ax Title of    Testimony of Alan J. Auerbach, Robert D. Burch Professor of Economics and Law Director, Burch Center for Tax Policy and Public Finance, University of California, Berkeley, Before the House Ways and Means Commitee: Tax Reform, Growth and International Competitiveness

June 9, 2005

109th Congress

Tax Reform, Growth and International Competitiveness

Testimony before the Committee on Ways and Means , U.S. House of Representatives

 

by

Alan J. Auerbach

Robert D. Burch Professor of Economics and Law

Director, Burch Center for Tax Policy and Public Finance

University of California , Berkeley

June 8, 2005

Mr. Chairman and Members of the Committee:

I am pleased to have this opportunity to offer my views on the effects of the tax system on economic growth and international competitiveness and the role that tax reform can play in promoting both. Today, I will discuss the design of effective tax incentives and how the current tax system measures up. I will also highlight one potential goal that tax policy cannot achieve. Finally, I will draw some conclusions regarding the design of fundamental tax reform and how potential reform plans should be evaluated.



WHAT IS COMPETITIVENESS?

The United States has experienced massive trade deficits in recent years, with net imports in the first quarter of 2005 approaching 6 percent of GDP. This has led to concerns that U.S. producers may be losing their ability to compete in the global marketplace. But how should one define competitiveness?

The simplest definition of competitiveness, perhaps, involves a comparison of domestic and foreign costs of production; U.S. producers are "competitive" if their costs of production are lower. But this measure of competitiveness suffers from two logical problems. First, the costs of foreign producers, measured in U.S. dollars, depend on the exchange rates between the dollar and the respective foreign currencies. A low enough dollar would make all U.S producers competitive according to this definition, even with no changes in domestic or foreign production techniques or local labor costs. Thus, U.S. producers could be competitive even in our weakest industries or, at a different configuration of exchange rates, not competitive even in our strongest. Second, the most fundamental economic concept of international trade - Ricardo's principle of comparative advantage - dictates that a country will export the goods and services in which it is relatively competitive, and import the goods and services in which it is not - even if it is extremely efficient at producing goods and services in the latter category. By definition, a country cannot have a comparative advantage in producing everything, and thus it cannot have a uniform cost advantage. Thus, a comparison of domestic and foreign costs, though perhaps intuitively appealing, is not a useful measure of competitiveness.

A better approach to measuring competitiveness relies on a comparison of productivity, the amount of goods and services produced per worker hour. Higher productivity means a higher attainable standard of living, for over the long term a society cannot consume more than it produces. Our objective, as a society, should be a high and rising standard of living. International trade is only a means to achieving this ultimate objective; strong exports or a trade surplus are not ends in themselves.



THE DETERMINANTS OF PRODUCTIVITY

Productivity per worker hour depends on a range of factors, many of which are influenced by taxation. We should consider the impact of taxation on human capital, tangible capital, and intangible capital - the three types of capital that increase productivity by augmenting basic, unskilled labor. All three types of capital are important determinants of productivity and policies to foster capital accumulation should be designed with all three types in mind (not just tangible capital). Our present tax system does not treat the different types of capital uniformly, thereby creating uneven incentives for capital accumulation.



Tax Policy and Human Capital

The costs of accumulating human capital through education, on-the-job training and workplace experience receive relatively favorable treatment under current law. A primary cost of education and training is forgone earnings - the earnings given up by staying in school or by accepting a lower current salary in order to gain valuable job experience - and this cost is effectively tax-deductible, because income and payroll taxes are avoided when earnings are forgone. The progressivity of the tax rate schedule offsets the benefit of being able to expense costs immediately, for those who accumulate human capital will have higher future incomes and thus face higher marginal tax rates. But there is another aspect of the "success" tax that is often ignored, that a progressive rate structure also provides insurance against income uncertainty. Success may be the result of education and hard work, but it also depends on luck: by choosing the right field of study, working for a successful employer, or avoiding debilitating illness. Indeed, recent research suggests that moving to a less progressive rate structure might be undesirable, if one takes the reduction in this insurance function into account. ¹

The marginal tax rate reductions of recent legislation, then, have offsetting effects on incentives for human capital accumulation, increasing after-tax investment returns, but also increasing the uncertainty of these returns. However, there is another important aspect of recent policy that discourages human capital investment - the fact that we are on an unsustainable fiscal path. That is, there is a significant risk of large future tax increases. The decision to invest in human capital relies on a very long time horizon, since the returns to education and training may be realized only after several years. The prospect of higher future taxes and the uncertainty regarding the form and level of these taxes each discourage human capital investment. A tax system's instability, in itself, discourages human capital investment (and indeed, other types of investment as well), and it is difficult to maintain a stable tax system when there is a large and growing imbalance between taxes and spending.



