Capital, or their affiliates may own the securities discussed. This blog is for informational purposes only. Nothing should be construed as investment advice. Please read our Terms and Conditions for further details.