“…In some cases, time in school may be a normal good while subsistence consumption is an inferior good, in which case child labor will be associated with poverty (Ray, ; Dammert, ), but evidence elsewhere suggests that this association is not universal (Bhalotra & Heady, ; Ersado, ; Oryoie, Alwang, & Tideman, ). Moreover, Del Carpio, Loayza, and Wada () show how positive shocks to wealth can reduce low‐skill child labor but increase high‐skill child labor, while Sarkar and Sarkar ( ) outline a theoretical model that shows the conditions under which child labor can persist while poverty declines. Nevertheless, the review by De Hoop and Rosati () indicates that exogenous increases in household wealth do tend to reduce child labor prevalence in most settings.…”