Nearly every day, the two major presidential candidates speak about the economic good or ill that stems from the 2001 and 2003 tax cuts. Besides the war in Iraq, few matters so divide the candidates and their supporters as their view of the wisdom of enacting substantial tax cuts in 2001 and again two years later. Indeed, many pundits believe that the election may well turn on whether or not the electorate believes the President's core economic policy is working.

President George W. Bush argues that these two important changes in U.S. tax law turned the tide of recessionary forces, supported the U.S. economy during the dark days following September 11 and the corporate scandals, and now explain a large part of the country's current prosperity and rates of high employment.

Just as vigorously, Senator John F. Kerry (D-MA) condemns the tax cuts for being overly generous to high-income taxpayers, draining revenues from the federal government at a time of war and on the eve of the baby boom retirement, and producing record federal budget deficits. Senator Kerry particularly endorses this last claim, and he is especially galled by the evaporation of budget surpluses that President Bill Clinton handed to his successor.

Both candidates have proposed additional changes in current tax law that reflect their views of how the 2001 and 2003 tax cuts affected the economy and federal finances. This report joins the debate over current economic policy by estimating how each candidate's tax proposal would affect the economy and the government's finances. This report finds that:

•  The Kerry tax plan slows economic activity until 2011, when it generally adopts the Bush approach of permanent tax cuts. Even so, the Bush plan consistently outperforms the Kerry plan.
• The two plans reflect sharply different approaches to tax policy: President Bush relies on supply-side tax changes while Senator Kerry focuses much of his attention on demand-side policy moves.
• Senator Kerry's greater reliance on targeted tax policy changes yields the unintended consequence of producing a tax cut for high-income taxpayers after 2011.

The candidates' tax plans join slightly over 30 related pieces of legislation in the U.S. Congress that currently await legislative action. These proposals range across the entire spectrum of initiatives, from making certain elements of the 2001 and 2003 tax cuts permanent to repealing them for specified classes of taxpayers to proposals for adding tax credits and closing tax loopholes.

The welter of claims and counterclaims about the Bush tax cuts and the candidates' new tax proposals present real problems for voters and taxpayers. Without a common metric against which to measure the effects of both proposals, voters and taxpayers may never obtain a good idea about which one is better for the economy and the federal government's financial future.

Basic Findings

This report employs just such a common measuring tool to assess the economic and fiscal prospects of the two plans. Center for Data Analysis (CDA) analysts used CDA tax models and tax information from other sources as inputs to Global Insight's U.S. Macroeconomic Model, one of the most widely respected forecasting models.^[1] Comparison of the likely economic and fiscal effects of these two competing plans is greatly facilitated by using the same economic model to evaluate both approaches. Among this report's findings are:

•  Stronger job growth under the Bush plan. The Bush tax plan leads to significantly stronger employment growth between 2005 and 2014 than is likely under the Kerry plan. In 2009 (or halfway through the 10-year period), the U.S. economy under the Bush tax plan would likely enjoy 288,000 more jobs than under the current-law baseline. The Kerry tax plan would cause total employment to fall by 202,000. By 2014, total employment under the Bush plan is projected to be 995,000 higher than it would be without the plan, while the Kerry plan is estimated to bring about an employment gain of 658,000 after he generally adopts the Bush policy of making the tax cuts permanent. Both plans would affect the economy most in the last four years of the forecast period, or 2011 through 2014. This is the period when, under current law, all of the 2001 and 2003 tax law changes disappear.
  
• Stronger economic growth under the Bush tax proposal. The Bush tax plan would lead to consistently stronger economic activity between 2005 and 2014 than would be likely under the Kerry plan. In 2009, the gross domestic product (GDP) is $39.3 billion higher with the Bush tax plan than it would be without it. In 2009, GDP under the Kerry plan is $2.4 billion higher than the baseline. By 2014, GDP is $105.4 billion higher under the Bush plan and $82.7 billion higher under the Kerry proposal.
  
• More spending money after taxes under the Bush plan. Under the Bush plan, the model shows that disposable income for a family of four is $872 higher than the baseline in 2009 and $3,904 higher in 2014. Under the Kerry tax proposals, disposable income for four persons would be $340 higher than the baseline in 2009 and $3,448 higher than the baseline in 2014.

The Bush Tax Plan

In 2001 and 2003, President Bush signed into law two tax cuts that saved taxpayers billions of dollars: the 2001 Economic Growth and Tax Reduction Reconciliation Act (EGTRRA) and the 2003 Jobs and Growth Tax Reduction and Reconciliation Act (JGTRRA). Due to the political complexities of the budget process, these tax cuts expire between now and 2014. Instead of letting these tax policy changes expire, President Bush proposes that they be made permanent.^[2]

Individual Income Tax. President Bush proposes making many of the individual income tax components from the EGTRRA permanent,^[3] including:

• The doubled child tax credit ($1,000 per child)
  
• The expanded dependent care credit ($3,000 per dependent, up to $6,000);
  
• Marriage penalty relief;
  
• The earned income credit expansion for married joint filers;
  
•  The 10 percent tax bracket (the lowest tax bracket); and
  
• The reduction in the marginal tax rates from 39.6 percent to 35 percent, 36 percent to 33 percent, 31 percent to 28 percent, and 28 percent to 25 percent.

