UC Berkeley

This dissertation is about how firms use incentives and information in internal personnel and management practices, in particular relating to hiring and innovation. In the first chapter, I study competition between workers inside of firms. Why do firms use incentives that encourage anti-social behavior among employees? Rank-based promotion schemes are among the most widespread forms competition and incentives, despite encouraging influence-peddling, sabotage and anti-social behavior. I study a natural experiment using rich administrative data from a large, white collar firm. At the firm, competitors for promotions depend partly on dates-of-hire. I utilize the date-of-hire assignment as a source of exogenous variation in the intensity of intra-worker competition. I use the firm's multidimensional timestamped productivity logs as ``time diaries'' to study the amount, character and allocation of output across tasks. I find that competition has significant incentives for effort and efficiency -- as well as lobbying- and sabotage- like behaviors -- without affecting the quality and innovativeness of output. I also find that employees facing high competition are more likely to quit and join other companies, particularly higher-performing employees. Lastly, I show that competition induces workers to differentiate and specialize by concentrating effort into a smaller set of tasks. These results show that while workers respond to incentives from competition, they also seek to avoid it through sorting and differentiation strategies. The productivity gains from differentiation and specialization may partly explain the common use of these incentives by firms.

In the second chapter, I study how firms use social networks in hiring. Using personnel data from nine large firms in three industries (call-centers, trucking, and high-tech), we empirically assess the benefit to firms of hiring through employee referrals. Compared to non-referred applicants, referred applicants are more likely to be hired and more likely to accept offers, even though referrals and non-referrals have similar skill characteristics. Referred workers tend to have similar productivity compared to non-referred workers on most measures, but referred workers have lower accident rates in trucking and produce more patents in high-tech. Referred workers are substantially less likely to quit and earn slightly higher wages than non-referred workers. In call-centers and trucking, the two industries for which we can calculate worker-level profits, referred workers yield substantially higher profits per worker than non-referred workers. These profit differences are driven by lower turnover and lower recruiting costs for referrals.

In the third and final chapter, I study the use of betting markets inside of firms. Despite the popularity of prediction markets among economists, businesses and policymakers have been slow to adopt them in decision making. Most studies of prediction markets outside the lab are from public markets with large trading populations. Corporate prediction markets face additional issues, such as thinness, weak incentives, limited entry and the potential for traders with biases or ulterior motives – raising questions about how well these markets will perform. We examine data from prediction markets run by Google, Ford Motor Company and an anonymous basic materials conglomerate (Firm X). Despite theoretically adverse conditions, we find these markets are relatively efficient, and improve upon the forecasts of experts at all three firms by as much as a 25\% reduction in mean squared error. The most notable inefficiency is an optimism bias in the markets at Google. The inefficiencies that do exist generally become smaller over time. More experienced traders and those with higher past performance trade against the identified inefficiencies, suggesting that the markets' efficiency improves because traders gain experience and less skilled traders exit the market.