My favorite moments from the Berkshire Hathaway annual meeting Q&A are seldom the “set piece” answers, where Warren Buffett and Charlie Munger give well-rehearsed answers to questions you’ve heard before. Rather, it’s the the improvised statements that occur to them on the fly that usually stay with me. This year it was Mr. Munger who made an impression when he mentioned that he typically makes enough money on the float on his taxes to pay his entire tax bill.

He was most likely referring to personal taxes payable upon selling shares of Berkshire Hathaway that he’d held for decades, but the point applies equally well to Berkshire itself, which as of 12/31/2012 reported $44.494 billion of “Income taxes, principally deferred” on the liability side of its balance sheet. This mammoth figure is properly thought of as a source of leverage, as Berkshire itself explains in its owner’s manual:

“Berkshire has access to two low-cost, non-perilous sources of leverage that allow us to safely own far more assets than our equity capital alone would permit: deferred taxes and float…Both of these funding sources have grown rapidly…Deferred tax liabilities bear no interest…Neither item, of course, is equity; these are real liabilities. But they are liabilities without covenants or due dates attached to them. In effect, they give us the benefit of debt—an ability to have more assets working for us—but saddle us with none of its drawbacks.”

Put another way, the leverage provided by deferred tax liabilities allows a given increase in asset value—from an increase in the value of an equity security held by Berkshire, for instance—to produce a greater increase in equity value than it would without leverage.

And unlike most things at Berkshire, the advantage of deferred tax leverage gets better as the company grows. Consider a $10,000,000 equity investment made today, fully financed by equity. For this investment to produce a 15 percent increase in book value this coming year, the asset would have to appreciate by a little over 23 percent, assuming a capital gains tax rate of 35 percent. But if this asset were to compound at 15 percent annually for 25 years, as Coke has more or less done since Berkshire bought it, then in order to produce a 15 percent increase in book value the year after that, the asset would need only increase by about 15.25 percent. The difference between an asset that has to appreciate by 23 percent to get you where you want to go and one that only has to appreciate by 15.25 percent is more than the difference between the average return on the S&P and average return on T-bills.

This arithmetic (which I hope I have right) may go a long way towards explaining why Buffett bought Coke in the first place in 1988. Back then it was growing at 15 percent year after year while still paying out more than half of its earnings as dividends. He must have known that such growth could not persist forever, else Coca-Cola would have taken over the entire world economy (along with yoga instructors and hedge fund managers, of course). But he also knew that, come the day when Coke’s growth slowed down, the built-up deferred tax leverage would make up for some of the slowdown.