Building Arks

March roundup

Wed, 01 Apr 2026 21:15:48 GMT

A summary of my activity this month.

Trades

Quite a lot of activity this month as my energy positions outperformed the rest and I recycled some capital. These were mostly small trades.

Reduced Transocean, bought more Howard Hughes, bought more Pershing Square.
Bought Saga Plc.
Sold Noble.
Bought more Ashmore.
Sold Cheniere.
Reduced Peyto.
Bought more Blackstone, KKR, and Apollo.
Bought more Ashmore.
Subscribe now

Articles

Howard Hughes Holdings review.
Brookfield investor day summaries.
Uber review.
Vail Resorts first look.
Cheniere Energy review.

What I found interesting this month

LongYield on Alibaba.
Grow or Die on Uber.
Cayucos Capital on Jardine Matheson.
Future Cognitive Capital on Microsoft.
Subscribe now

Thanks for reading - if you enjoyed reading this please like and restack, and do get in touch if you have questions.

Pete

Disclaimer: This post is for informational and educational purposes only. Building Arks is not licensed or regulated to provide any financial advisory service and nothing published by Building Arks should be taken as a recommendation to buy or sell securities, relied upon as financial advice, or treated as individual investment advice designed to meet your personal financial needs. You are advised to discuss your personal investment needs and options with qualified financial advisers. Building Arks uses information sources believed to be reliable, but does not guarantee the accuracy of the information in this post. The opinions expressed in this post are those of the publisher and are subject to change without notice. The publisher may or may not hold positions in the securities discussed in this post and may purchase or sell such positions without notice.

Review: Howard Hughes Holdings - Ackman's Berkshire?

Wed, 01 Apr 2026 14:30:50 GMT

Summary

What it does: real estate company becoming a diversified holding company.

Elevator pitch: very long term real estate assets are starting to generate cash. One of this generation’s great investors is deploying it. This combination might make a unique compounding vehicle.

Mental model: moat, value (read about my mental models here).

Valuation and potential returns: potential 3x in 5 years.

Exchange and ticker: NYSE, HHH.

Stock price and market cap: $63, $3.8bn.

Do I own it? Yes.

IR website: here.

Introduction and: why now?

Howard Hughes stock has gone absolutely nowhere for 15 years. Why should the next 15 be any different? The answer is simple. The company is starting to generate quite a lot of cash, and will likely do so for decades, and they’ve come up with a very interesting way to reinvest it.

HHH’s core business is in real estate - specifically, Master Planned Communities (MPCs). These assets often take 50-60 years to build out. They consume cash initially but as revenues rise and investment falls they pass a cash flow tipping point. HHH’s MPC business is past the tipping point and is starting to generate significant amounts of cash: I estimate about 50% of the $5.6bn GAV (gross asset value) will be sold within 10 years and 80% within 20 years. The problem is, there are only a few good MPCs nationally, so reinvesting cash into new MPCs can only be done opportunistically - it’s not a strategy.

So what to do with the cash? The obvious answer would be opportunistic buybacks and dividends, with the balance between the two determined by how fully valued the stock is. And right now that would be a good answer, because management insists the stock is undervalued. But it’s not a great long term strategy: eventually the stock will be fully valued, and while dividends return value to shareholders, they don’t create it.

Enter Bill Ackman’s Pershing Square Capital Management. Pershing has been involved with HHH for years, and in 2025, HHH hired Pershing Square to build a diversified holding company. This is riskier than the buybacks and dividends strategy, but if done well will create more value, because it means HHH has a very long reinvestment runway. It can now invest its cash into:

Buying a portfolio of minority stakes in listed companies for value. This runway is not all that long if it is done on HHH’s balance sheet - eventually they’d be classified as an investment holding company, which would bring a regulatory burden they don’t want. However by buying an insurance company and putting the minority positions onto the insureco balance sheet, HHH can redeploy any amount of MPC profits into this portfolio.
Buying control positions in private businesses.

As part of the deal, PSCM injected $900m of new capital into HHH by purchasing common stock at $100 per share. This cash will help fund the acquisition of Vantage, an insurance company. The balance of the investment will be funded by Vantage preferred stock issued to Pershing Square Holdings (PSH). Vantage’s shareholder equity will be invested into a portfolio of common stocks similar to Pershing’s existing portfolios. Future MPC cash flows will be used to pay down the Vantage prefs, and once that is done they will be invested into either control positions in high-quality private businesses or into additional insurance company equity (and therefore listed stocks), depending on where the best opportunities are. The structure looks like this:

This strategy looks incredible on paper, and might actually work. Pershing Square’s investing credentials are clear. But it will not be easy: a lot comes down to Pershing Square’s execution, and whether they are worth their fat fee.

Valuation

The table below shows my sum of the parts (SOTP) for today, which I think serves as a reasonable bear case, and a bull case for 2030. The column in-between shows cash flow for each segment, which builds up to the 2030 cash balance. Figures are in millions of dollars.

Let’s dive into the detail.

Subscribe now

Insurance

In December 2025, Howard Hughes announced the acquisition of an insurer named Vantage, which was founded in 2020 by Carlyle, Hellman & Friedman, and the founding management team.

Insurers make two types of profit (or loss). First, insurers try to charge more in premiums than they pay out in eventual claims. And second, in between collecting the premium and paying the claim, they invest their customers’ money - called float - and keep the returns.

There are two broad strategies in insurance. One is to underwrite programmatically, selling large volumes of less profitable insurance business. This results in high leverage (measured as premiums/equity and assets/equity) and as a result, regulators insist that the assets are invested conservatively. Weak insurance profits and conservative investing produce mediocre returns on assets, but the leverage helps turn this into an adequate return on equity.

The second strategy is to write less insurance, being pickier about the risks taken and the prices offered. This results in higher margins, and also in lower leverage, which has several advantages:

It reduces the impact of investment volatility and adverse reserve developments on regulatory solvency.
It gives flexibility to increase the volume of premiums written during periods of strong insurance pricing.
It improves customer perception of counterparty risk. Customers want an insurer whose cheque always clears.
It allows for optimised asset allocation.

It is this last point that can turn this type of insurer into a compounding machine. This second strategy was pioneered by a company you may have heard of: Berkshire Hathaway. The strategy lowers risks in two ways: it writes far less premium volume per dollar of equity than most insurers, and as a result has lower assets/equity; and it invests its float (i.e. its loss reserves) in cash and short-term Treasurys with little to no duration or credit risk. Because it takes so little risk, regulators allow it to invest its shareholder equity into a portfolio of common stocks. The combined returns from insurance margin, plus interest on float, plus returns on well-selected common stocks, achieved with low leverage and risk, have been the stuff of investing legend - and that is what HHH and Pershing want to replicate with Vantage.

Prior attempts to “replicate Berkshire” by hedge fund managers have not worked. Pershing argue that these were structured principally as financing vehicles, designed to boost funds under management for the fund managers. By contrast, Pershing will charge HHH a fee on its market cap, but they will not charge Vantage a fee for managing its investment portfolio. Their incentive is therefore to create value, not to write additional premiums to increase float and therefore fees for the investment manager, regardless of the insurance returns. Indeed, HHH have been explicit that Vantage will not have any top-down goals for premium volumes: premiums will only be written when the business is judged to be profitable. I think there is a second difference, too. Hedge funds run long/short books, where the longs are typically high turnover and the shorts have uncapped downside that can destroy a fund if they are wrong. There is no fundamental reason why hedge funds should compound - it is all down to the skill/luck of the manager. But Pershing is no longer really a hedge fund. They invest long-only and long term in some of the largest, fastest-growing, deepest-moat companies on earth. While Pershing have an exceptional record of outperforming broader markets, a lot of the compounding simply comes from the holdings themselves.

Because Vantage plans to run with low leverage, float invested in Treasurys, and equity invested in common stock, we can split the valuation into two parts:

The insurance part, which includes selling insurance, paying claims, and interest on float.
The investment part, which is the common stock portfolio.

This is not how most analysts would do it. Most would include the interest on float in the investment bucket, and would then calculate an ROE for the whole company. However, the interest on float is directly related to the amount of insurance written. It is a core part of insurance profits, whereas the returns from investing equity into common stocks is not. The advantage of doing it my way is that we can value the common stock portfolio at market prices and then value the insurance operation standalone.

Why market prices for the common stock portfolio? Because it is easily replicable. Pershing Square have said that Vantage’s common stock portfolio will be transparent and invested similarly to their existing portfolios, in liquid stocks of large companies. This means the portfolio is easy to replicate, and should not be valued higher than market prices even if the returns are spectacular. Pershing disagree with this logic, arguing that it is impossible to replicate the exact timing of their trades, the exact weights in their portfolio, or their occasional highly asymmetric hedges. This is true, but it is clear from the valuation of Pershing Square Holdings, which trades at a persistent discount to NAV, that the market does not see things this way. (Equally, the Vantage portfolio does not deserve to trade at a discount - the PSH discount is justified by fees, which Vantage does not pay.)

Valuing the standalone insurance operation is trickier. Vantage is currently making a 97% combined ratio. This is what most industries would call a 3% EBIT margin: it means that for every $100 Vantage collects in premiums, it pays out 97% in claims and operating costs. Believe it or not, this is actually pretty good: insurance is a commoditised industry and margins are thin.

HHH say that Vantage’s combined ratio will improve over time. They argue that while its loss ratio is good, its expense ratio is not. This is because it is a young business that has not yet achieved scale, so it is not yet fully leveraging its roughly $350m in fixed operating costs. As it grows, its combined (loss + expense) ratio should improve further.

Unfortunately, HHH have acquired Vantage at the end of a long insurance upcycle. Because insurance is commoditised, it is also cyclical. When profits are available, capital enters the market, pushing prices down. When losses are made, capital exits the market, pushing prices up. The key drivers of these cycles are catastrophe losses and interest rates. When (for example) a hurricane wipes out capital in the insurance industry, prices (at least in relevant lines of insurance) are likely to rise. And when interest rates rise (or fall), capital seeking the best risk-adjusted returns leaves (or enters) insurance.

The increase in interest rates after covid, combined with various catastrophe losses, drove a hard market (upcycle) in insurance pricing from roughly 2021-2025. That hard market is now softening. This makes it harder for insurers to write profitable insurance, and is likely to pressure both premium growth and combined ratios across the industry.

The following table shows two possible valuation scenarios for Vantage. Both assume $1.5bn in shareholder equity, 2x assets/equity, and $350m in operating expenses, which is roughly right for 2q26 according to the deal deck and commentary. The scenarios are:

Vantage writes fewer premiums at a higher loss ratio and achieves zero insurance profit.
Vantage writes more premiums at a lower loss ratio and achieves insurance net profit equal to 5% of shareholder’s equity.

This allows us to frame Vantage’s valuation quite simply.

If Vantage’s shareholder’s equity is invested in common stocks, and we are valuing this portfolio at market price, then we are going to pay 1x shareholder’s equity for the common stock portfolio.

In addition, we are going to pay 10x insurance earnings. 10x is a 10% earnings yield, and if you add inflation you get a low-teens total nominal return. That seems fair for a commoditised business.

For now, while I wait to see how the cycle unfolds and whether Vantage is a good underwriter, I will use Scenario 1 and assume insurance profits are $0. My SOTP today therefore only values Vantage at 1x equity for its common stock portfolio. However if Vantage proves its worth then the insurance operation could easily produce a return on equity of 5%. At 10x, that justifies another 0.5x equity for a total multiple of 1.5x, which is what I use for my 2030 SOTP.

The magic of this model is that it might well produce 20% or 25% compounded returns. Pershing Square say they aim for 20% returns on their common stock investments, and insurance could easily add 5%. This is why excellent insurer-investors like Berkshire Hathaway and Fairfax Financial have produced such exceptional compounded track records over time. Before acquiring Vantage, HHH produced this useful table showing how common stock returns and combined ratios drive compounding (the underlying assumptions about leverage and asset allocation are broadly similar to mine):

Four final thoughts on insurance:

Remember: the common stock portfolio is replicable, so even if it does produce a 20% return, it should not be valued at more than 1x. That’s why I structure the valuation as 1x common stocks + 10x insurance ROE, rather than 10x total ROE.
This valuation includes nothing for AdVantage, which deploys 3rd party capital into insurance policies underwritten by Vantage. This effectively allows Vantage to monetise its underwriting skills via a fee-stream, rather than just by writing policies backed by its own capital. This is very valuable in upcycles but less so in downcycles, and I don’t want to ascribe value to it until we have seen how it performs through cycles.
The big risk not captured in my valuation is if Vantage have dramatically underestimated future claims. In this case shareholder’s equity will be overestimated, since it will be used to pay claims. What gives me some comfort here is that virtually all of Vantage’s book has been written during a hard market which offered good pricing. I can’t guarantee it, but there really shouldn’t be systemic problems in the insurance book.
HHH have said that post-close, they will invest more primary equity into Vantage to de-lever the balance sheet and reduce the effective purchase price (since primary capital will go in at 1x p/bv). I don’t model this because I don’t know the amount, but it doesn’t really change my valuation - the cash merely moves from one of HHH’s pockets to another.

Master Planned Communities
These are large plots of land that are steadily sold to build new suburbs. Howard Hughes adds value by securing permits for long term development plans, building basic infrastructure, building commercial real estate, and controlling supply to maximise profits over time. As these communities mature, the price of land for homes rises dramatically. Development can take decades, with negative cash flow initially and very strong cash flow at the end, when the infrastructure is complete and land values are high.

I like this business. It makes intuitive sense to me that by investing early in a place people are likely to want to move to, investing to improve the area, curating a lovely place to live and work, and monopolising the supply of land with a long term value-maximisation mindset, it is possible to make good returns. Howard Hughes’ MPCs are in affordable areas with low taxes and excellent demographics, and land values are rising steadily:

75% of current MPC profits come from Summerlin (22,500 total gross acres along the western rim of the Las Vegas Valley) which opened in 1990, and another 22% from Bridgeland (11,500 total gross acres in Houston) which opened in 2010. The other MPCs are The Woodlands and Woodlands Hills (Houston) which are smaller but are scheduled to ramp profits over the next few years, and Teravalis (33,800 total gross acres in Greater Phoenix, Arizona), which only started in 2024 and is targeted for completion in 2086. In maturity terms Teravalis is the baby of the portfolio, but it is a very big baby.

