 |
|
|
|
|
|
|
||
 |
||||||||
 |
||||||||
|
||||||||
|
|
|
|
|
|
||||||||||
 |
|||||||||||||||
|
August 2009, Vol. IX, Issue I The Laffer Curve: Understanding the Relationship The Laffer Curve shows the relationship between tax rates and tax revenues, with the insight that taxable income is not predetermined. The Laffer Curve is best known for demonstrating that, at a certain point, higher tax rates fail to produce more revenue, but the key insight is the much more modest point that changes in tax rates cause changes in taxable income, which leads to some level of revenue feedback. Politicians on both sides often exaggerate, with Republicans sometimes arguing that the Laffer Curve means that "all tax cuts pay for themselves." Some Democrats, by contrast, argue that tax policy has no impact on economic performance. This paper uses real world evidence to demonstrate that certain tax cuts can have a positive impact on economic performance and that "supply-side" tax cuts therefore do not "cost" the government much in terms of foregone tax revenue. This paper further explains how the Joint Committee on Taxation's revenue-estimating process is based on the untenable theory that changes in tax policy - even dramatic reforms such as a flat tax - do not effect economic growth. In other words, the current system assumes the tax rates have no impact on taxable income. Because of congressional budget rules, this leads to a bias for tax increases and against tax cuts. This paper explains that "static scoring" should be replaced with "dynamic scoring." By Daniel J. Mitchell When politicians in Congress consider legislation to change tax policy, they rely on the Joint Committee on Taxation (JCT) to estimate how the changes will cause future tax revenues to rise or fall. The JCT uses a methodology known as static scoring which assumes that changes in tax policy have no impact on overall economic performance. This greatly simplifies the revenue-estimating process. For example static scoring allows the JCT to assume (with some modifications) that a 50 percent increase in tax rates will cause tax revenues to rise by about that amount. Likewise, a 50 percent reduction in tax rates will cause tax revenues to drop by a similar magnitude. Simplicity, however, is not the same as accuracy. Static scoring is widely criticized because there is considerable real-world evidence that tax policy does affect overall economic performance – which means that simplistic revenue-estimating models generate misleading numbers. An increase in tax rates, for instance, is likely to cause a reduction in taxable income because of both decreased incentives to earn income and increased incentives to hide income from tax authorities. This means a 50 percent jump in tax rates, for instance, will not generate a similar increase in tax receipts. Simply stated, it is unrealistic to calculate the revenue impact of changes in tax rates (especially a major change) without also estimating likely changes in taxable income. This approach – looking at the relationship of tax rates, taxable income, and tax revenue – is known as dynamic scoring. It is more challenging because there is considerable disagreement about the degree to which changes in tax rates affect incentives to earn income and disincentives to report income. But if policy makers want data that is closer to the truth, dynamic scoring is the right approach. Outside of the technical tax journals, the debate between static scoring and dynamic scoring usually is characterized as a fight about the Laffer Curve. Reviewing the Theory The Laffer Curve is a visual illustration of a dynamic scoring model. Named after Art Laffer, an economist who served as an advisor to Ronald Reagan, the Laffer Curve assumes that taxable income changes when tax rates are altered. And at some point, the reduction in taxable income from higher tax rates becomes large enough to completely offset the higher tax rate.This is the point where tax rates exceed the "revenue-maximizing" level. This means that there is a parabolic (though presumably not symmetrical) relationship between tax rates and tax revenue, unlike the linear relationship assumed by static scoring. Figures 1a and 1b compare the Laffer Curve with the relationship between tax rates and tax revenue assumed by a pure static-scoring model.
Larger Figure
Larger Figure In simple terms, the Laffer Curve is based on the presumably uncontroversial proposition that government won't collect any revenue if tax rates are zero. That's Point A. But neither will government collect any money when tax rates are 100 percent, which is Point C. After all, who is going to work if politicians seize every penny you make? There is a revenue-maximizing tax rate, identified in the illustration as Point B. The Laffer Curve became a part of the public policy debate during the 1970s because the top federal tax rate, which was then 70 percent, was so high that it discouraged the people in that tax bracket from engaging in productive behavior and it encouraged them to figure out ways of hiding income from the IRS. Proponents of the Laffer Curve explained that governments could lower the rates and actually collect more revenue since people would have more incentives to both earn additional income and to report that money to the IRS. In other words, the 70 percent tax rate was so high that America was on the wrong side – or downward sloping side – of the Laffer Curve, somewhere between points B and C. This meant, of course, that it was possible to lower the tax rate and collect more tax revenue, as demonstrated in Figure 2.