Key elements of JGTRRA would also be made permanent. The taxation on dividends and capital gains would continue to decline to 0 percent for filers below the 25 percent bracket. Taxpayers in higher rate brackets would pay lower taxes on dividends and capital gains. The tax rate on capital gains for these taxpayers declined from 20 percent to 15 percent under JGTRRA, which also changed the treatment of dividend income. Dividends are no longer considered ordinary taxable income, but instead receive the same treatment and taxation as long-term capital gains.

An important element of the Bush plan is its treatment of high-income taxpayers. Unlike Senator Kerry, the President does not use income as a test of whether or not a taxpayer is eligible for the 2001 and 2003 tax cuts. Thus, taxpayers with income in the top two income tax brackets are treated the same as taxpayers with incomes below that amount when the 2001 and 2003 tax cuts are made permanent.

Estate Tax. President Bush also differs from Senator Kerry on "death taxes." The Bush plan calls for the permanent repeal of the estate and generation-skipping taxes in 2011, when these taxes are otherwise scheduled to return to their 2001 levels. Between now and 2011, the President continues current law, which calls for a steady drop in the tax rate culminating in a one-year repeal of estate and generation-skipping taxes in 2010.

The Kerry Tax Plan for Individual Income Taxes

Some Expiring Provisions from 2001 and 2003 Made Permanent. Senator Kerry retains several provisions of the 2001 and 2003 tax bills:

• Marriage penalty relief;
  
• The doubled child tax credit ($1,000);
  
• The earned income credit expansion for married couples;
  
• The 10 percent tax bracket;
  
• Reductions in all regular income tax rates except the top two; and
  
• The tax cuts on dividends and capital gains for taxpayers with incomes below $200,000.

Increased Taxes on High-Income Taxpayers. Senator Kerry has described his plan as "rolling back" the tax cuts for filers with over $200,000 in income. While there is some uncertainty about the details of his plan, Kerry campaign materials indicate that the tax rates in the top two brackets would increase to their pre-EGTRRA levels.^[4] Tax rates on capital gains and dividends would also increase, but the higher rates appear to begin when taxable income exceeds $200,000. When total taxable income is less than $200,000, capital gains and dividends would be taxed at the current rates. Only the portion of capital gains and dividend income that exceeds $200,000 would be taxed at pre-EGTRRA rates.

Because the second highest bracket is below $200,000, the Kerry plan appears to add a special tax bracket for dividends and capital gains. All regular income in the second highest bracket would be taxed at 36 percent. The portion of capital gains income that, when added to regular income, is at or below $200,000 would be taxed at 15 percent. However, the remaining amount of capital gains in this bracket would be taxed at 20 percent. A similar split occurs for dividend income. The portion of dividend income that, when added to other taxable income, is less than or equal to $200,000 would be taxed at 15 percent. The remaining amount of dividend income in the bracket would be taxed at 36 percent.^[5]

The Kerry plan would tax all regular income in the top tax bracket at 39.6 percent. All dividend income in this bracket would also be taxed at 39.6 percent. The top rate on capital gains income would be 20 percent.

The Kerry campaign anticipates that tax increases in the top two brackets would raise substantial revenue that would pay for other tax cuts and new spending programs. However, compared to the current-law baseline, the Kerry plan would actually reduce receipts after tax year 2010 because the current-law baseline already takes into account the increase in top rates beginning in 2011. After 2010, there is a net reduction in tax liability for taxpayers in the top two brackets. Revenue in these years would be less than currently projected because taxpayers in the top two brackets benefit from rate reductions in the other tax brackets.

Child and Dependent Care Tax Credit. The Kerry plan would increase the maximum qualified expenses by $2,000 per child to $5,000 for one dependent and $10,000 for two or more.^[6] This change would increase the minimum credit by $400 for one dependent and $800 for two or more. In addition, the plan allows some taxpayers to take a credit in excess of their tax and thereby receive a refund.^[7]

Higher Education Tax Credit. Senator Kerry has proposed a College Opportunity Tax Credit, a new higher education tax credit that is an expanded version of the existing HOPE tax credit. Currently, the HOPE Credit allows taxpayers to take a nonrefundable tax credit equal to 100 percent of the first $1,000 of qualified education expenses plus 50 percent of the next $1,000. This tax benefit permits single filers with incomes below $41,000 and married filers with incomes below $83,000 in 2003 to claim a maximum credit of $1,500 for each student who had $2,000 of qualified expenses.^[8]

In addition to increasing the maximum credit by $1,000 to $2,500, the Kerry education tax credit could be claimed for up to four years of undergraduate study rather than the existing two years. The proposed credit is "refundable," which would allow taxpayers to receive a benefit even if they do not owe federal income tax. Based on available information about the plan, it appears that the existing income phaseouts would also apply to the new credit, as would the possibility of a reduction due to the alternative minimum tax (AMT).^[9]

Health Tax Credits.^[10]The Kerry tax plan offers an incentive for small businesses to purchase health insurance through pools that offer health plans, similar to the Federal Employees Health Benefits Program (FEHBP). Health insurance premiums would be paid by the workers and their employers. The plan uses tax credits to reduce the costs to employers.

