Introduction
Mr. Chairman and members of the committee, it is a great privilege to have the opportunity to appear before you today. My name is Eric Engen. I am a resident scholar at the American Enterprise Institute in Washington, D.C. where my research focuses on the effects of tax and budget policy on the economy. Prior to joining AEI, I was a senior economist and section chief at the Federal Reserve Board of Governors.
My testimony provides perspectives on some of the tax policy reforms discussed in the Economic Report of the President and proposed in the President’s recent budget.[1]
In particular, I focus on the investment and saving incentives provided by the tax relief for corporate earnings and the tax-free saving accounts.
My principal conclusions are as follows:
The taxation of capital income, sometimes at very high marginal rates, in the U.S. tax system stands in marked contrast to the implications of optimal tax theory in the economics literature, which has generally concluded that the optimal tax on capital income is zero. Capital taxes reduce saving and investment, and a smaller capital stock reduces the productivity and wages of workers.
The proposal to remove the "double taxation" of corporate earnings would lower the cost of investment for firms and increase the after-tax returns to savers that hold corporate equity, thus stimulating capital formation, boosting the productivity of workers, and raising wages and income.
Exempting corporate profits from personal income taxation reduces the tax incentives for corporations to retain earnings instead of paying dividends. Higher dividend payouts would help improve the allocation of corporate capital and assist stockholders in monitoring corporate managers.
Exempting corporate profits from personal income taxation reduces the tax incentives for corporations to finance investments with debt instead of equity. Less corporate debt reduces the probabilities of default and bankruptcy in an economic slowdown and thus would lower the risk premium included in the cost of financing corporate capital.
The United States has the second highest corporate tax rate among its economic competitors and is one of only three countries in the OECD that does not provide dividend tax relief. Although the corporate tax rate would still remain relatively high, eliminating the double taxation of dividends would improve the competitiveness of U.S. firms in the global economy.
The President’s proposals for expanding tax-free savings accounts would continue the trend seen for more than twenty years of moving the personal income tax towards a consumption tax base.
The Retirement Savings Accounts (RSAs) would be the most likely to increase personal saving as it would significantly increase the contribution limits associated with Roth IRAs, which RSAs would replace. The higher the contribution limits for these accounts then the greater the economic incentives for households to increase saving.
The Employer Retirement Savings Accounts (ERSAs) typically do not increase the contribution limits that employees will receive from similar types of current plans-401(k) and 403(b) accounts, for example-and thus would not be expected to increase marginal saving incentives more than current plans. However, the simplification and lower compliance costs of ERSAs should increase the availability of employer-based retirement accounts, particularly for small-business employees.
The effects of Lifetime Savings Accounts (LSAs) on aggregate personal saving are harder to estimate and more likely to be mixed, although they would probably be the most popular owing to their lack of withdrawal restrictions. Particularly in early years after the introduction of LSAs, households would have the incentive to merely shift assets from currently taxed bank accounts and mutual funds into an LSA. For lower- and middle-income households, eventually existing assets would be exhausted and marginal incentives to increase saving would start to become effective. For higher-income households, this process would take longer.
Background
Capital income, which reflects the returns to saving and investment, can face substantial rates of taxation, particularly if generated by corporate businesses. The federal corporate income tax rate for most corporations is currently 35 percent, and state corporate taxes add, on average, another 4 to 5 percent to the effective corporate tax rate. When corporate income is delivered to shareholders, it then often faces combined federal and state personal income tax rates on dividends that can exceed 40 percent. Thus, the overall marginal tax rate on distributed corporate income can easily be over 60 percent.[2] Even if corporate earnings are retained but ultimately dispersed to shareholders through the redemption of stocks that give rise to capital gains, which are typically taxed at a 20 percent rate in the personal income tax, the tax bite on the return from investment in corporate capital is still quite sizable. The high rates of taxation on capital income in the United States stand in marked contrast to the implications of optimal tax theory in the economics literature. Numerous economic studies have concluded that an optimal tax system in most scenarios will not include a tax on capital.[3] This conclusion reflects the highly distortionary effects of capital income taxes over long time periods-a distortion that "explodes" or "compounds" even with a small capital income tax.
