The AEA RCT Registry will be down for maintenance on Tuesday, April 20th, from 9pm EDT to 11pm EDT to perform necessary upgrades. The site will be in maintenance mode during that time. We apologize for any inconvenience this may cause.
Expansion of Small Firms and Job Creation: Evidence from Uganda
Initial registration date
January 22, 2016
March 27, 2021 4:09 PM EDT
This section is unavailable to the public. Use the button below
to request access to this information.
University of Southern California
Other Primary Investigator(s)
University College London
University College London & Institute for Fiscal Studies
London School of Economics
London School of Economics
Additional Trial Information
This project explores the link between skill-creation and Small and Medium Enterprises (SMEs) expansion in developing countries. We do so by evaluating a number of labour market interventions involving both young unemployed workers and SMEs in Uganda. On the labor supply side, we study the impact of a formal in-class vocational training program (VTP) on improving the labour market outcomes of young workers, and contrast this to the impact of informal apprenticeships. On the labour demand side, we test various interventions designed to exogenously ease such constraints related to credit, wages, and information about available skilled or unskilled workers, within the same empirical setting. As the project deals with both the demand and supply sides of the labor market, we are able to study how workers and SMEs match to each other, and in particular how the matching process is facilitated by SMEs having information about individuals that are willing to work, or information about individuals that are available to work and have been recently skilled. In addition, this project seeks to understand the indirect spillover effects of such interventions on firms in the social, business and market networks of directly treated firms. In an extension to this project, we track the original sample of workers over a longer period to study the longer-term impacts of the interventions on the labor market outcomes of workers after eight years, and to compare the effectiveness of formal vocational training and informal apprenticeships in building resilience to the economic shock caused by Covid-19.
This project combines interventions targeting the labor supply side as well as the labor demand side of the labor market. On the labor supply side, three types of interventions are being tested among a sample of unemployed young workers: (i) scholarships to attend a formal vocational training course; (ii) a job-matching service with potential employers; (iii) wage subsidies for potential employers to hire and train the young workers matched to them through informal apprenticeships. On the labor demand side, four interventions are being tested on a sample of SMEs operating in urban areas of Uganda: (i) a job-matching service with potential trained and untrained young employees, (ii) wage subsidies to hire and train the matched employees through informal apprenticeships; (iii) wage subsidies to hire and train any employee chosen by the entrepreneur outside the job-matching service; (iv) information about credit products available in the market and explicitly targeted to SMEs.
Intervention Start Date
Intervention End Date
Primary Outcomes (end points)
On the worker side, the main outcomes of interest are: skill development, employability, employment, occupational choice, wages, job stability/tenure/quality; job-search; additional training gained; changes in expectations and aspirations; other measures of well-being such as consumption; knowledge transfer to others in the workers’ social network and to other workers in the hiring firm. On the firm side, the main outcomes of interest are: growth of capital stock, input usage, output; profitability; changes in employment and wages offered; spillovers to other connected firms. We collect these outcomes in multiple survey rounds in the post-intervention period. In an extension to this project, we collect additional data eight years after the interventions, to study the long-term impact of these interventions and their efficacy in building resilience to the Covid-19 shock. In these additional survey rounds we collect again information on our key labor market outcomes, in particular: employment, occupational choice, wages; job-search; changes in expectations and aspirations. We collect these outcomes again immediately before, during and after the strict Covid-19 lockdown put in place by the Ugandan government in March 2020.
Primary Outcomes (explanation)
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
The experimental design combines interventions on both the worker and the firm side. On the worker side, applications were invited from unemployed young workers for scholarships covering six-month vocational training courses at registered institutions in Uganda. Over-subscription to the scholarship program was exploited to randomly allocate applicants to five treatment groups and one control group: workers in Group 1 and 2 were offered the vocational training scholarship; workers in Group 2 were also matched with potential employers after the end of the training. Group 3 was just offered the matching service without the training scholarship. Group 4 was matched with potential employers, who were also offered a six-month wage subsidy to hire and train the worker through informal apprenticeships. Group 0 was not offered any service and served as control.
On the firm side, SMEs were recruited in a number of urban areas across Uganda by means of a census. Firms identified in the census were allocated to six treatment groups and one control group. The experimental design on the firm side mirrors the one on the worker side: SMEs in Group 2 were matched with workers who received the six-month training at the vocational institutes. Group 3 was matched with untrained applicants to the training scholarship. Group 4 was matched with untrained applicants to the training scholarship and was offered a six-month wage subsidy to hire and train them. Group 5 was offered a six-month wage subsidy to hire and train any new worker of their choice from outside the job-matching service. Group 6 received promotion of a new credit product offered by BRAC NGO and specifically targeted at SMEs. Firms in Group 0 were not offered any intervention and thus served as control. An additional sample of network firms identified as being in the network of directly treated firms at baseline was added to the study to examine any spillover effects of the interventions.
Experimental Design Details
Randomization done in office by a computer (using STATA).
Individual level randomization was conducted on both the worker side and the firm side. So the unit of randomization is the individual applicant (on the worker side) and the individual firm (on the firm side).
Was the treatment clustered?
Sample size: planned number of clusters
Worker side: 1714 workers
Firm side: 2306 firms
Sample size: planned number of observations
Worker side: 1714 workers. Firm side: 2306 firms
Sample size (or number of clusters) by treatment arms
Group 0 (Control): 451 workers
Group 1 (Training): 390 workers
Group 2 (Training+Matching): 307 workers
Group 3 (Matching): 283 workers
Group 4 (Matching+Wage Subsidy): 283 workers
Group 0 (Control): 512 firms
Group 2 (Training+Matching): 256 firms
Group 3 (Matching): 513 firms
Group 4 (Matching+Wage Subsidy): 257 firms
Group 5 (Wage Subsidy): 256 firms
Group 6 (Credit): 512 firms
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Power calculations were conducted for the main outcomes on both the worker and firm side of the experiment. For details, please refer to Table 1 in the Supporting Documents and Materials section.
INSTITUTIONAL REVIEW BOARDS (IRBs)
UCL Research Ethics Committee (Note: this IRB covers the Covid-19 related extension)
IRB Approval Date
IRB Approval Number
UCL Research Ethics Commitee
IRB Approval Date
IRB Approval Number