Tax Policy and Intangible Capital

Intangible capital investments generally face more favorable tax incentives than do investments in human capital. Not only can investors write off such investments immediately, but qualifying investments are eligible for the Research and Experimentation credit. Because the immediate write-off, by itself, effectively eliminates the tax on new investments, the combination of immediate write-off and the R&E credit delivers an outright subsidy to investments in intangible capital. Does such favorable treatment have any justification? The answer is yes, because many intangible investments have the capacity to convey positive "spillovers" to others in society. Put another way, if investors cannot capture all the benefits of their investments, they need some sort of compensation from the rest of society - those who also gain from these investments - and a tax subsidy is therefore fair. The right level and scope of the subsidy are difficult to determine, however, for it is hard to identify the social payoffs to different types of investments in intangible capital.

It should also be remembered that intangible capital depends on more than research and development investment, and is influenced substantially by government policy beyond the realm of taxation. Our ability to achieve high levels of production relies on the clarity and force of our legal system, the rationality of our regulations, the flexibility of our employment relationships, and so forth. We should view tax policy toward intangible capital as but one component of a portfolio of policy tools.



Tax Policy and Tangible Capital

Incentives to invest in tangible capital, primarily plant and equipment, are affected by myriad tax provisions. In thinking about the nature of the provisions, it is useful to distinguish four dimensions in which tax provisions may vary:

1. Are they broad or targeted?

 

2. Are they temporary or permanent?

 

3. Are they aimed at savers - those who supply the funds for investment - or at the companies that actually spend the funds?

 

4. How do they apply to capital assets already in place, as opposed to prospective investments?

Broad versus Targeted

One of the key principles of tax design is that taxes should have as broad a base as possible, so as to permit the lowest possible tax rate for a given revenue requirement. A broad-based, low-rate system promotes efficiency by limiting the distortions of economic decisions, and may also be simpler and easier to administer than alternative systems.

Why deviate from this appealing norm? For example, why implement an investment tax credit for investment in equipment, as we did prior to 1986? One argument is that the favored investment provides significant social spillovers, basically the same rationale as for the favorable treatment of investments in intangible capital. But there is no convincing empirical support for the argument that certain types of equipment investment generate such positive spillovers, so this argument is weak. A second defense of targeted provisions is that they offset existing benefits available to other investments. For example, prior to the changes introduced by the Tax Reform Act of 1986, some argued that real estate investment enjoyed a variety of tax benefits, so that the investment credit simply leveled the playing field. While such an argument may have had merit, at least before 1986, it is extremely difficult to get such offsets right, as the tax provisions one must weigh in seeking balance are typically quite different.



Temporary versus Permanent

No tax provision is truly permanent, because each year brings the possibility of new tax legislation. But some provisions are explicitly temporary. The bonus depreciation provisions introduced in 2002 and enhanced and slightly extended in 2003 had this characteristic; indeed, bonus depreciation has now expired.

Why adopt temporary provisions? The strongest potential justification is as a stabilization device, to dampen swings in the economy. For example, bonus depreciation was introduced while the economy was still recovering from the 2001 recession, which had been characterized by a steep drop in equipment investment. Historically, adjustments of the investment tax credit were often motivated by similar concerns. There is plenty of evidence that fluctuations in investment incentives altered the path of investment, but no evidence that such effects served the purpose of stabilization. ² Very short-lived provisions, such as the bonus depreciation scheme, are especially tricky to implement because they provide more powerful incentives for longer-lived assets, allowing investors to lock in "cheap" capital for a long time. The variations in incentives across assets depend on how temporary the provisions are, making the tailoring of provisions to provide uniform incentives extremely difficult in a dynamic tax environment.



Saving versus Investment

With the existence of international capital flows, saving and investment can occur in different countries. That is, savers in the United States can supply funds for foreign investment, and businesses in the United States can attract funds for their investment from abroad. If U.S. capital accumulation is our goal, is it U.S. saving or U.S. investment we should be encouraging?

The argument for encouraging U.S. saving is straightforward, for increasing saving expands national wealth, whether this wealth is invested abroad or at home. An increase in U.S. saving would also contribute to a reduction in our trade and current account imbalances, for, as discussed further below, the share of our domestic production not claimed by domestic investment or government and private consumption is available for export.

