The demand side of the shadow economy: essays on informal consumption

The focus of this thesis is the study of the demand side of the shadow economy. To achieve this, the informal consumption of Peruvian families located in urban areas is studied. This is possible thanks to a household survey collecting information on where people acquired their goods. The main contribution of the research is that it identifies an unexplored area in the literature with limited theoretical discussion and few empirical applications. Information about why people purchase from informal markets will supplement wider knowledge of labour allocation on informal opportunities. The thesis uses an Almost Ideal Demand System in order to verify some demand properties of informal consumption: income and price elasticities, the existence of linkages between working and purchasing decisions and explore the effects of bargaining on expenditure allocation. Four robust results are encountered. First, the inferiority of informal consumption is rejected. Formal and informal expenditure are classified as normal, but income responses on the latter (necessity) are lower than on the former (luxury). Second, there are linkage effects between working and purchasing in the informal and formal sectors. These effects are stronger for informal consumption and among the self-employed. Linkages are also not equally applicable across all goods. Better results are found within quasi-substitutes with leisure. Third, formal and informal food consumption reveals elastic demand curves and imperfect substitution between them, with higher compensated own-price and cross-price elasticities for formal markets. Fourth, household members bargain in their allocation decisions across markets, with females’ decisions being closer to less-informal purchasing baskets. This result is clearer in the case of food consumption. Public policy recommendations based on these results are derived, where it is found that formalization policies will need to take into account their negative distributional effects.