DaTA aND MEthoDoLogY

Empirical analysis

We use a large number of micro datasets from African countries, spanning the period of 2005 to 2020, to document empirical facts on labour market outcomes in poor economies. We focus on three separate employment sectors: (i) subsistence sector, which includes own-account workers and unpaid family members,  (ii) private sector, which includes formal and informal firms, and (iii) public sector (such as national and local government) and state enterprises. We report key patterns in the selection into the public sector, the differences in  labour earnings between the private and public sectors, and workers' mobility between them.

Data SouRCES

Most datasets are part of the World Bank’s Living Standards Measurement Surveys (LSMS) programme and can be obtained here. A number of datasets have been obtained from their corresponding country's statistical office. The list of the micro datasets with the access links is below:

Definitions 

We focus on three sectors of employment, based on the main activity of a respondent in the last 12 months:

To compare monthly labour earnings across countries and over time, wages are deflated using Consumer Price Index (CPI) with 2010=100 and converted into international dollars using 2010  private consumption PPP conversion factor (LCU per international $).

Education groups are split into 6 levels depending on the highest qualification achieved: (1) none/illiterate/less than primary, (ii) primary, (iii) lower secondary, (iv) professional or vocational diploma (with or without a completed secondary degree),  (v) upper secondary, and (vi) post-secondary. 

Data Download

You can view and download the aggregated summary statistics on employment composition and wages by sector, gender, and education here. For a number of countries that have panel data, you can also find average transition rates between sectors between survey waves.


Economic Model

As part of this project, we develop and estimate a theoretical framework to better understand what drives private sector productivity, wages and job creation and how these phenomena are affected by public sector size and wages. Our starting point is the search and matching model with two-sided heterogeneity and firm entry costs in the private sector and the subsistence (or home) sector as in Rud and Trapeznikova (2020). We then include the exogenous public sector in the spirit of Gomes (2018) and Albrecht et al. (2018). We use the model to derive the equilibrium allocation of workers across the three sectors and show how public sector wages and vacancies interact with the private sector productivity, wages and employment through (mis)allocation of talent. 

The key mechanisms we propose are the following: First, the presence of high barriers to entry or labour market frictions can constrain firm entry and have a negative effect on job creation, productivity and wages in the private sector. Second, the existence of a (possibly too) high wage premium in the public sector could lead to misallocation of talent in this economy, away from more dynamic sectors. The main reason for this result is a tendency of highly educated workers to search for a job in the public sector, thus depriving private firms with vacant positions of more productive workers. Our preliminary results suggest that these negative effects are larger when the entry costs or other entry barriers are higher.  These mechanisms suggest that removing barriers in either sector can improve aggregate productivity through a reallocation of workers to more productive opportunities, amplifying the standard effect of generating more firm entry. 

We plan use the model to simulate the effects of counterfactual policies -- such as an increase in public sector employment or wages --  on private firms' job creation and productivity.