Employers who participate in the FEHBP-like pools and fund 50 percent or more of the cost of premiums would receive a 25 percent refundable tax credit. Businesses could also treat the premium as a deductible expense. The net cost to an average employer has been estimated at 25 percent of the overall premium.^[11] However, the proportion of the credit that could actually be used depends on the employee's income. Employers can claim the full credit for premiums for single workers whose income is 150 percent of poverty or less. The value of the credit declines as the employee's income increases, until it reaches zero for single employees whose income is 300 percent of poverty ($28,179 in 2003).^[12]

Newly covered employees could pay their premium through a deduction in their wages. Workers might also see a slower increase in their wages as employers offset the premium payments and increased administrative costs. Alternatively, business owners could pay for the new premiums by hiring fewer workers, reducing their profits, or passing the costs on to consumers through higher prices on goods and services.

The plan also allows employers who are already offering health insurance to shift coverage to the FEHBP-like pools. The new tax credits would tend to reduce existing health care expenses for these employers. However, the plan would also increase costs for these business owners by imposing an application fee equal to 10 percent of the premiums.

Workers who are eligible for unemployment insurance benefits would be allowed to purchase an employer-provided FEHBP-like plan; alternatively, they could purchase it on their own if none is provided by their former employer. The plan would also provide tax benefits to workers aged 55 to 65 who purchase a health policy, are not covered by Medicaid, and do not have access to employer-provided insurance. All others who are uninsured could purchase an FEHBP-like health policy and claim a tax credit that would cap their payments at an amount that varies between 6 percent and 12 percent of their income. The cap is phased out for incomes greater than or equal to 300 percent of the poverty level.

Death Tax. The Kerry plan would not repeal the death tax, but it would raise the unified exemption level to $2 million (or $4 million per couple) by 2005 and thereafter. However, Senator Kerry does not lower the top rate from the current rate of 48 percent. This is a higher exemption level than was passed under EGTRRA for the year 2005, the same as for the years 2006ñ2008, and lower than for the year 2009. The Kerry plan, however, would make exceptions: The plan raises the exemption to $10 million per couple for returns containing a family-owned business or farm.

Kerry Plan for Business Taxes

Reducing the Corporate Tax Rate. The Kerry plan would reduce the corporate income tax by 5 percent, dropping the marginal rate from 35 percent to 33.25 percent.

Tax Repatriation Holiday. Senator Kerry plans a one-time tax break for companies that repatriate foreign income. Instead of paying the full tax on these profits, companies would pay a special 10 percent tax on profits that they bring back to the United States. The lower tax rate would apply only to repatriations that exceed a base amount that represents the normal amount of income that would have been repatriated without this tax change. The base amount would be determined by averaging the amounts that were repatriated in prior years. In addition, CDA analysts assumed that corporations would be permitted to reduce their U.S. tax liability by using a modified version of the foreign tax credit.

Partial Repeal of Tax Deferral on Overseas Income. The Kerry plan would also reduce the tax deferral of corporate income earned overseas. The objective is to tax profits from foreign subsidiaries in the same way that domestic profits are taxed even though domestic profits have not been subject to tax by foreign governments, which is the case with foreign-source income. The repeal is partial because the plan includes exemptions for some multinational companies that sell their products abroad. The tax plan allows these companies to defer income if they sell a product in the country in which it is produced. An estimated two-thirds of foreign income would qualify for deferral under the Kerry plan.^[13]

Targeted New Jobs Tax Credit. The Kerry tax plan includes a tax credit that would offset the employer's portion of the payroll taxes for new employees in certain business sectors. Small businesses, manufacturers, and businesses in outsourcing-related industries would receive a credit toward the amount of payroll taxes that they pay if their payroll taxes increase due to more employees. This credit is designed to boost hiring and employment in these sectors and would expire after two years.

Conventional Analysis of the Candidates' Tax Plans[14]

Revenue Effects Before Accounting for the Economy. Using conventional estimating methods, the Bush plan is estimated to reduce revenue by $1.1 trillion over a 10-year period from fiscal year (FY) 2005 to FY 2014. (See Table 1.) About 60 percent of this reduction comes from tax reductions associated with supply-side economic incentives. These incentive effects result from changes in the marginal tax rates on ordinary income and capital income.

In contrast to the Bush tax plan, the revenue effects of the Kerry proposal are smaller and a greater proportion is associated with demand-side economic effects. Based on conventional estimating techniques, the Kerry tax plan would reduce revenues by an estimated $686 billion over 10 years. (See Table 2.) Of this amount, over 75 percent is attributable to demand-side tax reductions.

