In fiscal 2025 Berkshire earned $44.49B in operating earnings on revenue of about $371B, employed 387,815 people across 60-plus subsidiaries, and closed the year sitting on roughly $373B of cash and Treasury bills[1][8][33]. Then, on January 1, 2026, Warren Buffett stopped being its CEO for the first time in 60 years[4].

The genuinely open question is not whether Berkshire has been a strong long-run compounder — a 19.9% annual gain over 1965–2024 versus 10.4% for the S&P 500 settles that[10]. It is whether the next chapter resembles the last one. Greg Abel inherits unmatched financial firepower and a culture built to run without a founder — but also a company so large it has trailed the index for a decade, a cash hoard it has struggled to deploy, and a succession that has already cost it Charlie Munger and two senior lieutenants[12][13][26]. The evidence does not cut evenly across these questions: it leans toward the skeptics on size, toward continuity on succession, toward drag-so-far on the cash, and stays genuinely contested on the moat. Each section lays out both cases, and the closing weighing in Risks & the Forward View states each lean, its confidence, and what would flip it.

The decisive questions

Each links to the section that lays out the evidence on both sides.

The cash mountain

Cash, cash equivalents and Treasury bills, US$B, year-end (Q3'25 shows the intra-year record). The hoard nearly tripled in three years as Buffett sold stock and found little to buy — the single most-debated fact about Berkshire today.

How to read this

Nine sections, each built the same way: a neutral synthesis, a two-sided case-for / case-against ledger, sourced data and charts, and dated facts. Start with the question that interests you, or read in order from the Overview.

Berkshire is not one business but a holding company wrapped around an insurance engine. Buffett spent six decades turning underwriting float and subsidiary cash into a portfolio of wholly-owned operating companies (BNSF, Berkshire Hathaway Energy, GEICO, See's, Precision Castparts, Clayton Homes) and marketable stocks — a structure now run by Greg Abel[42][33].

What Berkshire actually is

Berkshire Hathaway is the corporate shell of a 19th-century textile company that Warren Buffett took over in 1965 and repurposed as an investment holding company[42]. The textile operations are long gone; what remains is a parent company in Omaha that owns insurance businesses outright, controls scores of non-insurance operating subsidiaries, and holds a large portfolio of publicly traded stocks. At the end of 2025 it employed 387,815 people across 60-plus major subsidiaries and generated roughly $371B of revenue[33].

The throughline is capital allocation. Insurance subsidiaries collect premiums up front and pay claims later, leaving Berkshire with float — money it can invest in the interim. Operating subsidiaries send their excess cash to Omaha. Buffett (now Abel) reallocates that cash into whatever offers the best risk-adjusted return: buying whole companies, buying stocks, or — increasingly in recent years — holding Treasury bills[49][50].

A six-decade timeline

Founding milestones, defining acquisitions, and the 2023–2026 transition.

Dates and figures: Berkshire history and overview[42]; 2025 scale[33]; Munger's death[25]; the $1T milestone[31]; the Apple sale[18]; Buffett's final letter and the Abel transition[28][4][5].

Earnings come from four places at once: insurance (underwriting plus the investment income on float), the BNSF railroad, Berkshire Hathaway Energy, and ~60 manufacturing/service/retail businesses. In FY2025 no single bucket dominated — manufacturing/service/retail and insurance investment income were the two largest, at $13.6B and $12.5B[21][35].

Where the FY2025 operating earnings came from

After-tax operating earnings by segment, US$B (total $44.5B). “Other” is mostly parent-company investment income and currency effects. Note that insurance contributes twice — underwriting profit and the investment income earned on float. Hover a slice for its share.

The three-part engine

1) Insurance supplies the fuel. GEICO, General Re, Berkshire Hathaway Reinsurance and others collect premiums today and pay claims over years. The gap is float — about $176B at year-end 2025 — which Berkshire invests for its own account. When underwriting roughly breaks even or better, that float is effectively free leverage[2][49].

2) Operating businesses generate cash. The wholly-owned companies — railroad, utilities, and the manufacturing/service/retail group (Precision Castparts, Clayton Homes, See's, Dairy Queen, Pilot and dozens more) — are run autonomously and remit excess cash to Omaha rather than empire-building locally[34][50].

3) A tiny head office reallocates it. Berkshire runs with an unusually small corporate staff; subsidiary CEOs have wide autonomy, and headquarters' core job is deciding where the consolidated cash goes next — acquisitions, stocks, buybacks, or Treasuries[50]. This is the lever Abel now controls.

