Warren Buffett's annual letter to shareholders, edited by the great Carol Loomis, is a yearly masterclass in how to think clearly. It's also highly educational: It has certainly informed my own thinking on many subjects, and nearly all of the time I end up agreeing with Buffett.
This year, Buffett has a longish section on the subject of share buybacks (check out page 7 of the PDF) which none other than Matt Levine has described as "straightforward and reasonable". So, if you want to know what to think about share buybacks, you should just ask Warren, right?
I'm not so sure - partly because Buffett has two weird idiosyncrasies which make him unreliable on this subject.
Before we get to Warren, however, it's important to set the parameters of the debate, which is never easy. The problem is that the three most common answers (Buybacks Good, Buybacks Bad, and Buybacks Good Only If The Share Price Is Low) tend to be answers to three different questions.
Buybacks Good is normally the answer to the question "If you were a shareholder, would you want your company to return excess profits via dividends or via buybacks?"
Buybacks Bad is the answer to the question "From an economic perspective, is it better to have companies reinvest their profits in growth, or is it better for them to spend that money buying back their own shares to boost their share price?"
And Warren Buffett's preference, Buybacks Good Only If The Share Price Is Low, is the answer to the question "If you are a long-term buy-and-hold shareholder, do you want your company to do buybacks?"
In other words, Buybacks Good and Buybacks Bad are making diametrically opposed assumptions. Buybacks Good is assuming that the company has excess profits which can't be productively reinvested to return something higher than the company's cost of equity; Buybacks Bad is assuming precisely that those excess profits can and should be productively reinvested back into the company. The disagreement between the two sides, then, really has nothing at all to do with buybacks, and everything to do with their priors on the question of whether corporations in general are capital-constrained, and what they might be able to achieve given more access to money.
Buffett is clear on this last question. "Both American corporations and private investors are today awash in funds looking to be sensibly deployed," he writes, adding that "I'm not aware of any enticing project that in recent years has died for lack of capital." (And yes, he's looking: he's in charge of $86 billion of cash just sitting there waiting to be spent.)
I'm inclined to agree with Buffett on this point. When people complain about, say, Apple (NASDAQ:AAPL) spending too much money on buybacks, are they really saying that Apple would be more innovative if only it spent more money on R&D?
The fact is that at most companies (and especially at pharmaceutical companies, where you tend to find the highest R&D budgets), there's really no evidence at all that marginal extra R&D spending gives you any marginal benefit. People love to look at R&D spending as a percentage of revenues, but the fact is that absolute numbers are important: Apple, for instance, is good at spending a certain amount of money on R&D, and finds it difficult to find attractive areas beyond that amount.
In an environment where there's more than enough capital to go round, the Buybacks Bad crew have, in general, done a pretty bad job of making the case that companies would be likely to create significant extra economic growth were they to hold on to the money and try to reinvest it, rather than returning it to shareholders. The law of diminishing marginal returns is a strong one, and there's always some point at which those returns fall below your cost of equity. It seems reasonable to assume that most companies are past that point right now.
What's more, the whole point of buying a stock, whether it's on the primary market or the secondary market, is that you hope to get back more money in the future than you're investing in the present. A company which never returns any money to shareholders is a company which, on a basic DCF calculation, has no value at all.
But this is the first place where Warren Buffett has a very unusual dog in this fight: Berkshire Hathaway (NYSE:BRK.A) (NYSE:BRK.B) has paid just one dividend in its entire existence (10 cents in 1967), and has only done the tiniest of stock buybacks. As a huge industrial concern with massive profits and cashflows, it's pretty much unique in keeping all its profits to itself.
Buffett can get away with this kind of behavior, partly because he has unquestioned control over Berkshire Hathaway, and partly because he's a legendary investor: his shareholders stick with him precisely because they think that he can make better capital-allocation decisions than they can. But that's true of very, very few other companies. Let's say you were a shareholder of, say, a profitable coal mining company in the 1950s, rather than a profitable textile manufacturer like Berkshire Hathaway used to be. Warren Buffett managed to close down the textile factories and pivot into lots of other businesses, but in the case of the coal mines, you would have wanted to be dividended the profits, so that you could invest them elsewhere, rather than just see your investment go slowly to zero.
It's important, then, to understand that behind his skepticism of buybacks, Buffett has an unwarranted skepticism of returning money to shareholders at all. That's just silly. At any given point in time, some companies are and should be growing, while other companies are and should be shrinking. The latter group has no good reason not to give their profits to their shareholders, and indeed the former group should do so too, insofar as they've already spent all the money they can usefully reinvest into their businesses.
Or consider a company like Apple, which grows enormously and eventually becomes mature and very valuable. There are good tax and other reasons why it might make sense to have a certain amount of debt on its balance sheet, but it never needed to borrow much money (or, when it did need to borrow money, it couldn't do so.) So it basically just does a debt-for-equity swap, issuing new bonds and buying back shares. That's a corporate finance transaction, and companies should at the very least have the option of doing such things, especially when interest rates are low and debt is cheap.
But Buffett never talks about interest rates when he discusses share buybacks; instead, he obsesses about "intrinsic value". Here's his logic:
Consider a simple analogy: If there are three equal partners in a business worth $3,000 and one is bought out by the partnership for $900, each of the remaining partners realizes an immediate gain of $50. If the exiting partner is paid $1,100, however, the continuing partners each suffer a loss of $50. The same math applies with corporations and their shareholders. Ergo, the question of whether a repurchase action is value-enhancing or value-destroying for continuing shareholders is entirely purchase-price dependent.
This makes sense right until you realize that when you're talking about a public company, the valuation is the purchase price. If an exiting partner gets paid $1,100, that's because her shares are worth $1,100 on the open market. If the other shareholders think that the company's "intrinsic value" is less than $1,100, then they shouldn't be holding those shares!
Buffett knows this, of course: he's an investor to his toenails. But he has another agenda, which he reveals later on in the letter:
Berkshire, like most corporations, nets considerably more from a dollar of dividends than it reaps from a dollar of capital gains. That will probably surprise those of our shareholders who are accustomed to thinking of capital gains as the route to tax-favored returns.
Most corporations, of course, don't own enormous baskets of dividend-paying stocks. But what Buffett is saying here is that he's the foremost member of one of the smallest clubs in the world: individuals who pay lower taxes on dividends than they do when companies do share buybacks.
For the overwhelming majority of buy-and-hold investors, on the other hand, buybacks are clearly better than dividends for tax purposes. Instead of making you richer by paying you taxable income, they make you richer by increasing the value of your shares - and you don't need to pay tax on that increase in wealth unless and until you sell those shares.
So I'm not entirely sure that I buy Buffett's insistence that companies should only do buybacks if they have a clear idea of their "intrinsic value". Sometimes, returning money to shareholders is the right thing to do: it's a crucial part of the market's capital-allocation process. When companies do it, they have a choice between dividends and buybacks. And shareholders in general prefer buybacks to dividends. If that's what they want, then give them buybacks! Especially because setting a dividend creates expectations for the future which might be hard to meet.
I don't think that buybacks are always good. If a company has ample opportunity to invest in growth, then it should do that. But if it's shrinking, or if there are other good reasons to return money to investors, then buybacks should nearly always be an option. Even if that doesn't make Warren Buffett quite as happy as a dividend would.