The argument for encouraging U.S investment, as opposed to U.S. saving, is less obvious, unless one subscribes to the view that such investment generates large social spillovers, a view that I have already dismissed. In the end, though, there is less distinction than there may appear to be between policies that encourage saving and policies that encourage investment, because investment and saving tend to move together. ³ Thus, policies that encourage saving encourage investment, and policies that encourage investment encourage saving.



New Capital versus Old Capital

Some people seem to believe that provisions that benefit capital necessarily benefit investment, but capital and investment are distinct. Provisions may reduce the tax burden on existing capital more or less than they reduce the tax burden on new investment. If stimulating investment is our aim, then we should seek to reduce the tax burden on investment. Although reducing the tax burden on existing capital at the same time may be unavoidable, doing so may actually discourage saving and investment, for example, by increasing asset values and stimulating the consumption of goods and services.

Historically, proposals and actual policies have varied considerably in their relative treatment of old and new capital. The following diagram illustrates this variation, with provisions favoring old capital more toward the left, and those favoring new capital more toward the right.

 

A reduction in the corporate tax rate lowers the tax burden on both new and old capital, though it may help old capital more to the extent that depreciation deductions are more favorable for new assets and shield more of their income from taxation. An investment tax credit, on the other hand, provides no benefit at all to existing capital. Moving in the other direction, a reduction in the capital gains tax rate favors old capital substantially more than new capital. A cut in the capital gains tax rate reduces the tax burden not only on future income from existing assets but also on these assets' past income - gains that have already accrued but have yet to be realized.

Focusing tax provisions on new capital makes the provisions more efficient at achieving their goal, but there are limits to the feasibility of doing this. For example, an incremental investment tax credit, applicable to investment in excess of some base level, would be an even more cost-effective way of encouraging investment. But though the incremental ITC has occasionally been proposed, the determination of its base investment level is complex and involves potentially perverse side effects of the type that have plagued the R&E credit over the years.

The way a tax provision is implemented also affects the extent to which it favors old or new capital. For example, phasing in a reduction in the corporate tax rate rather than adopting the reduction immediately concentrates more of the overall benefit on new capital. Because investment decisions are forward looking, a prospective corporate tax cut may be almost as effective at stimulating investment as an immediate one. But the phase-in procedure will save revenue and limit windfalls to capital already in place, the source of current income. By this logic, the dividend and capital gains tax reductions of 2003 get things almost exactly backward. If investors believe that the scheduled sunset after 2008 will occur, then the tax reductions will have almost no impact on the incentive to invest while still providing substantial tax benefits to the owners of existing assets.

Finally, it is important to look at substance rather than form in estimating the extent to which a tax provision favors new or existing capital. Provisions labeled "savings incentives" appear aimed at the generation of new savings, and hence targeted at new capital. Any provision that provided a tax benefit only for new saving would indeed be an incentive to save. But many tax incentives for "saving" are available to individuals who simply transfer existing assets into a qualifying account. Such asset transfers do not constitute new saving, and the associated tax benefits represent windfalls to existing assets. Legislators should keep this caveat in mind when considering the expansion of such schemes.



Measuring the "Bang for the Buck"

As just discussed, capital income tax provisions vary in the extent to which they provide windfalls to existing capital rather than incentives for new investment. For a given investment incentive, a provision that limits windfalls will have a lower revenue cost and hence a higher "bang for the buck" - a greater investment stimulus per dollar of lost revenue. But comparing the revenue lost by different provisions is tricky because the provisions may vary in their timing. Provisions that provide "front-loaded" incentives may appear more costly over a short revenue window than comparable provisions that are "back-loaded." A familiar recent example is the comparison between traditional IRAs and Roth IRAs. The two types of accounts offer similar tax benefits to depositors when calculated over the long term, but the tax benefits of the Roth IRAs are received by a depositor over time (through the exemption of earnings from tax), while the traditional IRA delivers its benefits through an immediate tax deduction. Thus, over a short period, say a ten-year budget window, the Roth IRA will appear less expensive and seem to have a bigger bang for the buck than the traditional IRA even though this is not the case.

Another example comes from the Tax Reform Act of 1986, which reduced the corporate tax rate while repealing the investment tax credit. As discussed above, the investment tax credit is focused more on new capital than is a corporate tax rate reduction, so trading in the ITC for a tax rate cut seems like a move in the wrong direction, at least in terms of potential bang for the buck. But the ITC is a front-loaded incentive, which makes it appear more expensive over a short time period than a corporate tax rate reduction with a similar impact on the incentive to invest. As the 1986 reform aimed at short-term revenue neutrality, the process was biased against the ITC. ⁴



WHAT TAX REFORM CANNOT DO

There are, of course, many things that tax reform cannot accomplish, but there is one objective so closely tied to tax reform in the minds of some that it deserves explicit attention, lest reform efforts be misdirected. Tax reform cannot reduce the trade deficit, at least not through the tax treatment of exports and imports.