Dynamic Analysis of the
Candidates' Tax Plans^[15]

Table 5 and Table 6 (Appendix 2) contain year-by-year (in fiscal years) results for key economic indicators from the CDA's dynamic analysis of the Bush and Kerry tax plans. All figures reported here are adjusted for inflation and referenced relative to projected U.S. economic performance under current law (the baseline).

The Bush Tax Plan. The CDA analysis found that the Bush tax plan would:

• Expand output. GDP averages $38.0 billion higher than the baseline during the first six years (through 2010) and an average of $111.3 billion per year thereafter. The early stimulus to GDP comes from accelerating demand-side provisions such as the $1,000 child credit, which operates mainly by boosting total consumption in the model. By 2011, the full effects of the Bush supply-side policy changes are evident: The GDP growth rate increases by nearly half of a percentage point in 2011 alone, principally as a result of retaining the improved incentives to work, save, and invest that were enacted in 2001 and 2003 but are scheduled to expire at the end of 2010.
  
• Increase employment. The employment level is higher than the baseline by 155,000 jobs in 2005, then 430,000 in 2006. The peak job increase over the baseline is 1.34 million additional jobs in 2012. The Bush plan, with its heavy emphasis on expanding the supply of labor and capital, raises the employment level by an average of 624,000 jobs per year during 2005ñ2014. Over the same time period, the average unemployment rate would be reduced by only 0.2 percentage points, mainly because lower tax rates cause labor force participation to rise almost as fast as employment.
  
• Increase disposable personal income. If the Bush tax plan is enacted, aggregate personal income is projected to average $58.0 billion above the baseline during the 2005ñ2010 period and $274.2 billion per year thereafter. After-tax income for a family of four would average $1,848 per year higher than the baseline according to the model.

The real key to the Bush plan is the reduction of tax burdens on capital, which should enhance total investment in new equipment. The Bush plan accomplishes this reduction by making permanent the tax cuts on capital (estate tax repeal, the lower tax on dividend income, and lower marginal tax rates generally) that are currently scheduled to expire.

If Congress permits these tax cuts to expire, per capita capital costs can be expected to rise, thus reversing the positive economic effects of their recent decline. The experience of the most recent quarters following JGTRRA shows that investment is expanding by over twice the historical average following the 2003 dividend and capital gain tax reductions.^[16]

Reductions in per capita labor costs are nearly as important to the economic results shown on Table 5 as the falling capital costs starting in 2011. Taxpayers received a reduction in the tax cost of additional labor in 2001. That meant that the trade-off between labor and leisure (where taking less leisure occurs when labor costs fall) changed in favor of labor. Not only does the fall in the costs of working tend to increase the supply of labor hours in the economy, but it also calls workers out of non-employment settings (e.g., home, school, and retirement) and into full-time or part-time employment.

The growth in household income that additional labor and lower taxes brings also stimulates household demand for goods and services. As Table 5 shows, household consumption expenditures rise significantly if the Bush economic plan is implemented. Not only does the growth of demand expand the level of gross domestic product and the income shares associated with a growing GDP, but it also boosts state and federal revenues. As noted on Table 5, federal revenues grow by $266.2 billion above the conventional revenue estimates due to increased economic activity. The Bush tax plan enjoys a feedback effect on federal revenues from increased economic activity of 23.5 percent over 10 years.

In sum, the simulation results indicate that the Bush plan would provide a strong stimulus to both the demand and supply sides of the economy, resulting in rapid GDP and employment growth with no significant inflationary pressure. Publicly held debt would grow by $987.1 billion over 10 years from 2005 through 2014, but long-term interest rates would increase by only 0.1 percentage point as a result. On balance, the simulation results demonstrate the positive impact that lower tax rates could have on the incomes of everyone residing and working in the United States.

The Kerry Tax Plan. Dynamic simulation of the Kerry tax and economic proposals, applying the same methodologies used in modeling the Bush plan, yields the following findings:

• Negligible impact during 2005-2010. The Kerry plan slows the U.S. economy, principally in employment and capital growth, and does nothing to improve GDP. The Kerry plan increases the $11 trillion annual GDP during the period by an annual average of $7 billion. Non-residential investment, inflation, and interest rates are also essentially unchanged during the period before 2011. The Kerry plan begins to exert its effect after 2010, when the Kerry proposal to extend most provisions of the 2001 and 2003 tax cuts to taxpayers with incomes under $200,000 takes effect. In other words, the Kerry tax proposals have their greatest economic effect only after making the 2001 and 2003 tax cuts permanent for most taxpayers, which happens late in the forecast period.
  
