Ever since Warren Buffett observed that his tax rate is unfairly low, President Obama keeps finding ways to get mileage out of his friend's lament. One of the president's recent proposals is to dramatically increase the tax rate on corporate dividends for upper-income shareholders.
It may seem that Buffett has no dog in this fight, since the company he runs, Berkshire Hathaway, pays no dividends anyway.
But that's not the whole story. Before 2003, when dividends were taxed at much higher rates, companies like Berkshire had a handy rationale for holding on to their shareholders' money indefinitely, claiming this was the most tax-efficient investment policy.
Lowering the dividend tax to 15% weakened that argument, and since then public companies started paying more dividends and investors have valued them more highly. Berkshire is one of a few megacorporations that still pay no dividends at all; since its chairman is a financial superstar, the company has become Exhibit A for the claim that dividends really don't matter.
But what if Berkshire shareholders actually would be better off if the company paid dividends? Since Buffett has placed himself squarely on the side of a tax policy that favors his own view of the matter, that question is of interest not only to Berkshire's shareholders but to Congress and the voters.
With that in mind, let's take a look at the Sage of Omaha's latest annual letter to his shareholders. As usual, it leads off with his favorite metric — a chart showing Berkshire's annual increase in book value dating from the present company's beginning in 1965.
Berkshire's record over the entire period is legendary, and even for the past ten years — well after the company had become too large to sustain its early rapid growth — the average increase in its book value still exceeded the market performance of the S&P 500.
But what if we focus on Berkshire's own market value rather than its book value?
Especially for companies that don't pay dividends, this is surely an equally important figure, since the only way shareholders can realize tangible gain is by selling shares. So, let's suppose that in 2003 (the year the dividend tax was lowered), a buy-and-hold investor had purchased the stocks that Berkshire buys for itself — all dividend payers, incidentally — and continued to match Berkshire's portfolio over time as its components were publicly reported.
Despite the fact that those purchases would have considerably lagged Berkshire's own buys at presumably better prices, the investor's total return would now be considerably higher than if he'd bought Berkshire Hathaway itself.
But haven't Berkshire investors relied on Buffett's astuteness in picking that basket of stocks in the first place? And haven't they also relied on Buffett's ability to buy entire companies and convert them to Berkshire subsidiaries, and on Buffett's skillful management of the core business of property and casualty insurance?
The answer is yes to all three questions, and for many years Buffett's expertise earned a market premium for Berkshire stock.
Lately, however, that hasn't been true — the S&P 500 far outperformed Berkshire over the past three years, even if we don't count (as we should) an S&P index fund's reinvested dividends.
Berkshire's recent underperformance isn't because the Oracle of Omaha has lost his touch, but because the (always forward-looking) market knows he can't go on forever, a point which Buffett often makes himself. Nor does the market believe that Buffett's investment skill translates into successor-picking skill. History offers few examples where such efforts work as planned. Last year, one possible successor — David Sokol — became an embarrassment.
So, if Berkshire's no-dividend policy has dampened the price of its stock, can that be reversed by using an alternative method of returning cash to shareholders: the corporate share buyback? The selling shareholders get the cash they seek and, since fewer shares are outstanding, the remaining shareholders end up with a greater proportion of the equity.
Like other companies whose stock is disfavored in the market, Berkshire did indeed embark on a share repurchase program. The company was prepared to repurchase shares at up to 110% of book value, and often there were opportunities to buy the stock even cheaper.
However, as Buffett writes in his current letter to shareholders, "I have mixed emotions when Berkshire shares sell well below intrinsic value. We like making money for continuing shareholders... We don't enjoy cashing out partners at a discount."
That's a fair summary of why share repurchases aren't such a good substitute for regular dividends. Done shrewdly, repurchases favor the remaining shareholders over the departing ones; done carelessly, the reverse is true.
Buffett writes at length of the admirable share repurchase program of IBM, a stock in Berkshire's portfolio. The market undervalued IBM for years, making its share buybacks that much more valuable for the remaining stockholders.
That unrealistically low price didn't displease Buffett at all, since he was under no pressure to sell at those prices.
In fact, Buffett is under no pressure to sell ever. Again, his shareholders' letter: "More than 98% of my net worth is in Berkshire stock, all which will go to various philanthropies." That makes his investment horizon infinite.
With no desire for current income from any Berkshire shares, how well does he relate to shareholders whose needs are more typical?
Buffett has said that he doesn't think he should be paying a lower tax rate than his secretary, but apparently he also doesn't think that his secretary, once retired, should live off the income from Berkshire stock.
So now Obama proposes a tax rate on dividends that would be vastly different for Buffett and for his secretary — if Berkshire ever did pay dividends.
The trouble is that the new tax rate would likely have a greater effect on whether dividends get distributed at all than on how they might be taxed if such distributions occurred.