HHH give us 3 ways to think about valuing the MPC business. The first two are earnings before tax (EBT), which is a P&L measure, and MPC Net Contribution, which is a cashflow measure. To get from EBT to Net Contribution, the big adjustments are that

Cost of sales is added back. This is what HHH paid to buy the land, so it is a noncash charge in the current period.
Proceeds from MUD and SID bond collections and sales are added back. These are bonds issued by municipalities to HHH to subsidise some of the development expenditures.
MPC development expenditures are deducted - this is the investment to make the land saleable (some of which earns MUD and SID bonds which can be collected or sold in future periods).

The following table shows EBT and Net Contribution. Net Contribution has lagged EBT by an average of $75m per year over the last 4 years due to MPC development expenditures. Eventually, Net Contribution will exceed EBT as communities mature and less development spend is needed. We are not there yet, but it is clear that profits and cash flows are growing and guidance is for that to continue, although 2025 was a bumper year on the sale of a particularly large plot in Summerlin so there is a step down in 2026:

I assume that MPC Net Contribution averages $400m per year over the next 5 years. For the purposes of my 2030 valuation, this means $2bn gets deducted from the MPC valuation and added to the holdco cash balance.

The third valuation measure is Gross Asset Value (GAV), which is a discounted cash flow. The table below shows % of GAV and planned sell out dates for each of the MPCs. By 2036, all of Bridgeland residential and The Woodlands, and Woodlands Hills will have been sold, along with perhaps half of Summerlin. That accounts for around 50% of today’s GAV. By 2046, Bridgeland commercial, Summerlin, and some Teravalis will have been sold, accounting for another ~35% of today’s GAV. By MPC standards, these timeframes are short: we are well into the monetisation phase. This data is from September 2024, which is the last breakdown of GAV that I can find:

The most recent GAV is $4.4bn, from 4q25. This excludes Teravalis because it is from a slide showing GAV growth excluding the acquisition of Teravalis in 2021. Teravalis (including Floreo) had a GAV of $944m 18 months ago. Given the high discount rates on these assets, their GAV is likely $1.2bn today, making a total YE25 GAV of $5.6bn. The segment also has a little over $400m in net cash, but this includes $400m of MUDs. HHH has a habit of selling MUDs at discounts of 20-25%, so I reduce segment net cash by $100m. MPC NAV is therefore $5.6bn + $0.3bn = $5.9bn today.

Should we trust a company-provided GAV? As you might expect, the key inputs to the DCF are aggressive, but defensible:

Commercial land price growth rate 3.5%, fractionally above probable inflation as population density rises.
Residential land price growth rate between 5% and 7% depending on the asset. This is in-line with historical data.
Discount rates of 7-9% for the more mature assets, 15% for Floreo, and 20% for Teravalis. The weighted average in 2024 was 9.6% and this will rise slowly as Floreo and Teravalis grow in the mix.
Cash margins between 75% and 97% - selling land you already own is a very cash generative business!

If I was going to question one assumption, it is the 7% residential price growth rate at Summerlin. This asset accounts for a large part of the overall GAV, and while 7% is in-line with historic performance, extrapolating it for the 17 years until sales are complete is not cautious. Offsetting this is the fact that Teravalis, a monster asset, is discounted at 20%; if it works out it will drive huge GAV upside. In fact, GAV is likely to rise despite accelerating land sales. At 20%, the Teravalis GAV will be $7.4bn in 10 years if everything goes to plan. That exceeds the GAV of all the MPCs today.

Overall I think the GAV HHH provide represents a full, but not crazy, valuation for the MPC segment.

Strategic Developments

This segment develops high-rise condos for sale. The majority of what it is building is at Ward Village in Hawaii, but it is also building the Ritz Carlton Residences, an ultra-luxury condo development in The Woodlands, which might become a template for further developments in the MPCs.

Earnings from Strategic Developments are lumpy. It takes 3-4 years to sell a tower, and the profit is booked at the end; on top of that, a couple of the Ward Village towers are affordable homes sold at cost. That’s why HHH booked $282m of segment EBT in 2024, when it had a high-value tower to sell, and -$14m in 2025, when it had an affordable tower to sell.

Nevertheless, the economics of Strategic Developments are attractive. Most of the towers are pre-sold, which reduces risk. Between deposits and construction loans, HHH has to put up very little equity to get a tower built. They have a decent record of delivering towers on-budget. And they already own the land, so the cash margins are decent. HHH guide to a 25-30% gross profit on sales. After net interest, I estimate 15% pretax margins and 12% net. That’s in Hawaii, where construction costs are high, because labour is tight and materials have to be shipped.

HHH has already delivered 8 towers at Ward Village, 2 of them affordable. It has 2 more under construction and 3 in pre-development. These last 5 are 71% sold on average with $3.5bn in revenue already under contract, which implies $4.9bn in total eventual revenues. In practice revenues will probably be slightly higher, since the final units usually go for the highest price.

The Ritz Carlton Residences are 76% pre-sold with $37m in deposits taken and $380m of revenue under contract, implying total sales of $500m. Working through build cost and likely interest profile I think net margins are likely to be slightly higher here - I assume 15%.

All of these developments will be complete by 2030. $5bn in revenues at 12% and $500m at 15% implies $675m in net profit over 5 years. I discount this to $500m for today’s valuation.

For the 2030 valuation, two things are relevant. First, in early 2025, entitlement changes at Ward Village made another 2.5-3.5 million square feet of development possible after 2030. That’s at least equivalent to the 5 towers currently in development and pre-development. Second, the success of the Ritz Carlton Residences, which generated $250m in future sales in their first week of marketing, is something that HHH might be able to replicate as they build out their MPCs. I therefore assume Strategic Developments, having generated $675m of cash flow between 2026 and 2030, is still worth $500m at the end of 2030.

Subscribe now

Operating Assets

This segment owns commercial real estate that Howard Hughes has built in its MPCs. These assets generate rent, but they also make the communities better, which drives residential land price inflation - so while this segment should be valued on its rental profits alone, a significant part of the return on investment from building these assets turns up in MPC segment profits.

The key metric for this segment is net operating income (NOI). This is rental income less cash operating expenses - effectively ebitda for real estate. It’s worth noting that once interest and depreciation are deducted the segment produces slightly negative EBT. One way to think about this is that the value is effectively in the depreciation (which generates cash flow and shields the business from tax) and the underlying land, with an option on the true asset life being longer than that assumed for depreciation purposes.

Roughly half of the NOI is from office, and the other half is a mix of retail, commercial, and multi-family. There is obviously debate about what assets like this are worth, especially office. My view is these are advantaged assets because they are located in thriving, growing residential communities and because Howard Hughes controls nearly all the commercial real estate in their MPCs. They rarely sell, and nobody else can build without their permission, so there is little risk of speculative overbuild. The downside is that HHH cannot sell a controlling stake, at least not until the MPCs are complete, although they have speculated that they could sell a minority stake in Operating Assets which would set a valuation marker and produce cash for reinvestment.

HHH reported $276m of operating asset NOI in 2025, and the company says stabilised NOI for existing assets and those under development will be $361m, with the vast majority of the uplift coming from existing assets that are not yet mature. The company does not say when stabilised NOI will be achieved, but the asset-by-asset disclosures suggest that it should take 3-5 years. For today’s valuation I assume the $262m in 2025 NOI is capitalised at 8%, and for the 2030 valuation I capitalise the $361m in stabilised NOI at 7%, on the basis that more mature assets in more mature communities should be worth more.

Notably, neither valuation assumes any franchise value in Operating Assets - yet it is highly likely HHH will continue to create value by building new assets.

Roughly, segment NOI less depreciation and interest is zero. With no EBT, there should be no income tax to pay. I therefore assume segment free cash flow (FCF) is equal to depreciation - roughly $170m per year - and add 5x$170m in cash flow from Operating Assets to the 2030 cash balance, less $30m for additional equity needed to complete assets under construction. I do not grow FCF with NOI, which should be conservative. In addition, I model 2030 segment net debt at 62% LTV (the company guides to 60-65%). The increase in debt from $2.5bn today generates an additional $689m of cash for HHH.

Holdco debt

HHH break out debt by segment in the 10k. There is $2.04bn of debt at the holdco level. The Vantage deal will cost $2.1bn, and HHH had $1.28bn of cash at the holdco level at YE25. I assume $1.2bn of this is spent, and therefore $0.9bn of Vantage prefs will be issued to PSH, for a total of $2.94bn in holdco debt and prefs by mid-2026.

Some notes:

HHH claim the MPCs are unencumbered; I think this is wrong: both the Operating Assets and Strategic Developments carry as much debt as they can bear. The MPC segment has very little, but the holdco has quite a lot. In effect, the MPC debt is simply at the holdco level.
The debt is well-structured: in terms of interest rates, 96% is fixed or swapped or capped, and after a $1bn bond issue in February 2026, ~60% is due after 2030.
HHH has the option to buy back the Vantage pref at a price linked to the growth in Vantage shareholder’s equity. I assume these prefs are repaid in 3 equal annual instalments (consistent with guidance) and grow the balance by 15% per year because I assume Vantage shareholders’ equity grows at this rate for my 2030 valuation.

Subscribe now

Overheads

HHH guide to $82-92m in overheads in 2026. I assume $90m and capitalise this at 10x. I also assume overheads inflate at 3% per year, which feeds into my cash projection and my capitalised overheads figure for 2030.

Overheads includes the fixed component of the Pershing Square fee, which is $15m per year inflated by the Personal Consumption (PCE) index.

A couple of key items are not included in overheads:

Deal fees, which could be significant if HHH pursues a lot of deals.
Stock based comp. I prefer to think about this in terms of potential dilution rather than the Black Scholes based P&L measure. There are 59m shares outstanding. No options were issued in 2025 and at year end there were 68k options outstanding, so options don’t represent a big dilution risk. 300k RSU’s were issued in 2025 at an exercise price of $77. Assuming RSU issuance continues at this pace and all are exercised, shareholders will suffer 0.5% dilution per year. I can live with that, especially considering Pershing has an incentive to keep stock based comp within sensible bounds (see below).

Performance fees - and Pershing Square’s alignment with minority shareholders

Overheads does not include the variable part of the Pershing fee, which is dependent on the share price. Specifically, Pershing gets 1.5% of the market cap over a threshold. The threshold is calculated by multiplying a fixed share count of 59.4m by a reference share price of $66.15, which inflates with the PCE index. In effect, Pershing can only increase its fee by increasing the share price above inflation. And notably, Pershing:

cannot increase its fee by causing HHH to issue more shares.
can increase its fee by causing HHH to buy back shares below intrinsic value, if doing so increases the share price.

I think HHH directors did an extremely poor job of negotiating the reference price. HHH has argued for years that their real estate business is worth over $100, but now HHH will pay a fee for any share price increases above $66. If the market merely comes to agree with HHH about the value of the real estate business, Pershing will collect a performance fee. That is not right. The reference share price should have been at least the $100 that Pershing themselves paid to inject new capital into HHH in 2025, and ideally $118, which was the NAV that HHH published in 2024 - and even this would have ignored NAV accretion in 2025.

However annoying this is, the fact is that the reference price is $66, and we must assess value on that basis. For me, the key is that there is strong alignment between Pershing and minority investors:

the variable fee clearly incentivises Pershing to increase the share price significantly and sustainably ahead of inflation.
the fixing of the reference share count incentivises Pershing to make (or influence the HHH board to make) shareholder-oriented decisions around share issuance, buybacks, and stock-based comp.
PSH and PSCM between them own 47% of HHH. Ackman and his team own 90% of PSCM and 27% of PSH. Indirectly, they own a lot of HHH. They want the share price to compound.
HHH is reputationally deeply significant for Ackman. He has often spoken of his admiration for Buffett and desire to build one of the great investment track records. To compound an insurance holding company for decades is to be compared to the Master. I doubt much motivates him more (except fees).

Pershing Square have several advantages over other fund managers. In particular, their assets under management is almost entirely permanent. Having managed both permanent and temporary capital, I can attest to the sad fact that capital duration affects manager psychology. It is much easier to focus on the long term when you have permanent capital. In addition, Pershing’s ability to attract talent is significant and their analyst tenure is impressive, which speaks to the culture at the firm. Finally, their performance is phenomenal - the chart below shows gross performance, which is relevant to Vantage, since it does not pay fees:

In short, while I think the HHH directors have done minorities a disservice with the reference price, I think Pershing Square is an excellent and aligned partner.

For the variable fee, my SOTP assumes:

$0 today, since the share price is below the reference price.
A $120 share price on average over the next 5 years, driving a total cash outflow of $225m.
That in 2030, the share price equals the NAV, maximising the variable fee, and that the fee is capitalised at 10x.

Subscribe now

Tax

Within my NAV, insurance, MPCs, and strategic developments are valued net of tax, and the operating assets make no earnings before tax so they pay no tax. I therefore do not capitalise future tax into the NAVs.

I do, however, deduct tax from the cash flows as follows: $400m in MPC EBT, less approximately $225m in holdco costs and interest, gives $175m in taxable profits each year, at a 22% rate.

Holdco cash

YE25 holdco cash was $1.28bn and I assume $1.2bn is used for the Vantage acquisition, so my starting holdco cash balance is $80m.

I exclude restricted cash. This is mostly condo deposits so including it risks double counting cash flow from Strategic Developments.

I then add the cash flows from each of the segments over the next 5 years to get to the 2030 cash balance. The striking possibility is that even after buying Vantage, HHH might have $1.7bn to deploy into new acquisitions over the next 5 years, before considering leverage.

Subscribe now

Conclusion

First, a confession. HHH is a complex beast, with 4 quite different businesses and an unusual partnership and fee stream to consider. In addition, while the company produces a lot of disclosure, it isn’t presented as clearly or as consistently as I would like. To research HHH is to make multiple choices about what matters, and what does not; what merits more time, and what does not. I may well have missed things, and I apologise if I did.

As I write HHH trades at $63 per share. Recall that HHH thought their NAV was $118 per share in 2024 and Pershing paid $100 per share in 2025. According to my maths:

The stock has 36% upside to my current SOTP.
The stock has ~200% upside of 204% to my 2030 SOTP, a CAGR of 25%.

Here’s a reminder of my SOTP:

And here is a summary of the assumptions:

I don’t expect the stock to rise 36% to its current NAV any time soon. It has traded at a discount for years. But I do think the 2030 NAV is realistic. Not conservative, perhaps, but realistic. I assume good performance from each of the assets, but not spectacular. Vantage shareholders’ equity could easily compound faster than 15%. MPC NAV might not beat my assumption, but with 50% of today’s GAV being sold in the next 10 years the duration of the cash flows will shorten considerably, reducing risk and helping the stock trade closer to intrinsic value. I assume operating assets reach their guided stabilised NOI, but not that any more assets are built. I assume future strategic developments achieve similar economics to current ones, not better, and that the Ritz Carlton Residences success can’t be replicated elsewhere in the MPCs. And I assume $1.7bn of accumulated cash sits on the balance sheet, undeployed.