Economic growth and a higher standard of living in the United States are ultimately achieved by increasing the productivity of U.S. workers. Increased productivity requires investment, which is funded by saving. Thus, the economic burden of capital income taxes is not just born by high-income capital owners. The lower level of capital accumulation that results from high capital income taxes also has adverse effects for workers. Less capital makes workers less productive. If worker productivity is lower, then wages are lower.[4] Therefore, even if workers owned no capital-a description that is increasingly less appropriate for households in the United States, more than half of whom own corporate stocks-then workers would still be better off with no tax on capital because their wages would be higher.[5]
When compared to our primary economic competitors, such as countries in the OECD, the United States has a relatively high corporate income tax rate and, unlike most of these competitors, does not provide relief for the "double taxation" of corporate income.[6] The combined U.S. federal and local corporate income tax rate is almost 40 percent, second only to Japan, while the average corporate income tax rate for other OECD countries is closer to 30 percent. Moreover, the United States is one of only three OECD countries that do not have provisions in its tax code for some relief from the double layer of taxation of corporate dividends. Switzerland and Ireland do not provide dividend tax relief but their corporate income tax rates are among the lowest in the OECD-21 percent and 12.5 percent, respectively.
The global economy is expanding rapidly. It is vital to the growth of the U.S. economy for U.S. businesses to be internationally competitive. Higher taxes in the United States on the returns to corporate capital inhibit the competitiveness of U.S.-based companies in foreign markets. As financial markets become more global, U.S. investors may tend to be more willing to invest in foreign-based rather than U.S.-based companies. Mergers may be more likely to be set up as a foreign acquisition of a U.S. corporation. Transactions where a foreign subsidiary acquires a U.S.-based parent company may become more frequent. The high rates of taxation on the return from corporate investment can tend to make the United States a relatively unbecoming location for the headquarters of a multinational corporation, which can, in turn, cause U.S. multinationals share in the global market to shrink.
President’s Proposals for Tax Reform
1. Reduction of the tax on corporate earnings
There are several different methods in which relief could be provided for the double taxation of corporate dividends in the United States. One would provide a shareholder credit for corporate taxes paid. When a corporate shareholder receives a taxable dividend, the shareholder would be entitled to a credit against their taxes for the corporate taxes effectively paid on the dividend income. Many countries that have tax relief for double taxation of dividends use a form of the shareholder credit. However, the Treasury Department advised against this approach in a 1992 report because of the complexity of actually implementing the shareholder credit.[7] In its report, Treasury recommended instead that dividend tax relief could be better implemented if a shareholder was allowed to exclude from gross income the dividends received from a corporation. The President’s proposal is consistent with Treasury's earlier assessment that this dividend exclusion framework is simpler than a shareholder credit, and could be implemented with less structural change to the tax code.[8] It also accounts for the fact that about half of dividend payments are currently not taxed, and thus removes the economic distortion of disproportionate taxes on dividends with a smaller reduction in federal revenues.
The President’s proposal not only removes the double taxation of corporate earnings distributed to shareholders, it also removes the double taxation on corporate earnings that are retained. The retained earnings of a corporation should be reflected in an increase in the value of corporate shares, which when sold generate taxable capital gains. Although the advantages of deferral and preferential tax rates mean that the second layer of taxation on these capital gains is smaller than the tax on dividends, the tax is still positive. If the President’s proposal only exempted dividends from personal taxation then dividends would be tax-advantaged compared to retained earnings. However, under this proposal, the tax treatment of all corporate earnings is equal.
There are a number of economic benefits that could be attained by having corporate earnings taxed only once. Taxing corporate earnings only once would lower the cost of investment for firms and increase the after-tax returns to savers that hold corporate equity, thus stimulating capital formation, boosting the productivity of workers, and raising wages and income. Although economists are not in complete agreement about the effect of dividend taxes on investment, typically the empirical results suggest that a reduction in the tax on dividends would markedly increase investment.[9]
The effect of reducing the tax on corporate earnings can also potentially raise stock prices. A fundamental determinant of the value of a share of corporate equity is the present discounted value of all future after-tax dividend payments. Thus, a reduction in taxes on dividends could lead to higher corporate stock values. However, if the reduction in dividend taxes stimulates new investment, then some of this new investment may be done by new firms that enter into markets and compete away the profits of existing firms, thus inhibiting increases in stock prices. The more (less) new investment then the less (more) likely that stock prices rise.