Larger Figure Using a hypothetical example, imagine that people in the highest tax bracket are willing to report $100 billion of taxable income when the rate is 70 percent, but they are willing to report $300 billion when the tax rate is 28 percent. As shown in Table 1, this means the lower tax rate generates more revenue. This is why proponents of the Laffer Curve in the 1970s argued that lower tax rates were a win-win proposition. Conservatives would be happy because lower tax rates would boost growth, and liberals would be happy because politicians would have more money to spend.
Larger Table With this understanding of the basic concept, it is useful to consider some of the implications of the Laffer Curve – and also to dismiss some of the myths. 1. Notwithstanding the previous example, and notwithstanding the exaggerated claims of some politicians, the Laffer Curve does not mean that all tax cuts pay for themselves. Indeed, it is only in very rare cases that this happens. There's pretty good evidence, as will be discussed later, that tax collections from the rich rose when Reagan cut the top tax rate from 70 percent to 28 percent. There's also lots of data showing that reductions in capital gains tax rates have increased tax receipts, largely because taxpayers easily can avoid the levy by holding onto assets when the rate is too high, but they are willing to sell assets and pay a tax when the rate is reasonable. But in most cases dealing with personal and corporate tax rates, nations are on the left side, or the upward sloping side of the Laffer Curve. This doesn't mean that lower tax rates are a bad idea when a nation is between points A and B. The economy will improve and taxable incomes will rise. But the increase in taxable income will not be enough to offset the effect of the lower tax rate. There will be revenue feedback, in other words, but not enough to make a tax cut self-financing. 2. The amount of revenue feedback varies depending on how you cut taxes. Lower marginal tax rates (often referred to as supply-side tax cuts) on productive behavior are likely to have the largest impact. Examples include income tax rate reductions, capital gains tax rate reductions, and dividend tax rate reductions. These policies generate Laffer Curve effects because they reduce the tax penalty on productive activity and people respond by working more, saving more, and investing more. The result is more taxable income. The actual level of revenue feedback depends on the situation, of course. Other tax cuts, though, such as expanded credits, deduction, and exemptions, are unlikely to have any significant impact on incentives to engage in productive behavior. This is because the marginal tax rates on additional increments of work, saving, and investment presumably are unchanged. This doesn't necessarily mean these are bad tax cuts. It just means that they don't lead to meaningful changes in taxable income, so there is little or no revenue feedback. 3. The revenue-maximizing point on the Laffer Curve is not ideal policy. As shown in Figure 3, the tax rate that maximizes revenue on the Laffer Curve is higher than the tax rate than maximizes growth. The growth-maximizing tax rate will be somewhere on the upward sloping section of the curve – though it is important to state that even a very low tax rate is bad for growth. The reason the growth-maximizing point is greater than zero is because that low tax rate (at least hopefully) is financing core public goods that help a market economy function – things such as rule-of-law, public safety, and honest courts. In other words, taxation is harmful but the net effect of overall fiscal policy is positive. Once tax rates climb above this level, however, the economy suffers because of both the higher tax rate and the increased burden of government spending. But tax rates still are not at punitive levels, so government can collect more revenue until tax rates approach Point B. As a theoretical concept, the Laffer Curve is indisputable. Even John Maynard Keynes wrote that "a reduction of taxation will run a better chance than an increase of balancing the budget."2 But we also know that some people on both sides of the debate exaggerate. Yes, there are a few tax cuts that may pay for themselves, but the vast majority of tax cuts are not in that category. And it's also true that there are some tax cuts that generate zero revenue feedback, but those also are rare cases. The same is true, of course, for tax increases. A few types of tax rate increases lose money. Others, though, are capable of generating considerable revenue. It all depends on the tax that is being raised. As a general rule, tax increases usually collect more money, but not as much as predicted by static revenue-estimating models.
