EVENTS

Date:	Friday, June 2, 2006
Time:	8:55 AM -- 3:00 PM
Location:	Wohlstetter Conference Center, Twelfth Floor, AEI 1150 Seventeenth Street, N.W., Washington, D.C. 20036

May 2006

Corporate Income Taxation and the Economy

According to the most recent statistics available from the Organisation of Economic Co-operation and Development (OECD), the U.S. combined corporate income-tax rate was second highest among the thirty OECD countries (39.3 percent in 2005). However, as a percentage of gross domestic product (GDP), the revenue yield of the corporate income-tax system was fourth lowest (2.1 percent of GDP in 2003).

The five economic research papers presented at this conference analyze the effects of corporate taxation on compliance costs, tax revenues, wage levels, economic growth, and innovation.
 
Corporate Income Taxation in a Modern Economy

Kenneth Judd
Hoover Institute

Economists have three main criticisms of the corporate income tax: it shifts capital from the corporate sector to less productive noncorporate uses, inhibits productivity and economic growth, and discourages equity financing in favor of debt financing. Taking a more dynamic view of the economy adds to the case against the corporate income tax. One new way to approach the corporate income tax is to look at its treatment of different types of assets with varying levels of aggregate risk. The corporate income tax gives preference to certain safe asset classes, such as debt and owner-occupied housing, compared to equity. This can be thought of as taxing intermediate goods, which imposes very high efficiency costs on the economy with little gain in revenue to compensate.

Another important new consideration is the effect of the corporate income tax in markets in which goods are sold above marginal cost, such as those with firms that need to recover the costs of research and development. In these instances, the marginal excess burden of the corporate income tax will be greater than in cases of perfect competition. By the same token, tax credits for investment and research and development in these cases yield large benefits.

Finally, future work on corporate taxation should adopt a lifecycle view of businesses. Because corporations mainly represent successful businesses whereas ventures that lose money remain partnerships taxed at the individual rate, the government taxes the winners more than it allows the losers to deduct. All of these innovations in studying the corporate income tax suggest that its effects are more detrimental than previously thought.

Robert Carroll
Department of the Treasury

Kenneth Judd’s paper makes an important contribution to the dynamic analysis of tax policy and has implications for several debates currently taking place in Washington. It begins with the fundamental insight that taxing income is tantamount to taxing an intermediate good, which tends to have a cascading effect over time. The paper also makes the point that when firms have market power, the higher prices they charge consumers already function as a kind of tax, and so introducing a corporate income tax on top of this creates an even greater distortion than under the commonly assumed model of perfect competition.

The paper has implications not just for the corporate income tax but also for the very notion of taxing capital income. The double taxation of corporate profits through the investor-level tax on capital gains and dividends bears part of the responsibility for the inefficiencies that stem from allocating too much capital to the noncorporate sector. Inclusion of imperfect competition and risk in models of the effects of tax policy may indicate larger benefits from lowering taxes on dividends and capital gains as well as state taxes. With respect to the life-cycle view of businesses, it is not only the corporate income tax that penalizes enterprises for their success, but also state taxes after the firms exceed the level of exemption.

Again taking a broader view, increasing the extent to which household assets are given consumption tax treatment would yield substantial benefits. It would be interesting to apply the paper’s ideas to issues of fairness in the distribution of the tax burden and to the choice between lowering statutory corporate tax rates and increasing the expensing of new investment. Also, it is important to work out the ramifications of taxing different types of capital investment at different levels and the specifics of how to optimally subsidize research and development.

Thomas Barthold
Joint Committee on Taxation

Professor Judd lays out four basic inefficiencies arising from taxing corporate capital. The first is the differential taxation of corporate versus noncorporate capital, which leads to productive inefficiency. Secondly, the corporate income tax discourages risky ventures and entrepreneurship. Additionally, the corporate income tax exacerbates existing distortions when applied to firms with market power. Finally, the tax inhibits innovation. An important result of the paper is that it is not simply high statutory tax rates but also variance among the rates that apply to different types of capital that is damaging. Two factors in the laws that contribute to this variance are interest deductions and depreciation allowances that do not accurately model true economic depreciation. Because of the difficulty of solving the latter, one might have to resort to simply lowering rates. Also, what about tax legislation that lowers rates only on some types of corporate activity, thus increasing variability? Are such measures on balance good or bad?

Another consideration is that in some instances variation can be useful, such as keeping asymmetry between individual and corporate rates to protect revenue and, as the paper suggests, subsidizing research and development (R&D) to correct for monopolistic competition. This latter proposal faces the difficulty that there are many ways to spend money to create rents in addition to R&D, including advertising and finding advantageous locations, that the paper seems to imply should also be subsidized. There would also be practical problems in terms of defining what constitutes research.

