LabFam seminar series: Statistical discrimination at young age: new evidence from four decades of individual data across 56 countries
Speaker: Joanna Tyrowicz, FAME|GRAPE University of Warsaw, and IZA
Statistical discrimination offers a compelling narrative on gender wage gaps during the early stages of the career. Expecting absences related to child-bearing and child-rearing, the employers discount productivity to adjust for the probable losses such as costs associated with finding substitutes, leaving customers, etc. If that is the case, lower and delayed fertility should imply lower discount in wages, and consequently reductions in the gender pay gap among entrants. We put this conjecture to test against the data. We provide a novel set of estimates of adjusted gender wage gaps among youth for 56 countries spanning four decades. We estimate that postponing childbirth by a year reduce the adjusted gap 2 percentage points (15%). We show that this estimate is consistent with statistical discrimination, but for some countries the estimates of AGWG imply that either statistical discrimination is not accurate or taste-based mechanisms are also at play.