Berkshire Hathaway has paid a dividend exactly once in its modern history, in 1967, and Warren Buffett spent the following six decades joking that he must have been in the bathroom when the board voted. The company pays nothing today, and the question searchers are really asking — whether that changes now that Greg Abel runs the company — got an answer in 2026.
Abel wrote in his shareholder letter that Berkshire will not pay dividends so long as more than one dollar of market value for shareholders is reasonably likely to be created by each dollar of retained earnings, with the board reviewing the policy annually. That is Buffett’s formulation, restated verbatim by his successor. Verified August 2026.
The facts
| Item | Detail |
|---|---|
| Current dividend | None |
| Last dividend paid | 1967 |
| Stated policy (2026, Abel) | No dividend while >$1 of value is likely per $1 retained; reviewed annually |
| Cash and short-term investments | Approximately $370–400bn |
| Capital returned instead | Buybacks — the largest quarterly repurchase since 2021 under Abel |
The argument for paying nothing
The logic is a single test, and it is more rigorous than most dividend policies ever get.
This analysis sits within our best dividend ETFs hub.
A dollar of earnings can be handed to shareholders, who will pay tax on it and reinvest what remains at whatever return they can find. Or it can be retained and deployed by the company. Retention is correct if and only if the company can turn that dollar into more than a dollar of market value — which requires having somewhere genuinely good to put it. For most companies, most of the time, the honest answer is no, which is why most mature companies pay dividends.
Berkshire’s answer has been yes for sixty years because of an unusual structure: insurance operations that generate float — premiums held before claims are paid — deployed into wholly-owned operating businesses and a large public equity portfolio, with capital moved between them by a central allocator without the tax friction a shareholder would face doing the same thing. That arrangement compounded capital at rates that made retention obviously correct.
The tax point deserves emphasis for a European reader. A dividend is taxed on receipt; retained earnings that grow the share price are taxed only when you sell, and only on the gain. For an investor in a high-dividend-tax jurisdiction, retention is a genuine, quantifiable advantage rather than an ideological preference — it is the same argument the accumulating-versus-distributing calculator makes about fund share classes, applied at the company level.
Building a dividend income stream? Start with the free portfolio sheet.
The Dividend Portfolio Starter Sheet: yields, payout calendars and the withholding-tax checklist for EU investors.
The cash question, and why it now matters
The counter-argument has grown harder to dismiss as the cash pile grew. Roughly $370–400 billion sitting in cash and short-term Treasuries is not compounding at Berkshire’s historical rate — it is earning the policy rate, which any shareholder could earn themselves in a money-market fund without paying a conglomerate to hold it for them.
The retention test is explicit about this: retain only while a dollar retained is likely to become more than a dollar of value. Cash earning the risk-free rate does not obviously pass that test, and the size of the pile is itself evidence that the company has not found enough places to deploy it at attractive returns. That is the strongest version of the dividend case, and it does not depend on disliking the policy.
Berkshire’s actual answer has been buybacks rather than dividends, and under Abel the company executed its largest quarterly repurchase since 2021 alongside significant new equity investments. That is the intellectually consistent response: a buyback returns capital only when management judges the shares undervalued, whereas a dividend commits to returning capital regardless of price. It is a more discriminating tool, and it is taxed more favourably for most shareholders.
Whether a dividend eventually arrives is a genuine open question. The board reviews the policy annually, the cash keeps accumulating, and the arithmetic gets harder to defend each year. What can be said with confidence is that the 2026 letter did not signal a change, and that the new management chose buybacks and deployment over distribution when it had the opportunity to do otherwise.
What this means for an income investor
If you are building a portfolio to produce cash, Berkshire is not a holding for that job, and no amount of quality changes that. Its role, if any, is as a compounding equity position — a diversified conglomerate with an insurance engine and an unusual capital allocator, held for total return.
The synthetic-dividend approach works here and is worth stating properly, because it is the correct answer to “I want income from Berkshire.” Selling a small percentage of the position annually produces cash on exactly the same economics as a dividend, with two differences: you choose the timing and amount rather than the board, and in most European jurisdictions you are taxed on the gain portion of what you sell rather than on the full amount received. For a long-held position with a low cost basis that is worse than it sounds; for a recently-bought one it is considerably better than a dividend would have been.
The clean alternative: hold a dividend fund for the income sleeve and Berkshire, if you want it, in the growth sleeve. Trying to make one holding do both jobs is how income portfolios end up owning things that do not pay.
Reader scenarios
“Will Berkshire ever pay a dividend?” Possibly, and the cash pile makes it more plausible each year. The 2026 letter restated the existing policy without signalling a change, and the board reviews it annually.
“Is a buyback as good as a dividend?” Better in most European tax situations, and better in principle because it is only executed when management judges the shares cheap. Worse if you need predictable cash, because it produces none.
“Should I own it at all?” As a growth holding, it is a defensible single stock — a diversified conglomerate rather than a single business. The single-stock ceiling of 2–3% still applies, and it is already inside every US index fund you own.
Quick-start checklist
- Accept there is no dividend and that the policy was restated in 2026, not softened.
- Understand the retention test — it is a standard, not a preference.
- Note the cash pile is the strongest argument against the current policy.
- If you want income from it, sell a small slice annually rather than waiting.
