Per-capita income in many sub-Saharan African countries, such as Chad and Niger, is less than 1/30th of that of the United States. Most economists and social scientists suspect that this is in part due to institutional failures that stop these societies from adopting the best technologies. A particularly interesting historical example comes from the di®usion of railways in the nineteenth century. While railways are regarded as a key technology driving the industrial revolution, there were large lags in its di®usion. For example, in 1850 the United States had 14,518km of track, Britain 9,797km and Germany 5,856km, in the Russian and Hapsburgh Empires there were just 501km and 1,357km, respectively (all data from Mitchell, 1993). Why do societies, as in this example, fail to adopt the best available technologies? One answer is that existing powerful `interest groups ' block the introduction of new technologies in order to protect their economic rents, and societies are able to make technological advances only if they can defeat such groups. Economic monopolies may be one example. A monopolist might wish to block the intro-

We build a model of child labor and study its implications for welfare. We assume that there is a trade‐off between child labor and the accumulation of human capital. Even if parents are altruistic and child labor is socially inefficient, it may arise in equilibrium because parents fail to fully internalize its negative effects. This occurs when bequests are zero or when capital markets are imperfect. We also study the effects of a simple ban on child labor and derive conditions under which it may be Pareto improving in general equilibrium. We show that the implications of child labor for fertility are ambiguous.