Effects of tax rules on dividend policy

Tyler Cowen asks some rather interesting questions concerning Microsoft’s decision to declare a $3 per share “special dividend” (since Microsoft has more than 10 billion shares outstanding, this translates into more than $32 billion in cash, thus representing the largest corporate dividend payment in history). Specifically, 1) would these dividends have happened without the Bush tax cuts, and 2) does Microsoft fear that Kerry will win and raise taxes on dividends? Professor Cowen’s answers to these questions are “maybe not” and “probably” respectively. The Wall Street Journal corroborates Professor Cowen by noting that “…the company was clearly concerned with the possibility that John Kerry might be elected President and carry out his promise to return dividends to their former status as ordinary income (thus raising the dividend tax back to the nearly 40% Clinton-era top rate from today’s 15%).

That dividend policy is sensitive to tax rules is empirically borne out by a new working paper authored by Raj Chetty and Emmanuel Saez entitled “Do Dividend Payments Respond to Taxes? Preliminary Evidence from the 2003 Dividend Tax Cut“. Chetty and Saez note “The individual income tax burden on dividends was lowered sharply in 2003 from a maximum rate of 35% to 15%, creating a unique opportunity to analyze the effects of dividend taxes on dividend payments by U.S. corporations.” They find, among other things, that 1) the fraction of publicly traded firms paying dividends began to increase in 2003 after having declined continuously for more than two decades, and 2) firms that were already paying dividends prior to 2003 raised their dividend payments significantly after the tax cut became law.

A long standing theorem in finance is that any time a firm’s shareholders can find more highly valued uses of capital than the firm, then excess cash should be returned to shareholders. Indeed, the Washington Post quotes Wharton finance professor Jeremy Siegel as saying that “Cash that’s just sitting around gets discounted”. However, this theorem implicitly assumes that there are no tax asymmetries. The most famous tax asymmetry in corporate finance is that debt related income is only taxed at the personal level, whereas equity related income is taxed at both the corporate and personal levels. At the margin, this tax asymmetry compels firms to be more highly leveraged than they otherwise might be, and also causes firms to avoid generating cash distributions for their shareholders. Another important tax asymmetry which existed until last year was that cash distributions through share buybacks were more tax-efficient transactions than cash distributions through dividend payments. While the 2003 dividend tax reform doesn’t address the double taxation issue, it does significantly reduce the tax penalty for cash distributions to shareholders. Furthermore, the tax code is now neutral about the form of equity-related cash distributions, whereas before it favored share buybacks over dividend payments.