My standard hurdle for investments is 15%, which compounded is a double in 5 years. Here we can buy a solid set of assets, trading at a discount to a relatively conservative current NAV, with realistic line of sight on making 3x our money over 5 years and compounding after that, because MPC GAV will continue to grow and because HHH now has a virtually infinite reinvestment runway.

Could things go wrong? Could the Vantage balance sheet be littered with huge unforeseen claims? Could the MPCs suddenly lose momentum, perhaps because of large unforeseen demographic shifts? Yes. Of course. But investing is about embracing uncertainty, deciding what risks you want to take, and ensuring you are being paid to take them. On that basis, I am happy owning HHH at $63 per share.

Links to previous work

Thanks for reading - if you enjoyed reading this please like and restack, and do get in touch if you have questions.

Pete

Update: Brookfield investor day summaries

Building Arks — Thu, 19 Mar 2026 13:26:46 GMT

With my larger holdings, I manually summarise the transcripts from earnings calls and investor days. Obviously it would be quicker to get AI to do it, but I just don’t think you get the nuance that way, or the same clarity on what is changing over time and what isn’t.

Here are my summaries of the last Brookfield investor day transcripts. I hope they’re useful.

BN - real estate and Wealth Management are huge opportunities

19% CAGR over 30 years, or 27x.
Plan value today is $180bn or $68 per share. This is 16x average annual earnings over the next 5y and 10x average annual DE.
- “I’m quite confident we can sell all the businesses for probably premiums to those numbers.”
- $82bn is public holdings.
- $12bn is funds managed by BAM, compounding at 15%. Will realise net $5bn over 5 years.
- $34bn is PV of carry, valued as net unrealised carry + annual targeted carry x10. (They also do a DCF which produces the same result at an 8.5% discount rate.) They’ll convert $6bn to cash over the next 3 years and $25bn over the next 10, net.
- $26bn is real estate.
- $26bn is BWS at 15x annualised DE.
- Expect 25% DE and 16% plan value CAGR for next 5 years.
- 5y DEPS target $6.95.
Will generate $53bn of cash flow over the next 5 years, I think including asset sales. $28bn will go to dividends and to grow BWS, $25bn is free to be reinvested.

Generating 15% returns is easier than it was due to scale and capabilities.
Nimble within two guiderails: value investing, and driving returns through operating capabilities.
“Once we understand an industry, we move fast and we deploy at scale. We’re not early, but once we understand it, we move fast and we deploy at scale.”

Subscribe now

Real estate
- Sold $5bn of balance sheet assets over the last year and refinanced $16bn.
- Plan value $26bn and DE $730bn. Will sell $24bn over the next 5 years, reinvest $10bn, and generate $3bn in annual dividends, leaving 2030 plan value $15bn and DE $640m.
- In the last year, balance sheet investments have signed 15msf of leases at 11% rental spreads. This will drive NOI over the next 2 years as leases roll. 5y NOI LFL CAGR should be 4%.
- Now splitting assets into 3:
  - Super-core is 60% of invested equity. 34 irreplaceable and complex assets that they can keep reinvesting in and will own forever (but may sell stakes). 90% of this is office and retail with 95% occupancy and 47% LTV. 10% is resi and mixed use.
  - Core-plus is 25% of invested equity. 57 assets of equal quality to supercore, but with business plans that are complete or being completed and which will be sold. The performance metrics are almost identical to supercore: 27 office assets have 44% LTV and 94% occupancy.
  - Value add is 15% of invested equity. 95 assets generally in secondary markets where the plan is to reposition and sell.
- Office: 2 years ago only 5% of companies required employees to be in the office all the time. Now it’s over 50%. 2 years ago the average work requirement was 2.5 days per week. Now it is almost 4 days. But nothing has been built for years, and existing supply is being converted to other uses. Very little space is available for new leases, so the premium for new office space averages 60% globally. In NYC it is over 100%.
- Retail is driven by consumer spending. US retail spending is 140% of 2019 levels but there is virtually no retail construction - since 2015, US retail construction has focused on convenience and not shopping destinations. “This directly impacts retailers who are starved for more space.”
  - Alderwood Mall in Washington is 99% occupied with NOI up 50% since 2021. They’ve just refinanced at 5.9%, having recently redeveloped an old Sears box into 76ksf of retail space plus 328 apartments. The new retail space is 100% leased and generates $4m of NOI, while the apartments are 96% leased and have just been sold at a 4.7% cap rate.
  - Fashion Place in Salt Lake City is 100% leased and generates $1000psf in sales. Replacing underperforming brands has driven sales up 20% since 2021 and they have just refinanced at 5.4%.
- Flatt: “I’ve seen [this cycle] 5 times before. The psyche got hurt more this time, but the fundamentals are actually way, way better as we come out of the bottom of the market”.

Brookfield Wealth Solutions
- $135bn of assets, $14bn of equity, $1.7bn of DE ($2.1bn next year).
  - 3 big recent transactions have been at an average of 1.8x book, which implies $26bn of value or $20 per share of BN.
  - Also generates $300m of fees to BAM on $30bn of assets invested in funds.
  - Including the UK acquisition they will have $180bn of assets, $30-35bn of inflows, $10bn of outflows, and $20-25bn of net inflows. Add some M&A and they can reach $350bn of assets and $5.5bn of DE in 5 years, reaching a 17% ROE and valued at 12-15x.
- “Investment-led insurance” - by investing in high-yielding private assets, you can have low leverage and lots of cash. This combination is great for risk, returns, and regulators. Most insurers operate at 10-20x leverage. Brookfield operate at 7-8x and they have $50bn of assets in cash or cash equivalents. All their businesses have A ratings and 2 have had upgrades under Brookfield ownership due to retained capital.
- “We think we can build a business that can compound capital for many decades to come” driven by aging and Brookfield’s ability to invest at scale.
  - Today 1 in 6 Americans are over 65. By 2050, it’s 1 in 4. Same all over the developed world. This plus the lack of DB pensions drives a $7tn retirement deficit.
  - Aging drives demand for wealth products.
    - Pensions and Sovereigns is a $22tn market.
    - Today US retirement accounts have $40tn (401ks, IRAs, and employer DC accounts). By 2040 this will be $100tn.
    - 20 years ago 7% was in annuities; now 10%, and likely going to 15%.
    - Trump’s recent Executive Order opens retirement accounts to private assets.
  - Brookfield can offer high return private investments in the accumulation phase, and then competitive annuities for the decumulation phase.
- Initially thought they would do this as a reinsurer and then realised the value of distribution. American National and American Equity previously did $5bn of annual annuity origination combined. Now they do $25bn with more to come - but they scale back when rates aren’t right.
- Regulation is “an incredible moat”. They work hard with regulators “to make sure our disclosure is at the top end of their expectations and to default to overcommunication…the more we work with regulators, the more that moat widens and gives us protection and creates scarcity value”.
- Massive value created by buying insurers at or below book, selling noncore lines for 1.6x, and then selling all their long dated bonds just before inflation took off.

Subscribe now

BAM

FRE: $2.7bn today → 17% CAGR to $3.59/share by 2030.
- DE: 18% CAGR. Base case 2x in 5 on $1.2tn FRE-bearing capital, “inherent operating leverage”, rapidly scaling complementary products.
- Additional levers get to >20% DE growth.
- Exceeded last 5-year plan: FRE-bearing capital $563bn vs $510bn target; FRE $2.7bn vs $2.6bn.
- Carry drives growth after 2030.
- 87% of fee-bearing capital is long-term or permanent in nature, growing to 92% by 2030 — structural earnings visibility most peers can’t match.

Past 12 months: raised $97bn, deployed $135bn, monetised $75bn, returned $50bn - record monetisations despite low M&A volumes.
Competitive position.
- Owner-operator model, co-investment alignment, and boots-on-ground operational expertise are the core differentiators.
- Scale creates a self-reinforcing loop — platform attracts capital, capital enables better deals, deals build the track record.
- Brookfield capital goes in first, clients come alongside.
- Cross-platform information advantage: Westinghouse underwritten because of renewables expertise; Canadian housing underwritten because of real estate presence; Chemelex surfaced because of infrastructure knowledge.
- Compounding expertise from repetition - hundreds of renewables deals, deep infra experience.
Growth levers
- Products: >70% of fee revenue from strategies launched in the last decade. Lots more to come following the pattern flagship → complementary → retail.
- Institutional share of wallet; although BAM is in 39 of the world’s top 50 investors, they only have 2% of their alts portfolio.
- Partnerships: $40bn of go-forward deployment from bilateral partnerships announced in 2025 alone. Barclays partnership seeding financial infrastructure fund; sovereign AI partnerships seeding AI infrastructure fund. Partnerships are multi-dimensional — tenants, offtakers, pension managers, co-investors.
- Individuals: private wealth raising $10bn this year, up 50% y/y. 60,000 private wealth clients vs zero 5 years ago. 401(k)/retirement markets are “early, early, early innings” this opening is not in their base case.
- Europe institutional: expect to raise 3x 2024 levels in 2025 — “infancy.”
- Family capital also barely started.
Outlook by business segment
- Infrastructure: massive multi-year investment need. Spinning out dedicated AI infrastructure fund, which is differentiated because half is yielding, not greenfield, with sovereign and hyperscaler counterparties on 20yr+ contracts. US data centre vacancy <2.5%; 10GW of lease capacity added in US this year alone.
- Renewables & Transition franchise now generates >$400m annual revenue, built in <5 years from zero. Largest transition strategy globally.
- Real Estate: — fundamentals strong, supply near-zero, capital structures normalising, transaction activity recovering and rate cuts will accelerate this. Sold $3.7bn, deployed $6bn in last 12 months. “Rents going through the roof” in next 5 years. Buying multifamily at 5.7% caps (vs historically selling at 3.7%), with pathway to >7%. 20‑yr opportunistic track record: 21% gross / 17% net on $44bn equity.
- Private Equity: 26% average gross IRR over 25 years across 6 vintages — “best performing PE track record in the market.” Wealth, structured equity, and financial infrastructure products on the way.
- Credit & Castlelake: $250bn FBC today vs $100bn 5 years ago. Private credit is still very early and asset-backed finance is “taking off.” Capital markets business growing — 2-3 years to materiality but credit growth could accelerate that significantly.
AI and productivity
- “Robotics and AI can soon change a lot of processes… margins will expand dramatically over time.”
- “We are possibly in one of the greatest investment booms in history… leading to some of the greatest productivity advances we will ever see.”
M&A: 7 meaningful additions in 5 years (Oaktree, LCM, 17Capital, Primary Wave, Pinegrove, Castlelake, Angel Oak). Options allow buying ~$250m of incremental FRE over time.

BIP

L5y FFO/unit CAGR 10%
- L5y understated true performance: FX headwind cost ~2% annually; rate headwind cost 2-3%. Adjusted for both, underlying growth was ~14%.
- Targeting return to 14%+ long-term rate over next 5 years.
- Distributions: target higher end of 5-9% target range, without increasing payout ratio (currently 67%, down from 78%).
2025 to date: deployed $2.1bn ($700m organic, $1.4bn new investments); recycled $2.8bn at 20% IRR / 4x MOIC - annual record and “exceptional” given these were regulated/contracted assets. Already at $3bn recycling target. Private market transactions achieved 15x EV/EBITDA — 3-4 turns above where BIP trades publicly.
Competitive position
- Full-cycle strategy: deploy at 12-15% hurdle, crystallise value through recycling, maintain liquidity to go on offence in dislocations. COVID, rate shock, and the trade war produced some of BIP’s best acquisitions.
- Cross-sector learning:
  - Intel semiconductor deal assembled from data centre contracting + LNG project finance + corporate partnership governance knowhow.
  - Hotwire identified as two businesses (development platform + stabilised portfolio) at signing - value not recognised by sellers.
- 100% of new investments funded internally over last 3-4 years - recycling makes BIP self-funding.
- Private market buyers paying 10-11% cost of equity for BIP’s mature assets; BIP redeploying into 15-17% returning opportunities - spread is core compounding engine.
Growth levers
- Organic backlog 4x larger than 5 years ago - highest-returning capital deployment.
- Recycling program 5x larger, now $2-3bn annual run rate going forward.
- AI infrastructure: $500m annual BIP deployment targeted.
- Macro turning: rates stable/falling and USD softening for first time in years.
AI buildout
- 1-2% of US GDP, larger than the fiber build-out, approaching railroad-era scale, analogous to electric grid build-out in duration and necessity.
- Power density per rack is 10x non-AI and forecast to rise 5-10x within 10y.
- Data centre inventory sold out.
- AI factories are DCs plus $30m/MW of chips - 4x the capital intensity of standard cloud.
Recent acquisitions
- Colonial Pipeline: largest US refined products system, 50% of East Coast demand, 5,500 miles Houston-NY, 9x EBITDA with FERC-regulated inflation-linked tariffs and 90%+ utilisation. Bought during trade war noise with limited competition. Upside from operational improvement.
- Hotwire: bulk fibre to HOAs, immediately recycled stabilised assets to lower-cost capital. Growth via homebuilder partnerships and Brookfield ecosystem.

Subscribe now

BBU

Adjusted EBITDA $2.7bn, 5y CAGR 16%.
- EBITDA margins doubled from 12% to 24%.
- Adjusted EFO/unit: $3.65 → $6.90.
- NAV: $28 in 2020 → $54 today.
- Capital recycling: $2bn target for 24 months achieved in 12; confident of another $2bn over next 24. The sale of 3 partial interests at an 8.6% discount to Brookfield’s evergreen fund, if reinvested in buybacks at market prices, adds $2.50 in NAV/unit.
Long term record
- 26% gross / 20% net IRR over 25 years, driven by margin expansion, not multiple expansion or financial engineering.
- Since IPO: 25 investments monetised, $8bn of proceeds at BBU share, in line with carrying NAV and above target returns.
Brookfield ecosystem ($1.1tn AUM, 400 companies, 30 countries) is the information advantage
- Cross-portfolio best practices identified and deployed at scale.
- AI use cases now live across 200+ applications; best practice in one company rolled to all 400.
- AI value creation office: 30 dedicated people sharing lessons.
- Buying industrial companies that benefit from AI as a productivity tool.
  - Clarios: custom ML algorithm on 30yr-old facility saved 60% more than expected, now rolling across all Clarios plants and comparable BBU businesses.
  - Sagen: ML models on 25 years of proprietary data — mortgage underwriting automation up 1.5x, better house price prediction, improved fraud detection.
Financial infrastructure is a new priority vertical.
- $4tn market opportunity.
- Strategy: buy asset-light software and services businesses (not banks), unshackle them from conglomerate parent constraints, digitise/automate, and deploy operational expertise.
- Target characteristics: scale, distribution, technology-enabled, regulatory moat.
- Network + Magnati combined into #1 payments provider in Middle East/Africa, processing >60% of regional payments.
- Barclays payments carve-out: building “the largest startup you can imagine” — 3,000 people, every process being rebuilt from scratch. Target: $200m EBITDA uplift.
- Now have leading payment infrastructure in 60+ countries.
- Ranjan: “we can quadruple the EBITDA of these businesses soon after buying them.”

