Give Me Shelter

The responsiveness of taxpayers to changes in marginal tax rates has become perhaps the most central issue in public finance, and nowhere is the debate more heated than at the very high end of the income distribution. Yet it seems there is little direct evidence on the rich and their money. Data used to study tax responsiveness of the rich – based on actual tax returns and limited government data – has suffered from serious flaws which make analysis difficult. Now, by compiling one of the most comprehensive data sets ever gathered on the rich, assistant professor of economics Austan Goolsbee at the University of Chicago Graduate School of Business has uncovered striking evidence that challenges conventional wisdom on how the rich respond to taxation.

A common view among economists is that people with high incomes are more responsive to tax changes than people who earn less. The rich presumably have more discretion over the form their income takes -- in salary, stock options, non-taxable perks, capital gains and dividends -- than those workers who depend primarily on wages and salaries. High-income Americans can choose how they will receive their taxable income and how they want that income to be labeled, enabling them to reduce their tax bills. Raise their taxes, and the rich will respond by cutting back on efforts to bolster their businesses and their own incomes. This damages the economy, hurting both rich and poor.

Several studies imply that since high-income people respond to changing tax rates by reporting less income, raising their taxes is an inefficient way for the government to raise revenue. Many economists speculate, then, that the Laffer Curve -- which states that tax revenues will rise when tax rates are lowered -- may apply to high income people, even though it does not hold true for the overall population. If you lower tax rates for the rich, they will not hide their money; they will work harder, invest more, and ultimately generate revenue. Everyone will benefit.

In 1981 President Ronald Reagan and Congress banked on this principle and cut the maximum federal income tax rate from 71 to 50 percent. In Reagan's second term, the rate fell further, to 28 percent. While the economy and stock market boomed throughout the 1980s, subsequent research shows that this surge mainly benefited wealthy Americans. Huge deficits were created.

Bill Clinton was elected president in 1992 with the promise to raise taxes on high-income Americans. In 1993, Congress raised the marginal tax rate from 31 percent to 39.6 percent on income greater than $250,000. How did the rich respond to this increase? Goolsbee notes that upon hearing of the tax increase in 1992, the wealthy ran for cover, and taxable income fell significantly in 1993. In order to assess the true nature of this response, however, Goolsbee needed to collect data which could distinguish between permanent changes in a rich person's taxable income and temporary effects that are due to timing alone.

"The revenue implications of tax cuts hinge on whether taxable income changes are changes to the form of compensation or to the timing of compensation," says Goolsbee. "But this has been largely untestable up to now."

Goolsbee collected data for more than 10,000 of America's top executives by looking at proxy statements for companies listed in Standard & Poor's 500-stock index, the S&P MidCap 400 and the S&P SmallCap 600. These proxy statements, which by law must publish the compensation of a company's top five executives, include complete information not only on wages and salaries of top executives, but on other forms of compensation such as bonuses, stock options, and even perks. Between 1991 and 1995 the average taxable income of these 10,000 executives, earned at their firms in 1992 dollars, was $852,000.

Goolsbee found that although these executives make up only 1 percent of America's one million highest-income taxpayers, they accounted for 21 percent of the total change in the taxable income of that group after the 1993 tax hike. One executive alone -- Michael Eisner, chief executive of Walt Disney Company -- was responsible for more than 2 percent of the fall in taxable income among the top one million taxpayers in 1993. The reason is that Eisner cashed in almost $200 million of stock options on November 30, 1992 and none in 1993.

"The general theory is that if there is a big response of rich people to taxation, it means that a progressive tax code is extremely expensive. It doesn't generate a lot of revenue, but it does generate a lot of inefficiency," says Goolsbee. "It leads people to put their money in inefficient forms or spend their time hiding their money, rather than doing things to help the economy."

"So people think that it is counterproductive and inefficient to tax the wealthy, that a progressive tax structure is bad for the economy. But I believe that claim is based on evidence that is wrong," adds Goolsbee. "Essentially, I show that those big responses are short-term responses, and when you look at long-term responses, taxing the rich generates about the same inefficiencies as taxing the middle class or anyone else."

Although there was a pronounced decline in taxable income after the 1993 tax hike, this decline was a temporary shift in the timing of compensation. According to the data, high-income executives saw their average taxable income drop significantly from 1992 to 1993, falling by $179,000, or almost 16 percent. But looking more broadly at 1991 to 1995, however, this 16 percent drop followed a tremendous rise of 27 percent, or $242,000, from 1991 to 1992. In addition, the 1993 drop was not sustained.

"The raw data show clearly that the change to taxable income in response to taxation was comprised of a substantial increase in the exercising of options in 1992 followed by a dramatic decrease in 1993," says Goolsbee, "which is highly suggestive of a simple timing shift. The majority of this action is concentrated among executives at the top of the income distribution. They simply juggled their income between years." Executives without stock options show six times less responsiveness to taxation.

It is clear that taxing the rich can lead to dramatic shifting of taxable income in the years immediately surrounding a tax change. Tax changes may allow many to avoid taxation for a short period of time. Once the dust settles, however, the total reduction in their taxable income may be modest. Goolsbee believes the efficiency loss for the Clinton tax increase as estimated in recent tax responsiveness literature may be significantly overstated.

"If you look over a longer period, beyond 1992 and 1993, you see a pretty modest effect of taxes on the rich," he adds. "Which is not to say, 'We ought to increase taxes on the rich.' I am not stating that. What I demonstrate is that taxing the rich has just as much inefficiency -- almost exactly the same inefficiency -- as taxing anyone else. About 1/4 of the revenue generated from rich people's taxes is burned in inefficiency, and that's almost exactly what we've estimated to be the inefficiency generated from other taxes."