Warren Buffett: pulls out a financial statement and sets it on the table "In financial services, reputation IS the product. One scandal and decades of trust evaporate. But Brookfield presents us with something more interesting than a trust question — it's an intelligibility question. Bruce Flatt just told us on the earnings call that a million dollars invested thirty years ago is worth two hundred and eighty-five million today. That's a 19% annual compound return, which puts Brookfield in rarefied company. But here's what's nagging at me — if this is truly one of the great compounding machines of our generation, why does the market price it at a 35-40% discount to Blackstone and KKR? Markets can be wrong, but they're not stupid for thirty years running. So let me pose the central question to this group: Is Brookfield Corporation a genuinely great business obscured by accounting complexity, or is the complexity itself the tell — the thing that should make us walk away?"

Charlie Munger: "That's exactly the right place to start, and I'll answer directly: the complexity is both the feature and the bug, and distinguishing between the two is the entire analytical challenge. Let me invert your question. What would have to be true for Brookfield to be a bad investment from here? Three things: the GAAP-to-distributable-earnings gap would need to reflect genuine overstatement rather than consolidation artifacts, the $502 billion in debt would need to harbor correlated risks that management has obscured, and the 49% share dilution would need to signal that management prioritizes empire-building over per-share value. Now, I've studied holding company structures for sixty years — from Berkshire to Leucadia to Loews to Jardine Matheson — and Brookfield most closely resembles what Berkshire looked like in the 1980s: a business where the GAAP statements were almost useless for understanding the economic reality because the consolidation of insurance float, operating businesses, and investment portfolios produced numbers that bore no resemblance to the underlying earning power. The difference is that Charlie Munger and Warren Buffett were running Berkshire, and we could verify the pieces ourselves. With Brookfield, Bruce Flatt is asking you to trust his supplemental disclosures in a way that requires more faith than I'm typically comfortable extending."

Dev Kantesaria: "Charlie, I want to challenge the Berkshire analogy directly, because I think it's flattering but fundamentally misleading. Berkshire in the 1980s was a holding company with an insurance float and wholly-owned operating businesses — but the operating businesses were things like See's Candies and Nebraska Furniture Mart, which generated enormous free cash flow on minimal invested capital. Brookfield's operating businesses are hydroelectric dams, toll roads, fiber networks, and container terminals — every single one of which is capital-intensive, interest-rate sensitive, and requires continuous reinvestment just to maintain current capacity. When I apply my toll booth test to Brookfield at the consolidated level, the answer is unambiguous: no one is forced to allocate capital to Brookfield's funds. Institutional investors can choose Blackstone, KKR, Apollo, Ares, or a dozen other capable managers. There is no mandatory checkpoint here. The pension fund in Ontario doesn't wake up and say 'I literally cannot deploy this capital without paying Brookfield's toll' — they have a menu of thirty alternatives. That's not inevitability; that's competition for a commodity product called 'alternative asset management.'"

David Tepper: "Dev, I hear you on the toll booth framework, but you're applying a framework designed for monopolistic software businesses to a fundamentally different animal, and it's producing a false negative. Let me reframe what Brookfield actually is. You said institutional investors have thirty alternatives — technically true, but practically misleading. When the government of Qatar wants a single partner to build $10 billion in data center infrastructure across three continents, the list of organizations that can actually execute that is maybe four or five firms on earth. When Microsoft needs a partner to source 10 gigawatts of renewable power for their AI training clusters, they don't run an RFP through thirty managers — they call Brookfield because Brookfield actually operates the physical assets, not just the financial structures around them. Bruce mentioned partnerships with NVIDIA, Microsoft, JPMorgan, and multiple sovereign governments on the call. That's not a commodity product — that's a capabilities moat built over four decades of owning and operating real infrastructure. The switching cost isn't contractual; it's competence-based. You can technically hire someone else, but when you're deploying billions into physical assets across thirty countries with different regulatory regimes, you don't switch partners because their fee is 10 basis points lower."

Mohnish Pabrai: "Tepper's point about the competence moat is exactly right, and I want to add something that I think this group is underweighting — the self-reinforcing nature of Brookfield's fundraising flywheel. They raised $112 billion in a single year. Let that sink in. That's more than most asset managers have in total AUM. And here's the Darwinian dynamic: because Brookfield is now one of the largest allocators on earth, they see deals that literally no one else sees. A $15 billion infrastructure platform in India doesn't get shopped to a $20 billion fund — it gets shown to Brookfield, Blackstone, and maybe two others. Scale begets deal flow, deal flow begets returns, returns beget fundraising, and fundraising begets scale. This is the flywheel that Jeff Bezos talks about at Amazon, applied to alternative asset management. And unlike Amazon in 2005, this flywheel has been spinning for thirty years with documented results."

