“…Attempting to correct for ability bias and measurement error, most articles estimate returns to schooling in an Instrumental Variables (IV) framework using compulsory schooling laws and educational background variables as instruments. Estimates range widely, from large positive returns (Harmon & Walker, 1995;Oreopoulos, 2006;Wang, 2013, Winters, 2015, through moderate positive returns (Angrist & Krueger, 1991;Devereux & Fan, 2011;Sakellariou & Fang, 2016), to no returns to schooling at all (Grenet, 2013;Meghir & Palme, 2005;Oosterbeek & Webbink, 2007;Pischke & von Wachter, 2008;Stephens & Yang, 2014). The conventional practice in the literature is to consider education as a continuous variable measuring the number of years of schooling.…”