• Large impact after 2010. The Kerry plan would make many elements of JGTRRA and EGTRRA tax cuts permanent rather than let them expire during 2009ñ2011, thus extending some aggregate supply-side benefits of lower marginal rates on labor and capital income. In addition, these permanent provisions stimulate disposable personal income and personal consumption. In inflation-adjusted terms, both consumption and personal savings rise relative to the baseline during 2011ñ2014 by an average of $111.6 billion and $150.3 billion, respectively. Real disposable income per family of four would be $3,030 higher per year in 2011ñ2014. The net result is that GDP grows above the baseline by $63.6 billion in 2011, a sixfold increase over the impact in 2010. The average GDP improvement over the baseline during 2011ñ2014 is $82.3 billion.
  
• Employment seesaws. Employment surges after 2011 due to the factors discussed above, but the Kerry plan reduces potential employment growth in prior years. For the first six years (2005 to 2010), average total employment is below the baseline by 126,000. That average annual amount is equal to a 7.1 percent drop in the average annual increase in forecasted employment of 1,769,000 between 2005 and 2010. Once the permanent tax cuts become effective in 2011, employment levels rise by an annual average of 682,000 jobs over the baseline.
  
• Expanding budget deficit. The Kerry plan cuts tax revenue and results in a $637.8 billion increase in publicly held debt over the period 2005 through 2014. While this growth in federal government debt held by the public is 65 percent as large as that produced by the Bush plan (which comes in at $987.1 billion over 10 years), increasing debt has a limited impact on the economy according to the model. Long-term 10-year Treasury bond rates rise by slightly more under the Kerry plan than they do under the Bush plan (both during the full 10 years and after 2010), but that increase is due principally to lower productivity growth under the Kerry plan than under the Bush plan.

The Kerry plan, like the Bush plan, stimulates additional economic activity from 2011 onward by making the 2001 and 2003 tax cuts permanent for taxpayers with incomes below $200,000. That is, the Kerry plan reaps economic benefits once it embraces some of the supply-side tax reductions. As a result, the Kerry plan yields additional federal revenues beyond the conventional estimates. Increased economic activity reduces the 10-year change in revenues by $121.5 billion, or by 17.7 percent of the conventional costs. This means that the Kerry plan reduces federal revenues by $564.4 billion instead of $685.9 billion over the 10-year period 2005 through 2014. The Bush plan, on the other hand, results in an economic feedback in additional revenues of $266 billion, which reduces the static cost of the Bush plan by 23.5 percent.

While a portion of the Kerry tax proposals bears a strong resemblance to the Bush proposal (specifically, in dealing with permanency of the 2001 and 2003 tax legislation), Senator Kerry relies elsewhere on targeted tax increases and reductions to achieve his policy ends. For example, the Kerry two-year jobs credit is designed to boost short-term employment by extending a tax credit to small businesses. However, the two-year jobs credit Kerry proposes is economically dubious, although CDA analysts give it the benefit of the doubt by assuming it generates thousands of net new jobs. The jobs credit will be both expensive and inefficient because companies will be paid for new hires even when those hires would have happened anyway.^[17]

The jobs credit is emblematic of an approach that uses temporary rate reductions, credits, and deductions to boost household and business purchases. While tax policy changes like the education and health care tax credits in the Kerry plan boost short-term output and employment, they leave unchanged the incentives to work, save, and invest. By leaving long-term costs of working and investing the same after the temporary tax cuts as they were before, they also leave workers and investors with no tax-related reason to change their behavior. Thus, the growth rate of the economy remains virtually unaffected by demand-side tax cuts. As in the Bush plan, only the supply-side changes in the Kerry plan affect output and employment in a significant and sustained way.

The Kerry tax plan begins with two tax policy changes that are not well calculated to boost employment or the rate of economic activity. First, the tax increases on taxpayers with high incomes may well choke off the high rate of investment growth of recent years. This investment growth is one factor-perhaps even the major factor-behind the remarkable increases in per-worker productivity. Indeed, productivity growth and the income and wealth effects that stem from such growth may be the chief victims of the Kerry tax increase.

The second move, targeted demand-side tax cuts, fails to counter the economically dulling effects of the tax increases. Not until 2011 does the Kerry plan begin to produce economic gains comparable to the Bush plan, and then it does so only by adopting the key element of the Bush plan-making the 2001 and 2003 tax cuts permanent.

Conclusion

This analytical comparison of the competing tax plans underscores the common objectives and sharp policy differences between the two candidates. On one hand, President Bush and Senator Kerry share a desire for expanded economic activity and for achieving income goals: President Bush argues for giving all taxpayers a tax cut while Senator Kerry would essentially exclude upper-income taxpayers from most tax cut benefits.

On the other hand, the two candidates have advanced distinctively different approaches to tax policy. President Bush devotes the largest part of his 10-year tax reduction to changing the incentives to work and invest. Senator Kerry, however, devotes a much larger share of his tax cut to supporting the demand or consumption side of the economy. This emphasis is evident in the number of targeted tax cuts designed to reallocate tax cuts toward household and government spending and away from high-income taxpayers who also own the majority of resources available for investment. These supply-side and demand-side differences account for the distinctive economic and fiscal effects of the two plans.

