Cato Institute
1000 Massachusetts Ave, NW
Washington, DC 20001-5403
Phone (202) 842 0200
Fax (202) 842 3490
 Contact Us
Support Cato

Cut the Dividend Tax

by Alan Reynolds

December 5, 2002

Alan Reynolds is a senior fellow with the Cato Institute.

A recent Wall Street Journal feature says, "The White House is leaning toward proposing a sharp reduction of taxes that individuals pay on corporate dividends." But the article also claimed that, "Treasury Secretary Paul O'Neill appears skeptical about a dividend-tax cut, in part because he worries about the cost and thinks other tax breaks might be more effective in stimulating economic growth." Those two worries could really be thought of as one -- namely, which tax changes give the most "bang for the buck."

Comparing the expected benefits of tax changes with the estimated revenue loss is certainly superior to the Business Roundtable's strategy of advertising the impact of a tax package by the amount of money it fails to collect. If the Rountable's $300 billion "stimulus package" really boosted economic growth (as parts of it would), then it wouldn't cost $300 billion. One of the most vital items in that package -- taking the top four tax rates down to 25 percent to 35 percent next year -- involves a static revenue loss of a mere $18 billion a year for three years. That easily passes the "bang for the buck" test.

As for the boldest new proposal -- slashing the tax on dividends -- Treasury Secretary O'Neill worries too much about grossly exaggerated estimated revenue losses, while the benefits are being too narrowly defined. We should never gauge the economic impact ("bang") of any incremental tax reform solely in terms of short-term economic performance ("stimulus"). If the government had paid more attention to the long run a few years ago, the recent wave of unsettling corporate bankruptcies need not have been as traumatic as it has been. Bad tax policies contributed to excessive corporate debt, and better policies today could reduce future systemic risks.

The costs to the Treasury of cutting the individual tax on dividends would be negligible in the longer run, much smaller than the official bean counters will ever admit. But the main reason is not the reason Treasury Secretary O'Neill has been hearing. "According to a White House study," says the Journal article, "eliminating the tax could boost economic activity enough to allow the government to recoup as much as 40 percent to 50 percent of the cost -- estimated at up to $25 billion a year." A figure of "up to $25 billion" would be small change for a government that collects about $2 trillion a year. But the amount that $25 billion might be reduced by a stronger economy is nonetheless a secondary issue if the initial estimate is itself wildly off base.

So far as I can tell, nobody is really expecting the government to totally eliminate the tax individuals pay on dividends. I would gladly settle for paying the same 20 percent tax on dividends as we pay on capital gains. And that simple change would involve very little loss of revenue, if any. The reason is simple: Corporations would have strong incentives to pay out more of their earnings as dividends to stockholders and much less as interest to bondholders. Since dividends would still be double-taxed, unlike interest, replacing interest with dividends increases the portion of corporate income subject to taxation. There will also be other revenue gains arising from (1) taxable individuals bidding dividend-paying stocks away from tax-exempt institutions, and (2) individuals switching from tax-deferred and tax-exempt investments into assets that pay dividends.

Unlike dividend payments, corporate interest payments are deductible. The more corporations rely on debt rather than equity (stock), the less tax they owe. But heavy reliance on debt increases the risk of bankruptcy. Massive bankruptcies have secondary effects on other companies and their workers that further shrink tax collections. Moreover, the revenues Treasury loses from huge corporate interest deductions are not matched by taxing interest received by individuals, because a large portion of that interest (and dividends) goes to tax-exempt institutions. Investors in high tax brackets invest in tax-exempt bonds and real estate or stock expected to generate capital gains.

Because dividends are not deductible to the corporation, dividends received should likewise not be taxable to individuals. Tax symmetry of that sort would make corporations indifferent between financing new investments with debt or equity. But corporations also finance investments with retained earnings, which is intended to result in capital gains. As long as reinvested earnings are double taxed by the capital gains tax, it makes sense to tax dividends at the same rate. It also makes sense to provide that relief at the individual level, because too many corporate deductions can leave them with no taxable income against which to write off the cost of buildings and machines.

Investors facing a 20 percent tax on dividends would pay more for stocks with a promise of paying good dividends. Higher stock prices, in turn, would make it more attractive for companies to finance new investments by selling stock rather than bonds (or borrowing from banks). In short, the ratio of debt to equity would fall. And that has two very important consequences.

First of all, corporations with less debt and more equity would become far less vulnerable to bankruptcy during cyclical downturns. Companies can't stop paying bondholders and banks just because business turns bad, but they can and do cut their dividends. Future recessions might well be milder and recoveries stronger, which would also be good news for government budgets.

Second, if tax rates on dividends were more reasonable, more corporations would pay them more generously and more investors would gladly pay taxes on them. Rather than shunning a personal tax of 37.5 percent on dividends, many investors would pay 20 percent on a greatly increased flow of dividends. Somebody ought to remind the folks who estimate tax receipts that 20 percent of something usually adds up to more than 35 percent of nothing. Then perhaps Secretary O'Neill could quit fretting about the IRS being shortchanged and instead focus on the enormous and lasting benefits.

This article originally appeared on Creators.com on December 5, 2002.