WAGE ADJUSTMENT: The Data and Regression Specification 6

Posted by Kathryn Schwartz on September 11, 2014
WAGE ADJUSTMENT

Average overtime wage and employment elasticities have remained consistently negative through out the sample period, with a slight absolute value increase.

The relative size of exchange rate effects on industry activity appears to be highly correlated with industry-specific characteristics other than its external orientation. In table 4 we report the results of correlating the estimated elasticities reported in Table 3 with four different industry characteristics: average industry markup during the sample period, unionization rates, percent of workers without a college degree, and industry capital-labor ratios.

A clear pattern of correlation exists between our estimated exchange rate wage and employment elasticities and industry markups and the level of workers’ education. High markup industries have significantly higher estimated elasticities of total and overtime wage to exchange rates and significantly lower employment and overtime activity exchange rate elasticities. Industries with lower shares of non-college-degree workers also have higher elasticities of total wages and overtime wages to exchange rates.

w6749-13
However, the elasticity of total employment in those industries is also significantly higher. We do not find a clear relationship between employment and wage responses to exchange rates in an industry and the industry’s capital-labor ratios or unionization rates. This strong correlation between these specific industry characteristics and exchange rate elasticities suggest that issues such as the degree of education of the labor force, type of education and competitive pressures in the industry might result in quite different mechanisms for the adjustment of labor markets.

We further explore cross-industry differences in their adjustment to exchange rate movements by re-estimating equation (13) separately for each industry. We computed for each of the 20 two-digit SIC manufacturing industries similar specifications to those reported in Tables 1 and 2. The limited number of observations and degrees of freedom in each of these regressions lead them to have very little power. As a consequence, the parameter estimates are extremely noisy and should be viewed with a great deal of skepticism. Although we present the parameter estimates on exchange rate terms from these industry-specific regressions (Appendix Tables 2 and 3), we do not have much faith in them.

In the vast majority of cases, the hypothesis of an insignificant exchange rate effect could not be rejected for the exchange rate terms for permanent exchange rate changes on regular wages and employment. In regressions on overtime responsiveness to actual changes in exchange rates, the final rows of these tables show that the coefficients on the non-interacted and interacted exchange rate terms are seldom significant in explaining either basic wages and employment or overtime activity.

Only some of these industry-specific measures are reassuring. In particular, there is a strong correlation between the estimated response elasticities for industry wages, overtime wages, and overtime employment reported in Table 4 and those implied from the industry specific regressions. However, the two elasticities approaches give nearly uncorrelated results for the two employment measures. Electronic Payday Loans Online

Tags: , ,