Float reached ~$176B at year-end 2025 (up ~$5B)[2]. But 2025 was a softer underwriting year: net underwriting earnings fell ~20% to $7.3B, GEICO's pre-tax underwriting profit slipped to $6.8B from $7.8B, and Q4 underwriting profit dropped 54%[20][21]. Cheap, growing float on one hand; a softening insurance cycle on the other.

How float works

An insurer is paid premiums now and pays claims later. The money it holds in between is float: it belongs to policyholders eventually, but in the meantime the insurer invests it. If underwriting at least breaks even, the float costs nothing — and Berkshire has grown it from a few hundred million dollars in the 1970s to about $176B today[49][2]. That pool is a major reason Berkshire could buy stocks and businesses at the scale it has.

The 2025 numbers

Berkshire's re/insurance operations earned $9.5B in pre-tax underwriting profit in 2025 (down from $11.4B), or $7.3B after tax — a ~20% decline[20]. GEICO, the auto insurer that is usually the underwriting workhorse, made $6.8B pre-tax versus $7.8B a year earlier[20]. In the property-casualty reinsurance business overseen by vice chairman Ajit Jain, Berkshire deliberately wrote less premium as competition pushed prices down, and Abel signaled it will keep doing so while the soft cycle lasts[22]. Separately, insurance investment income — the return on the float and the cash pile — contributed $12.5B of operating earnings, cushioning the underwriting dip[21].

Berkshire ended 2025 with about $373B in cash and Treasury bills — down from a $381.6B Q3 record but still the largest in its history[8]. It got there by selling, net, roughly $190B of stock across 2022–2024 (about $143B in 2024 alone)[53]. Bulls call it dry powder for the next downturn; bears call it a confession that Berkshire can't find enough worth buying[9].

The concentrated portfolio

Top disclosed equity holdings, US$B (Q4 2025 13F). The book is famously concentrated: a few large, cash-generative franchises dominate. Hover a bar for detail.

Beyond the top three, the Q4 2025 13F showed Berkshire roughly tripling its Alphabet stake, opening a new position in The New York Times (~$352M), cutting Amazon by ~77%, and adding to Chevron[17]. The Apple position — Berkshire's biggest — was trimmed ~4% in Q4 and remains the anchor at roughly $57.8–62B[16].

Why so much cash?

The hoard grew from ~$109B (end-2022) to ~$168B (end-2023) to ~$334B (end-2024) before peaking at $381.6B in Q3 2025[51][8]. Two forces drove it: Buffett sold stock — most dramatically ~two-thirds of Apple in 2024, which he tied partly to the prospect of higher future capital-gains taxes — and he found little to buy at prices he considered sensible[18][19]. The cash isn't idle: parked in Treasury bills, it earns billions in interest that flow into operating earnings. But it also earns far less than equities did over the same stretch[9].

What the market pays for the business ex-cash

A back-of-envelope walk (illustrative; simple subtraction over the cited figures). Take the ~$1.05T market value (June 2026)[31], strip out the ~$373B of cash and T-bills[8], and roughly $677B remains for everything else Berkshire owns. Set that against FY2025 operating earnings of $44.49B[1] minus the $12.5B of insurance investment income — the bucket where the interest on the cash and float lands[21] — leaving ~$32B from the operating businesses and underwriting. The market is paying about 21× the ex-cash earnings, and ~36 cents of every dollar of Berkshire's market value is currently a Treasury-bill portfolio. The split is approximate (investment income includes float-funded returns, not just T-bill interest), but the shape of the conclusion is robust: what Abel does with those 36 cents matters more to the next decade than anything the other businesses do at the margin.

Abel inherited the role with Buffett's blessing (“Greg will be the decider”) and moved fast: a $6.8B acquisition of homebuilder Taylor Morrison, a $10B Alphabet investment, and resumed buybacks — about $16.8B deployed in his first week[4][5]. But the transition also thinned the bench: Munger died in 2023, and investment manager Todd Combs and CFO Marc Hamburg both departed around the handover[25][26].

The new leadership map

Greg Abel — former CEO of Berkshire Hathaway Energy and vice chairman for non-insurance operations — is CEO, with authority over capital allocation[4][7]. Ajit Jain remains vice chairman for insurance. The investment portfolio, long run by Todd Combs and Ted Weschler, lost Combs in December 2025 when he left to advise JPMorgan's Jamie Dimon; longtime CFO Marc Hamburg retired after ~40 years[26][27]. Warren Buffett, 95, stays on as chairman and says he comes to the office but is “going quiet” — no longer writing the annual report or speaking at length at the meeting[28].