BEP

FFO/unit: 10% organic annual growth. Building blocks:
- inflation escalation (~2%, $150m incremental FFO over 5 years)
- re-contracting at higher prices ($100m)
- asset optimisation (1-2%)
- development pipeline execution ($385m, ~5%).
Competitive position
- Only platform combining large-scale low-cost wind/solar with irreplaceable baseload — hydro, nuclear (Westinghouse), and now batteries (Neoen).
- Hyperscalers need 99.999% availability; BEP can deliver both cheapest power and reliability, most peers can only do one.
- Hydro: largest private portfolio globally. More strategic than ever - dispatchable, 24/7, scarce. Selectively recycling non-core assets (First Hydro UK, Maine portfolio) while reinvesting in core (increased Isagen stake). Google framework locks in long term contracted cash flows and enables up-financing; it is also is a “hunting licence” for US hydro M&A.
- Nuclear: Westinghouse serves ~2/3 of world’s nuclear fleet, supplies ~half of all reactors globally = global scale + deep US expertise = ideal positioning for new builds, life extensions, and accelerating global investment. Still “very early stages of a long sustained investment cycle.”
- Discipline as an edge: deliberately avoided offshore wind (subsidy-dependent, construction risk), residential solar (unproven economics), and highly leveraged structures.
- “Clean portfolio, track record and balance sheet” entering the upswing. Highest credit rating in sector and financing terms improving - 10-year notes in March at lowest coupon in 5y, tightest spread in 20y.
Growth drivers
- Data centres alone drive 8-10% annual US power demand growth through 2050.
- Deployment: $9-10bn over next 5 years across development and M&A - step up from prior targets.
- Half is now proprietary organic development.
- Target adding 10GW annually by 2027, sustained from there.
- Capital recycling now a core, recurring part of the model. Each portfolio company now recycles assets as they mature. Target >$2bn of asset sales annually.
- M&A: pipeline >$100bn enterprise value. BEP wins on certainty of capital and execution speed, not price. Renewables regulatory uncertainty in US market is creating attractive entry points. $25bn of partner capital available alongside BEP’s own $4.7bn - total equity firepower ~$30bn.
- Batteries: costs down 90% since 2010, further declines expected. Neoen acquisition makes BEP global leader - 1.5GW operating, ~50GW development pipeline. Fastest-growing segment within BEP. Firms up intermittent renewables to meet reliability demands, particularly for datacenters.
- Wind and solar development are nearly the entire pipeline in key data centre markets. Contracting activity with hyperscalers up ~100% in last 2 years. Microsoft and Google partnerships - only two such framework agreements in the industry.
- Opportunity to consolidate distributed generation: smaller renewable assets, typically solar, sited close to or at a commercial or industrial customer’s premises, with the power contracted directly to that offtaker. These ease grid constraints and deliver cheap power.
Capital recycling
- Platforms: Saeta Yield, India platform, Luminace majority - 26% IRR, 2.6x MOIC.
- Individual assets: First Hydro, Shepherd’s Flat, Maine hydros - 18% IRR, 3x MOIC.

Subscribe now

Links to previous work

Thanks for reading - if you enjoyed reading this please like and restack, and do get in touch if you have questions.

Pete

Review: Uber in 20 years

Tue, 17 Mar 2026 16:44:08 GMT

Summary

What it does: Orchestrates mobility.

Elevator pitch: Uber is a dominant 3-sided network that allows customers to order transport and hamburgers.

Mental model: Scaler, moat, asymmetric payoff.

Valuation and potential returns: 30x non-GAAP net income, but potential to compound in the teens for 20 years.

Exchange and ticker: NYSE, UBER

Stock price and market cap: $78, $154bn.

Do I own it? Yes.

IR website: here.

Tl;dr

This is a thought piece about Uber’s TAM and competitive position in 10 and 20 years. The underlying product should become far cheaper and therefore the market far larger. Autonomous vehicle (AV) supply is likely to fragment, meaning many providers will need Uber to scale and compete. AI personal assistants might come to dominate demand, creating a risk of disintermediation, but there are strong practical and regulatory reasons why assistants will not bypass Uber. Uber should probably engineer its take rate down to help grow the market and reduce the risk of disintermediation, but even so there should be plenty of space for revenue growth and operating leverage. The resulting skew between more bearish and more bullish investor outcomes is attractive.

Introduction

Investing requires a balance between attention to detail now and a broader, more general understanding of how the long term future might unfold. This piece attempts to imagine how Uber’s competitive moat might evolve over 10-20 years in the face of two interrelated threats: the growth of AV platforms, and disintermediation by AI.

For a more traditional introduction to Uber, packed with details on current market shares and financials, I recommend this by :

Hated Moats

Uber Technologies: Deep Dive Analysis

Overview & Positioning…

3 months ago · 43 likes · 5 comments

For long term investors in Uber, I think there are 5 things that really matter:

Potentially explosive growth in the total addressable market (TAM).
Whether the autonomous vehicle systems market fragments or is winner-take-all.
Whether AI can disintermediate Uber.
What happens with take rates.
The upside/downside skew.

Let’s ride.

Subscribe now

TAM explosion

AVs will massively expand Uber’s TAM, because they will massively reduce the cost of transport.

Why?

AV drivers will be much cheaper than human drivers. This reduces the cost per mile.
AV technology enables a wholesale shift from personal use toward rideshare. This will dramatically increase utilisation of the global car fleet: personal cars get used about 5% of the time, whereas cars on Uber achieve 3-4x that, and without driver fatigue AVs will achieve more. Since depreciation is a significant part of total cost and is fixed, higher utilisation of the global fleet further reduces cost per mile.
AVs can be owned in fleets. Part of the cost of individually-owned cars is personalising them. Colour, trim, and specification choices all add complexity and cost. Fleet-owned cars can be mass-produced to exactly the same design, further reducing cost per mile.
AVs will be safer than cars driven by humans and as discussed, the cars will cost less. This brings down the cost of insurance. In addition, AVs will eventually bring down the cost of regulatory compliance, broadly defined. No more sexual assaults or risk of drivers being classified as employees. All these things further reduce cost per mile.
Electric vehicles are more expensive to buy than infernal combustion vehicles, but less expensive to run and maintain because they have 1/10th the number of moving parts. As utilisation rises the fleet will naturally shift towards EVs, further reducing cost per mile.
Large numbers of autonomous electric vehicles (AEV) solve one of the world’s great problems today: renewable electricity is cheap on a levelised cost basis but intermittent, and when you include the cost of storage or backup it’s expensive. EVs are batteries on wheels. An AEV can constantly compute the highest-value use between charging, discharging, and mobility. This will require significant investments in local grids, but the societal benefit is compelling. AEVs will get paid for balancing grids, further reducing cost per mile.
AEVs will have AI inference chips on board. These chips will be idle when the car is not driving. This idle time will represent a large distributed compute capacity. It won’t be the most efficient compute, but the marginal cost will be tiny, so it’ll get used. AEVs will get paid for distributed compute, further reducing cost per mile.

Uber’s user base is relatively small today. Its annual active user count is roughly half Spotify’s monthly active user count, despite the fact that Spotify arguably has more competition (Apple, YouTube, and Amazon are all huge in music streaming). Why? Because Spotify is free. Uber’s user base is fundamentally limited by the high marginal cost not of its service, but of the product it distributes. Most people can’t afford to use taxis and food delivery very often, and many not at all. The scope for TAM growth is absolutely massive: taxis and ridehailing likely account for less than <1% of distance travelled by humans on land, whereas personal cars likely account for >70%. As cost per mile comes down, I would expect total distance travelled to rise and ridehailing to gain share. For many people, I expect ridehailing to replace car ownership entirely, eventually. The potential for TAM expansion is enormous: if I am right the user base will explode and frequency will rise.

The above focuses on mobility, but the same applies in delivery. The human driver represents about half the cost of last-mile delivery. Robotics will transform this. Uber has partnerships with Nuro, Lucid, and Serve to introduce AVs and robots for delivery. The impact on total cost will not be as significant as it might be for mobility, because the underlying cost of the hamburger being delivered won’t change, but reducing the cost of delivery itself will expand the market.

These changes have a number of broad societal benefits, too. Driving becomes safer. Cities become more liveable: without personally-owned cars, inner-city parking can be eliminated and roads can be re-designed with larger walkways and cycleways. Carbon intensity is reduced. I think it is likely that, in time, AV will receive both regulatory and popular support.

Pushbacks:

“I want to keep my car. I like driving”. Do you, though? Really? I get that taking a Ferrari round Silverstone or going off-roading in a Land Rover is fun. But driving to work? Driving with your kids screaming in the back? These things are not fun. Given the choice most people would far rather sit in the back, do some work, make a call, watch a film, play with the kids, or just stare out of the window.
Specific use cases. You might use fleet AEVs on Uber for your commute, but not for heading out of town at the weekends, and definitely not for your annual half term ski holiday drive from London to the Alps. But I think that as cost per mile falls, as fleet AEVs become ubiquitous, and as Uber innovates around specific use cases and availability in less-dense areas, you’ll book AEVs to do exactly those things. That half term ski trip? You’ll book 2 AEVs 6 months in advance for the whole week, and it’ll still be cheaper than owning a car.

I am not arguing any of this will happen soon. This is a multi-decade, high-level prediction. But I think the logic is compelling, and that the change will accelerate as the cost benefits become clearer.

Subscribe now

AVs - fragmentation vs dominance

Uber will benefit from TAM expansion if it is the demand aggregator for multiple AV types and fleets. It will not benefit if two or three AV manufacturers dominate the market and build their own demand platforms.

In this, I feel pretty confident. I think fragmentation is likely for a variety of reasons:

The market is huge. Huge markets attract competition.
None of the up-and-coming AV manufacturers has a pre-existing advantage: a product that is miles ahead, ownership of all the data or all the customers, or vast amounts of sunk capital that others have to match to catch up.
There’s no real barrier to entry in car manufacturing, a famously commoditised industry.
There’s no real barrier to entry in support infrastructure (charging, etc.).
There’s no real barrier to entry in developing AV systems. This is more controversial, but it is already clear that multiple players are developing AV datasets and/or the ability to train AV models in virtual worlds. Also, the challenge does not lend itself to having one winner: in the long run it’s not about being the best, but about being good enough. I think many AV systems will eventually meet that hurdle.
Several powerful entities are highly motivated to ensure no one AV system dominates, including Uber, NVIDIA, regulators (for competitive reasons), and governments (for security reasons). NVIDIA’s full stack AV offering and Uber’s Autonomous Solutions are great examples of how these two companies, at least, can be agents of fragmentation - not just beneficiaries of it.

Assuming the market does fragment, Uber is a huge beneficiary. Utilisation will be the holy grail of AV economics and it will not be possible for every player to develop a demand platform - they will need an aggregator. Uber is positioning itself as a plug-and-play monetisation platform for AV. It offers instant utilisation, dispatch, operating and finance partnerships, customer support, and localised compliance. As a result it has a rapidly growing array of partnerships with AV developers and legacy car manufacturers, all aimed at launching AV fleets in multiple cities worldwide over the next few years. These solutions may not be the best or cheapest initially, but Uber lets them monetise quickly and reinvest to improve. For details on Uber’s partnerships (and links to some interesting models), I recommend this by :

Platform Aeronaut

Uber Is Quietly Winning the AV Rideshare Setup

If 2025 was the proof point that consumers will actually take autonomous rides at scale, 2026 is starting to look like the year the strategic map gets redrawn. For the last few years the AV debate has mostly been framed around who has the best self-driving technology. That still matters of course. But increasingly that is the wrong question for investor…

23 days ago · 27 likes · 1 comment · Thomas Reiner

Uber takes a lower percentage of gross bookings for AVs (the target appears to be around 20%) than it does for human drivers (around 30%). However, it also has lower driver incentive and insurance costs. In the article below, estimates that Uber’s “real net revenue” might be as much as 65% higher with an AV driver than it is with a human one. I would add that if sales, marketing, R&D, G&A, and D&A costs remain the same, all of that drops to the EBIT line. Applying this assumption to 2025 financials implies that shifting to AV could nearly double EBIT (assuming only mobility benefits - if the other segments do, too, EBIT could triple).

Fundamentally Sound

On Uber: AVs, Take Rates, and Fragmentation

One of the primary arguments against Uber as an investment can be summarized in two words: AV disruption…

3 months ago · 152 likes · 4 comments · Manu Invests

Uber’s economics also work for AV fleet operators. Uber takes 20% of their revenue. In return, it delivers higher revenue and lower costs than they can generate on their own:

With Uber, fleets don’t need to acquire customers. My guess is that most startups would need to invest >>100% of revenues for years to match the utilisation that Uber can provide on day 1 - something investors simply will not support in a market with so many players and no evidence of winner-take-all characteristics.
Uber estimates it generates 30% higher utilisation than standalone AV operators are achieving, and believes that the gap will grow from here because the standalone operators are currently running small fleets in very high-demand areas, with free publicity and novelty value on their side. These ideal utilisation conditions fade as fleets grow and AV expands into less dense areas.