William W. Beach is John M. Olin Fellow in Economics and Director of, Ralph A. Rector, Ph.D., is a Research Fellow and Project Manager in, Rea S. Hederman, Jr., is Manager of Operations and a Senior Policy Analyst in, Alfredo B. Goyburu is a Policy Analyst in, and Tim Kane, Ph.D., is Research Fellow in Macroeconomics in the Center for Data Analysis at The Heritage Foundation.

Appendix 1
Methodology

Overview

This appendix discusses how Center for Data Analysis analysts at The Heritage Foundation performed simulations to estimate the economic and fiscal impact of the Bush and Kerry tax proposals. CDA analysts assumed that all the proposals except the Kerry higher education and health care tax credits would take effect on January 1, 2005, and that the Kerry education and health care credits would take effect on January 1, 2006.^[18] In some cases, information currently available about the proposals did not provide enough details for CDA economists to conduct a quantitative analysis. In such cases, they made assumptions on how the proposals would be implemented.

Economic Models

CDA analysts used a version of the Global Insight (GI) U.S. Macroeconomic Model^[19] to analyze the macroeconomic fiscal and economic effects of each candidate's tax proposal. The model was adjusted so that its baseline fiscal and economic projections would be consistent with projections from the January 2004 Congressional Budget Office (CBO) budget and economic report.^[20] The CBO baseline forecast assumes that current law will be unchanged during the 10-year budget window. For example, the baseline assumes that tax-law changes resulting from EGTRRA will expire after 2010 because of the sunset provisions contained in the law. As a result, the CBO baseline serves as a neutral point against which to compare the effects of the two tax proposals.

CDA analysts also used the CDA personal income tax microsimulation model to estimate the change in year-to-year federal revenues for most of the individual income tax proposals. The model simulates the effect of tax law changes for a representative sample of taxpayers. Data for these taxpayers are extrapolated or "aged" to reflect detailed taxpayer characteristics through 2014. The data are aged so that they are consistent with the CBO baseline forecast from the GI model. For purposes of this analysis, the microsimulation produced conventional revenue estimates. In addition, some behavioral changes resulting from the change in capital gains and dividends tax rates have not been included. (Other forms of tax minimization behavior were included.)

Revenue estimates were calculated by comparing estimated federal receipts under current law to the estimated revenues that would be collected assuming that the candidate's proposals were adopted and the economy under the new law did not differ from the CBO baseline forecast. In general, CDA analysts converted the calendar year static revenue estimates, including those produced by the microsimulation model, into annualized quarterly estimates for use in the GI model.

Changing Regular Tax Rates

Revenue Estimate. The average effective personal income tax rate variable in the GI model was adjusted to produce static revenue estimates equal to those generated by the microsimulation tax model.

Economic Effects. Changes in marginal personal tax rates alter the after-tax return on the marginal dollar of labor income. Microeconomic theory suggests that increases in the marginal after-tax return on labor also increase the incentive to work and, therefore, labor force participation. CDA analysts simulated how changes in personal income tax rates would affect work incentives by estimating the amount that the labor force participation rate in the model would change in response to the individual income tax proposals.

A meta-study performed by the Congressional Budget Office found elasticity estimates ranging from 0.1 to 0.2.^[21] In other words, a 1 percent increase in after-tax labor compensation could cause a 0.1 percent to 0.2 percent increase in labor force participation. For this simulation, a 0.15 percent adjustment (measured as a share of the overall baseline labor compensation) was used to estimate the change in labor force participation.

Changing Capital Gains and Dividend Tax Rates

Revenue Estimate. CDA analysts used the microsimulation tax model to estimate differences in collections resulting from changes in tax rates on capital gains and dividend income for each candidate's proposal. The revenue effects of changing the capital gains and dividends tax rates were simulated in the GI model by adjusting the average personal income tax rate.

Economic Effects. Although the capital gains and dividend tax rates are applied to individual income, these proposals change the tax rate on income generated by the corporate sector of the economy. However, the GI model lacks a variable that measures personal taxation of corporate income. To simulate this provision, CDA analysts incorporated the economic effects of changes in personal taxation of corporate income by adjusting the top federal tax rate on corporate income without altering the average tax rate on corporate income. This approach allowed the revenue change resulting from each candidate's proposal to be represented accurately as a change in personal income tax collections; yet it also allowed the model to capture the effect that each tax change proposal would exert on the after-tax return to capital in the economy.

Separate percentage changes in capital gains and dividend tax rates were calculated by dividing the estimated revenue differences by the appropriate tax base. A weighted average of the change in the two rates was computed to represent the overall increase or decrease in taxation of corporate income resulting from personal income tax proposals. This weighted average took into account the share of each type of corporate income in total corporate income, the proportion of each type of corporate income that is taxable as individual income, and the estimates of the effective tax rate on each type of corporate income.^[22] Changes in this tax rate have a direct effect on the after-tax return to capital in the economy.^[23]

Extending Expiring Provisions and Creating New Tax Credits

CDA economists used the microsimulation tax model to estimate differences in collections resulting from making permanent the new 10 percent individual income tax bracket, the higher child tax credit, the expanded earned income credit for married joint filers, and marriage penalty relief.

