The Effects of Supplemental Income and Labor Product on Metropolitan Labor Cost Differentials

Regio~lal variations in labor costs attract a great deal of attention because of their poterttial to affect the regional distribution of eco~lomic activity. Because of the tnajor role that labor costs play in total production costs, regional differences in labor costs Inay translate into dramatic regional differences in profitability. Profitability, in turn, is a major determinant of a~hether or not existing firms will expand and of whether or not nemr firms are likely to Icrate in a given region. Most studies of regional variations in labor costs are based solely on data for payroll per employee or wages. Efowever, the measure of labor costs that is most relevant to the profitability of a firm is the total cost of the labor needed to produce its output. There are at least two steps involved in getting from simple wage data to esti~wates of labor costs per unit of output. First, nonwage income to workers ("supplemental income") must be added to wages to get total labor costs per hour. In 1977, the value of t~leasurable supplen~ental illcome paid to manufacturing workers was, on average, about 20 percent of the value of wages paid to those workers. This percentage showed considerable mriation, however, among regio~is and states. (See Garofalo and Fogarty [I9841 .) By 1982, supplemental illcome hati increased to nearly 25 percent of wages. Second, labor costs per hour r~lust be scaled by the amount of output per hour that is attributable to labor inputs, as opposed to other inputs, if the objective is to measure labor costs per unit of output. The amount of output generated by a unit of labor input varies for dif-9 fere~lt mrorkers and for different production processes. Labor productivity mrill be different among workers possessillg different skills or other per-sotla1 characteristics. Similarly, productivity will vaqr for a single worker according to the amount of other factors of productiort (machinery, energy, or raw materials) used in a particular production process. The research described here has been d~rected toward incorporating these two corrections to the raw wage rates to obtaiii a more accurate measure of labor costs per urnit of output. The supplemetital i~lcome estimates are a direct exte~ision of work clone at the state level by Garofalo and Fogmy. The strateby employed to control for labor productivity differs significantly from that used in most other studies. Researchers interested in analyzing regional labor productivity patterns face hard decisions …