The trade deficit, by definition, equals the amount by which domestic investment spending plus non-investment purchases by government and households exceed domestic production, i.e., GDP; we must import the excess of what we spend over what we produce domestically. Unless we reduce domestic investment spending - clearly not an objective of tax reform - the only ways to reduce the trade deficit are to increase domestic production or to reduce spending by government and households. Either one of these changes - increasing production or reducing non-investment spending - amounts to an increase in U.S. national saving, so reducing the trade deficit (assuming we don't sacrifice domestic investment) requires an increase in national saving. Thus, tax reform must increase national saving in order to reduce the trade deficit.

Tax reform can increase national saving by expanding the economy's productivity, and hence the size of potential GDP, as well as by encouraging saving directly, encouraging households to spend less and save more of their current income. (Other government policies, too, can reduce the trade deficit, notably deficit reduction policies that reduce government spending and, through tax increases that reduce household disposable income, reduce household spending.) But tax reform cannot increase national saving simply by its treatment of exports and imports.

Some consumption tax proposals, such a value added tax or a retail sales tax, incorporate border adjustments - they are not imposed on exports, but are imposed on imports. Other proposals do not include border adjustments; they do not relieve the tax on exports and do not tax imports. In comparing the approaches with and without border adjustments, some have argued that inclusion of border adjustments would reduce the trade deficit by making our exports cheaper abroad and our imports more expensive at home. But, with no other differences in economic fundamentals in the United States or abroad, border adjustments will simply strengthen the dollar, putting importers and exporters in the same competitive positions no matter which approach is adopted. ⁵ Ironically, the stronger dollar that border adjustments would induce would also provide windfalls to foreigners holding dollar-denominated assets - the very ones who have financed our recent trade deficits.

In summary, tax reform can help the trade balance, but only indirectly, through its impact on national saving. The decision whether to include border adjustments as part of a tax reform package should hinge primarily on other factors, such as simplicity, compliance costs and ease of administration.



IMPLICATIONS FOR FUNDAMENTAL TAX REFORM

On the basis of the preceding discussion, I offer the following guidelines for tax reform design.

1. In thinking about the sources of productivity, keep all forms of capital, not just tangible capital, in mind.

 

2. With few exceptions, avoid narrowly targeted tax provisions. Such measures may be justified in exceptional cases, but more often are a source of complexity and distortion.

 

3. Focus on incentives. Provisions formally associated with "capital" or "saving" do not necessarily encourage capital accumulation very much. Providing windfalls to existing assets does not stimulate investment or saving.

 

4. Take future revenue consequences into account. Proposals can be tailored to minimize short-run revenue costs or maximize short-run revenue gains with little impact on the proposals' economic effects or long-run revenue consequences. The attractiveness of these proposals should not hinge on such cosmetic alterations.

 

5. Pay close attention to the design of transition provisions. Phase-in provisions can be an effective method of focusing tax benefits on investment rather than on existing assets. But the piecemeal application of tax incentives can also introduce opportunities for tax arbitrage, where individuals receive tax benefits simply by transferring existing assets to achieve tax-favored status.

 

6. Do not conceive of tax reform as providing simple solutions to the U.S. trade deficit.

 

7. Remember that the objectives of tax reform will be undercut if the overall fiscal system is unstable. Tax reform and broader fiscal reform are complementary.

¹ Shinichi Nishiyama and Kent Smetters, "Consumption Taxes and Economic Efficiency in a Stochastic OLG Economy," National Bureau of Economic Research Working Paper No. 9492, February 2003.

² Alan J. Auerbach and Kevin Hassett, "Tax Policy and Business Fixed Investment in the United States," Journal of Public Economics 47(2), March 1992, pp. 141-170.

³ Martin Feldstein and Charles Horioka, "Domestic Saving and International Capital Flows," The Economic Journal 90(358), June 1980, pp. 314-329.

⁴ Indeed, the one-year revenue effects of these two tax provisions were roughly offsetting. See Alan J. Auerbach and Joel Slemrod, "The Economic Effects of the Tax Reform Act of 1986, Journal of Economic Literature 35(2), June 1997, pp. 589-632."

⁵ See Alan J. Auerbach, "The Future of Fundamental Tax Reform," American Economic Review 87(2), May 1997, pp. 143-146.