Estimates for the revenue differences resulting from the Kerry health care tax credits were based on calculations by Professor Kenneth E. Thorpe of Emory University.^[24] Estimates for the revenue differences resulting from the Kerry higher education tax credit were based on calculations by Steve Robblee, Simone Berkowitz, and Isabel Sawhill of the Brookings Institution.^[25] For each of these proposals, CDA analysts modified the average effective per

Revenue Estimate. The revenue effects of the tax repatriation holiday were estimated by first projecting total stranded profits using independently produced estimates.^[26] Next, calculations were made to estimate the amount of offshore profits that would qualify for treatment under the Kerry plan and the amount that could feasibly be repatriated within the time period allowed by the plan.^[27] This amount was then reduced to account for estimated profits that would normally be repatriated within the year even without the proposal. The net amount of repatriated profits resulting from the Kerry tax plan was then adjusted by adding back the amount of foreign tax paid on these profits. The amount after the add-back represents income subject to the federal corporate income tax before the foreign tax credit is deducted. A base dividend amount-an estimate of corporate profits earned abroad that would normally have been repatriated within the year, even in the absence of the proposal-was subtracted from the net amount.

The effect on corporate income tax liabilities before credits was computed by multiplying taxable income (including the add-back) by the special 10 percent tax rate. The tax before credits was then reduced to take into account foreign tax credits used against U.S. corporate tax liabilities incurred on repatriated profits. The amount of these credits was estimated by assuming a weighted average foreign tax rate of 20 percent and an adjustment factor equal to the ratio of the repatriation tax holiday rate divided by the top corporate statutory tax rate. Estimates were also made to reflect the fact that some of the additional tax from repatriated profits reflects a timing decision. That is, some of the profits would eventually have been taxed within the budget window. This adjustment used the same assumptions as those used to estimate profits repatriated as a result of the tax holiday. The adjustment lowered the estimate of corporate tax collections slightly as compared to the baseline receipts for fiscal years 2007 through 2014.

Because the domestic corporate tax base is not affected by the proposal, CDA analysts accounted for the change in tax collections by modifying a variable in the GI model that takes account of differences between unified budget receipts and tax revenues collected on income flows as defined in the National Income and Product Accounts.

Economic Effects. CDA researchers took account of the additional after-tax income resulting from the repatriation holiday by estimating how corporations would use the amounts. Following the results of a survey taken by the Bank of America relating to a similar proposal,^[28] CDA analysts assumed that about 30 percent of the additional income would be used for non-residential investment, approximately 30 percent would be used to reduce corporate debt, and the remaining portion would be used in asset transfers and financial adjustments that are not taken into account in the model. Slight reductions in non-residential investment and slight increases in corporate debt were made for 2006-2014 to reflect the loss of repatriated profits assumed to be included in the baseline projections.

Partially Repealing the Deferral of Overseas Income

The Kerry proposal to partially end deferral of corporate income earned abroad would alter the income subject to the domestic corporate tax rate but not necessarily the amount of domestic profits. After consulting with representatives of Global Insight, CDA researchers simulated the revenue effects of the proposal by altering the corporate tax base equation in the GI model. The adjustment increased the corporate income tax base, and therefore the tax liability, by increasing the portion of income earned abroad that is included in the tax base. The estimated change in the portion of income included in the tax base was derived from calculations performed by Martin Sullivan in his Tax Notes analysis of the Kerry corporate tax change proposals.^[29]

Enacting a New Jobs Credit

Revenue Estimate. The Department of Labor's new series on business employment dynamics indicates that from 1992 through 2003, an average of 8.1 million new private-sector jobs were created on a gross basis every quarter, while 7.7 million were lost on average.^[30] CDA analysts used the gross number of private-sector jobs created to estimate the number of jobs generated under the baseline forecast at newly opened establishments and manufacturing establishments that are expanding their workforce.^[31] In addition, CDA analysts estimated the number of jobs at expanding small businesses.^[32] An adjustment was made so that jobs at expanding small businesses that are engaged in manufacturing would not be double-counted.

Because the credit reduces the cost of labor to business, there is the expectation that additional jobs will be created. Economists express the relationship between a percentage reduction in labor costs and the percentage of increased demand for labor as the wage elasticity of demand, which the literature suggests can range between -0.4 and -0.5.^[33] For example, with an elasticity of -0.5 and a 10 percent reduction in labor costs, the quantity of labor demanded would rise by 5 percent.