Larger Figure The Laffer Curve in Practice The previous section discussed the Laffer Curve from a theoretical perspective and used a hypothetical example to show how the government could collect more money with a 28 percent tax rate than with a 70 percent tax rate. Are there real-world examples, though, of the Laffer Curve being so strong that the government actually collected more money after reducing tax rates? The IRS periodically publishes a document known as "The Statistics of Income," which provides considerable detail about tax revenues. Looking at the IRS numbers for 1980 and focusing on the returns showing taxable income above $200,000, there are nearly 117,000 rich people.3 These taxpayers – the ones hit by the 70 percent marginal tax rate – reported more than $36 billion of taxable income that year and the IRS grabbed more than $19 billion of that amount.4
Larger Table So what happened in 1988, when the top tax rate had dropped to 28 percent? The IRS numbers are remarkable. The number of rich people jumped to nearly 724,000 and they reported nearly $353 billion of taxable income above $200,000.5 The government's slice of that much-larger pie was more than $99 billion, five times as much revenue as was collected when the top tax rate was 70 percent. While those numbers illustrate a powerful Laffer Curve effect, a few caveats are appropriate. Some of the additional revenue is probably the result of population growing by about seven percent during that period. There also was inflation of about 44 percent during those eight years. And perhaps most important, even if Reagan hadn't reduced the top tax rate, there would have been some increase in the number of rich people and the amount of income they reported. Reagan's reduction in the top tax rate might not have been responsible for the rich paying five times as much. It might not even have been responsible for the rich paying three times as much. But almost surely this was one of the rare cases when a lower tax rate does "pay for itself." There are other examples of what could be called Extreme Laffer Curve responses. Ireland used to have a 50 percent corporate tax rate. That corporate rate in 1985 collected tax revenues equal to 1.1 percent of GDP.6 By 2006, as the chart shows, the tax rate was down to 12.5 percent and revenues were 3.8 percent of GDP.7 And what's really remarkable is that GDP was more than three times bigger – and that's after adjusting for inflation. So Ireland's government is getting a much bigger slice of a much bigger pie even though the tax rate is much lower. Actually, the insight of the Laffer Curve is that the government in some cases can get a bigger slice because the tax rate is lower.
Larger Figure Another example is Russia, which used to have a so-called progressive tax system with a top rate of 30 percent. This wasn't too surprising. After all, Karl Marx was one of the first advocates of penalizing successful people with higher tax rates. But in a dramatic reform, Russia implemented a 13 percent flat tax in 2001. Did this result in less revenue? Not according to the data. Receipts from the personal income tax have skyrocketed, jumping from 174.5 billion rubles in 2000 to 930.4 billion rubles in 2006.8 The chart shows that inflation-adjusted personal income tax revenues have been growing by an average of nearly 19 percent annually.
Larger Figure It's time, once again, for some more caveats. To help illustrate how the Laffer Curve works, this paper has cited extreme examples. These are a few of the rare instances where tax rate reductions result in more revenue, what might be called the "Extreme Curve Effect". But as we discussed earlier, the vast majority of tax cuts don't give the government more money. Instead, the revenue feedback is more modest, meaning that the growth in taxable income is not enough to compensate for the effect of the lower tax rate. We'll call this more likely result a "Regular Laffer Curve Effect." And in a few cases, where tax cuts aren't designed to improve incentives to earn taxable income, there is no revenue feedback at all. It is also worth mentioning that the Laffer Curve works in reverse. If politicians increase tax rates – especially if they do something really destructive such as boosting the top tax rate and punishing entrepreneurs and investors, people will earn and report less taxable income. A recent example comes from the state of Maryland, which imposed a higher tax rate on the so-called rich in 2008. This tax increase supposedly was going to generate an additional $100 million for the politicians. Instead, tax collections from rich people fell by more than $100 million.9 To be sure, not all of that lost revenue can be attributed to the higher tax rate. The deteriorating national economy presumably probably deserves most of the blame. But it also is true that the higher rate led to behavioral changes consistent with the Laffer Curve, including less incentive to earn income, more incentive to hide income, and (perhaps very significant) a financial reason