Subscribe now

Nevertheless, some players will go it alone and build their own demand platforms. Successful standalone players will have two key things in common: deep pockets, and cheap customer acquisition costs as a result of existing brand awareness and legacy apps. In the west, Tesla’s Robotaxi and Google’s Waymo are the obvious candidates. For these players, going it alone may be a rational choice: saving 20% of revenue in the long run justifies the investment to acquire customers today. But even in these cases, selling spare capacity on Uber to maximise utilisation will probably make sense. And even advantaged standalone players do face two challenges:

They will grow city by city, and it will take years to match Uber’s coverage. Uber is using that time to become the demand aggregator for dozens of other AV fleets.
Users simply don’t want too many ridehailing apps. My home in London is in a ridehailing blackspot. I often have to search on 3 apps to get a ride, and I absolutely hate it. It is stressful and often expensive, because if I get distracted I can easily end up with two rides and a cancellation fee. I recognise that competition is necessary, but I would far rather consolidate my usage onto one app.

My guess is the vast majority of demand will settle onto a maximum of 3 apps per area. The network effects in this business are primarily local, so not everywhere will have the same winners, but Uber is very well positioned. Certainly if Waymo and Tesla came to London at scale, I’d stop using FreeNow and Gett instantly - but not Uber.

For these reasons, as long as consumers use apps to hail rides, I am pretty confident Uber will be a dominant player. And therein lies the problem. What if consumers don’t use apps?

Can AI disintermediate Uber?

recently wrote an excellent article on how hard it was for Uber to build its network, and therefore how hard it would be to disrupt.

The Dutch Investors

Why Uber is close to being undisruptable

Last month, we analyzed Uber for our premium members. That meant speaking to former employees, reading Super Pumped, and reading its annual reports. During this process, we found some interesting facts that we wanted to share with you…

a month ago · 32 likes · The Dutch Investors

As an Uber shareholder, reading this felt like slipping into a warm bath of confirmation bias after a long day. But unfortunately, I can imagine how it might be wrong.

Uber’s network connects millions of drivers with hundreds of millions of consumers. The marketplace is incredibly fragmented. That’s why the network was difficult to build. But even in a fragmented AV world, supply will consolidate down to a few AV fleets. And if AI personal assistants (AIPA’s) get better - which they certainly will - the demand side could consolidate too.

In other words, if the future of ridehailing is AV and “Hey, Siri”, there might only be a few dozen major fleet providers on one side of the network and 3-4 AIPA’s on the other. That makes the network much easier to replicate. If the AIPA’s develop direct links to the fleets, they could bypass Uber entirely. This worries me much more than the AV transition because fewer people seem to be focused on it.

Notably, Uber cannot counter this threat with “hey Uber”. The operating system providers do not allow the creation of always-on wake words at the OS level, partly because they gain a huge advantage by keeping this functionality to themselves, and partly because there are data security implications of always-on listening so they can’t let everyone do it. There is a workaround in Android, but it requires running a foreground service with persistent notifications, isn’t optimal for battery performance, and is rarely used. So in practice “hey Uber” is only usable when the app is already open, which won’t beat “hey Siri” in terms of user experience.

Nonetheless, I think Uber has a number of potential defences.

The first is that the AIPA’s might route demand through Uber rather than bypass it, and indeed they already do - you can access Uber’s services through Siri, Alexa, and Google Assistant today. There are two very good reasons for AIPA’s to do this:

To reach their full potential AIPA’s will need to offer great services across a wide range of activities, not just mobility. The best way to do this is probably not to take everything in house, but to have specialist providers who excel in each area.
If AIPA’s start disintermediating the orchestration layers in major industries, they will become dominant consumer gateways with extraordinary power across large areas of the consumer economy. I highly doubt that competition regulators will allow this.

Subscribe now

The second defence is that to disintermediate Uber, AIPA’s would have to orchestrate mobility themselves, dynamically predicting demand, optimising routes, booking rides, negotiating prices, and supporting customers. They would have to do this well enough to offer better economics to fleets than Uber, despite the fact that the fleets would have to take the lead on building support infrastructure partnerships and navigating complex and varied local regulations, things Uber currently helps with. Uber’s vast datasets, deep experience, and emerging partnerships in these areas represent a deep moat. For more on what Uber does under the hood, I recommend this by :

Future Cognitive Capital

[$UBER] Why Almost Everyone Still Misunderstands Uber

Why Almost Everyone Still Misunderstands Uber…

3 months ago · 43 likes · 7 comments · Hidden Market Gems

A third defence is that Uber is actually a 3-sided marketplace. While supply and demand for mobility might get consolidated, it’s harder to imagine the same happening to delivery. AIPAs might be able to replicate a significant part of delivery by connecting to major chains, but a large part of delivery will always be fragmented amongst thousands of mom & pop restaurants and shops. Replicating these relationships would be costly.

The fourth defence is time. Consolidation down to a few players on either side of the network is going to take years. Major braking factors are the ramp-up of AV vehicle production, the optimisation of AV designs to bring down costs, the slow creep of regulatory acceptance, and the expansion of AV outside of major city centres. In the meantime, mobility is clearly going to be a hybrid of AV and human, which is much harder for the AIPA’s to replicate.

This combination of defences gives Uber a deep moat, full of crocodiles. Is it unbreachable? Theoretically, no - very few moats are unbreachable in the fact of a deep-pocketed and determined competitor. But it is a huge deterrent. It seems very likely that the easiest thing for AIPA’s will be to route demand through Uber, and that Uber has time to adapt, deepening its relationships with AIPA’s, ensuring it is the default API for mobility, and - perhaps - reconsidering its take rate strategy.

Take rates

As cost per mile falls and TAM expands, there are three possible paths for take rates.

Uber’s take rate could stay the same in absolute dollars. Markets might celebrate this in the short term but it would be a disaster in the long: eventually Uber would represent the majority of the cost of a ride, and network participants will be motivated to bypass Uber.
Uber’s take rate could stay the same in percentage terms. This would be very good for Uber’s profits, since we expect the TAM to expand explosively and many of Uber’s operating costs are somewhat fixed.
Uber’s take rate could come down in percentage terms. This would be less good for profit growth but phenomenal for Uber’s competitive sustainability.

This might not be a popular view, but I will be very bullish if Uber finds intelligent ways to manage its take rate down as gross bookings inflect upwards. Of course, I will not be bullish if this happens chaotically under intense competitive pressure. But win-win solutions, in which (for example) Uber offers lower take rates to fleets that dramatically expand low-cost supply in strategic areas, would be a very good outcome. If it does this, Uber will combine three of my favourite competitive moats:

Network effects, which Uber already has.
Scale economies shared, in which economies of scale are shared with customers in the form of lower prices, which drives more demand and more economies of scale which are also shared, creating a powerful flywheel.
The cheap critical component, in which some critical component is so cheap that nobody even thinks to replace it or compete with it, but it is still priced to give excellent returns to its provider.

Subscribe now

Valuation and skew

I think investing is about embracing uncertainty. Many investment writeups predict what will happen. My view - especially at a time of rapid and accelerating technological change - is that we simply don’t know, and we need to be comfortable with that. What we can do, however, is frame different long term futures and assess risk-reward skew.

This table shows what would happen to Uber’s 20 year revenue CAGR under very different gross billing and take rate scenarios:

The next table takes the revenue CAGR range in yellow in the table above, and shows the resulting 20-year non-GAAP net income CAGR under different margin assumptions:

Some thoughts on the inputs:

As discussed above, I would like to see take rates trend down over time.
Gross billings grew 19% in 2025 without any real help from falling cost per mile. Over the next 20 years I would expect cost per mile to fall hard in real terms; as a result total distance travelled will grow ridehailing will take a much larger share. I can only guess at the impact, but I think gross billings could easily grow +/-10x under those conditions. I don’t have a strong view on whether mobility outgrows delivery, so I apply the growth to consolidated gross billings.
Net margins. In 2025, Uber reported a 10% non-GAAP net margin. There are 4 reasons to expect this to increase: 1) as discussed above, AV economics may drive up to 2-3x greater EBIT for the same level of gross billings; 2) Uber has significant fixed costs, creating strong operating leverage as revenue grows; 3) AI should enable productivity increases, and 4) freight is currently almost 10% of revenue but lossmaking - this will presumably either scale to profitability or be closed.

This is obviously not a sophisticated analysis - indeed, I imagine it gives the detail monkeys among my readership the yips - but I find it useful for framing skew.

In the lower scenario, with 5% revenue CAGR and 15% margins, net income compounds at 7%. Uber currently trades at 30x non-GAAP net income, so the multiple would almost certainly fall in this scenario, but the deployment of free cash flow into buybacks and dividends would help. Investor outcomes would be poor, but not terrible.

In the middling scenarios, with 5-10x growth in gross bookings, 10-15% take rates, and 20-25% non-GAAP net income margins, net income compounds in the low teens for 20 years. In this scenario, I would expect FCF deployment to more than offset any multiple compression.

In the more optimistic scenarios, 11% revenue growth and 30% net margins generates 17% non-GAAP net income CAGR. With buybacks and dividends the compounded returns could be spectacular.

In short: I like the skew.

None of these scenarios represents the absolute worst or absolute best possible outcomes. In the worst case, Uber might fail completely, in which case investors will lose their money. On the other hand, the top table suggests revenue could grow faster than 11%, driving non-GAAP net income compounding at over 20% before FCF deployment, in which case investors could make 40-80x their money. This is the beauty of equity investing: the downside is capped, but the upside is not. I don’t plan on either of these last two scenarios, but if I had to pick, I’d say the second is more likely than the first.

Again: I like the skew.

Links to previous work

Thanks for reading - if you enjoyed reading this please like and restack, and do get in touch if you have questions.

Pete

First look: Vail Resorts.

Building Arks — Tue, 10 Mar 2026 17:35:29 GMT

I recently came across a reference to Vail as being cheap on every metric. Also, I have a high level thesis that if AI makes people more productive, they’ll have more more money and more time. That bodes well for expensive discretionary experiences on a 10-20 year view, and Vail fits, so I thought I’d have a look.

Ticker: MTN

IR site.

What it does

Vail operates 42 ski resorts globally. It generally leases the land and owns the resorts themselves - the ski infrastructure and associated lodging, dining, and retail real estate.

These could reasonably be described as irreplaceable assets, but the irreplaceable bit is the location, not the resort, so one long term question I have is whether leases can be renewed on good terms.

Summary figures for the last full year:

$860m of ebitda, $300m of net income, $320m of FCF.
$3bn of net debt and operating leases.
Negative tangible asset value.

Profitability for the current full year is guided lower on poor snow conditions.

The vast majority of ebitda comes from Mountain operations, which break down like this:

Subscribe now

Outlook

After several years of acquiring assets and then covid, Vail needs a reset to deliver visitor growth.

The CEO has been in place since May 2025, but had previously been CEO for 14 years, so he knows the business well. He discussed his views on the 4q25 call. According to him Vail has:

Been over-reliant on email marketing and has not sufficiently developed other channels.
Been over-reliant on call to action marketing at the expense of building an emotional connection to the brand and specific resorts.
Lost focus on Lift sales as they have grown Pass sales - they can grow both.
Failed to capitalise on strong engagement with their app, which doesn’t have checkout functionality, so conversion is low.
Not optimised the price/benefit relationship across thousands of products.

All of these problems feel solvable over a multi-year timespan.

Notably, a lot of people enter and exit the industry each year - skiing for more or fewer days, or skipping a year or two. This means there is a large opportunity to increase utilisation within the ski demographic, not just by attracting new skiers.

Subscribe now

Valuation at a share price of $134, using last full year data

16x P/E.
9x ev/ebitda.
6.7% free cash flow yield.

I’d describe this as reasonable rather than cheap, but it would get cheap quickly if the company can grow revenue at high incremental margins.

Areas for more work

Lease renewals.
Deeper dive into financials, especially incremental margins and returns on capex.
Get more comfortable with capital allocation. The company is paying a large dividend and buying back stock; I worry it should be reducing leverage.

Conclusion

I thought I would be intrigued enough to buy a tracking position. On reflection I am not. I think these are fantastic assets and I can imagine a future in which rising disposable income and management efforts drive significant revenue growth at high incremental margins. But given the debt and the risk around lease renewals, I need to do a lot more work before I risk capital.

What do you think?

Subscribe now

Links to previous work

Thanks for reading - if you enjoyed reading this please like and restack, and do get in touch if you have questions.

Pete

Review: Cheniere Energy - LNG export major

Wed, 04 Mar 2026 15:46:54 GMT

Summary

What it does: liquefies natural gas for export.

Elevator pitch: well-managed and cash generative infrastructure company with contracted cash flows, strong growth at high incremental returns on capital, and a large buyback programme.

Mental model: value.

Valuation and potential returns: DCF suggests 13% IRR to 2030 with relatively low risk.

Exchange: NYSE, ticker LNG

Stock price and market cap: $246, $52bn.

Type of holding: this is one of my trades, at a 1-2% position.

IR website: LNG IR

Introduction

Cheniere owns two LNG liquefaction assets in Louisiana and Texas. These complex, multi-billion assets cool natural gas into liquid form so it can be shipped worldwide.

I have owned Cheniere continuously for 10 years and have got the sizing more right than wrong over time. I first bought the stock in the mid $30s in 2016, and then sold quite a bit in the mid $60s in 2018. After a lot of progress in the business I added a little in the high $50s in late 2019, and then I made it a 10% position in the low $30s when markets collapsed in 2020. At that point the stock was trading for 3x distributable cash flow if you looked 18 months forward. I sold most of my stock in 2022 at $120-150, 4-5x the price I’d paid 2 years earlier, making it one of my most personally significant investments ever. I largely missed out on the rally from $150 to $250 in 2024/25, but the big percentage gains had been made by then. Finally, I added at $190 in December 2025, and that position is up 30% in 2.5 months.

My early gains were not without risk. The company was developing two new, highly complex, multibillion LNG facilities, and the potential for cost overruns or other problems was significant. It was also highly levered, so the potential downside was great. However, Cheniere is superbly managed. They have consistently under-promised and over-delivered. Today, Cheniere is a materially de-risked company: it is investment grade, with copious cash flows, strong growth at high returns on incremental capital, and significant buybacks.

Let’s dive in.

Why do we need LNG?

Oil is liquid. This makes it easy to move around. As a result, global oil prices are pretty efficient - once you adjust for quality and transport costs, there is basically one global price of oil.

Natural gas is different. It is, well, a gas. This makes it very hard to ship. You can’t put it on a truck or a ship in its natural state. You can put it down a pipe, but pipes are a bitch. They’re expensive, unpopular, and vulnerable, and there aren’t enough of them. As a result, gas prices vary wildly from place to place. Where there is local oversupply, prices are very low or even negative (where gas is a by-product of drilling for oil and there is no local demand). Where there is local undersupply, prices are much higher. This creates incredible arbitrage potential and has incentivised the development of the LNG market. By cooling gas into its liquid state, you can ship large volumes over long distances fairly efficiently.

