A close look at the relationship
With President Obama reluctant to tout Obamacare or the 2009 fiscal stimulus, tax increases on the rich have at times seemed like the only idea he is willing to defend. At the heart of the president’s passionate and persistent calls for them is an idea of fairness, namely, that because America’s highest earners have benefited greatly from the genius of our system of government, they have a special obligation to pay more than they do at present.
This idea is widely shared by the public at large. A recent survey from the Pew Research Center found that 44 percent of Americans believe that a tax increase on households earning more than $250,000 would both boost the economy and make the tax code fairer. In contrast, only 22 percent believe that it would harm the economy, and only 21 percent believe that it would make the tax code less fair. Moreover, 64 percent of Democrats believe that such a tax increase would be a boon to the economy, so the politics of President Obama’s stance make perfect sense. What is less clear is that tax increases on the rich would actually encourage economic growth or significantly reduce the deficit.
The president often couches his advocacy for tax increases in the language of commonsense deficit reduction, and it is certainly true that restoring the Clinton-era tax rates across the board would dramatically increase federal revenues. But the president hasn’t called for restoring Clinton-era tax rates across the board. Rather, he has called for the preservation of the vast majority of the Bush-era tax cuts — the cuts that apply to households with incomes of less than $250,000 and individuals with incomes of less than $200,000, which together account for 80 percent of the revenue loss. As a result, while President Obama’s tax proposal says his plan will raise more than a trillion dollars over the next decade, it would fall about $2.35 trillion short of a full restoration of the Clinton-era tax rates.
The president has also called for curbing the use of itemized deductions by high-earning taxpayers, and Obamacare has raised the Medicare tax for individuals earning more than $200,000 and created the Unearned Income Medicare Contribution, a tax on investment income. The end result is that the rich will pay substantially higher taxes than they did during the Clinton years. There is a good deal of evidence to suggest that the president would be willing to go even farther in raising taxes on high-earning taxpayers, including his praise of the Gang of Six deficit-reduction plan, which was proposed in July of last year and would hike taxes by $2.3 trillion relative to current policy.
It is, of course, possible that President Obama doesn’t intend to raise taxes only on the rich. A number of his current and former advisers, including former White House budget director Peter Orszag, have called for tax increases on middle-income households as well. Noam Scheiber, a senior editor at The New Republic, suggests in his book The Escape Artists that the president has at various points found these arguments persuasive. But if we take the president at his word, he intends to raise taxes on the rich alone, primarily through sharp increases in effective marginal tax rates.
Would such a tax increase actually improve our growth prospects? Or would it badly undermine them? The standard left-wing position on taxes is that they don’t act as a brake on economic activity, particularly when levied on “millionaires and billionaires.” Liberal legislators and pundits — including President Obama — have referred to the Clinton economy to demonstrate that tax increases are not incompatible with growth. Conservatives, by contrast, tend to be more concerned with the negative impacts of taxation, but they often worry more about the short-term effects of higher taxation on job creation or struggling families than about long-term growth.
The problem with liberals’ invocations of the Clinton economy is that we don’t know how much the economy would have grown in the 1990s with lower taxes. Liberals seem to assume that the growth rate would have been the same, but they provide no reason to believe this. George H. W. Bush’s and Bill Clinton’s tax hikes were relatively modest, and so cannot be taken to demonstrate that raising taxes does not reduce growth. In addition, the Nineties economy benefited from an unusual circumstance in the form of a tech bubble. And in 1997, President Clinton and the Republican Congress approved a deep cut in the taxation of capital income, a measure that may well have mitigated the impact of tax increases. These factors make it difficult to reach conclusive judgments.
Meanwhile, conservatives aren’t necessarily wrong to argue that high taxes diminish freedom, harm working families, stifle job creation, and fuel the growth of government; but the more fundamental problem is that they discourage valuable economic activity and leave us all poorer — especially in the long run. The actual magnitude of these effects is an empirical question, and while there remains substantial disagreement among scholars, many important findings suggest a sizable role for taxation in influencing economic behavior.
