Why is the 2003 tax cut working so much better than the 2001 tax cut? Why is the economy performing better, for instance, and why are tax revenues growing faster than projected today compared to what happened after the 2001 tax legislation? The answer is that not all tax cuts are created equal. Tax cuts based on the Keynesian notion of putting money in people's pockets in the form of rebates and credits do not work-and these are the tax cuts that dominated the tax legislation approved in May 2001. Supply-side tax cuts, by contrast, do improve economic performance because they reduce tax rates on work, saving, and investment. And since lower tax rates on productive behavior dominated the May 2003 legislation, it is hardly surprising that the economy has responded positively.

As the following comparison indicates, the 2001 and 2003 tax cuts yielded significantly different results:

• Economic growth since the 2003 tax cut has averaged nearly 4.4 percent on a yearly basis, compared to just 1.9 percent in the period following the 2001 tax cut.[1]
   
• Net job creation since the 2003 tax cut has averaged more than 150,000 per month, compared to declining job numbers in the period after the 2001 tax cut.[2]
   
• Tax revenues have grown by an average of more than 6 percent annually since the 2003 tax cut, compared to falling tax collections after the 2001 tax cut.[3]

To be sure, tax policy is only one of many government policies that impact economic growth. Moreover, exogenous factors such as the terrorist attack in 2001 influence economic performance. So it would be wrong to attribute all of the good news since May 2003 to the supply-side tax cut, just as it would be incorrect to blame the Keynesian tax cut for all the job losses and economic weakness between May 2001 and May 2003. Furthermore, not every provision of the 2001 tax cut was economically misguided and not every component of the 2003 tax cut was based on sound economic policy.

To reiterate, it is not enough merely to cut taxes. Tax reductions only benefit the economy if the "price" of engaging in productive behavior is reduced. Keynesians argue that rebates and credits boost growth by injecting purchasing power in the economy, but this simplistic analysis fails to realize that government withdraws an equal amount of purchasing power from the economy when it borrows money to finance the rebates and credits.

An examination of the major provisions of the two tax bills underscores the difference between Keynesian tax cuts and supply-side tax cuts. The 2001 tax cut, for example, included many provisions that had no positive impact on economic performance because of poor design. Other tax provisions in 2001 were based on supply-side principles, but implementation was postponed-which meant a concomitant delay in the pro-growth impact. A review of the major provisions in the 2001 bill illustrates the problem:

1. Tax rebate - No pro-growth impact. Depending on household status, taxpayers received $300 to $600 checks. These checks did nothing to improve economic performance because the rebate was not tied to economic activity.
    
2. Child tax credits - No pro-growth impact. Households received an additional tax credit of $100 per eligible child. The provision did nothing to improve economic performance because the credits were not tied to economic activity.
    
3. Ten percent tax bracket - Negligible pro-growth impact. The legislation created a 10 percent tax rate for income below $6,000 for individuals and $12,000 for married couples. This lower rate had a very modest pro-growth impact by reducing the marginal tax rate on people who earn very modest incomes.
    
4. Income tax rate reductions - Minor pro-growth impact. Income tax rates were lowered one percentage point for years 2001 to 2003, but the bulk of the income tax reductions were postponed until 2004 and 2006. This had the unfortunate effect of postponing the lion's share of the pro-growth impact.
    
5. Death tax repeal - Minor pro-growth impact. Immediate death tax repeal would have a very large pro-growth effect, but this is not what was approved in 2001. Instead, the legislation included a provision to gradually phase out the death tax. The death tax rate did drop immediately from 55 percent to 50 percent, but the rate will remain at least 45 percent until 2010, when it finally drops to zero (albeit for only one year). This delay in repeal means that the bulk of the economic benefit will not occur until 2010-and even that presupposes that the repeal is made permanent.

The 2003 tax cut was not perfect, but most of the major components were based on supply-side principles. And because the legislation focused on good tax policy, it generated much better economic results. A review of the major provisions in the 2003 bill illustrates the link between good policy and good results:

1. Accelerated child tax credits - No pro-growth impact. The child tax credit was increased to $1,000 per eligible child. This provision did nothing to improve economic performance since the credits were not tied to economic activity.
    
2. Reduced tax on new business investment - Significant pro-growth impact. The legislation immediately reduced the "depreciation" tax on new business investment and also expanded the amount of investment that small businesses could "expense" in the year when the cost was incurred, meaning no tax on new investment.
    
3. Accelerated income tax rate reductions - Significant pro-growth impact. The 2001 legislation included income tax rate reductions, but most of those reductions were postponed until 2004 and 2006. The 2003 tax cut made those lower tax rates effective immediately.
    
4. Reduction in double-taxation of dividends and capital gains - Significant pro-growth impact. The legislation included provisions that reduced the double-tax on both dividends and capital gains to 15 percent. These reductions took place immediately, thus ensuring no incentive to postpone pro-growth activity.
    
5. Payment to the states - Moderate anti-growth impact. The 2003 tax legislation included a $20 billion spending increase to subsidize state government spending. This provision resulted in a transfer of resources from the productive sector of the economy to the government.

Some politicians argue that taxes should be higher, but if they really want more money in Washington, they should argue for more tax cuts like the ones adopted in 2003. Tax revenues have risen much faster than inflation since the 2003 tax cut was enacted. The results for 2005 have been especially impressive. Revenues to date for the current fiscal year are up more than 13 percent compared to the previous period.

This does not mean that tax cuts "pay for themselves," but it does mean that the right kind of tax policy-lower tax rates on work, saving, and investment-will lead to faster economic growth. And faster economic growth means more income for the government to tax. In other words, the best way to generate tax revenue is to expand the "tax base."

But the purpose of good tax policy is not to give politicians more money to waste. Pro-growth tax cuts should be implemented to boost economic performance and expand individual opportunity. Indeed, to the extent that pro-growth tax cuts generate more income and a larger tax base, any additional revenue should be used to finance further reforms to bring America closer to a simple and fair flat tax.

Daniel J. Mitchell, Ph.D., is McKenna Senior Research Fellow in the Thomas A. Roe Institute for Eco­nomic Policy Studies at The Heritage Foundation.