Fracking transformed the US from being short gas to being very, very long gas. Gas prices plummeted as a result, creating a huge incentive to export gas. LNG is the only way to do this at scale.

Subscribe now

Cheniere’s journey and reinvestment economics

Through the 1990s and the 2000s it was more or less consensus that the US would need to import gas in the future. Cheniere was originally conceived as an LNG import terminal. Once it became clear that fracking was a paradigm shift, Cheniere pivoted to producing LNG for export. Thankfully the sites it owned were perfect, and some of the equipment could be repurposed, but liquefaction plants are huge projects and it took over a decade to achieve production at scale. It was an extremely impressive decade: from scratch, Cheniere built two of the biggest LNG complexes on earth, and they delivered each piece ahead of time and within budget. That’s an incredible achievement and evidences one of the key planks of the investment case: management is exceptional.

Cheniere produces 50-53mtpa today, with projects under construction to get it to 60-63mtpa, projects in advanced planning to get to 71-75mtpa, and potential beyond that to get to 100mtpa. Cheniere has two great advantages when it comes to return on incremental capital deployed:

It is building brownfield, which reduces costs. By expanding current facilities Cheniere can leverage current assets and contractor workforces, and deliver additional capacity at a construction cost to ebitda ratio of 6-7x. By my maths this translates into a 20% free cash flow yield to equity, which for assets that will likely run for 60 years is superb.
Cheniere has been shipping LNG for 10 years and has developed an enviable reputation for reliability. Some customers are prepared to pay a premium for this. Cheniere discussed this on the 4q25 earnings call: they are signing 25-year contracts with world class counterparties with a tolling fee of more than $2.75/mmbtu, when they estimate the market rate is below $2.50. If the spread is 30c, Cheniere are getting 12% more ebitda per $1 of capex than the average.

Cheniere’s business model and commodity sensitivity

Cheniere aims to sell 95% of its production under long term contracts, making its cash flows highly predictable. These contracts are priced to provide Cheniere with a return on its investment. They do not have exposure to commodity prices and tend to be with investment grade counterparties, so I think it is reasonable to view the profits from these contracts as low risk.

The other 5% of production is sold by Cheniere’s marketing arm, CMI, which captures the spread between whatever it can buy gas for in the US and whatever it can sell LNG for overseas. These profits are therefore “at risk”, but the risk skews to the upside because Cheniere doesn’t have to produce these volumes if the spread isn’t attractive. As long as CMI locks in its purchase and sales prices at the same time it can’t lose money, but when global LNG prices spike it can make lots.

One of the counterintuitive outputs of this business model is that in the long term, Cheniere might benefit more from low LNG prices than high ones. Although CMI will makes less money at low LNG prices, Cheniere is primarily in the business of deploying capital into brownfield growth under 20-year contracts at 20% FCF yields on equity. To do more of this, it needs more demand for LNG. And for that, it needs low prices.

Subscribe now

LNG market outlook

The LNG market has had an “interesting” few years. The late 2010s saw a rapid expansion of global supply and rapid demand growth out of Asia as a result. A dearth of projects in the early 2020’s meant less new supply, and there were projections for a tighter market. Then Russia embarked on its appalling invasion of the Ukraine. In response, Europe switched rapidly from buying Russian gas to buying LNG, tightening the market rapidly and pushing global LNG benchmark prices to very unusual levels. In effect, Europe outbid Asia for volume: Asian buyers tend to be price sensitive, so Asian LNG demand has barely grown since 2021.

Over the last couple of years the outlook has shifted again. European demand has become the new normal, and the market has largely adapted to it. A wave of LNG liquefaction projects over the next few years will add significantly to supply. And Asia has continued to invest in regasification and gas distribution capacity, which stands waiting to absorb new supply at lower prices.

To a great degree, this doesn’t affect Cheniere, which has 95% of its production through 2035 already contracted. But it’s worth knowing, because there will be headlines about an LNG supply glut. The tl;dr is: yes there is a lot of supply coming, but it is likely to stimulate Asian demand. Says Cheniere: “we’ve always been of the view that moderate prices are good for this industry.”

Growth projects

The following projects take Cheniere to 60-63mtpa of production in 2028, and 71-75mtpa in 2030, and 100mtpa thereafter:

Subscribe now

Legal & capital structure

Cheniere owns its Corpus Christi asset outright but it owns its Sabine Pass asset through a partnership which is separately listed (CQP). As a result

Cash flow growth is not linearly correlated to production growth - if the growth is at Sabine Pass, Cheniere only owns 51% of it.
You have to adjust for minority interests when calculating an ev/ebitda ratio. The lazy way to do this is to add the part of CQP’s market cap that Cheniere does not own to the EV. This doesn’t help you distinguish between the valuations of Cheniere and CQP, but it is good enough for a sense-check.

As it happens, the majority of Cheniere’s growth is at Corpus. Corpus has 21mtpa of capacity now, with 9mtpa under construction, and potential for 24 beyond that, giving a total of 54mtpa and 33mtpa of growth. Sabine Pass has 30mtpa now, with none under construction and potential for 20mtpa, giving a total of 50mtpa and 20mtpa of growth. This doesn’t affect the economics, but Cheniere owns a higher percentage of the growth at Corpus than at Sabine.

On the capital side, like most infrastructure companies, Cheniere is levered. On a consolidated basis it finished 2025 at 3.3x net debt/ebitda, including operating leases but excluding restricted cash. The debt structure can be seen at page 19 here. The debt is well-laddered, and has been coming down in both absolute and relative (to ebitda and asset) terms. Growth is funded 50/50 equity/debt, which is a very effective way to de-lever in relative terms. Cheniere had 5 ratings upgrades in 2025 and has had many more over the preceding years.

Management

Key to the investment case is the management team. It is incredibly hard to create a business like this from scratch, and to build multiple multibillion-dollar assets ahead of time and underbudget, but this team has done so consistently. They have also communicated with the market consistently and clearly, and they have implemented a highly rational capital allocation plan that has allowed them to:

Grow the business from 0 to 50 mtpa.
De-lever the business, both by paying down debt and by funding ebitda growth with equity, and achieve investment grade.
Reduce the share count from 257m at the peak in 2018 to 210m now, per TIKR.

Management have also had some luck, of the sort nobody wants. Over the last few years Cheniere has grown production fast and ahead of schedule. As a result, the company has had more cargoes to sell at spot than planned, just when spot prices were high because of the war in Ukraine. This has generated excess cash flow which Cheniere has used to de-lever and buy back shares. I hope they do not get that kind of luck again, although as I type LNG prices have spiked on the Iran war.

Risks

Russian gas might come back. I regard this as a low risk; Russian gas production has mostly been redirected to Asia rather than curtailed. If the political situation changes dramatically Europe might start importing more Russian gas, but they have already built LNG regas infrastructure and will remain an opportunistic LNG customer. And in the end, lower prices serve to grow the gas market.

Long term margins. Cheniere aim to sign contracts (and market spot gas) at margins of $2.50-3 per mcf. This has increased from $2-2.50 when they started out, which implies that they have been able to increase margins as the cost of building assets has risen. However, if LNG liquefaction capacity gets overbuilt in the future, margins will come down when contracts begin to roll off. In the worst case, LNG liquefaction might end up like oil refining, which is highly commoditised and done on a spot basis with no long term contracts. I think this is possible, but not soon: as long as buyers need to incentivise new capacity, long term contracts will be needed. I see this as a 2050 risk.

Feedgas. Cheniere is basically built on the shale revolution. That’s what supplies the gas that Cheniere cools and sells. It is clear that the rapid growth phase of North American shale is over. That in itself is not a huge problem: there is plenty of gas and it remains cheap by global standards. However, some analysts such as Goehring & Rozencwajg argue that gas production will now enter a long decline. That might well make Cheniere’s product uncompetitive, pressuring margins once contracts start to roll off. I think G&R get a lot right, but I think their argument is more compelling at $50 oil and $2 gas than it is at $65 oil and $3.50 gas. Said another way, I don’t think it takes a huge uplift in the gas price to incentivise more drilling, and I don’t think the shales are so exhausted that drilling won’t work.

Hurricanes. Cheniere’s assets are on the Louisiana and Texas coasts, potentially in the direct path of a hurricane. However management have known this since day 1. The assets are built to withstand hurricanes, and procedures are in place.

Subscribe now

Impact of the Iran war

If the war escalates and Qatari LNG exports are impaired in any serious way, LNG prices could stay high. This benefits Cheniere but not transformatively since most of its volumes are contracted (and most of what isn’t contracted has already been forward sold for 2026).

Equally, if the war ends tomorrow, Cheniere might immediately trade back to the low $230s, where it was prewar.

In short, I don’t think the war is hugely relevant to Cheniere unless it escalates and lasts for a long time. I very much hope that is not the case but if it is, Cheniere will benefit from being a reliable supplier well away from the war zone.

Valuation

These are long lived assets. LNG facilities can easily run for 50 years and I think well maintained ones with consistent feed gas in a mature market can likely last significantly longer.

Cheniere uses a bespoke metric called distributable cash flow, or DCF. This is basically ebitda less maintenance capex, interest, and cash tax. It overstates long term owner earnings in 2 ways: first, these are new assets so maintenance capex is fairly low; second, Cheniere is still investing hard which creates tax allowances. However DCF is an accurate measure of the cash the company will produce over the next 5-10 years, so I think it is a reasonable valuation tool. Cheniere’s latest DCF guidance is laid out on page 13 here.

Once projects currently under construction are completed in 2028, the company will produce 60-63mtpa and $21 of DCFPS assuming the middle of the DCF guidance range and no further buybacks. However, management consistently beats guidance and they are buying back shares. Assuming the top end of the guidance range and a share count of 200m, 2028 run rate DCFPS reaches $23.5.

Once the company has completed its $10bn authorised buyback and the CC and SPL Expansion projects, all due by 2030, DCFPS guidance increases to $30. I think they will beat this. Again, guidance uses the middle of the range for DCF but tend to beat guidance, so I think we can use the top end. Also, I don’t think they have accounted for all of their FCF. On the 2q25 call the company guided to $25bn of DCF from 2h25 to 2030, and $7.5bn of equity investment to get to 71-75mtpa of production in 2030. Of this, $2.4bn of DCF was to be produced and $1.4bn of capex was invested in 2h25, leaving $22.6bn of DCF and $6.1bn of capex for 2026-2030. This nets to $16.5bn of FCF over those 5 years. The dividend will consume ~$2.5bn leaving $14bn. Guidance assumes a $10bn buyback, leaving $4bn unaccounted for. Either the buyback will be bigger than guided, or the company will keep growing production.

Finally, the 2030 guidance assumes $10bn of buybacks at $285 per share. This is well above the current share price, which is a sensible assumption for a management team issuing guidance. However as an investor I’m equally interested in how many shares the company could buy back at the current price. If all FCF (after dividends) is used to buy back shares at the current price the 2030 share count will be 157m, not the 175m that management assume. With that share count, and using the top end of the DCF range, we get nearly $36 in DCFPS.

To summarise:

The implied multiples seem reasonable to me for a long-lived, highly contracted, cash-generative asset. I also run a simple DCF which assumes that all FCF after dividends is used for buybacks at the current price, and at the end of 2030 the stock is valued at 12x 2031 DCFPS, which seems fair given they will still have a runway for 33% growth (75mtpa to 100mtpa) at advantaged economics.

With a DCF built this way, the IRR is doubly sensitive to the share price because it affects both the starting point and the buyback. The alternative is to guess what price the buyback will average. I like my method because it tells me the IRR if the shares remain at the current price, but it is a matter of preference.

My DCF puts the 5-year IRR at today’s share price of $246 at 13%. At $230, the IRR is 15%; at $290 it is 8%. For me, it is a comfortable hold at the current price.

Subscribe now

Links to previous work

If you enjoyed reading this please like and restack, and do get in touch if you have questions.

Pete

February roundup

Fri, 27 Feb 2026 21:09:12 GMT

Once a month I’ll pull together my trades and articles into one email.

Trades

Sold some of my $AKO.A, Embotelladora Andina, on valuation grounds.
Bought and then after more work bought more. I think Uber is going to see a huge expansion in its TAM as AV technology advances.
Bought more $HHH, Howard Hughes Corp. While I don’t like the fees, I do like the businesses, and I think this could compound strongly for many years.
Sold some of my oil rig basket after a strong run.
Sold some $BCH, Banco de Chile, on valuation grounds.
Subscribe now

Articles

Brookfield - assorted thoughts on things I don’t think get covered enough, and a postscript.
IRSA - value in Argentine property.

Thanks for reading - and please get in touch if you have questions.

Pete

Review: Brookfield

Building Arks — Wed, 25 Feb 2026 15:22:54 GMT

Summary

What it does: Brookfield is a leading global alternatives/private asset manager.

Elevator pitch: Brookfield is a phenomenal franchise with deep moats in a growing industry. The company has great management, the stock has delivered 19% CAGR for 30 years, and there are multi-decade tailwinds in private asset creation - yet the stock is cheap both on an asset basis and on future cash flows.

Mental model: moat, value.

Valuation and potential returns: trading at 6.5x target 2030 distributable earnings. I expect the stock to compound at 15-20%.

Exchange: NYSE, TSX.

Stock price and market cap: $45, $101bn.

Type of holding: I own this in my capital allocator bucket. I intend to own it permanently unless it becomes obviously overvalued. It is 12% of my portfolio.

Introduction

I have owned Brookfield almost continuously since 2008 and in size since 2018. Over that time I have developed confidence in certain aspects of the company that don’t always get attention. This note discusses some of those. It is not a primer - others have written excellent primers, some of which are linked below.

I use the term “Brookfield” to refer to the whole organism; when I want to reference a specific entity, I use the ticker. I own BN.

Slide snapshots are from the BN 2025 investor day deck.

Quick overview and further reading

Brookfield Corporation (BN) owns a controlling stake in Brookfield Asset Management (BAM). BN also invests its balance sheet capital into investments sourced by BAM. BAM is one of the world’s leading private asset managers. It invests client money into a diverse array of assets for a fee. These assets broadly form “the backbone of the global economy” - real estate, energy generation and transmission, data infrastructure, transport, logistics, and much more. For more on what Brookfield is and does, here are a couple of good recent pieces:

Rock and Turner’s 3-part analysis of the alternatives industry and major players (part 3 is on Brookfield).
Rijnberk’s Brookfield review.

