In a recent column, I discussed some of the support for supply-side economics at the World Bank, the International Monetary Fund, and in academia. I would like to extend my discussion to some emerging research on labor supply that further supports the supply-side view of tax policy.
For decades, most economists took the view that taxation has little if any effect on work. This followed from the experience of World War II, where tax rates rose very sharply in all countries with little falloff in work effort. Also, shortly after the war, the Harvard Business School did a major study of the impact of taxes on corporate executives, which found that taxes had a minuscule impact on their hours of work.
Obviously, when people feel that their lives are at stake, they will make sacrifices that they would never make under normal circumstances, especially if they are necessary for only a limited time. Therefore, a study of labor supply during World War II tells us virtually nothing about how people react to high tax rates during peacetime.
Similarly, an examination of corporate executives tells us little about how average people behave, especially if the analysis is limited to work hours. Executives get a lot of perks and a psychological kick from being the boss, have flexible (if long) hours, and can probably adjust their income to compensate for tax changes in ways average workers cannot.
Nevertheless, the view that tax rates have no impact on labor supply was taken as given until the early 1980s. Economist Jerry Hausman of M.I.T. was the first to break ranks. In a path-breaking paper for the Brookings Institution in 1981, he showed that “the current tax system does significantly reduce labor supply.” He found that the labor supply of married males was 7 percent less than it would be if the same revenue were collected from a flat-rate tax system.
In the years since, economists have devised many new statistical techniques for measuring labor supply. They have also had a number of significant changes in tax structure over this period, providing data and opportunities to study the interaction between taxes and work. And better international data allows for comparisons between countries. The result is a growing consensus that taxes affect labor supply much more than previously assumed.
A study last year by economists Daniel Aaronson and Eric French of the Federal Reserve Bank of Chicago found that previous studies had understated the impact of progressive tax rates on labor supply in the U.S. by 10 percent for men and twice that for women.
Earlier this year, Syracuse University economists James Ziliak, Thomas Kniesner, and Douglas Holtz-Eakin (now director of the Congressional Budget Office) found that the labor response to tax changes was twice the standard estimate. Their research suggests that we could increase our living standard by 20 percent if we just collected the same revenue from a flat-rate system.
In 2001, economists Marco Bianchi, Bjorn Gudmundson, and Gylfi Zoega published a paper in the prestigious American Economic Review examining an interesting natural experiment in Iceland. Owing to a major tax reform, workers in that country in effect got a one-year reduction in their tax rate to zero. The economists found that during this time there was a significant jump in the labor supply of both men and women. After tax rates returned to normal, that supply fell back to its previous level.
In his Richard T. Ely lecture to the American Economic Association in 2002, economist Edward Prescott of the University of Minnesota concluded that almost all of the difference in living standards between the U.S. and France is accounted for by the impact of taxes on work. He notes that while the capital/output ratio is about the same in both countries, French workers work 30 percent less, due entirely to the much heavier French tax burden on labor. Prof. Prescott concluded that if France had the U.S. tax system, the French standard of living would immediately rise by 20 percent.
This result is confirmed by an examination of data from the Organization for Economic Cooperation and Development. I looked at total yearly work hours and taxes as a share of the gross domestic product in various countries. The data show clearly that the heavier the tax burden, the fewer the hours worked. In Mexico, the U.S., and Japan (where the tax burden is less than 30 percent of GDP), workers put in more that 1,800 hours a year on the job on average. In France and Belgium, where taxes take more than 45 percent of GDP, workers spend fewer than 1,600 hours on the job because it just isn’t worth the effort once taxes are deducted from their pay.
Common sense tells us that people work less when their after-tax reward is reduced. It’s good to see that economists finally agree.