January 2005
Taxes and the Job Market
At a January 14 AEI conference, Steven Davis, a professor of international business and economics at the University of Chicago's Graduate School of Business, presented his research into the long-term effects of different tax rates in various countries. Davis found that taxes affect work activity directly through labor supply, labor demand, and also indirectly through government spending, which is determined by available tax revenue such as unemployment benefits or Social Security. Davis found that higher tax rates on salaries and on household spending can lead to less work time, more time spent on unpaid work at home and on leisure, a larger untaxed underground economy, a smaller national output, and less employment in industries that rely heavily on low wages and low skilled labor. Davis's findings suggest that differences in the tax rates among major developed economies are a primary reason for large international differences in working hours and in types of available jobs.
Steven Davis
University of Chicago
I believe that taxes on labor income and consumption expenditures have several negative, long-term effects on the economy. I believe taxes on labor and consumption lead to tax avoidance and evasion on many different levels. Taxes on these activities lead people to reduce their number of work hours and revert to untaxed, non-market production activities, such as leisure and household activities and greater use of underground markets.
In order to study the long-term effect of different tax rates, we related the use of these non-market activities to tax rates in different countries. Our study looked at nineteen countries, using mainly OECD data. The size of the shadow economy was estimated using to different models: the Currency Demand Model and the Electricity Model.
The regression analysis of the data showed a correlation between the variables. Our study found that higher tax rates resulted in a decreased number of hours worked per employee. We also found that an increase in rates resulted in an increase in the percentage of GDP held by the shadow economy. A unit standard deviation tax hike leads to 122 fewer market hours worked per year, a 4.9-percentage drop in the employee-to-population ratio, a 3.8-percent increase of the shadow economy’s share of the GDP, and a 10- to 30-percent decrease in value-added and employment shares in several industries.
The tax differences affect labor through a couple of different paths: direct effects on labor supply, shifting away from demand of workers with elastic supply, and through government programs that discourage labor supply.
Kevin A. Hassett
AEI
I contend that there is a correlation between the labor market differences among countries and relative tax rates. I believe this is because higher taxes move employees away from market production activities and towards shadow production, or those labor efforts that are not observed by tax authorities. In order to investigate the correlation further, I am currently working on a paper to look at the possible relationship between the differences between countries and capital. We believe a link may exist between the differences and the use of and taxes on capital; high taxes on labor and capital may be correlated. Capital flows that increase labor demand should increase hours worked per employee. Indeed, it could be that capital taxation is more responsible for labor supply movements, given the high sensitivity of capital and a possible correlation between capital flows and labor supply.
Lee Price
Economic Policy Institute
This paper is interesting, indeed it is provocative to a social Democrat who would positively respond to some policy implications that this paper presents. Moreover it addresses the readily observed notion that Europe works less.
There is however, a fundamental problem with the study. First, it is important to consider that countries with higher taxes maintain significant institutional differences from low- tax countries beyond their respective tax rates. For example, Europe has customs and laws on working hours that directly effect working hours per adult. Additionally, employer expectations may differ among countries. These are factors that affect labor that are not explained by taxation. Second, the study relies on data from a limited number of countries. Commendably, the study is honest about the limitations of the data used.
There are also some elements that should have been addressed but were not. For example, there has been a change in market versus non-market labor or of women as they have moved out of jobs at home to join the labor market. Part of this is explained by changes in social norms over time, making the effects of tax rates harder to gauge. Moreover, it is hard to parse this element out of the data and distinguish it from movement in the labor market that may be explained by taxation. One possible method of addressing this concern would be to regress data from men and women separately.
There are some additional concerns that the paper did not address sufficiently. First, the paper does not adequately address government spending. The paper essentially treats government spending as pouring money down a black hole, yet people are receiving services from spending. Second, the United States and Europe have enjoyed comparable productivity gains in the past. From this productivity dividend, Europe chose leisure while the United States chose more output. Without a valuation of leisure or government spending, it cannot be flatly said that Europe made a bad move.
AEI intern Michael Wilson and AEI staff assistant Gordon Gray prepared this summary.