Robert Vinall: "Mohnish, I'm sympathetic to the flywheel thesis, but I want to probe something that nobody's addressed yet: the reinvestment runway at the per-share level, which is what actually matters to us as shareholders. The platform is clearly growing — AUM up from $100 billion to a trillion over a decade, fee-related earnings growing 22% on 12% AUM growth, all the operating leverage you'd want. Beautiful. But in 2025, weighted average shares outstanding jumped from 1.512 billion to 2.247 billion — a 49% dilution event. That means distributable earnings per share of $2.27, which would have been $3.57 on the prior share count. When I evaluate reinvestment runways, I'm obsessive about per-share compounding, not aggregate growth. A business that grows earnings 11% in aggregate but dilutes shares 49% has destroyed per-share value in that year, regardless of how impressive the top-line fundraising numbers look. I understand this was the BNT corporate simplification, not a traditional equity raise — but the economic reality for an existing BN shareholder is that their claim on future earnings was mechanically reduced by a third. I need to see 2026 and 2027 demonstrate genuine per-share compounding on this new, larger base before I can underwrite the reinvestment thesis."

Pulak Prasad: "Robert raises the critical evolutionary question: is this organism growing stronger or just growing larger? In biology, there's an important distinction between an organism that scales efficiently — think of a blue whale, where metabolic efficiency improves with body size — and one that merely accumulates mass until it collapses under its own weight, like a beached whale. Brookfield has survived since 1899 — two world wars, the Great Depression, the Asian financial crisis, the GFC, COVID. That 126-year survival record is genuinely remarkable and tells me the organizational DNA is resilient. But here's what concerns me from an evolutionary perspective: the balance sheet went from $391 billion in total assets in 2021 to $519 billion in 2025, with consolidated debt exploding from $122 billion to $502 billion. Even if 90% of that debt is non-recourse and ring-fenced at the project level, the complexity of managing refinancing across hundreds of entities in thirty countries during a liquidity crisis is itself a survival risk. Brookfield navigated 2008 brilliantly, but it was a $30 billion company then. Can the same institutional reflexes operate at $500 billion?"

Warren Buffett: "Pulak, that's the question that keeps me up at night. Let me push the group into the financial evidence now, because I think the numbers tell a more complicated story than either the bulls or bears want to admit. Look at the GAAP net income trajectory: $3.97 billion in 2021, $2.06 billion in 2022, $1.13 billion in 2023, $641 million in 2024. That's a decline of 84% over three years while the platform was supposedly scaling to record levels. Management says distributable earnings grew 11% to $5.4 billion — but GAAP says earning power collapsed. One of these narratives is profoundly wrong, and our investment thesis depends entirely on which one we believe."

Charlie Munger: "And this is where the incentive structure becomes critical. Show me the incentive and I'll show you the outcome. Management designs the distributable earnings calculation themselves. They decide what gets added back — the $9.7 billion in depreciation, the fair value adjustments, the minority interest allocations. When management's compensation and public narrative depend on a metric they themselves define, calculate, and present without independent audit attestation, I apply maximum skepticism. I'm not saying they're dishonest — Flatt's 30-year track record suggests the opposite — but I'm saying the structure creates the incentive for optimistic interpretation, and the 4:1 gap between management DE and GAAP net income is wide enough to drive a truck through. The GAAP ROIC has been stuck at 2.9% to 4.1% for an entire decade. Book value per share has been essentially flat at $18.38 to $18.53 from 2021 to 2024. If distributable earnings were truly $5.4 billion, you'd expect something to show up in the audited statements — an improving ROE, rising book value, positive free cash flow. Instead, free cash flow has been deeply negative every single year: negative $13 billion, negative $31 billion, negative $23 billion, negative $22 billion, negative $21 billion. That's $110 billion in cumulative negative free cash flow over five years. Where did the money go?"

David Tepper: "Charlie, the money went into assets — that's the whole point of the business model. Every dollar of negative free cash flow is a dollar deployed into infrastructure, renewable energy, real estate, or insurance investment portfolios. This isn't a SaaS company burning cash on customer acquisition with no tangible backing — these are physical assets generating contractual cash flows. The $9.7 billion in annual depreciation is being applied to hydroelectric dams that have been generating power for fifty years and will generate power for another fifty. Depreciating a dam built in 1970 at its historical cost basis tells you nothing about its current earning power or replacement value. This is exactly the same distortion that made Berkshire's GAAP numbers useless in the 1980s — you were depreciating insurance company bonds and wholly-owned businesses at historical cost while the economic value compounded. I'm not saying Brookfield is Berkshire, but the accounting distortion is structurally identical. And here's what the bears miss: operating cash flow has been consistently positive and growing — $7.9 billion in 2021, $8.8 billion in 2022, $6.5 billion in 2023, $7.6 billion in 2024, $11 billion in 2025. The business generates enormous cash; it just reinvests more than it generates because the return opportunities exceed the cost of capital."