The economic theory on this topic, known as the life-cycle theory of taxes, holds that taxes discourage economic activity by making it less rewarding. Higher tax rates alter the various economic decisions people make over the course of their lives — from the number of hours they work, to when they enter or leave the work force, to whether and how they invest, to whether they go to college, to whether they take big economic risks in the hope of a spectacular payoff. Collectively, these decisions have the potential to lower overall income.
One key metric for judging the impact of taxes is the elasticity of taxable income (ETI). This number expresses the degree to which taxable income decreases in response to tax increases. Suppose a taxpayer earns $100 that is taxed at the top marginal rate of 35 percent. If the government then hikes this rate to 40 percent, and the taxpayer doesn’t change his behavior, the government will collect $5 more from him than it did before. But in reality, the taxpayer may decide to work less or make more of an effort to find tax deductions, so that next year he reports less income. The ETI is a measure, calculated through a complex formula, of just how much less income taxpayers report. In this example, at an ETI of 0.4, the government will raise only $1.50 in additional revenue. The higher the ETI, the less money the government raises.
Economists come to different conclusions when they try to calculate the ETI for federal taxes. Emmanuel Saez (a Berkeley economist who has supported higher tax rates) and Jonathan Gruber (an MIT economist who has done the same) have estimated a value of 0.4. This figure is typical of such research, though economists have found a range of values. Estimated ETI values for top-end earners tend to be higher, at 0.5–0.6.
A key limitation of this research is that most studies estimate the effects of taxation over the short term and consider only those taxpayers who are in the work force. It seems reasonable to expect that the long-run impacts are larger than what researchers have found.
There is a close link between the ETI and the Laffer curve, the inverse-U-shaped relationship between tax revenue and tax rates that implies taxes above a certain rate discourage so much economic activity that they cost more money than they bring in. The ETI can be used to calculate this turning point. An ETI of 0.4 implies a revenue-maximizing rate of 61 percent, while an ETI of 0.8 implies a revenue-maximizing rate of 44 percent.
Although few economists believe that the U.S. is above the revenue-maximizing rate, and that therefore cutting taxes would increase tax revenue, maximizing revenue should not be the sole consideration in deciding tax rates. Higher tax rates carry costs even when they raise revenue for the government. As long as the elasticity of taxable income is higher than 0, any increase in taxation will result in a lower measurable amount of economic activity and reduce the amount of reported income. Additionally, the closer we are to the top of the Laffer curve, the harder it is to raise tax revenue by increasing taxes. The top of the Laffer curve is best regarded as the upper limit of feasible tax rates rather than as the optimal rate.
Another key finding from economists is that taxable income tends to be more responsive than overall income to tax changes. In other words, most of the revenue lost from higher taxation is a result of taxpayers’ finding more deductions rather than of their actually working less. One important implication is that simultaneously eliminating deductions and lowering tax rates — an approach that has been embraced by the co-chairmen of the Bowles-Simpson deficit commission, the Domenici-Rivlin debt-reduction task force, and the Romney-Ryan campaign — could substantially increase the efficiency of the tax system while yielding comparable revenue.
It would be a mistake, however, to completely ignore tax rates’ impact on the amount of work people do — that is, on actual economic activity. In the research on this topic, an important distinction is made between people who remain in the work force but change the number of hours they work and people who enter or leave the work force entirely.
Many liberal advocates of higher taxes point out that taxes tend not to change the behavior of people who are already working, at least not immediately. This should not be surprising — after all, very few people will want to quit working entirely in response to tax changes, and most workers are tied to a specific schedule in their current jobs.
The picture changes, however, in the long run. Raj Chetty, an economist at Harvard, has suggested that people slowly adjust their work hours downward when taxes rise — for example, by finding less demanding jobs. Chetty’s research suggests that a 10 percent increase in taxes could lead to a 5 percent decrease in hours worked.
