This is the sixth in this year’s job market series
Labor misallocation is believed to be a key driver of differences in income across countries (Hsieh and Klenow 2010). However, the causes of this misallocation are not always well understood and there is little evidence on what interventions can improve the allocation of workers in the economy. These issues are particularly important in Sub-Saharan Africa, where worker mobility from low to high productivity sectors is often limited (McMillan, Rodrick and Verduzco-Gallo 2014).
My job market paper provides new experimental evidence from Ethiopia showing that subsidizing job applications can reduce inefficiencies in the allocation of workers’ talent.
Jobseekers in Ethiopia apply for very few jobs: is this good or bad for firms?
In many labor markets around the world (including the academic job market for economists!) getting a job can be time consuming and expensive.
In Addis Ababa, the context of our study, a jobseeker who wants to secure a job at a formal firm has to visit the firm multiple times to deposit application materials, complete screening tests, and attend interviews. To make these visits jobseekers typically have to pay high transport costs and have to take time off informal income-generating activities (Abebe et al. 2017). Hence, perhaps unsurprisingly, jobseekers only apply for a small subset of all available job openings.
A job vacancy board in Addis Ababa
What implications do these costs have for employers and for the overall allocation of talent? One view is that application costs actually help employers to screen out unsuitable candidates. This would be consistent with the role that application costs are believed to play in other contexts, for example the targeting of social assistance programs (Alatas et al. 2016). On the other hand, if talented workers struggle to find the time and money required to complete job applications, high application costs will limit the firm’s ability to make good hires and will force high-ability jobseekers to spend inefficiently long times in unemployment or in the wrong jobs.
Reducing application costs with financial incentives
In partnership with a local employer, we evaluate the effects of offering a small monetary incentive (worth 4.5 USD) to the people who apply for a clerical job and compare it to a second intervention in which the wage offered for the job is doubled (raising the expected value of the application by 105 USD). We randomize these two treatments over the sample of individuals who call to inquire about the jobs, after seeing them advertised in newspaper inserts and job vacancy boards. Applying for the job requires attending a formal testing session, which we use to collect rich data on the cognitive ability, non-cognitive ability and relevant work experience of all candidates. These measures are reliable predictors of work performance (Hoffman, Kahn and Li 2017).
Financial incentives attract better applicants
We find that the application incentive improves the quality of the pool of applicants. Cognitive ability is higher at the top, at the mean, and at the bottom of the distribution. The magnitude of the increase in applicant quality is substantial. The number of top applicants, that is, those with cognitive ability above the 90th percentile of the distribution in the control group, doubles. This is similar in magnitude to the effect of doubling the wage.
Is this a better pool of applicants also in the eyes of local employers?
To answer this question, we run a second experiment with a sample of recruiters working in firms that have open vacancies for clerical workers. Recruiters are shown the anonymized CVs of a subset of applicants from the first experiment and choose who they would like to invite to make a new job application at their firm. This allows us to validate our measures of ability.
We confirm that the application incentive attracts better workers: applicants from the incentive group are more than twice as likely to be chosen by recruiters, compared to control group applicants.
How do local recruiters rate the applicants from the three experimental groups?
Talented low-income jobseekers drive the effects of the intervention
The talented applicants attracted by the application incentive are individuals who often have little work experience, are not formally employed and are female. These are groups who have, on average, worse outcomes in the labor market and lower incomes. Thus, the application incentive does not increase quality at the cost of attracting individuals who have better outside options. On the contrary, this intervention mostly taps from the pool of talent of low-income jobseekers who stand to benefit the most from the job.
To explore this point further, we use the exogenous variation generated by the experiment to structurally estimate a model of job application decisions. We find that the average cost of making a job application amounts to up to 13 percent of the monthly wage. This cost is heterogeneous (poor people find it harder to pay transport costs and to give up informal income-generating activities) and is positively correlated with jobseeker quality. According to the model’s estimates, up to 30 percent of individuals in the control and the high wage groups are unable to apply for the job because of credit constraints. These constraints are relaxed by the application incentive.
Subsiding job applications can improve the allocation of talent
We calculate that this intervention has positive returns for the average firm in this market. Using a Mincerian estimate of the returns to cognitive ability, and taking into account the cost of the incentive and of the additional recruitment efforts, we estimate that the internal rate of return of the application incentive is 11 percent. This substantially higher than the returns that can be obtained by increasing the wage.
What is particularly important is that these gains are generated by those jobseekers who have the worst prospects in the labor market. Thus subsiding job applications can improve the allocation of jobseekers’ talent, benefitting both workers and firms. This has important implications for the allocative role that can be played by interventions that help the poor to participate in the labor market, through subsidies or quotas.
Finally, you may be wondering: if application incentives have positive returns for firms, why do employers not commonly use them? In our second experiment we also elicit recruiters’ forecasts of the effects of this intervention. We show that recruiters substantially underestimate the positive effect of the incentive on the quality of applicants. Informing firms about these benefits may thus be a cost-effective intervention in this context
Stefano Caria is a post-doc at the University of Oxford. You can read the full paper here.