to move to a state with a low or zero tax rate on income. One of the most classic – and tragic – examples of the Laffer Curve had nothing to do with income tax rates. In 1990, as part of President Bush's surrender of his read-my-lips-no-new-taxes promise, he agreed to a so-called luxury tax on yacht purchases. This punitive tax supposedly was going to make rich people pay more money, but they bought fewer boats, or at least they bought fewer boats in the United States, and the government collected far less money than projected. But that's just the beginning of the story. Lots of boatyards lost business, so they generated less income for the government to tax. And a lot of middle-class workers in those boatyards lost their jobs, meaning not only that they had less income to tax, but also that some of them started relying on government handouts, so it was a lose-lose situation for the budget.10 Disentangling all these different effects is not easy, but it's quite likely that the luxury tax was a net revenue-loser for the government – a reverse case of the "Extreme Laffer Curve Effect." Based on the evidence, it is clear that tax policy affects behavior. Yet amazingly, the revenue-estimating process for tax legislation ignores this fact. Regardless of the magnitude of the reform, the revenue-estimators assume changes in tax policy have no impact on the economy and no meaningful impact on taxable income. As a result, this bizarrre approach creates a bias for bad tax policy and higher tax rates. Fixing Washington's Bizarre Revenue-Estimating Methodology The Laffer Curve shows how pro-growth changes in tax policy lead to more taxable income, which means that tax rate reductions generate revenue feedback. Conversely, we know that tax rate increases will hurt economic performance. And since this translates into less taxable income, it means tax hikes, at the very least, do not raise as much money as politicians expect. This is common sense for people in the real world. Business owners know they would lose customers if they doubled their prices, so they would never assume that this is a realistic way to double revenue. Likewise, entrepreneurs always look for efficiencies, because they understand they can increase total profits if they can lower prices and attract more buyers. Unfortunately, the revenue-estimating system in Washington lacks these common-sense rules. When Congress debates tax legislation, it relies on the Joint Committee on Taxation to prepare official revenue estimates. This Committee, which is controlled by politicians from the tax-writing committees, assumes that changes in tax policy have zero impact on the economy's overall performance. The JCT even admits on its website that "the Joint Committee staff assumes that a proposal will not change total income and therefore holds Gross National Product fixed."11 Let's look at a couple of examples to understand what this really means. If Congress is debating a bill to double income tax rates, the Joint Committee on Taxation will assume that the economy's growth is unaffected even though such a proposal would have a crippling impact on incentives to work, save, and invest. The Joint Committee on Taxation would even assume there is no "macroeconomic" impact if the internal revenue code is put in the shredder and replaced by a simple and fair flat tax. It doesn't matter that growth has expanded and that more jobs have been created in the countries that have adopted a flat tax. The JCT ignores real-world evidence and instead relies on simplistic models. Defenders of the status quo claim that the JCT does dynamic scoring, but this is only true in the very limited sense that the models incorporate what are called microeconomic effects, such as people using more tax preferences to protect their income when tax rates increase.12 But this is like guessing who won a baseball game by looking at the score in the first inning. Yes, it's part of the answer, but only a tiny fraction of the information needed. Here's a real-world example. Back in 1989, the ranking Republican on the Finance Committee, Senator Bob Packwood of Oregon, sent a letter to the JCT, asking how much tax revenue would be raised if the government confiscated every penny of income above $200,000. The JCT's response was stunning. As the Senator explained in his November 14 floor statement, the JCT estimated that this 100 percent tax rate would collect $104 billion in 1989, rising to $299 billion in 1993.13 And when Senator Packwood asked the bureaucrats whether this was realistic, they gave him the same revenue estimate, but included a footnote stating "that these estimated taxes do not account for any behavioral response."14 So why does this odd system exist? There are two explanations. The charitable explanation is that dynamic scoring is difficult. If you ask five economists to estimate how much faster the economy will grow under a flat tax, for instance, you'll probably get six different answers. So how, then, is the Joint Committee on Taxation supposed to measure revenue