A different operating style?

Abel signaled he may depart from Buffett's famously hands-off model. On the Taylor Morrison deal he said he expects to “unify our site-built homebuilding operations into a combined platform” with Berkshire's existing Clayton Homes — exactly the kind of cross-subsidiary integration Buffett avoided[6]. Analysts read this as Abel's “strength as an operator” translating into more consolidation for efficiency than Berkshire saw under Buffett[47]. On current evidence the read leans toward sharpening, at low confidence: Buffett publicly praised the speed and smoothness of the first deal[6], and the consolidation targets efficiency in operations Buffett left alone[47] — but the cost, if any, will surface where it can't be quickly measured: in whether family sellers keep choosing Omaha without an auction[43].

The governance clock

The May 2026 annual meeting — the “Woodstock for Capitalists” — was the first in six decades not led by Buffett as CEO, a closely watched confidence test that Abel presided over[46].

Berkshire's durable advantages are structural: near-zero-cost float, permanent capital that never forces a sale, and a reputation Munger called “a seamless web of deserved trust” that wins deals without auctions[37][38]. The pressure on those edges: private equity has bid up acquisitions, and a cheap S&P index fund is a direct substitute for owning Berkshire as an investment[43][13].

Five Forces — Berkshire's capital-deployment arena

An unconventional framing: rather than one product market, we rate the forces on Berkshire's actual competitive game — sourcing and deploying capital. “High pressure” means a tougher environment for Berkshire. Click a force for the rated basis.

Positioning — permanence vs. operating involvement

Where Berkshire sits among the ways investors get diversified, professionally-allocated exposure. The x-axis is the permanence of the capital; the y-axis is how hands-on the owner is. Hover a point for the basis. Abel's signaled shift nudges Berkshire upward on involvement.

The two genuine moats

Permanent capital + float. Because Berkshire funds itself with shareholder equity and ~$176B of insurance float rather than redeemable funds or short-term debt, it can hold through panics and buy when others must sell — the source of several of its most-cited historical deals[49][37]. Reputation and culture. For a family owner who wants a permanent home and autonomy, “sell to Berkshire” has been a category of its own — a soft moat that let Buffett acquire without competitive auctions[43][38].

Two true facts in tension. Over 1965–2024 Berkshire compounded at 19.9% a year vs the S&P 500's 10.4% — turning $100 into a fortune[10][11]. But over 2015–2025 the index returned ~304% against Berkshire's ~234%, and 2025 (Buffett's last year as CEO) is on track to lag the market by double digits[13][12].

Operating earnings, FY2021–FY2025

After-tax operating earnings, US$B — Berkshire's preferred performance measure, which strips out volatile mark-to-market swings in the stock portfolio. FY2025 dipped ~6% on weaker insurance underwriting. (Net earnings, which include investment gains, are far more volatile: $66.97B in 2025 vs $88.99B in 2024.)

The 60-year scorecard

Compound annual gain, 1965–2024 — Berkshire per-share market value vs the S&P 500 with dividends, from Buffett's own performance table.

The recent reversal

Total return over the most recent decade, 2015–2025 (%). For the first time, the index out-ran Berkshire over a full ten years — the empirical core of the “too big to outperform” debate.

On valuation, Berkshire traded around 1.44× book value in June 2026, close to its ~1.43× twenty-year average and a touch below its 12-month average of ~1.51× — a modest premium, not an extreme one[32]. Book value per share, a rough proxy for intrinsic value, compounded at an estimated ~18.1% over 1965–2025 vs ~10.5% for the S&P 500[48].

What the price assumes

As of June 2026, a ~$1.05T market value[31] against $44.49B of FY2025 operating earnings[1] works out to roughly 23–24× trailing operating earnings (derived: market cap ÷ operating earnings; illustrative), while the ~1.44× book multiple sits almost exactly on its ~1.43× twenty-year average[32]. Read together, the market is pricing continuity: no post-Buffett discount, but also no credit for the ~$373B being deployed brilliantly. At its own historical-average multiple, the price assumes Berkshire keeps compounding book value at something like its recent pace — not its 1965-vintage ~18% pace[48] — while operating earnings, which fell ~6% in FY2025[1], stabilize. The closest structural peer, Markel, trades nearer ~14× forward earnings[39], though Berkshire's headline multiple is inflated by a low-yielding cash mountain the peers don't carry. That is the bar both cases are measured against: the bull needs deployment plus a re-rating off an average base; the bear needs the “Buffett premium” to have been larger than this very ordinary multiple admits[44].

