Download the Study

Skills, Tasks and Technologies: Implications for Employment and
Earnings

What are the Implications of Skills, Tasks, and Technology on Earnings?

Income Distribution, June 2010

In an exam­i­na­tion of the returns to skills and the evo­lu­tion of earn­ings inequal­ity, Daron Acemoglu and David Autor argue that despite past suc­cesses of this model, empir­i­cal devel­op­ments of the last three decades ought to be under­scored.  These devel­op­ments include (1) sig­nif­i­cant declines in real wages of low skill work­ers, par­tic­u­larly low skill males; (2) non-monotone changes in wages at dif­fer­ent parts of the earn­ings dis­tri­b­u­tion dur­ing dif­fer­ent decades; (3) broad-based increases in employ­ment in high skill and low skill occu­pa­tions rel­a­tive to mid­dle skilled occu­pa­tions (i.e., job polar­iza­tion); (4) rapid dif­fu­sion of new tech­nolo­gies that directly sub­sti­tute cap­i­tal for labor in tasks pre­vi­ously per­formed by moderately-skilled work­ers; and (5) expand­ing off­shoring oppor­tu­ni­ties, enabled by tech­nol­ogy, which allow for­eign labor to sub­sti­tute for domes­tic work­ers spe­cific tasks.

Acemoglu and Autor argue that to account for recent changes in the earn­ings and employ­ment dis­tri­b­u­tion in the United States and other advanced economies, and also to develop a bet­ter under­stand­ing of the impact of tech­nol­ogy on labor mar­ket out­comes, it is nec­es­sary to sub­stan­tially enrich the model by which we under­stand returns to skills and the evo­lu­tion of earn­ings inequal­ity.  They doc­u­ment changes and argue that a task-based frame­work, in which tasks are the basic unit of pro­duc­tion and the allo­ca­tion of skills to tasks is endoge­nously deter­mined pro­vides an alter­na­tive framework.