The Compliance Cost of Taxing Business

Joel Slemrod
University of Michigan

There are two components of the cost of collecting a tax: the administrative costs of operating the tax collecting agencies and the compliance costs borne by those subject to the tax. The latter are ten to twenty times larger in the case of the income tax. Based on surveys of businesses, compliance costs for the corporate income tax turned out to be over $2 billion a year for the largest companies, approximately 2.7 percent of revenue collected from firms of this size. This number is actually relatively small because big companies already have elaborate accounting infrastructure in place. In general, compliance costs relative to size are lower for big businesses. Compliance costs also vary among sectors, and complying with tax laws on income from abroad is disproportionately costly.

Compliance costs represent resources that could have gone to more productive use, though in many cases they are partially composed of purely voluntary attempts to minimize a firm’s tax burden. One the one hand, in terms of the incidence of compliance costs, sector-to-sector differences might discourage activity in particular areas. On the other hand, these differences may be appropriate if the true cost of extracting revenue from a sector really is higher.

Businesses, for their part, say that the biggest cost savings would come from equalizing the definitions of income used for tax and financial statement purposes, and from creating uniformity in corporate tax laws among states and between the state and federal levels. Though it has been suggested that tax reforms such as a flat tax or value-added tax could yield reductions in compliance costs, businesses surveyed were skeptical of this, and past experience does not necessarily support this conclusion. In particular, the transition period could generate severe complexities. While the complexity of the tax code might serve some legitimate ends, it could likely be simplified to a certain extent without sacrificing other priorities.

Rosanne Altshuler
Rutgers University

The main source of compliance costs is complexity, yet simplification raises equity and efficiency concerns. Another important issue about which we know little is comparative compliance and its implications, for example, for an add-on VAT. Foreign source income is the source of a large proportion of corporate income tax compliance costs, so simplifying its tax treatment would bring substantial benefits. Other important goals for reform include identifying parts of the tax code that induce tax-planning and scrapping the corporate alternative minimum tax. A lack of details about how a consumption tax would work causes uncertainty and skepticism among businesses and renders comparative tax evaluation difficult.

Dividing businesses into categories by size provides important insights into the sources of complexity. For medium-sized businesses, three important causes of compliance spending were depreciation rules, AMT, and R&D credit rules. Taxing similar income at different rates also provides opportunities to reduce corporate tax bills. The President’s Advisory Panel on Federal Tax Reform suggested trying to equalize the tax treatment of corporate dividends and corporate debt, broadening the tax base, and eliminating special breaks and provisions, which would go a long way toward reducing compliance costs.

Mark Mazur
Internal Revenue Service

Simplifying tax compliance is important because compliance costs siphon productive resources from the economy. Studying taxpayer behavior on this issue can help the government design tax rules, forms, and instructions in a more optimal way. There are three main types of compliance burden: pre-filing, the cost of maintaining a tax department and seeking legal and accounting advice; filing, the cost of actually submitting the tax; and post-filing, which includes auditing and appeals procedures. A distinction needs to be made between accounting that a firm needs to run its business and extra costs necessitated by federal and state/local tax codes, which can be difficult to separate out. Some amount of burden will be inevitable.

Average burden for firms is about $1 million, though this varies by industry due to different amounts of tax expenditures in the tax code. This is why oil and gas firms, for example, tend to have higher compliance costs. The survey data in the paper do a good job of providing a rough estimate of the magnitude of the compliance burden and set the stage for a comparative burden analysis.

Developments in the Taxation of Corporate Profit in the OECD Since 1965: Rates, Bases, and Revenues

Michael Devereux
University of Warwick

Over time, the average corporate tax rate in OECD countries has decreased, while tax revenues from corporate sources as a percentage of GDP have gone up. In the late 1980s and early 1990s, most OECD countries significantly lowered their tax rates, and since then the average rate has held on a fairly steady, slightly downward trend. The European Commission is concerned that this across-the-board lowering may be due to competition
for investment. The weighted average of present discounted value of depreciation allowances has also basically held constant after deceasing in the late 1980s. The unweighted average of corporate tax revenues as a share of GDP has risen over the same timeframe. Looking at the OECD--excluding the United States--taxable profit rose a great deal as a proportion of GDP in the 1990s. There is a generally negative relationship between the corporate tax rate and corporate tax revenue.