Exempting corporate profits from personal income taxation reduces the tax incentives for corporations to retain earnings instead of paying dividends.[10] Higher dividend payouts would help improve the allocation of corporate capital because this proposal would remove the "lock-in" effect caused by the current tax incentives that make it easier for a firm to keep and reinvest corporate earnings. Instead, there would be no tax disincentive for corporate earnings being reinvested in capital with the highest expected return, whether it is in the same firm or in another business venture. Moreover, a higher payout of dividends would assist stockholders in monitoring corporate managers. Dividends can only be paid with "real" earnings and thus corporate managers could not hide behind a tax code that discourages dividends while they generate only "paper" profits.
Exempting corporate profits from personal income taxation reduces the tax incentives for corporations to finance investments with debt instead of equity.[11] Less corporate debt reduces the probabilities of default and bankruptcy in an economic slowdown and thus would lower the risk premium included in the cost of financing corporate capital.
2. Tax-free savings accounts
Retirement Savings Accounts. Of the three saving account proposals, RSAs would be the most likely to increase personal saving as it would significantly increase the contribution limits associated with Roth IRAs, which RSAs would replace. The profusion of asset shifting seen when IRAs were initially introduced on a universal basis over two decades ago has probably been exhausted. The higher contribution limits allowed by these accounts would result in a greater number of savers facing increased marginal incentives for saving.[12] These marginal incentives for saving would tend to be more prevalent for lower- and middle-income households.[13]
Employer Retirement Savings Accounts. ERSAs typically do not increase the contribution limits that employees will receive from similar types of current plans-401(k) and 403(b) accounts, for example-and thus would not be expected to increase marginal saving incentives beyond those available in current employer-based plans. However, the simplification and lower compliance costs of ERSAs should increase the availability of employer-based retirement accounts, particularly for small-business employees. The costs and complexity of setting up retirement plans are frequently cited by small business owners as reasons for not offering a pension plan to their employees.
Lifetime Savings Accounts. The effect of LSAs on aggregate personal saving are harder to estimate and more likely to be mixed. Particularly in early years after the introduction of LSAs, households would have the incentive to merely shift assets from currently taxed bank accounts and mutual funds into an LSA. For lower- and middle-income households, eventually existing assets would be exhausted and marginal incentives to increase saving would start to become effective. For higher-income households, this process would take longer. Moreover, because LSAs have no withdrawal restrictions then these accounts would be more likely to attract saving done for more short-term purposes than retirement saving, such as precautionary saving. Models of household saving typically imply that precautionary saving is less sensitive to changes in the after-tax return than longer-term retirement saving.[14] For this reason, even if an LSA provides a marginal tax incentive for a household to save more, it may not induce as much increased saving as an RSA or ERSA which focuses on retirement saving.
In general, the RSA and ERSA proposals are merely extensions of recent trends to increase the contribution limits to tax-favored retirement accounts and to simplify the provision of employer provided retirement accounts. Essentially, increasing the contribution limits for retirement accounts moves the personal income tax increasingly towards a consumption tax base as it exempts an increasing share of capital income from taxation. LSAs would be a much newer saving vehicle and would be a significant step further towards a consumption tax. LSAs would probably be quite popular since they do not have withdrawal restrictions. However, partly because the contribution limits prohibit households that save greater amounts from having a marginal incentive to save more, estimating the impact of LSAs on saving would be more speculative.
Another important issue in evaluating the potential net saving effects of these accounts is the possible interaction with household borrowing-in particular, tax-deductible mortgage borrowing.[15] To the degree that households that own a house essentially use increases in tax-deductible mortgage debt to essentially finance tax-favored saving, without having to reduce spending, then net personal saving does not increase. A complete shift to consumption tax treatment at the personal income tax level entails exempting all capital income received from taxation and also not allowing interest paid for borrowing to be tax-deductible. Both of these features are important for getting the full benefits of increased saving by switching to a consumption tax. To the degree that these saving accounts limit contributions, and since the tax-deductibility of mortgage interest remains available, then the positive effects on personal saving from these accounts are less than what would be expected from a complete switch to a consumption tax base.