On scale, nothing is close: Berkshire's ~$1.05T market value dwarfs Markel (~$28B), Fairfax (~$45B) and Loews (~$19B), the “mini-Berkshires” that copied the float-plus-conglomerate model[31][39]. But all of them — including Berkshire at ~1.4× book — trade near or only modestly above book, and the true benchmark, the S&P 500, beat Berkshire over the last decade[32][13].

Scale — Berkshire stands alone

Approximate market capitalization, US$B, early-to-mid 2026. The insurance-led conglomerates explicitly modeled on Berkshire remain an order of magnitude smaller. Hover for detail.

Reading the table

Sources: market caps and the $1T milestone[31]; Markel valuation and activist pressure[39][41]; Fairfax structure[40]; Berkshire's book multiple[32]; decade return gap[13]. Figures are point-in-time and approximate; valuations move daily.

The bear case is rarely about insolvency — Berkshire is famously over-capitalized. It is that the company is now too large to compound as it once did, sits on cash it struggles to deploy, is led by a new team without Buffett's record, and will gradually lose the concentrated voting control that buffered it[14][8][29].

The main risks

1) Size caps returns. Buffett himself wrote that “companies capable of moving the needle have been endlessly picked over,” and the 2015–2025 underperformance is the evidence[14][13]. A $1T company simply cannot find enough large, cheap opportunities to grow quickly.

2) Cash drag. A record ~$373B in Treasuries earns interest but lags equities; the longer it sits, the more visible the opportunity cost[8][9].

3) Key-man & execution risk. Munger's death, Combs' departure, the CFO's retirement and Buffett “going quiet” concentrate execution on a new team — and the “Buffett premium” in the stock is, by definition, exposed[25][26][28].

4) Governance. As Buffett donates ~$149B of stock, his ~30% voting bloc diffuses, potentially exposing the board to the activist pressure that conglomerates usually attract[29][30].

5) Valuation & substitution. At ~1.4–1.5× book, some analysts argue Berkshire is not cheap and could even warrant a discount to the sum of its parts, while a low-cost index fund is a cheaper way to own diversified equities[44][15].

Three scenarios to weigh (not predictions)

Scenarios are possibilities, not weights. Here is where the evidence actually sits today on each decisive question — the lean, the confidence, and the tripwires a reader can check.

The weighing

On whether Berkshire is now too big to outperform: the evidence leans yes (high confidence). The controlling evidence is Buffett's own admission that “companies capable of moving the needle have been endlessly picked over”[14] and the first full decade of trailing the index — ~234% vs ~304% over 2015–2025[13] — which outweighs the 60-year 19.9%-a-year record[10] because that record was compiled when Berkshire was a small fraction of its current ~$1.05T size[31]. The strongest surviving counter-argument: per-share book value still compounded at an estimated ~18.1% over 1965–2025 — the business keeps compounding even when the stock lags[48]. What would flip this reading: a rolling five-year total return back above the S&P 500 at the FY2028 annual report; or a negotiated acquisition large enough (above ~$50B) to show scale can still find scale. Pre-mortem: if this looks wrong in two years, the most likely reason is a market downturn that let the cash buy cheaply and flattered relative returns — or, on the other side, that the lean under-weighted how much of the index's decade was mega-cap tech Berkshire mostly doesn't own.

On whether the record cash is firepower or a problem: the evidence leans drag-so-far (medium confidence). The controlling evidence is the net sale of more than $190B of stock across 2022–2024[53] and Buffett's fruitless hunt for an “elephant” despite record liquidity[43], which outweighs the optionality case because an option left unexercised for three years while the index compounded carries a realized, measurable cost[13]. The strongest surviving counter-argument: Abel deployed $16.8B in his first week and resumed buybacks — the bottleneck may have been Buffett's late-career caution, not the structure[5][24]. What would flip this reading: cash below ~$300B at the FY2026 10-K with underwriting and operating earnings intact; conversely, cash back above ~$400B with buybacks paused again would harden it. Pre-mortem: if this looks wrong in two years, the most likely reason is a 2026–27 drawdown that let Abel run the classic Berkshire playbook at scale — or, on the other side, fast deployment at full prices (Taylor Morrison came at a ~24% premium[5]) that spent the patience without earning its reward.

