The Effects of Intensive and Extensive Margins of Dualisation on Protection for Outsiders in Developing Countries

Insider-outsider models of politics often analyze a single dimension of labor market dualization. I propose that labor market dualization has both intensive and extensive margins. The intensive margin refers to the differences in ex-ante protection between formal and informal labor. The extensive margin, in turn, refers to the group size of informal labor. These two dimensions affect the distribution of power between labor market groups. In autocratic regimes, insiders' power increases with the intensive margin of dualization independently from the size of informality (provided that formal workers’ power results from rigid of labor market institutions). In democratic regimes, however, outsiders' power increases with group-size, given that a large segment of informal labor is electorally attractive for political incumbents. I argue that these two margins of dualization influence the amount of ex-post protection (i.e., cash transfers) targeted to each labor market group. First, democratization is expected to expand protection for outsider informal workers. With a larger selectorte the probability that the median actor is an informal worker increases, and therefore democratization is likely redirect some government resources to social protection for outsiders. Second, this effect grows larger with both extensive and intensive margins of dualization. The larger the size of informal labor (extensive margin) the more pronounced is the effect of regime change on protection for outsiders. Interestingly, the adoption of programs aimed at protecting informal workers is also more likely where labor market regulations are more rigid (intensive margin). Large differences in ex-ante protection decrease horizontal mobility and reduce outsiders’ chances of accessing to a formal job. Hence, the government has larger incentives to adopt new forms of ex-post protection for informal workers as a means of building electoral support among outsiders. I test the argument on the adoption of conditional-cash transfers programs in a global sample of developing countries.