Notes
[1] I am testifying on my own behalf and not as a representative of AEI.
[2] This second layer of taxes on dividends can be avoided if the shareholder is tax-exempt, such as a non-profit organization, and are typically delayed until withdrawal if the dividends go to shares held in a tax-preferred retirement or insurance arrangement, such as a 401(k) or other pension plan, an IRA, or variable annuity. See William Gale, "About Half of Dividend Payments Do Not Face Double Taxation," Tax Notes (Nov. 11, 2002).
[3] Ken Judd, "Optimal Taxation and Spending in General Competitive Growth Models," Journal of Public Economics (1999) and "The Impact of Tax Reform in Modern Dynamic Economies," in K. Hassett and G. Hubbard (eds.) Transition Costs of Fundamental Tax Reform (2001), and Alan Auerbach and James Hines, "Taxation and Economic Efficiency," in A. Auerbach and M. Feldstein (eds.) Handbook of Public Economics, Volume 3 (2002) provide recent discussion and summaries of this literature.
[4] A recent National Bureau of Economic Research working paper by Casey Mulligan, "Capital Tax Incidence: First Impressions from the Time Series" (#9374, December 2002) finds evidence that the economic burden of capital taxes are shifted significantly on to workers over the long run.
[5] Greg Mankiw, "Commentary: Balanced-Budget Restraint in Taxing Income from Wealth in the Ramsey Model," in K. Hassett and G. Hubbard (eds.) Transition Costs of Fundamental Tax Reform (2001) presents an economic model that provides this result.
[6] Eric Engen and Kevin Hassett, "Does the U.S. Corporate Tax Have a Future?" Tax Notes 30th Anniversay Issue (2002) discusses more fully these issues concerning high corporate taxation in the United States relative to our economic competitors. Indeed, since that article was written, some European countries-such as Belgium, Italy, France, and Luxembourg, for example-have lowered their corporate tax rates even further than shown in the paper.
[7] Department of the Treasury, AIntegration of the Individual and Corporate Tax Systems: Taxing Business Income Once@ (January 1992).
[8] Indeed, for about a decade prior to its repeal in the Tax Reform Act of 1986, taxpayers were permitted a limited exclusion of dividends from gross income in the personal income tax.
[9] A seminal study by Jim Poterba and Larry Summers, "The Economic Effects of Dividend Taxes," in Altman and Subrahmanyam (eds.), Recent Advances in Corporate Finance (1985) found that dividend taxes had substantial negative effects on corporate investment. A recent study by Alan Auerbach and Kevin Hassett, "On the Marginal Source of Funds," Journal of Public Economics (2003), found a smaller but still economically significant effect of dividend taxes on investment.
[10] Jim Poterba, "Tax Policy and Corporate Saving," Brookings Papers on Economic Activity (1987) found that dividend payout rates are quite sensitive to changes in marginal income tax rates.
[11] John Graham, "Do Personal Taxes Affect Corporate Financing Decisions?" Journal of Public Economics (2002) and "Taxes and Corporate Finance: A Review" mimeo, Duke University (2003).
[12] Eric Engen and William Gale, "IRAs and Saving in a Stochastic Life Cycle Model" (1993) and "Do Saving Incentives Work?" Brookings Papers on Economic Activity (1994).
[13] Eric Engen and William Gale, "The Effects of 401(k) Plans on Household Wealth: Differences Across Earnings Groups" NBER working paper #8032 (2000).
[14] Eric Engen, "Consumption and Saving in a Life Cycle Model with Stochastic Earnings and Uncertain Lifespan" (1993).
[15] Eric Engen and William Gale, "Debt, Taxes, and the Effects of 401(k) Plans on Household Wealth Accumulation," (1997).
Eric M. Engen is a resident scholar at AEI.