Why private asset managers exist

They always have - we just used to call them investment banks. The economy needs capital to invest in innovation and asset creation. But traditional investment banking proved to be dangerous. Investing depositor money on a fractional reserve levered basis into on-balance-sheet risky investments carried a real risk of contagion and deflationary collapse. Over time many of these activities were regulated away from banks, with the last big bout of regulation coming after the GFC. This created the conditions for alternatives managers to thrive. Instead of issuing deposits, creating money via fractional reserve banking, and owning assets on their own balance sheets, private asset managers raise equity and debt capital from investors and own the assets in fund structures. The investors understand the risks they are taking better than depositors do, fractional reserve banking is at least somewhat removed from the equation, leverage is lowered, and the risk of contagion is much reduced. And it just so happens that managing private assets this way is an excellent business.

Fee pressure in private assets

Asset management businesses are scalable, asset light, and cash generative. However, managers of public assets have been subject to huge fee pressure for many years as a result of competition from low cost index funds. This has pressured manager economics and arguably made the industry worse for investors: with lower fees, the only way to make the business work is to take on more assets, which rarely helps performance.

If assets don’t trade on public exchanges, there’s no way to create an index or a low-fee index fund. This shields managers of private assets from excessive fee pressure. There is still competition, because there are several good private asset managers and many wannabes, but there is no race to the bottom. There are two things that I think are absolutely key to any investment in Brookfield. This is the first of them.

Subscribe now

TAM 1: long term growth in private assets

Arguably, and if done well, holding assets privately is a win-win-win:

It’s better for the manager, who earns a higher fee.
It’s better for the investor, who doesn’t suffer public market volatility.
It’s better for the underlying asset, which gets patient capital.

This is not a consensus view. Consensus has it that the average manager can’t outperform and so fee reduction is the name of the game. Some also argue that public markets are better for investors, because they offer immediate liquidity and more transparent pricing. But most investors don’t need immediate liquidity - their time horizon is measured in decades - and unfortunately public markets are highly volatile and humans are emotional. Investors often underperform their public market investments because they buy high and sell low. Peter Lynch’s Magellan Fund compounded at 29% from 1977 to 1990, yet the average investor in the fund lost money. I think illiquidity is actually an advantage for most investors, even at the cost of a slightly higher fee.

If I am right, and all participants have an incentive to take more assets private over time, then Brookfield’s TAM will keep growing.

TAM 2: life insurers’ desperate need for long term assets

Over the last 5 years Brookfield has built a significant insurance business. This primarily sells life insurance and retirement annuities - to grossly oversimplify, retirees in effect lend money to Brookfield in return for fixed payments. Brookfield then invest the money, and pocket the spread between the cost of funding and the return on assets. Aging and the growing need for retirement income are common knowledge amongst investors. What is perhaps less obvious is how hard it is to find appropriate income-generating assets in a world with structurally low real interest rates. Being able to source appropriate investments is a huge competitive advantage for any seller of retirement products. The kinds of assets Brookfield buys, builds, and operates are perfect, and I think they will build a huge business here over time.

Scale as an advantage rather than a disadvantage

In public asset management, scale is usually seen as a disadvantage. There are only so many mis-pricings, and the bigger you are, the harder it is to get meaningful exposure to them within a reasonably diversified portfolio.

I think the opposite is true for private asset managers. Private asset managers buy control, not slivers of equity. Control requires big cheques. And as a rule, bigger businesses are higher quality businesses - more established, with greater cash flows and deeper moats. To buy control in a big business requires a really big cheque, so paradoxically there is less competition for the biggest and best assets. Also, the bigger you are as a private asset manager, the better your relationships with investors and capital providers - and access to finance is the name of the game.

Subscribe now

Diversification and internal competition for capital

I like the fact that Brookfield is internally diversified, for three reasons.

First, I can concentrate my capital on the risk I want to take (Brookfield’s fundraising and investment sourcing franchise, and their culture and management skill) and still not be over-exposed to any one underlying asset class, country, or sector.

Second, internal diversification means there is significant competition for Brookfield’s own capital. Consensus has it that conglomerates are bad because they lack focus and are inefficient, and that’s a fair criticism of many. But as a manager of third party capital Brookfield is highly incentivised to optimise financial outcomes and move on to the next investment, and in cases like that I think internal competition for capital is a very good thing. Contrast that with “compounder bro” thinking, where the focus is on finding a business that does one thing extremely well with deep competitive moats. Get that kind of investing right and you’ll do very well. But those businesses can usually only invest their capital in one place. That is a recipe for disaster if the moat turns out to be more of a ditch - especially if the “investment” was in the form of buybacks at elevated compounder multiples.

Third, internal diversification makes it easier for the company to adapt and stay relevant. 50% of what Brookfield invest in today didn’t exist as widely-held institutional asset classes 15 years ago. I suspect the same will be true in another 15 years, as vast new institutional asset classes come into being (AV fleets? Humanoid robots? Datacentres in space?). This adaptability arguably gives Brookfield a better chance of thriving in the long term than any other potential moat could do.

(Almost) all economic conditions are good

Brookfield is exceptionally well-structured to benefit from cycles.

This is partly because it mostly invests in assets with stable cash flows. In many cases, these cash flows are actually contracted; in others, the stability comes from things like large aftermarket revenue streams or in the case of asset management from the fact that most of the assets are locked up in multiyear funds and can’t be redeemed on demand. Revenues are also largely inflation linked - again in some cases this is explicitly contracted, in others it is implicit but effective.

More importantly, because Brookfield is both a buyer and a seller, one or other of Brookfield’s activities can thrive in each part of the cycle. Growth is strong, liquidity is abundant, and markets are euphoric? Sell assets and reap carry. The opposite? Raise money, buy assets, and collect fees while you wait. Inflation and rates are rising? Sit tight and watch your revenues go up with inflation. The opposite? Refinance and up-finance assets to reduce interest costs and free up cash to return to investors or buy new assets. The key to long term returns is not interrupting compounding, and when the world falls apart, what stops me selling Brookfield is the knowledge that they are raising and deploying capital at phenomenal returns.

Subscribe now

For example, during covid, markets panicked about the future of office and retail real estate. Brookfield doubled down and bought out the minorities in Brookfield Property Partners (BPY) in 2021. Then inflation and interest rates rose, wiping out profits at BPY and depressing real estate valuations further. But Brookfield saw that high inflation would eventually lead to rent growth and new construction starts had stalled, creating a supply vacuum for a few years’ time. Brookfield went marketing and raised around $35bn into two new vintages of their flagship Real Estate Opportunities fund, BSREP IV and V, which closed in 2022 and 2025. Today, rental growth is strong, leverage is positive, liquidity is returning, and rates seem headed down, with asset prices likely to rise “substantially” (BN 4q25 letter).

Today, there are concerns in the market about private credit. There are likely bad credits in commoditised parts of the market, and credits exposed to software may be exposed to AI disruption. Brookfield has a large private credit franchise, which mostly specialises in asset-backed lending with little software exposure. But they also have Oaktree’s famous Credit Opportunities Fund, which goes hunting when there is blood in the water. This makes Oaktree a counter-cyclical business: it struggles to raise money and invest in the good times, but it cleans up in the bad times. Brookfield commented on their 4q25 call that troubles in private credit had increased activity at Credit Opportunities in the last couple of months. Let’s see how that pans out.

Debt structure

Brookfield is highly levered. But usually debt structure matters more than debt quantum, and Brookfield’s debt is exceptionally well-structured.

At the corporate level, BN and BAM are relatively lightly levered and most of the debt is long, low, and fixed: exactly what you want to juice equity compounding without much risk.

The vast majority of Brookfield’s debt is at the asset level, using debt that suits the asset - e.g. where a hydro power plant has a 20-year PPA, Brookfield will typically borrow amortising 20-year debt against that PPA, effectively securitising the PPA payments to generate immediate cash for reinvestment. Crucially, the asset-level debt is ringfenced: it is non-recourse to the company and it is not cross-collateralised against other assets. This means problems in one asset can’t infect others. This is the second thing that I think is absolutely key to any investment in Brookfield.

Finally, Brookfield structures some of its debt to create options. A great example is the debt within BPY when Brookfield took it private. At a headline level, BPY was highly levered. However the portfolio comprised two parts: core prime real estate that Brookfield intended to own forever, and redevelopment/turnaround assets intended for eventual sale. The core assets were levered around 50%, while the noncore assets were much more levered. I love this: the significant majority of BPY equity was in the best assets, relatively lowly levered, and carried low risk. The thin sliver of equity in the noncore assets was at great risk, but there wasn’t much of it and it was nonrecourse to the other assets. This meant that most of the risk in the noncore assets was borne by the lenders, but if Brookfield could engineer a turnaround at any of those assets, it stood to make a lot of money. The headline leverage looked scary, but in fact BPY owned a low risk core business and a quiver of free options.

Subscribe now

Levered inflation protection

The combination of inflation linked assets carrying appropriately structured debt means that Brookfield offers levered exposure to inflation. We live in a fiat world in which governments can’t resist printing money, especially in crises. That creates inflation. The first goal of any investment strategy has to be to preserve purchasing power. If your revenues grow with inflation, and you lever appropriately, your profits should grow a little faster than inflation. That alone is investing gold.

Brookfield as its own banker

While Brookfield doesn’t cross-collateralise its debt, BN does have a huge balance sheet which it can use to support subsidiaries when they need it. This mechanism is used relatively sparingly, but it is very valuable: for example, if an affiliate wants to complete an acquisition but financing conditions aren’t great, BN can lend bridge financing until the affiliate can issue debt on good terms. This adds an important degree of flexibility to Brookfield.

Consistent messaging/strategy

One of the things you notice when you own a stock for nearly 20 years is how much the “story” changes. I love it when stories don’t change. When management describes their business the same way year in, year out, I think it tells you something quite important about the durability of the business. I find it remarkable how consistent Brookfield’s messaging has been as they have methodically built out their franchise. That doesn’t mean that everything they have talked about has worked out - that would be freakish. But the messaging about key things has been very stable: how they think, what their strategic direction is, what kinds of assets they like, why demand for these assets is growing, what returns they target, how they achieve those returns, why investors like Brookfield, how they build their culture, how they handle succession, etc. And at a high level, they have absolutely achieved what they set out to achieve 20 and 30 years ago. I think this is a phenomenally well-managed business.

Subscribe now

Alignment

Senior management hold much of their net worth in Partners Value Investments, which in turn owns BN, BAM, and other Brookfield entities. Partners is separately listed (PVF.UN on the TSX) and importantly it is levered. In most companies, managers get paid in stock options. This gives them an incentive to swing for the fences: if it works they make out like a bandit, and if not they lose nothing. PVF arguably achieves the opposite: because it is levered, the senior managers at Brookfield will lose everything before minority shareholders do. They have a strong incentive to grow carefully and methodically while protecting the downside.

Second, the company is aligned with its clients. Both BN and BAM invest into funds managed by BAM. If I was a client, this would give me confidence, and I think it is an advantage over less well-capitalised firms. It also allows Brookfield to seed new funds to broaden their business.

Third party due diligence

Brookfield is a relatively difficult business to analyse. It owns consolidated, associate, and minority stakes in a huge variety of assets, which it buys and sells regularly. This makes for complicated financial reporting and lumpy earnings. Brookfield disclose a huge amount of information and in my view do a good job of helping investors understand what really matters, but a degree of trust is needed.

One thing that helps me build trust in Brookfield is 3rd party due diligence. Brookfield has relationships with 2,500 institutions who invest in their funds and coinvest directly in some of their investments. Some of these institutions will be relatively small and have limited capacity for due diligence. But some are monsters - huge professional organisations with deep capabilities that do extensive research and choose their partners carefully. The fact that they are willing to back Brookfield, and reinvest again and again in its funds and sometimes directly in the underlying assets, is telling.

Added to this is the fact that Howard Marks and Bruce Karsch chose to sell their life’s work to Brookfield. These men - and the people who work for them at Oaktree - are exceptional analysts and investors. Their choice to partner permanently with Brookfield - and to oversubscribe for shares rather than cash in the original transaction - is a useful indicator.

The fundraising and asset sourcing machine

Finally, Brookfield has transformed itself over the last 30, 20, and 10 years. 30 years ago it was an owner and operator of assets recovering from a major crisis (long story short: this is how Brookfield learned not to cross-collateralise its debt). 20 years ago it was well on the path to being a manager of 3rd party capital with its operating expertise as the backbone, but it was subscale. 10 years ago it had relationships with 250 institutional investors and was in the market with 4 strategies. Today it has relationships with 2,500 institutional investors and in 2026 it will raise money across 60 strategies. In addition it has 70,000 private wealth clients and 800,000 insurance policyholders, and all these numbers are growing. Across multiple asset classes it offers flagship funds, evergreen supercore and core funds, credit funds, geographical sleeves, co-investment opportunities, etc., meaning it offers something to suit the return, risk, duration, and diversification requirements of almost any investor. As they said on the last call, they “have built a business that can raise capital more consistently and deliver an earnings profile that is more predictable, more resilient, and better positioned to grow across economic cycles”.

Subscribe now

End note

Hopefully some of the above is useful. I think this is a phenomenal business with phenomenal tailwinds. BN’s 5-year plan calls for just under $7 of distributable earnings per share in 2030. Historically they have roughly hit their 5-year targets. The stock is at $45 today. If they deliver $7 in DEPS I don’t think it will stay there. I hope to own BN for another 17 years and expect to be very happy with the outcome.

Review: IRSA - cheap Argentine property cockroach

Wed, 28 Jan 2026 12:34:52 GMT

Summary

What it is: IRSA is an Argentine property company. Its main assets are a dominant mall portfolio and a premium development land bank. It is a cockroach that survived covid and some of the worst macroeconomic mismanagement on earth at the same time.

Why you should read this: IRSA is in pole position to benefit from Javier Milei’s reforms in Argentina. It trades at 1x an understated NAV that seems set to grow. It has a cash-generative and growing core business, a huge residential development project in the best possible location in Buenos Aires that will soon turn cashflow positive, an attractive land bank waiting to be developed, and an under-levered balance sheet. I believe it can reinvest internally and fund strong returns to shareholders for many years.

Potential returns: Bear case down 50%. Base case 15-20% for 5 years. Bull case 20-25% for 10 years.

Exchange: NYSE, ticker IRS.

Stock price and market cap: $18 per GDS (10 shares per GDS); market cap $1.4bn.

Year end: June 30th.

IR website: here.

Do I own it? Yes, at a 2% weighting, using moat and value mental models.