feedback? Another challenge is disentangling the effects of other policies. If politicians raise taxes and adopt protectionist policies at the same time, the economy will be hit pretty hard, and it would be difficult to figure out which bad policy deserves which share of the blame. During the 1930s, for instance, Presidents Hoover and Roosevelt both expanded the burden of government. Federal spending rose dramatically. Tax rates were increased. Government regulation and intervention became pervasive. And Hoover (though not Roosevelt) approved protectionist legislation. Not surprisingly, the economy remained weak during the entire decade and tax revenues were stagnant. Because so many bad policies were implemented at the same time, it would be extremely difficult to isolate the damaging impact of the higher tax rates. These are some of the reasons why dynamic scoring will never produce a 100-percent correct revenue estimate. But the key thing to understand is that it will produce an estimate that is much closer to the truth than static scoring. Let's now shift to the less-benign reason why dynamic scoring isn't being used. Simply stated, some people like the fact the current system is rigged against good tax policy. Congressional budget rules are designed to make it difficult – at least on paper – to approve legislation that increases the budget deficit. And since the JCT routinely over-estimates the revenues that can be obtained by raising tax rates and likewise exaggerates the revenues foregone when tax rates are lowered, static scoring tilts the playing field in favor of bigger government. This is why advocates of higher taxes and more spending are dogmatically opposed to dynamic scoring. Finally, it's worth noting that the Joint Committee on Taxation refuses to make its revenue-estimating model public. Instead, the JCT operates in a totally nontransparent fashion, even though taxpayers pay their salaries and finance their so-called model. Perhaps the defenders of the status quo are against transparency because there's no way to defend the static scoring in an open debate. Conclusion There is considerable evidence showing that the Laffer Curve provides valuable insights that could be used to significantly improve the revenue-estimating process. Modernizing that process is especially important now that lawmakers seem determined to impose higher tax rates on productive behavior. But if those policy makers understand that many of those tax-rate increases will not yield much revenue, perhaps they would reconsider adopting anti-growth policies. __________________________________ Daniel Mitchell is a Senior Fellow at the Cato Institute (cato.org), a free-market think tank located in Washington, DC. He also co-founded the Center for Freedom and Prosperity Foundation and serves as the Chairman of its Board of Directors. __________________________________ The Center for Freedom and Prosperity Foundation is a public policy, research, and educational organization operating under Section 501(c)3. It is privately supported, and receives no funds from any government at any level, nor does it perform any government or other contract work. Nothing written here is to be construed as necessarily reflecting the views of the Center for Freedom and Prosperity Foundation or as an attempt to aid or hinder the passage of any bill before Congress. Center for Freedom and Prosperity Foundation, the research and educational affiliate of the Center for Freedom and Prosperity (CFP), can be reached by calling 202-285-0244 or visiting our web site at www.freedomandprosperity.org.
End Notes: 1 This study is designed to accompany the Center for Freedom and Prosperity's Video Trilogy on the Laffer Curve, which can be watched by visiting http://www.freedomandprosperity.org/videos/laffercurve1-3/laffercurve1-3.shtml 2 John Maynard Keynes, The Collected Writings of John Maynard Keynes (London: Macmillan, Cambridge University Press, 1972). 3 http://www.irs.gov/taxstats/indtaxstats/article/0,,id=96981,00.html 6 http://www.oecd.org/dataoecd/48/27/41498733.pdf 8 Anne Ivanova; Michael Keen; and Alexander Klemm. The Russian 'flat tax' reform // Economic Policy, Vol. 20, No. 43, pp. 397-444, June 2005. 9 Laura Smitherman, Maryland plan to tax millionaires backfires: Top earners disappearing as economy withers, The Baltimore Sun, May 14, 2009 10 Joint Economic Committee, 1992 Annual Report, April 1992. 11 http://www.jct.gov/about-us/revenue-estimating.htm 12 For good descriptions of the current revenue estimating process, see joint hearing, Review of Congressional Budget Cost Estimating, Committee on the Budget, U.S. House of Representatives, and Committee on the Budget, U.S. Senate, Serial No. 104-1, January 10, 1995, and Jane G. Gravelle, "Dynamic Revenue Estimating," CRS Report for Congress, Congressional Research Service, December 14, 1994. 13 Congressional Record, November 14, 1989, p. S15534. 14 Letter to Senator Robert Packwood from Joint Committee on Taxation, November 15, 1988.
|
|||||||||||||||
 |
|||||||||||||||