Economists, including Princeton’s Richard Rogerson, have also found that tax rates can have large effects on individuals’ decisions to enter or leave the work force. And they have found that some workers are more tax-sensitive than others. Married women with children and people nearing retirement age tend to be particularly sensitive, since many such workers are on the fence about whether to participate in the labor force to begin with.
Some researchers have tried to measure the impact of tax rates on a variety of other life decisions, such as whether to start a business or go to college. Ideally, we could tally up all of the effects that tax rates have and arrive at a grand total. Unfortunately, that remains a challenge. There are many things that are still unclear about the overall effect of tax rates on GDP.
Still, we can get a general idea by looking at how different taxation policies play out in different countries. Edward Prescott, a Nobel laureate in economics, has suggested that the United States’ low tax rates encourage Americans to work more hours than their European counterparts, which in turn helps explain why the U.S. has higher GDP per capita than most European countries. To be sure, not all economists believe that taxes matter as much as Prescott has suggested, and it is difficult to definitively establish causality when so many other factors (e.g., regulations and unions) play a role. But Prescott’s research is nevertheless suggestive.
Whether or not taxes in general should go up is partly a political question rather than strictly an empirical one; the amount of tax revenue needed depends on the desired size of government. But as the U.S. struggles with the question of how to reduce its mounting debts and deficits — spending reductions, tax increases, or both? — there are some facts that should play a role in the discussion.
Until we reach the top of the Laffer curve, higher taxes bring in more revenue, and thus help to control our debt. But as we saw above, taxes also affect economic activity, and thus our GDP. So taxes affect the debt-to-GDP ratio in both the numerator and the denominator. Even if a tax increase raises revenue, it may lower GDP in a way that makes achieving long-run fiscal balance harder. This may help explain the finding of Harvard economist Alberto Alesina that it is easier to keep a balanced budget in the long run when adjustments are made through spending reductions rather than tax increases.
As we also saw above, however, we can improve our tax system by lowering marginal tax rates while eliminating some deductions. This would boost incentives for work and employment. The Tax Reform Act of 1986 succeeded in doing this, and it was passed with bipartisan support. The problem is that, much as barnacles stick to the hull of an aging ship, deductions and loopholes have riddled the tax code. The elimination of these inefficiencies represents low-hanging fruit that should be plucked for the sake of economic recovery.
In the long run, especially if higher tax revenue is needed to finance greater spending, tax policy should be designed to encourage rather than discourage economic activity. Relying more on consumption or property taxes and less on income and capital taxes would boost incentives for investment. Further reducing the marriage penalty could boost labor-force participation by women. Generally, the focus should be on keeping the tax base broad and tax rates relatively flat.
Finally, we need a stable institutional arrangement for deciding tax rates. Currently, they are set on a year-by-year basis, and Congress often does not finalize the new rate structure until close to the end of the year. This results in enormous uncertainty for taxpayers. The system sometimes yields momentary tax cuts — such as the Obama stimulus checks — but tax cuts that are not permanent do not meaningfully change incentives for economic activity. We need to reduce uncertainty and set the stage for long-run economic growth, as in 1986.
In this election, Democrats and Republicans are offering dramatically different visions of how to reform America’s broken system of taxation. President Obama has pledged to increase taxes on the rich, on the understanding that the gains from deficit reduction and increased public spending will outweigh any economic damage. Mitt Romney, in contrast, is committed to an overhaul of the tax code that broadens the tax base while lowering marginal tax rates.
Both candidates have left a great deal unsaid regarding how they intend to reconcile their various goals, and it is a near certainty that either would have to adjust his plans as economic circumstances changed. What we do know is that individuals respond to tax rates, and that the tax-policy choices we make in the coming months will have a great impact on our economic well-being for decades to come.
– Mr. Gupta is a Ph.D. candidate in finance at Columbia University’s Graduate School of Business.