- Keep the income sleeve and the growth sleeve as separate holdings.
Verdict
Berkshire pays no dividend because its stated test — retain only while a dollar becomes more than a dollar — has passed for sixty years, and Greg Abel restated that test unchanged in 2026 while returning capital through the largest buyback since 2021. The honest complication is the roughly $370–400 billion in cash earning the risk-free rate, which is precisely the condition the test says should trigger distribution. For a European income investor the practical conclusion is unambiguous: this is not an income holding, the tax treatment of retention genuinely favours you, and the correct way to extract cash from it is to sell a slice on your own schedule. Whether a dividend arrives is a live question — and it will be a signal about opportunity, not about generosity.
What Berkshire actually is now
Part of what makes the dividend question interesting is that Berkshire is no longer the thing most people picture. It is not a fund of public stocks with a famous manager attached; it is a conglomerate whose wholly-owned operating businesses generate the majority of its earnings.
The insurance operations sit at the centre — underwriting that has been consistently profitable, generating float that is effectively borrowed money at negative cost. Around them sit a railroad, a large regulated utility group, manufacturing, energy and retail businesses, and a public equity portfolio that is large in absolute terms and a shrinking share of the total.
That structure is why the retention argument held for so long: capital could be moved from a business generating more cash than it could reinvest into one that needed it, internally, without the shareholder-level tax that would apply if the money went out as a dividend and came back as a purchase. It is a tax-efficient internal capital market, and it is genuinely difficult to replicate at the individual investor level.
It is also why the cash pile is a harder problem now than it was thirty years ago. A company of this size needs very large opportunities to move the needle, and very large opportunities at attractive prices are rare. The scale that made Berkshire formidable is the same scale that makes deploying incremental capital difficult, and that tension is the honest core of the dividend debate.
Buybacks versus dividends, worked through
The choice between the two is not a matter of preference, and the arithmetic is worth stating for a European reader because the conclusion differs from the American one.
A dividend pays every shareholder the same amount per share regardless of price, and is taxed on receipt. In Germany, France, or most EU jurisdictions that means an immediate deduction of 25–30% before the investor has done anything.
A buyback reduces the share count, raising every remaining holder’s proportional ownership. No tax event occurs for the holder who does nothing. The holder who wants cash sells a slice and is taxed only on the gain portion — which for a position held at a profit is a fraction of the amount received.
The buyback is therefore more tax-efficient for a European holder in almost every case, with one important condition: it only creates value when executed below intrinsic value. A company buying back overpriced shares is destroying value for continuing holders, which is why Berkshire’s policy explicitly conditions repurchases on management’s judgement of price. That condition is what separates a disciplined buyback from the value-destroying programmes common elsewhere.
The general principle extends beyond Berkshire and matters for fund selection too: it is the same logic that makes accumulating share classes preferable during the building years. Deferred tax on gains beats immediate tax on income, unless a national allowance makes the income free.
Frequently asked questions
Does Berkshire Hathaway pay a dividend?
No. The company has paid one dividend in its modern history, in 1967, and pays nothing today.
Has the policy changed under Greg Abel?
No — his 2026 shareholder letter restated Buffett’s test: no dividends while more than a dollar of market value is likely to be created per dollar retained, reviewed annually by the board.
Why does Berkshire hold so much cash?
Approximately $370–400bn accumulated because attractive deployment opportunities were scarce. It is the strongest argument that the retention test is no longer clearly being passed.
How does Berkshire return capital?
Through share buybacks, executed when management judges the shares undervalued — including the largest quarterly repurchase since 2021 under Abel — rather than through committed distributions.
Is a buyback better than a dividend for me?
In most European tax systems yes, because you are taxed on gains when you choose to sell rather than on income when the board chooses to pay. It produces no cash without action on your part.
How do I get income from a Berkshire holding?
Sell a small percentage annually. The economics match a dividend, you control the timing and amount, and in most jurisdictions only the gain portion is taxed.
What is float and why does it matter?
Insurance premiums held before claims are paid — effectively borrowed money that has cost Berkshire less than nothing, and the engine behind decades of compounding.
Is Berkshire mostly a stock portfolio?
No longer. The wholly-owned operating businesses — insurance, railroad, utilities, manufacturing — generate the majority of earnings, with the public portfolio a shrinking share.
Why is the cash pile a problem?
Cash earning the risk-free rate does not obviously pass the company’s own retention test, and its size is evidence that attractive deployment opportunities have been scarce.
Are buybacks better than dividends for Europeans?
In almost every EU tax system, yes — no tax event for holders who do nothing, and only the gain portion taxed for those who sell a slice.
When does a buyback destroy value?
When shares are repurchased above intrinsic value. Berkshire’s policy explicitly conditions repurchases on management’s price judgement, which is what separates it from many corporate programmes.
Would a dividend be a bad sign?
It would signal that management no longer expects to create more than a dollar of value per dollar retained — an admission about opportunity, not an act of generosity.
Cited sources: Berkshire Hathaway shareholder letters, SEC filings, and The dividend ETF pillar. Verify current fund figures before investing. Educational content, not financial advice.
European UCITS/ETF investor with 11+ years building a global dividend and index portfolio, and founder of Financial Expert Class. I write every guide here from first-hand experience investing across EU brokers and tax regimes — not theory. More about me →