WAGE ADJUSTMENT: The Data and Regression Specification 4

Posted by Kathryn Schwartz on September 07, 2014

From Table 1 observe that permanent exchange rates have significant explanatory power with respect to industry wages in low-markup industries. A dollar depreciation increases the average wages in industries that are more heavily involved in export activity. As industry export orientation rises, so does the role of the dollar in slowing industry wage growth. Industry reliance on imported production inputs can reverse the stimuli to wages associated with dollar depreciations. All else equal, as an industry increases its reliance on imported inputs, dollar appreciations have a smaller role in wage growth restraint.

Indeed, in industries with imported input shares exceeding their export shares, appreciations could potentially accelerate wage growth. The same sign pattern of effects of exchange rates on wages appear for the full sample of industries and for the sub-sample including high-markup industries. However, for neither of them can we reject the hypothesis that the estimated exchange rate effects are jointly insignificant.

The effects of a permanent exchange rate change on industry employment measures are much weaker. All of the individual exchange rate coefficients for total hours are statistically insignificant, although the F-test rejects the hypothesis of these coefficients being equal to zero for high-markup industries. In the specifications explaining the total number of employees, the coefficients on the export terms are statistically significant for the full and high markup samples, as well as the coefficient with imported inputs.

However, this significance is most likely resulting from the colinearity among these exchange rate terms since for all samples the F-tests are unable to reject the null hypothesis of these coefficients being jointly different from zero. Given these results, the effects of exchange rate changes on these measures of employment activity in manufacturing industries is likely to be very small.

The rest of the regressors show consistently significant results for all the different specifications. Increases in the prices of other inputs are consistently negatively correlated with industry wages and positively correlated with industry employment. Industry employment, as expected, is high when the economy is in a boom, i.e., when the level of GDP is high. It is however, surprising that wages appear to decrease in periods of high economic activity.

Overtime wages and overtime employment both are more significantly affected by exchange rate movements than total wages and employment (Table 2). The F-tests easily reject the null hypothesis of no exchange rate effect in all cases, with the exception of overtime employment in high markup industries (which is only marginally rejected). Dollar depreciations significantly increase overtime wages in high-markup industries with large export orientation, and decrease overtime wages in those industries with a larger imported-input orientation. We observe an opposite sign pattern for low markup industries. The effects of exchange rates on overtime employment are more significant in low markup industries than in high-markup industries. Electronic Payday Loans Online

Tags: , ,