Most employees find their job through their social networks (Hofreiter and Banha, 2019), and the reliance on personal connections for employment is even more vital for low-income individuals and in high-poverty settings. 60-70% of individuals find work through their networks in South and Southeast Asia (Witte, 2021). Economic theory suggests five main hypotheses for why employers hire workers they know: 1) employers have better information about the productivity of potential employees that are part of their network relative to strangers; 2) employers are better able to monitor and incentivise high productivity from employees in their network because of the larger reputational costs of shirking for connected individuals; 3) providing a job to someone in their network provides private utility for the employer (either from warm-glow altruism or because they believe the employee will reciprocate); 4) employers find it more pleasant to work with individuals in their network; or 5) selection, in that individuals connected to employers are more productive than those not connected, potentially because connectedness is correlated with education and family wealth. This project aims to test which of these hypotheses explains the hiring patterns in Pakistan.
The project partnered with a construction firm which hires laborers for short-term and renewable contracts, and collected detailed hourly observations on the tasks completed by the labourer, such as the number of bricks/tiles placed, the volume of plaster mixed, and so on, for at least three full days for each worker. The research team also conducted a baseline survey with the labourers at the end of the first day, covering a variety of dimensions, including contractual arrangements, payment mechanisms, social ties, monitoring and reporting systems, job search time and processes, negotiation processes, and work experience. For the intervention of the study, they randomly vary three aspects of the hiring process: 1) information: whether the hiring manager receives information that the worker was an above-average performer from the baseline data; 2) bonus: whether the hiring manager received a bonus payment based on the productivity of the team he or she hires to work that week; and 3) monitoring: whether the hiring manager is also responsible for monitoring the employees on the worksite that day. They primary outcomes of interest are the average productivity of the hired team and what types of laborers the hiring manager chooses, in particular, the extent to which he or she hires from those in his social network.
The prevalence of network-based hiring is extremely harmful to marginalised and less-connected individuals, disadvantaging minority groups, low-income individuals, and those with smaller social and familial networks (Witte, 2021; Beaman, Keleher, and Magruder, 2018). This approach may also be less economically efficient if employers are frequently hiring low productivity, connected individuals or if employees devote substantial effort to developing social rather than human capital. Understanding which channels drive the social network premium can help policymakers better serve these marginalised employees.