Do Dividend Payments Respond to Tax Cuts? Does it Really Matter?

American Council for Capital Formation
August 2004

DR. MARGO THORNING
SENIOR VICE PRESIDENT AND CHIEF ECONOMIST
AMERICAN COUNCIL FOR CAPITAL FORMATION*

This special report describes the findings of two new studies on the impact of the 2003 tax reduction on dividends received by individuals. These two studies, “Do Dividend Payments Respond to Taxes?: Preliminary Evidence from the 2003 Dividend Tax Cut,” by Raj Chetty and Emmanuel Saez (NBER Working Paper No. 10572), and “Dividend Tax Cuts Deemed Effective,” by Frank A. Fernandez (SIA Research Reports, Vol. V, No. 8, August 2, 2004), measure the corporate response to the reduced dividend tax rates in the Jobs and Growth Tax Relief Reconciliation Act of 2003. The studies suggest that corporations have increased dividend payments in response to the cut in the dividend tax rate. The importance of the phenomenon is that it should stimulate more saving and investment.

DIVIDEND TAX CUTS OF 2003

One of the key provisions of the Jobs and Growth Tax Relief Reconciliation Act of 2003 (hereafter, the 2003 tax reform) was to reduce the individual tax on dividend income substantially. As authors Raj Chetty and Emmanuel Saez note, historically, dividend income has been considered ordinary income for federal individual income tax purposes and hence taxed according to the regular, progressive income tax schedule. The 2003 tax reform introduced favored treatment for dividends, starting retroactively at the beginning of 2003, whereby dividends are taxed at a rate of 15 percent instead of facing the regular income tax schedule with a top rate of 35 percent.¹ This tax change effectively gave to dividend income the same favorable tax treatment accorded to individual capital gains income (see figure 1).  Conservatives argued that such a change would greatly reduce the tax disadvantage of dividends and hence induce firms to pay more  dividends, ultimately fostering more investment and business activity. Liberals countered that such a tax cut would benefit mostly wealthy taxpayers who own a disproportionate share of total dividends paid out to individuals in non-tax favored accounts, with little or no economic benefits for the wide majority of American families.

The effects of dividend taxes on dividend payments and the investment behavior of corporations have generated much interest in the public economics literature.  However, the academic debate on this topic remains controversial. The “old view” on this issue, implicit among supporters of the 2003 dividend cut, says that dividend taxes reduce the net return on investment and hence reduce the supply of investment. Therefore, when taxes on dividends are cut, individuals are more willing to save and invest their money in stocks, spurring business investment, profits, and dividend distributions.   MIT Professor James Poterba and Harvard Professor Lawrence Summers (1985), using time series evidence from the United Kingdom, found that consistent with this view, dividend payments and investment were higher when the tax on dividends was lower. More recently, Poterba (2004) uses U.S. time series data since 1929 and finds a negative association between dividend payments and the dividend tax rates relative to taxes on capital gains. In contrast, Chetty and Saez note that the new view on dividend taxation, implicit among critics of the 2003 tax reforms, assumes that marginal investments are entirely financed by retained earnings rather than new shares issues. Under this assumption, the tax on dividends paid out does not affect investment decisions of firms, and profits and dividends paid out should not change either.  In this case, the dividend tax cut is irrelevant for corporate decisions and simply benefits individual investors by reducing their tax burden. As dividend income is very concentrated at the top of the income and wealth distributions, a tax cut on dividends should provide a tax break to the wealthy without expansion of investment and business activity.

The large change in the tax treatment of dividends in 2003 offers a unique opportunity to cast light on this debate, Chetty and Saez state. The authors use Center for Research on Security Prices (CRSP) data on dividend payments available through the first quarter of 2004 to test whether dividend payout policies changed significantly after the 2003 tax reform. Total regular dividends have surged by nearly 20 percent since the beginning of 2003, the point at which the lower tax rate was first proposed and ultimately retroactively applied. However, since the aggregate dividends series is very volatile and driven by unusually large payouts, the authors noted that it is difficult to make statistically robust inferences about the effects of the tax change without disaggregating the data further. They therefore divide the analysis of the response to the tax reform into three margins: (1) the extensive margin (initiations and terminations of regular dividend payments); (2) the intensive margin (increases or decreases in payment amounts by firms already paying); and (3) special dividends (one time distributions).

