Discretion in Hiring

Danielle Li (Harvard Business School)
Mitch Hoffman (University of Toronto)
Lisa Kahn (Yale)

Abstract: Who should make hiring decisions? Many firms rely on hiring managers to evaluate applications and make job offers. These hiring managers may be informed about a worker's quality, but their efficacy may be undermined by biases or bad judgement. The use of quantitative metrics such as job testing enables firms to limit these concerns, but potentially at the cost of ignoring valuable information. This paper examines whether firms should rely on hard metrics or grant managers discretion in making hiring decisions. We evaluate the staggered introduction of a job test across 131 locations of 15 firms employing low-skill service sector workers. We show that testing improves the match quality of hired workers, as measured by their completed tenure, by about 15\%. Further, when faced with similar applicant pools, we find that managers who exercise more discretion (as measured by their likelihood of overruling the test recommendations) systematically end up with worse hires. This result suggests that managers make exceptions to test recommendations because they are biased, not because they are better informed. In this setting, we find that firms can improve productivity by limiting managerial discretion.