Labor Taxes, Wage Setting and the Relative Wage Effect
01.01.2002
Giorgio Brunello, Maria Luisa Parisi, Daniela Sonedda
H24
Progressive taxation,wage determination,Italy
Economy and Society
Gianmarco I.P. Ottaviano
Public finance solutions to high unemployment in Europe have often been advocated during the past years. With unemployment concentrated among the young and unskilled, it has been suggested that the reduction of social security contributions for low wage earnings, financed by a carbon tax, could yield a double dividend, the reduction of unemployment and the abatement of pollution. A decline in average labor taxes reduces unemployment if it generates lower pretax wages. Pretax wages fall if real after tax income from unemployment and leisure is not affected or only partially affected by the change in average taxes. When unemployment benefits are not taxed in a unionized economy, lower average labor taxes reduce the replacement ratio, and unions are willing to accept lower pretax wages because the net income loss from employment increases. Changes in labor taxation do not necessarily require that average labor taxes vary. In principle, a switch from payroll to income taxes, given average rates, could affect wage pressure and unemployment. The empirical evidence to date, however, does not support this possibility.When labor taxation is nonlinear, another opportunity is to vary the degree of labor tax progressivity. Economic theory suggests that higher progressivity could reduce unemployment. Suppose that wages are bargained over by unions. When a union contemplates the possibility of a wage hike, it has to consider that say for a 1% increase in the after tax wage the pretax wage increases by 1/v, where v denotes the coefficient of residual income progression. This implies that the expected employment loss associated to the higher after tax wage is e/v, where e is the elasticity of labor demand. Since v<1 with progressive taxation, this loss is higher with progressive than with proportional taxation. It follows that, when labor markets are not perfectly competitive, a certain degree of tax progressivity can be desirable because it makes wage increases less attractive to unions, with positive consequences on the unemployment rate.This empirical paper adds to the existing literature additional evidence based on Italian data. We use both panel and grouped data to study the effects of average and marginal (payroll and income) tax rates on wage pressure, and investigate whether these effects vary by skill, age group and region of residence.Our empirical findings are summarized as follows:a) changes in average payroll taxes are not fully absorbed by offsetting changes in the after tax wage and affect both pretax wages and employment;b) the estimated effects of changes in average income taxes on pretax wages are mixed but on balance the evidence suggests that after tax wages do not fully offset these changes;c) higher tax progressivity increases pretax wages;d) there are significant differences in the relationship between labor taxes and pretax wages by age group but not by region of residence or skill.We also find that the estimated elasticity of pretax wages to changes in marginal tax rates, given average rates, is large and ranges between 0.919 and 1.131, depending on the specification being used. It is difficult to explain this large positive elasticity exclusively with the argument that the labor supply effect prevails over the wage moderation effect. Therefore, we add an additional mechanism, the relative wage effect, and describe how it can contribute to producing the above results. Briefly put, when individuals and unions care about relative wages and an increase in tax progressivity reduces the own wage via the wage moderation effect, this reduction translates into lower relative wages, which can only be avoided by increasing wage pressure.