FIGURE 1: MAXIMUM FEDERAL TAX RATES ON DIVIDENDS AND LONG TERM INDIVIDUAL CAPITAL GAINS

IMPACT OF 2003 TAX REFORM

The Chetty/Saez analysis shows the fraction of firms paying regular (monthly, quarterly, semi-annual, or annual) dividends started to increase precisely in 2003, after falling continuously for more than two decades. The fraction of regular dividend payers fell from approximately 60 percent in 1980 to a low of 20 percent in the fourth quarter of 2002, but has rebounded to nearly 25 percent in the past year. The fact that the decline in the fraction of dividend payers stops precisely in 2003 constitutes strong evidence that the 2003 tax reform induced more firms to start paying regular dividends. Of the 3,813 firms that are listed in the 2004 CRSP, 113 initiated regular dividends in 2003, in comparison with 21 in 2002, and an average of 22 in prior years. There is also evidence of dividend increases on the intensive margin. The authors examined the number of firms increasing (or decreasing) their payments by a significant margin, such as 10 percent or 20 percent. Regardless of the cutoff, they chose, Chetty and Saez find strong, robust evidence that firms were more likely to increase their regular payments after the reform. Again, these increases are widespread, occurring across all sizes of firms.

They found special (i.e., one time, non-recurring) dividends also increased following the 2003 tax reform. Among firms that were active in the 2004 CRSP data, 69 made special payments in 2003, compared with 35 in 2002 and an average of 66 in prior years. The total amount of special dividends paid in 2003, $1.9 billion, greatly exceeded the $520 million paid in 2002 and the prior years average of $696 million.

Corroborating the NBER results described above is an analysis by Frank Fernandez of the Securities Industry Association. Dr. Fernandez states that the number of companies that have initiated or increased dividends has grown since the middle of last year, the date that the dividend tax cut became effective. Of the approximately 7,000 publicly owned companies that report dividends to Standard & Poor’s Dividend Record, 895 reported dividend increases during the first half of 2004, a 13.6 percent increase compared to the same period in 2003, and a 19.8 percent increase from the same period of 2002. This is in addition to an increase in the number of companies that have initiated dividends and a decline in the number that halted or reduced dividends. Another study, using a smaller sample, found that 12.7 percent of 807 firms paid dividends for the first time in 2003, compared with on average only 2.8 percent of non-dividend paying firms that initiated dividends during the prior 10-year period.  For dividend-paying firms, 41 percent increased dividends in 2003, compared to an average of 35 percent in earlier years.

Fernandez concludes, as do Chetty and Saez, that most of the data supports the “traditional view.” The traditional view holds that the amount of dividends paid will rise as the tax burden on dividends relative to capital gains is decreased, and that lowering the dividend tax rate will increase the dividend payout ratio and incentives for real investment, ultimately fostering more investment and business activity.

WHY IT MATTERS: IMPACT OF LOWER TAXES ON DIVIDEND
AND OTHER CAPITAL INCOME ON ECONOMIC GROWTH

The NBER and SIA analyzes of the corporate response to tax cuts on dividends do not estimate the impact of these cuts on investment. However, Fernandez notes that the conclusion reached by the Treasury 12 years ago² still holds true: the long-term benefits derived from eliminating biases and distortions is roughly comparable to the costs generated by lost tax revenues and resultant higher fiscal deficits. Fernandez reasons that if the presumed long-term benefits of higher growth in incomes and jobs are considered, the balance tips well in favor of the dividend tax cuts A study by Dr. Allen Sinai, chief global economist, Decision Economics, Inc., prepared for the American Council for Capital Formation that estimated the impact of completely eliminating the tax on dividends and interest income by switching to a consumed income tax supports Fernandez’s conclusion. According to the Sinai study, if the U.S. had had such a tax system from 1991 to 2004, GDP, business capital spending, employment, and productivity growth would have been larger (see Table 1). The 2003 dividend rate cut was a step in the right direction in terms of reducing U.S. reliance on the taxation of capital as a revenue source.

CONCLUSION

Pressure is expected to rise on the next Congress to act to curb the sharp expansion of the federal deficit, and one likely target would be to reverse the dividend tax cut well before its scheduled “sunset” at end-2008. Such an outcome could have negative impacts on U.S. saving and investment. As Fernandez notes, to allow this pro-investor initiative to sunset would adversely affect the economy’s long-term outlook and reduce the attractiveness of the equity investments that help fuel innovation, economic growth and prosperity. The sunset provision on dividend tax-rate cuts—if accelerated—may exaggerate a decline in equity values and sharply reduce investor interest in shares and, perhaps, share prices as well in the months prior to the sunset.

Table 1: Economic Impact on the United States of Switching to a Consumption Tax in 1991
Expensing business investment, removal of the business and personal interest deduction,
and tax exemption of savings

Notes

1. More precisely, taxpayers in the bottom two income tax brackets (facing a regular marginal tax rate of 10% or 15%) face a new dividend tax rate of only 5%, while taxpayers in the top four brackets (facing marginal tax rates of 25, 28, 33, or 35%) face a new dividend tax rate of 15%. Taxpayers on the Alternative Minimum Tax schedule (flat rate of 28%) benefit as well from the reduced 15% tax rate on their dividend income.

2. Report of the U.S. Treasury Department, Integration of the Individual and Corporate Tax Systems, January 1992. See specifically Part V: Economic Analysis of Integration, Chapter 13: Economic Effects of Integration.