On whether Berkshire can thrive without Buffett: the evidence leans toward the machine surviving the man (medium confidence). The controlling evidence is $44.5B of operating earnings produced by 60-plus businesses that do not depend on anyone's stock-picking[1][35] and Abel's decisive, Buffett-praised first weeks[5][6], which outweighs the thinned-bench worry because the operating engine, not portfolio genius, now generates most of the value. The strongest surviving counter-argument: the bench genuinely thinned — Munger's death, Combs' exit to JPMorgan, the CFO's retirement[25][26] — and the ~30% voting bloc that shielded the board dissolves as Buffett donates his shares[29][30], so the culture loses its author and its shield in the same decade. What would flip this reading: another principal departure (Jain or Weschler) before the 2027 annual meeting; or a disclosed activist stake as the voting bloc thins. Pre-mortem: if this looks wrong in two years, the most likely reason is that Abel's operator instincts quietly eroded the autonomy that made sellers choose Omaha[47] — or, on the other side, that the lean under-credited how much value a hands-on operator adds by tightening units Buffett left alone.

On whether the machine still has a moat: the evidence leans toward a real but narrower moat (contested confidence). The structural half holds: ~$176B of near-zero-cost float[2] and permanent capital that never forces a sale[37][49] are facts, not narratives, and they outweigh the index-substitution critique on durability because no fund-based rival can replicate them. What deadlocks the other half — deal-sourcing — is that Buffett himself could not land a large deal in his final years despite record liquidity[43], so the evidence cannot distinguish a structurally eroded edge (private-equity auctions, index funds beating Berkshire over 2015–2025[13]) from one that was simply unexercised. The strongest surviving counter-argument: the skeptics' case that the deal-sourcing reputation was personal to Buffett and departs with him[47]. What would flip this reading: a negotiated, non-auction acquisition coming to Omaha under Abel by end-2027 would re-arm the sourcing moat; float falling below $176B through the soft reinsurance cycle[22] would weaken the structural one. Pre-mortem: if this looks wrong in two years, the most likely reason is over-crediting float that a soft cycle quietly shrank — or, on the other side, under-crediting a reputation that turns out to attach to the institution rather than the man.

Each item is sourced in the section above it; see Sources for the full list.

Method

Research proceeded by fan-out web search and direct fetching of primary and reputable secondary sources. Every URL cited was opened and read during the run; each claim was transcribed into a structured manifest with a source tier, a confidence level and a stance. The load-bearing figures — FY2025 operating and net earnings, segment earnings, and the ~$176B float — come from Berkshire's own February 28, 2026 earnings news release[1][21][35][2]. The cash pile, portfolio holdings, market caps, valuation ratios and peer figures are point-in-time third-party figures, labeled as estimates where appropriate.

Frameworks used

The analysis applies the Pyramid Principle for the answer-first Executive Summary; Porter's Five Forces, framed unconventionally around Berkshire's capital-deployment arena rather than a single product market; a 2×2 positioning map of capital permanence vs. operating involvement; a peer benchmark against the “mini-Berkshire” conglomerates and the S&P 500; an even-handed SWOT (woven through the sections); and bull/base/bear scenarios in the forward view, presented as possibilities to weigh rather than a prediction. BCG, Ansoff and a formal value-chain were skipped — a holding company of 60+ disparate businesses doesn't yield to them honestly, and an empty framework is worse than none.

Disclosed vs. estimated

Disclosed, high-confidence figures — operating earnings ($44.49B), net earnings ($66.97B), segment earnings, float (~$176B) and the Kraft Heinz/Occidental impairments (~$8.26B) — are from Berkshire's reported results[1][45]. The cash figure (~$373B year-end; $381.6B Q3 record) is reported by reputable secondary sources from the 10-Q/10-K[8]. Multi-year cash and operating-earnings history (2021–2024) come from secondary compilations and are labeled[51][52]. Market capitalization (~$1.05T), the price-to-book ratio (~1.44×), peer figures and 13F portfolio values are third-party, point-in-time, and move daily[31][32][17]. The 60-year and decade return comparisons are from Buffett's performance table and market-data compilers[10][13].

Neutrality & independence

This is an analysis with shown work, not a recommendation. Every section pairs the case for and against with sourced evidence; the Executive Summary states where the evidence leans and at what confidence, and the forward view closes with an explicit weighing — controlling evidence, surviving counter-arguments, and the tripwires that would flip each reading — rather than a buy/sell call. The Teardown is independent and not affiliated with Berkshire Hathaway, and this is not investment advice — no rating, price target, or recommendation to buy or sell any security. The achieved evidence mix (see the Sources) spans supporting, critical and neutral citations by design.

Cross-checked at build time by an automated link checker. A few primary filings (SEC EDGAR) and some outlets bot-wall automated fetchers; where that happened the equivalent figures here were taken from Berkshire's own news release or annual report, which were fetched and read. See Methodology & Limits.