This paper analyses the determinants of the size of the informal economy using cross-country regressions. Two sets of global data using indirect estimation techniques and the perception of business leaders for 109 countries as well as a regional set for Latin America based on direct data are used to estimate the size of the informal economies. Indirect estimation techniques arrive at higher estimates of the size of the informal economy than the perceptions of business leaders because they include not only the (fundamentally legal) activities of the informal sector, but also those activities which are illegal per se. Both kinds of estimate show strong regional differences in the size of the informal economies. Regressions on a set of indicators covering the intensity of regulations, taxes and the cost of establishing a business reveal that the intensity of labour regulations seems to be the most important factor in explaining the size of the informal economy in cross-country regressions using the rational behaviour approach. Socio-cultural indicators are only important in explaining the size of the informal economies in Latin America. Government efficiency is an important factor in explaining the size of the informal economy in world regressions. Regional regressions reveal that different aspects of governance dominate the relationship between government efficiency and the size of the informal economy in the different regions. Governments that seek to limit or decrease the size of the informal economy must therefore start from a country-specific analysis of the reasons why economic agents choose to conduct their business in the informal sector.