Quick history

IRSA has been listed in Argentina for over 70 years and on the NYSE for over 30. It is controlled by Cresud, also dual-listed, which has a 53% stake. Cresud is in turn controlled by the Elsztain family. IRSA is the product of a series of acquisitions, mergers, and internal development. None of these is particularly relevant to the investment case today. What is relevant is that I think the Elsztains have been strong value-oriented opportunistic capital allocators over time, and have both protected and created value despite operating in one of the world’s basket-case economies. Recent examples include the sale of half the office portfolio at strong cap rates in the 5-7% range, the acquisition of “turnaround” mall assets at higher cap rates, the repurchase of shares at discounted prices, and the payment of significant dividends. That said, as with any controlled investment, there is always a risk that the controlling party does something that is not in minority interests.

This article focuses on the mall portfolio and the development land bank. The other assets - mainly offices and hotels - are smaller and don’t move the needle.

Malls

Malls are a staple in much of Latin America. Despite the rapid growth of sophisticated ecommerce platforms like MercadoLibre, malls are thriving for a range of reasons:

They’re air-conditioned, comfortable, and safe, in countries where street shopping sometimes isn’t.
They never got overbuilt like they did in more developed markets.
For high-end foreign brands, they’re the natural gateway to new countries and cities.
They’ve done a good job of evolving their brands and experiences to meet modern needs.

IRSA dominates the Argentine mall industry. It has 9 malls with a 67% market share in the City of Buenos Aires, and 8 malls in other cities with another 1 in development. Income-producing gross leasable area (GLA) stands at 371,000 today and will grow to 458,000 over the next 2-3 years new and redeveloped assets come on-line. Occupancy at the stabilised assets is 98%. This business has two key economic attributes. First, as with most real estate, operating costs are low relative to rent, so once the asset has been built, cash margins are high. Second, 75-80% of revenues are rents, structured with an inflation-linked base rate and a % of sales above a threshold. The inflation protection is critically important in a high-inflation economy like Argentina’s. The % of sales provides exposure to the compounding of real GDP, which will be valuable if Milei’s reforms succeed. The final 20-25% of revenue comes from a variety of smaller income streams like key money, concessions, parking, and advertising, all of which I would expect to rise with nominal GDP over time.

The combination of high cash margins and inflation-linkage means that IRSA’s malls are cash generative under virtually all economic conditions. However, in the company accounts this cash flow is merged with a variety of lower margin and more volatile income streams: hotels, real estate development, property sales, and the change in fair value of investment properties, which can produce wild non-cash swings on the P&L. These income streams are all likely to be profitable over time, but to the casual observer they obscure the attractiveness of the malls business.

Land bank part 1: Ramblas del Plata

IRSA has a land bank to die for. The crown jewel is Ramblas del Plata (RdP), an enormous plot in prime Buenos Aires within easy reach of the central business district and sandwiched between the redeveloped historic dock area of Puerto Madero, an ecological reserve, and the waters of the Rio de la Plata (known to history as the last resting place of the Admiral Graf Spee, one of Hitler’s pocket battleships).

It took IRSA decades to secure development permits for RdP, but in 2023 they got the final green light for an 866,000 square metre development. This is big - depending on the final design, there will be 6-10,000 homes plus office and retail space.

IRSA is using a capital-light development model for RdP. Most of the land will be sold to developers for an up-front payment, a mortgage, and a percentage of eventual sales, which is around 25% initially and is expected to rise towards 30% as the project matures. The up-front payments fund basic infrastructure like roads and utilities, and the percentage of eventual sales is almost pure profit.

Of the 866,000 square metres permitted for building, 693,000 is saleable. This will be heavily weighted towards residential, and IRSA has guided to a sale price of $4-6,000 per square metre for residential, which is easily to verified on online property portals. The exact proportion of other uses hasn’t been set yet, but I am not sure it matters: premium office and retail fetch similar or possibly higher pricing.

If the average sale is at $5000 per square metre, and IRSA gets 27.5%, they’ll recognise $950m in share-of-sale revenues at gross margins approaching 100%. After tax and discounted, this is worth $350-400m today depending on the exact inputs used. That’s similar to the value IRSA assigns to RdP on its balance sheet, so the asset is not undervalued today, but I’m happy to earn the discount rate as it gets monetised. In addition, if Argentina’s economy goes into sustainable growth mode we could see development being accelerated and real estate prices rising in real terms, both of which increase the discounted value. IRSA may also make positive cashflow on the upfront payments.

The bottom line is that RdP is going to be a huge generator of cash for IRSA over the next 5, 10, and 15 years. As of the most recent quarter (1q26), plots permitted for 116,000 square metres of buildable area had been sold or swapped with developers and sales of completed properties to final buyers should start in 2028.

One potential risk is that Buenos Aires property prices are higher than peer cities in the region such as Santiago and Panama City. This is particularly true for premium property. The differences aren’t vast, and can probably be explained by local supply conditions - e.g. Buenos Aires has limited waterfront property close to the city centre and central business district. However, during the bad years of Argentine economy policy, capital controls made it difficult to get money out of the country, and investing in premium real estate was one of the few good ways for the wealthy to protect against inflation. It’s possible that as the economy normalises, rich Argentines diversify into other assets, and premium real estate prices moderate somewhat. This is worth watching, although I think it’s more likely that prices rise with economic growth and increased mortgage issuance (see below).

Land bank part 2: the rest

In addition to RdP, IRSA has an extensive collection of smaller assets that aren’t generating income yet. Data from the company website and IR deck suggest that by developing or renovating all of these, IRSA would add around 400,000 square metres of saleable or leasable space, not including the projects that will grow mall GLA to 458,000 square metres. I don’t attempt to value these assets precisely: there isn’t enough data and I can’t assess permitting risk, although some assets are already permitted; but collectively these assets provide significant optionality for a savvy management team to reinvest opportunistically if the economic conditions are right.

The 400,000 figure does not include two huge plots near Lujan which could support 3.5 million square metres of building, mostly residential and saleable. These assets have been held for years and don’t get discussed, so I am not sure how valuable they are. At 8am on a Tuesday morning Google Maps puts Lujan roughly a 1 hour and 15 minute drive from the upmarket/CBD areas of Buenos Aires, which is theoretically commutable, but my experience of Latin American cities is that traffic can be unpredictable and when it’s bad, it’s bad. Still, these assets would move the needle if they came to life.

Viva la Libertad, carajo!

Something has changed in Argentina. Argentina’s political history is complex, but the Peronists have been the dominant political party for most of the post-war period (and when they weren’t, it was generally because they were banned rather than unpopular). Peronism is a broad church, but it has leant heavily left economically, and Argentina has paid a high price in inflation, lack of investment, and rolling economic crises. A lasting change to free market policies would do wonders for Argentina’s people and asset values, so it is highly relevant that pro-market (non-Peronist) candidates have won 2 of the last 3 presidential elections.

Mauricio Macri won the 2015 election on a pro-market platform. Unfortunately pro-market reforms usually entail pain before gain. In a bid to lessen the pain, Macri opted for slow reforms. It didn’t work - the benefits didn’t arrive before the 2019 elections. If he’d reformed the economy more aggressively, Macri might have won again, but in the event he lost to a disastrous Peronist government that stoked inflation to over 200%.

Enter Javier Milei.

Milei became President in December 2023 and has transformed Argentina’s political scene in 2 short years. He is that rarest of things: a charismatic free-market economist. His administration has ripped off the band-aid, balancing the government budget in a few short months, bringing inflation down from >200% to 20-30%, and taking a chainsaw to Argentina’s endemic overregulation. In the first 6 months this was painful, but Argentina’s economy has since responded, accelerating to 4-5% growth. Poverty is down, Milei’s party recently won a significant victory in the midterm elections, and his approval rating is high. There are challenges ahead - not least accumulating sufficient fx reserves to float the currency - but Milei’s Argentina is going in the right direction.

A free market Argentina has immense potential. In the 1980s a team of economists trained under Milton Friedman in Chicago reformed Chile’s economy along strictly free market lines, and the result was a 30-year economic miracle, with rapid economic growth, a massive reduction in poverty levels, and net cash on the sovereign balance sheet. Argentina is not Chile, and 2 years is a long time in politics. But at the same time, I think Milei’s reforms are going to work. If they do, then going into the next election in late 2027 the economy will be growing, jobs will be being created, inflation will be down, and living standards will be rising. That’s a very good backdrop for re-election, and if Milei serves 8 years he might create a “new normal” in Argentine economic thinking. In Chile, the left won every election from 1988 to 2010, but they didn’t undo the Chicago reforms because they were working so well. (It was only when the next generation, who did not remember the bad old days, came to dominate politics that the reforms came under real threat - and even then, the changes have been fairly minor - the real damage has been to confidence.)

There is a chance Argentina is at the start of that journey. For its own sake, I hope it is.

The mortgage market

After years of low issuance and high inflation, Argentine private sector debt to GDP is around 8%. Within that, consumer debt is around 5% and mortgage debt is around 0.5%. This is incredibly low: in the US mortgage debt is worth 50% of GDP and in neighbouring Chile it is nearly 30%.

What banks need in order to lend with confidence is controlled inflation and stable interest rates. Banks have already started lending again in Argentina, and if conditions continue to stabilise lending will only accelerate. Sustained debt issuance over 5-10 years would likely have a significant impact on investment, job creation, consumer spending, and premium real estate prices. IRSA owns malls and premium residential development properties. It is hard to imagine a company better placed to benefit from a reflation of the Argentine economy.

Balance sheet

Side note: the Argentine peso is volatile against the dollar, so when converting balance sheet values to dollars you need to be careful to use the exchange rate on the balance sheet date. However you can think about IRSA in dollars: its assets are all either inflation-linked or dollar-denominated, and its debt is in dollars.

IRSA’s balance sheet is under-levered. As of September 2025, IRSA has $500m of debt and $190m of cash. Maturities are manageable, with the majority after 2032. Rental ebitda was $190m in 2025. This puts net debt at 1.6x historic ebitda, and rental ebitda will grow with real gdp and mall GLA growth. The loan/assets ratio was 9%, but I think assets are understated. Malls represent 43% of assets, and due to the lack of a liquid market for malls IRSA values its malls using a DCF. At year-end 2025 the discount rate was 11%, giving a valuation of 7.5x mall ebitda. In offices, where there is a liquid market, IRSA has been selling at cap rates of 5-7%. I suspect the lack of a liquid malls market is due to the scarcity of assets and IRSA’s dominance in the sector, and that if IRSA wanted to sell malls it could achieve cap rates well below 11%. In other words, I think IRSA carries its malls below market value and the loan/asset ratio is even lower than it looks.

Low leverage made sense in the highly unpredictable economic environment IRSA faced from 2019-2023, especially since IRSA did not have cash taxes to pay. It will not make sense if the economy continues to stabilise. Management have started increasing leverage, issuing a $300m 10-year dollar bond at 8% in the middle of 2025, and I expect this to continue slowly - potentially for many years as asset values rise. This means IRSA has 3 big sources of cash to fund landbank development and continued returns to shareholders: operating cashflow from the malls, development cashflow from (in particular) RdP, and rising leverage.

Two other balance sheet observations:

IRSA’s main liability is deferred tax, which is largely payable if they sell assets. However, much of their asset base is probably going to be held in perpetuity. If the deferred tax is never going to be paid, that value arguably accrues to IRSA’s shareholders, and book value could be thought of as being up to 40% higher.
IRSA owns a 29.9% stake in a government-controlled mortgage lender, Banco Hipotecario. This is separately listed and trades at 1.5x book value, putting IRSA’s stake at $150m. Banks in more stable emerging markets often trade at 3-4x book.

Valuation

IRSA trades at $18/GDS and a $1.4bn market cap. Using the most recent (1q26) balance sheet NAV of $1.3bn, that’s a multiple of 1.1x. As argued above, balance sheet NAV likely understates true value given the discount rate used in malls and the deduction of deferred tax payable on sale.

The bear case is that Milei loses the next election. Between 2020 and 2022 the stock traded in a range between $3 and $5. I don’t think it will go there again for 3 reasons. First, that period included covid. Second, by making freedom popular May have changed the political discourse in Argentina for a generation; if so, the only way for the Peronists to regain power will be to move to the centre. And third, when the stock hit its lows RdP wasn’t permitted, whereas by the next Presidential term it will be generating cash. If Milei loses I could see the stock halving, which puts me off making IRSA a large position, but IRSA is a cockroach: it’s survived Peronism before and would again. At that price it would be dirt cheap and I would likely slowly accumulate shares.

The base case is that Milei wins again, the economy grows for 5 years, mall cashflows rise, discount rates come down, RdP starts generating cash, development of the rest of the land bank accelerates, and leverage rises slowly. In this case NAV could double over 5 years, the premium to NAV could expand somewhat, and IRSA could pay significant dividends, for a 15-20% 5-year compounded total return.

The bull case requires that we are at the start of Argentina’s “Chile moment”. Real GDP grows at 4-5% for 10 years driven by rising corporate and foreign investment. This creates vast wealth and the middle class expands significantly. Mall ebitda compounds at 7-8%, inline with nominal dollar GDP. This combined with a reduced discount rate triples mall NAV, enabling IRSA to borrow an additional $1bn, pay a 10% dividend yield for 10 years out of debt proceeds and mall income, and grow its NAV by 50%. RdP produces $750m of after-tax cash flow, adding another $400m of NAV. IRSA leverages the $750m with another $750m of debt to build $1.5bn of leasable assets. The cap rate at cost is 8%, so this generates $120m in annuity ebitda. However the market rate for these properties is 6%, so the finished assets are worth $2bn, adding another $500m of NAV. With mortgage issuance booming Banco Hipotecario’s book value compounds at 15% in dollar terms and the stock trades at 3x book, adding another $1bn to NAV. In total, NAV triples over 10 years for a 12% CAGR, and Argentina’s stock market prices in a continuation of the good times so the valuation rises to 1.5x NAV. This gives a 15% 10-year stock price CAGR plus a 10% average dividend on the current share price. None of these are specific predictions, but I don’t think the assumptions are heroic either - rather, what produces exceptional returns is that IRSA could feasibly do all of these things at the same time.

Conclusion

The absolute downside is greater than I would like, especially given the difficulty of assigning a probability to the next election outcome. But overall this is the kind of investment I love: a solid and fairly high-probability base case, with a bear case I can live with and an exciting but not unrealistic bull case. I am happy to hold IRSA at the current price, and would likely add on a significant drop.