This paper tests the relevance of the tax-push hypothesis of wage formation for nine Western European countries. It presents a microeconomic theoretical model of optimizing union behavior from which a macroeconomic testable model of union wage setting is then derived. According to the model, the union optimally chooses the nominal wage so as to maximize an objective function that depends positively on the net real wage and employment.
The trade-off between the net real wage and employment is a function of both the weight assigned to employment relative to the net real wage and the progressivity of direct taxation. Two variables—the progressivity and the compensation effects—are shown to account for the influence (if any) of tax rates on wage formation. The progressivity effect—which may be positive, negative, or zero—explains the wage movements caused by changes in the marginal direct tax rate for a given average. The compensation effect—which may also be positive, negative, or zero—explains wage movements caused by changes in the indirect tax rate and in social security contributions as well as in the direct tax rate to the extent that the average indirect tax rate may vary without affecting the marginal tax rate. Empirical analysis is necessary to determine the sign and importance of these two effects.
The paper examines data for nine European countries during 1960-88. These seem to show three robust regularities as follows:
- In general, small open economies, such as Austria, Denmark, and the Netherlands, show zero compensation and progressivity effects.
- In larger, less open economies, indirect and social security tax rate increases tend to be transferred in the long run to the real labor cost, and a rise in direct tax rates raises the steady-state gross real wage.
- All countries show a weakening of the tax shifting onto real labor costs starting in the late 1970s or the early 1980s. The changing union attitude implied coincides with the introduction of some form of fiscal indexation and decreasing progressivity.
The paper thus indicates that the tax-push hypothesis is confirmed but that the tax push varies across countries, time periods, and different fiscal policies.