In Mozambique’s three largest cities (Maputo, Beira, and Nampula), informal businesses—those operating outside formal licensing and registration procedures—outnumber formal firms by a factor of 9 to 1. The World Bank’s Enterprise Analysis Unit recently published surveys of informal businesses in Mozambique, conducted in collaboration with the country management unit and colleagues from the Finance, Competition, and Innovation Global Practice. These surveys were designed to mirror the standard World Bank Enterprise Surveys, which cover the formal sector, but were tailored to better understand the unique conditions in which informal firms operate. Thanks to recent methodological innovation in sampling techniques, these surveys now also provide an estimate of the total number of informal businesses.
The surveys use stratified adaptive cluster sampling methods—a method commonly used in the field of biology—that allow researchers to efficiently study subjects that cluster near each other. In practice, the method is implemented as follows: first, we take a city such as the capital city Maputo, and divide it into 150 by 150 meter squares—each stratified based on the likely concentration of informal business activities (see the illustration below). Second, we randomly select a pre-defined number of squares for a full enumeration of all informal businesses in each. The process is adaptive in the sense that enumeration is expanded to all adjacent squares if the number of informal firms found in any square is above a pre-defined threshold. This method allows for the same unbiased precision as stratified random sampling (Thompson 1990) but it can be implemented at a lower cost and with reduced fieldwork. Informal businesses tend to operate in close proximity to one another, forming clusters of economic activity.
Figure 1: Primary Sampling Units (Squares) for Maputo