Dev Kantesaria: "David, you're making my case for me. A business that must continuously reinvest more than it generates — year after year, decade after decade — is by definition capital-intensive. That's the opposite of what I look for. When Visa processes a transaction, the marginal cost is essentially zero and the cash drops straight to free cash flow. When Brookfield builds a data center, they deploy $2 billion in capital, finance it with project-level debt, and pray that the contractual cash flows cover the debt service, the operating expenses, and leave enough for a return on equity. If interest rates spike, the financing costs eat the return. If the anchor tenant walks, the asset sits idle. That's risk, not inevitability. The 3.9% ROIC isn't just an accounting distortion — it reflects the fundamental reality that you're deploying massive capital for modest returns. The asset management fees may be capital-light, but the entity we're being asked to buy owns $500 billion in capital-heavy assets alongside those fees."

Mohnish Pabrai: "Let me bring us to the price question, because I think the intellectual debate about GAAP versus DE can go on forever. Here's my framework: what am I actually buying at $39.45, and what's the downside? Brookfield owns 75% of BAM, which trades publicly. That stake alone is worth roughly $26-27 per share. So at $39.45, I'm paying about $13 per share for everything else — the insurance platform with $145 billion in assets, the operating businesses across 30 countries, $11.6 billion in accrued carried interest, and optionality on the BN-BNT merger closing the conglomerate discount. Now, can those pieces be worth zero? In any realistic scenario, no. The insurance business alone has regulatory equity backing it. The infrastructure assets have contractual cash flows. So my downside to the BAM floor is roughly 33%, but my upside if the market re-rates even halfway toward Blackstone's multiple is 35-40%. That's a reasonable asymmetric setup — not the 10:1 I typically hunt for, but better than most things in this market. The 88 billion market cap is near my upper threshold, but the asset-backed downside protection makes this an exception."

Robert Vinall: "Mohnish, I like the sum-of-parts logic, but I'd caution against treating the BAM stake as a hard floor. BAM's value is itself a function of Brookfield's fundraising momentum — if the platform stumbles, BAM's multiple compresses too. These pieces are correlated, not independent. My preference is to wait for cleaner per-share financials. If management delivers $2.50-2.70 in DE per share on the new 2.25 billion share base in 2026 — proving the dilution was structural, not value-destructive — I'd buy below $38. That gives me a 20%+ margin of safety to a conservative $48 fair value using a 15x DE multiple, which respects both the platform quality and the opacity discount."

Pulak Prasad: "I land in a similar place to Robert. The organism has survived 126 years — that counts for something enormous in my framework. But the recent leverage expansion and share dilution represent a mutation event that needs observation before I commit capital. I'd buy below $37, which gives me sufficient cushion against the scenario where the post-merger share count introduces further complexity. The 'away from desk' test is illuminating here — could I own this for twelve months without checking? Honestly, with $502 billion in consolidated debt and a GAAP earnings trajectory that's declining 40% per year, I'd be nervous. I need more stability in the observable financials before I could sleep well."

David Tepper: "That's where we differ fundamentally. I've spent my entire career buying things that make people nervous — and the nervousness is the opportunity. At $39.45 and 17x distributable earnings, with a platform growing fee-related earnings at 22% and catalysts including the BN-BNT merger and $91 billion in realized asset sales already in motion, I'm buying now. The market is pricing in permanent opacity, and I'm betting the simplification closes that gap within eighteen months."

Warren Buffett: surveys the room "Let me try to pull this together, because we've had a genuinely illuminating debate that exposes the central tension in Brookfield better than any analyst report I've read. On the qualitative side, we have broad agreement that this is a real business with real competitive advantages — David's point about being one of four or five firms on earth that can execute $10 billion cross-border infrastructure mandates is not hand-waving, it's an observable fact confirmed by the NVIDIA, Microsoft, and sovereign government partnerships Bruce cited on the call. The fundraising flywheel Mohnish described — $112 billion raised in a single year — is genuine and self-reinforcing. No one in this room thinks Brookfield is a fraud or a house of cards.

But here's where the debate got truly interesting: the financial evidence is genuinely ambiguous in a way that separates this council more than almost any stock we've discussed. Charlie put his finger on the core problem — GAAP net income has declined 84% over three years while management claims distributable earnings grew 11%. Book value per share has been flat for four years. Cumulative free cash flow is negative $110 billion over five years. Dev is right that these are the financial signatures of a capital-intensive business earning below its cost of capital, and David is right that the accounting framework is structurally incapable of capturing the economics of a holding company that consolidates infrastructure assets alongside fee-light management businesses. Both things are simultaneously true, which is precisely why the market applies a 35-40% discount to peer multiples — not because the market is ignorant, but because the market can't resolve the ambiguity either.

Four of us — Charlie, Robert, Pulak, and myself — want to own this business but at a lower price, somewhere in the $36-38 range, which gives us a genuine margin of safety against the possibility that the truth lives closer to GAAP than to management's framework. David and Mohnish believe the current $39.45 already embeds sufficient safety given the BAM stake floor and the catalyst timeline. Dev walks away entirely on philosophical grounds that I respect even if I don't fully share. The majority view is Buy Lower — not because we doubt the business, but because when three different earnings metrics disagree by multiples of each other, humility about what we don't know should be reflected in the price we're willing to pay."