However, previous experience with earlier jobs tax credit laws suggests that this elasticity overstates the effect of such credits. This may be because the credits did not apply to the entire labor force and wages for all existing workers are unaffected. Moreover, the job credits brought about only a temporary reduction in the price of labor, causing employers to resist making permanent hires at levels above those already planned. As a result, hiring response expressed as an elasticity would likely be far closer to zero than -0.5. For example, the U.S. General Accounting Office (GAO) reports that the 1981 Targeted Jobs Tax Credit indicates that the credit "in all likelihood had zero net impact on the employment levels of the target group members."^[34]

Instead of using a zero response as could be indicated from the GAO study, CDA analysts used an elasticity near -0.1. The elasticity was applied to a typical qualifying job that pays wages of $28,500 annually. An employer with this job qualifies for a savings of $2,180 on the employer portion of the payroll tax. An additional assumption is that wages for qualifying employees constitute slightly more than 70 percent of the total cost of labor.

If fully taken up by employers, CDA analysts estimate that about 13.3 million jobs would be eligible for the new credit. This total includes qualifying jobs that would have been created even without the jobs tax credit and some that were created in response to the credit. The Kerry campaign has said that the program would be funded by revenues generated from the one-year corporate tax repatriation holiday, which it estimates at $22 billion.^[35] Thus, the campaign projects that the jobs tax credit would cost $11 billion per year for two years-roughly a third lower than the maximum potential cost.

CDA analysts also assume that the take-up rate by employers would be far lower than 100 percent. Employers would need to be aware of the credit, learn about its provisions, and produce, maintain, and process the necessary paperwork. Such awareness and take-up issues have limited the effectiveness of jobs incentive programs in the past,^[36] often because the government sets up high paperwork hurdles in order to limit program cost.

Due to limited awareness, compliance burdens, and non-applicability to firms without taxable profits, CDA analysts assumed a take-up rate of approximately 17 percent. This take-up rate implies that the credit would be claimed for about 2.3 million workers and would reduce revenues by about $5 billion per year for two years.

Economic Effects. CDA analysts computed the ratio of new jobs that potentially could be created as a result of the tax credit to the total number that potentially would qualify for the credit. If a similar ratio is applied to the estimated 2.3 million workers for whom the credit would be claimed, the new credit would raise the employment level by about 35,000 for two years. The variable in the GI model representing non-farm establishment employment was increased to reflect the net new jobs resulting from the jobs tax credit.

Modifying the Estate Tax

Revenue Estimate. CDA analysts used a Congressional Budget Office estimate of the revenue differences resulting from permanent abolition of the estate tax to simulate the revenue effects of the Bush estate tax proposal. This estimate assumed that the scheduled repeal of the estate tax would continue past 2010.^[37]

CDA researchers applied a set of equations to approximate the revenue differences caused by the Kerry estate tax proposals. Because the estate tax is a tax on wealth, not on income, CDA researchers reflected the revenue differences by adjusting a variable in the GI model that represents tax collections not related to income flows captured in the National Income and Products Accounts.

Economic Effects. CDA researchers estimated two behavioral effects associated with the estate tax proposals. First, CDA analysts estimated the economic effects of amounts spent on tax avoidance activities by owners of estates that might be subjected to the estate tax. It was assumed that tax avoidance spending by the individuals was a response to long-run rather than short-run changes in the estate tax law. Taking into account the longer time period reflects the assumption that tax avoidance activity related to the estate tax is optimized for the estate tax law prevailing in the year the estate owner expects to die, not for any year before that.^[38]

CDA analysts adjusted the GI model's price index for miscellaneous business services in order to reflect changes in the demand for estate tax avoidance services resulting from changes in the estate tax law. The business services price deflator variable was adjusted so that nominal spending on miscellaneous business services changed by 30 cents for every dollar of change in the long-run projection of federal estate tax collections.^[39] The inflation-adjusted consumption of miscellaneous business services was not assumed to change under either candidate's proposal for the estate tax.

A change in the cost of capital was also assumed to occur as a result of changes in the estate tax. Because it is a tax on capital, the estate tax increases the minimum rate of return sought by investors. This minimum return is assumed to be a factor in the decision to engage in new projects. All other things being equal, projects that do not have projected returns above the minimum will not be initiated.

Previous research indicates that if the estate tax had been repealed prior to 1997, the required return on investments would have fallen approximately 3 percent.^[40] CDA analysts reduced this estimate to reflect the 1997 reduction in top federal estate tax rates. The percentage was further adjusted in the simulation of the Kerry proposal to reflect the continued, albeit somewhat diminished, existence of the estate tax under his proposal. A variable in the model representing the 10-year Treasury bond rate was reduced to reflect a reduction in the minimum required rate of return on capital.

Federal Funds Rate

Variables in the GI model were set so as to allow actions by the monetary authority, as simulated in the model, to adjust the federal funds rate. With these settings, the federal funds rate tends to increase (decline) when the unemployment rate declines (increases), the Consumer Price Index increases (declines), or the Consumer Price Index accelerates (decelerates).

Employment Effects by State

Estimates for the number of new jobs by state were determined by taking the change in employment as determined by the CDA macroeconomic model and distributing them according to Bureau of Labor Statistics data on the level of state employment.^[41]

Appendix 2
Macroeconomic Modeling Results