What factors might affect taxable revenue to cause this relationship? Perhaps investment decreased, or maybe income shifted abroad or to the noncorporate sector. Aggregating the tax shifting model from firm to country is problematic because we do not observe “full” taxable income, which is affected by GDP and effective marginal tax rates, or the true “alternative” tax rate, which depends on both individual income tax rates and international tax rates. The models imply some sort of non-linear relationship between the tax rate and tax revenues and relatively low revenue-maximizing rates, but the results are not very robust. The relationship loses its significance with the inclusion of many other variables. There is not yet enough evidence to conclude that lowering tax rates raises more revenue via a Laffer Curve effect.

William Gale
Brookings Institution

It is clear that corporate tax rates have gone down as corporate tax revenues as a share of GDP have gone up, but what do we make of this? The three possibilities are a Laffer Curve in investment and business activity, income shifting either to the personal sector or to other countries, or some other factor unrelated to tax rates that inhibits raising revenues. One modification to the regressions in the paper would be to take only the minimum core sample to isolate the effects of variations in the sample size. Other potential problems include the lack of robustness for fixed effects by country, the possibly endogenous nature of tax rate changes, potential lag time in the effects of rate reductions, and the quadratic formulation of the model. The sensitivity of revenues to changes in the corporate tax rate might also be different in countries with a VAT. It seems that some relationship exists, but it is crucial that we discover what is driving it; if tax cuts generated real increases in economic activity, there would be different policy implications than if tax cuts simply caused revenue shifting.

Martin Sullivan
Tax Analysts

Michael Devereux’s research supports three conclusions. First, corporate tax rates are down. Since 2004, when the paper leaves off, the trend has continued. Frequently, tax cut legislation has been accompanied by base broadening measures. The model does a good job of establishing a Laffer curve relationship, but there is no basis for identifying a particular rate as revenue-maximizing. Also, since the unweighted average of taxable profits crossed the weighted average between 1982 and 2004, there is some evidence that profits have shifted from large, high-tax countries to smaller, low-tax ones. This and a general increase in before-tax profits seem to be likely explanations for the increase in corporate tax revenues. When making choices about where to invest, corporations look to the average tax rate and not marginal tax rate-related provisions like investment tax credits and accelerated depreciation allowances. Thus the statutory corporate tax rate is the most important component for international competitiveness.

Keynote Address

The Honorable Edward Lazear
President’s Council of Economic Advisers

The recently extended 15 percent tax rate on corporate dividends and capital gains, originally enacted in 2003, has been a significant boon to the economy. 5.3 million jobs have been created since passage of the tax cuts, productivity has risen, and unemployment has fallen. Evidence indicates that firms increased the amount of dividends paid out in response to the tax cuts, and the stock market rise can at least partly be attributed to the legislation and the ability of companies to raise capital more efficiently.

There are three important principles to keep in mind when designing a tax code. First, it should not give preference to current consumption over future consumption by penalizing saving. Second, it should not distinguish between different types of capital by, for example, taxing investment in housing less than investment in physical capital. Third, it should minimize variations in tax treatment of different forms of business, such as partnerships versus corporations, and different types of financing. Current law creates a much heavier effective tax rate on equity financing versus debt financing, at 36 and -6.4 percent, respectively.

Though the benefits from simplifying the tax code could be substantial, increasing the growth rate of the economy would add much more to GDP. One important aspect of stimulating growth is increasing productivity and thus wages. The important thing in achieving growth is to maintain a high rate of return on investment to attract the foreign capital necessary to sustain high levels of physical capital formation. The tax cuts helped erode the bias against equity investment and also reduced incentives for companies to retain earnings instead of investing in highly innovative venture capital projects. The legislation made it less disadvantageous tax-wise to become a corporation rather than a partnership. Finally, by lowing marginal tax rates for high earners, the tax cuts improved incentives to invest in human capital. Extending the provisions encouraging physical capital first was the right strategy because the elasticities there are higher. Extension of health savings accounts would help fix another distortion wherein employers have an easier time deducting contributions to health plans than individuals.

Taxes and Wages

Kevin A. Hassett and Aparna Mathur
AEI

Exhaustive data collection has made possible a look at the impact of corporate taxes on wages across countries. A lot of the very high incidence on workers comes from high capital elasticities--since wages are largely a function of capital, changing the tax rate can affect wages significantly. The objective was to take data from seventy-two countries and add tax variables to wage equations in order to see how much varying the tax rate impacts wages. The possibility that tax rates may be partly endogenous and that there might be a lag time for increases in the capital stock give reasons for caution. However, the study found strong links between corporate tax rates and wages across countries.

Controlling for other factors, a 1 percent increase in the top corporate income tax rate reduces wages by 0.8 percent. Other tax variables such as VAT or labor income taxes did not explain variations in wages. It seems that since capital is highly mobile, capital taxation will fall heavily on workers in the form of wage reductions. Additionally, there was a positive association between tax rates in neighboring countries and wages at home. If tax rates are comparatively high in nearby areas, then capital will flow in and boost wages. The magnitude of this effect is much greater in smaller economies.

Jared Bernstein
Economic Policy Institute

The research of Hassett and Mathur generated clear and plausible relationships between capital taxation, investment, and wages, with an enormous amount of data and rigorous testing. However, there are some concerns about all of the links in the paper’s reasoning. To what extent are investment decisions really based on tax considerations? How strong are the links between investment and productivity, and between productivity and wages? Certain simplifying assumptions, such as holding depreciation, inflation, and real interest rates constant over time, may be unrealistic. In addition to some other minor econometric quibbles, the exclusive focus on the manufacturing sector seems to limit the applications of the study for economic policy as a whole.

There is also strong correlation between high statutory corporate tax rates and high levels of capital formation, which seems to challenge the paper’s conclusions, but this may just reflect the fact that big countries have more investment and higher wages. Also potentially problematic is the significance of contemporary tax rates when the hypothesis of the authors implies that there should be a lag period before taxes affect wages. Other worries include the dearth of variation in top statutory rates in the data, the possibility of high tax rate countries with low wages nonetheless providing valuable government services, and the deviation between productivity and growth in real living standards.

William Randolph
Congressional Budget Office

Though the paper is groundbreaking work on the question of the incidence of the corporate income tax, it does not represent a final answer to the controversy. The elasticity estimated would lead to a much larger decline in wage income per dollar of additional tax revenue than one would think. Perhaps something else is going on--it is possible that countries lowered their tax rates during times of economic strength, rather than the tax cuts actually improving investment.

Corporate Income Taxes and Economic Growth

Dale Jorgenson
Harvard University

In the status quo, certain types of income, such as that from pensions, are taxed on a consumption basis, whereas corporate income is taxed twice. Consensus has developed on the need to reduce this distortion, but taxation of income from housing remains a big controversy. Rental housing is taxed at the individual level, but owner-occupied housing, due to deductions for mortgage interest payments and property taxes, is actually subsidized in the tax code. Another important consideration for tax reform is the fact that state and local taxes employ the same tax base and definitions as the current federal income tax. Finally, revenue neutrality should be an important constraint on any tax reform discussion.

There are a variety of consumption tax proposals, but there is a less radical way to fix the problems with corporate capital and housing taxation while keeping income as the tax base. No definitions, exemptions, deductions, or credits would change. Income would be divided into earned and unearned, with the former taxed at 11 percent and the latter at 31 percent. Marginal tax rates would decline, but progressivity would be maintained indirectly. Corporations would receive tax credits on equipment and structures, and new construction would be subject to an entirely new tax to pay for the other provisions. New houses would still be obtainable because lower corporate taxes would mean construction firms could charge less, but the heavy tax would level the playing field with other types of capital. This tax reform would add about twenty cents on the dollar to national wealth, which compares favorably with other tax reform proposals.

Jason Cummins
Brevan Howard

This proposal is ingenious in that it has the econometric analysis to satisfy the economic community but is also hardheaded in its political calculations, giving it a more realistic shot of getting passed. However, the credits for business will be difficult politically in an environment in which corporate profits are through the roof and capital’s share of income is at an all time high, even if wage earners would not be hurt. Also, the complicated array of tax credits necessary to equalize the tax burden on corporate and noncorporate capital relies on calculated effective marginal tax rates, which are not necessarily reliable for substantive real-world tax reform. It is difficult to accurately arrive at the proper depreciation rates, since current ones are based on outdated empirical studies. It would also be interesting to see a time horizon for when the welfare gains from the tax reform would materialize.

Douglas Holtz-Eakin
Council on Foreign Relations

It is apparent that there is tremendous opportunity to improve the current tax code, but trying to recover revenue with a bad tax or failing to reduce spending could make the situation much worse. The paper emphasizes the importance of the owner-occupied housing distortion and the welfare gains from fixing this intractable discrepancy. It also underlines the importance of capital taxation. Finally, it highlights how progressivity conditions the choice between income and consumption as the tax base. It is difficult, however, to understand why the tax reform proposals ended up being ranked the way that they did--for example, why there turned out to be a bigger welfare gain from a progressive labor income tax than from a progressive consumption tax. It would also have been more useful to model the tax reform proposals starting from the current tax code complete with the Bush tax cuts, rather than assuming the 1996 laws as the paper did.

AEI intern Matt Perlman prepared this summary.