Back

Why do managers matter for firm performance? Evidence has shown that there is a robust relationship between good management and firm performance. What are the managers doing? I provide evidence that a critical role of managers is efficiently allocating workers to jobs. 2/15

The panel data covers the population of white-collar workers in the firm: 200k workers, 30k managers in 100 countries over 10 years. The large firm embeds a rich internal labor market where workers can do hundreds of different jobs, e.g in R&D, HR, Finance, Sales, Marketing. 4/15

First, I identify successful managers by their own speed of promotion, as a revealed preference measure of the firm. I also show that this measure is correlated with alternative measures of manager quality, like manager fixed effects in workers’ outcomes. 5/15

Second, I exploit a natural experiment inside the multinational: middle managers rotate laterally across teams every 2 yrs to get exposure to different projects. I show that the rotations provide for exogenous variation in the workers’ exposure to different manager types. 6/15

Third, the research design is based on an event-study design: I follow workers’ career trajectories up to 7 years after the initial manager exposure. The plots confirm the absence of pre-trends. The objects of interest are: 1) gaining a good manager 2) losing a good manager. 7/15

I find that gaining a good manager causes workers to reallocate laterally within the firm, including task-distant transfers (e.g. from R&D to Finance). The moves are scattered throughout the organisation. At the same time, there is no impact on exit from the firm. 8/15

This leads to large and persistent gains in workers’ career progression and productivity. 7 years after the manager change, workers earn 30% more and perform better on objective performance measures (sales). 9/15

These effects are asymmetric in that the reverse transition, losing a good manager, has NO impact on workers’ career path. This provides more evidence on managers’ allocative role: once a good worker-job match is created, the gains persist a downgrade in manager quality. 10/15

In terms of aggregate firm productivity, I combine establishment-level measures of output per worker and costs per unit of output in the 100 countries. I estimate that the share of good managers within an establishment output per worker and costs (thus profits). 11/15

My results imply that the visible hands of managers match workers’ specific skills to specialized jobs, leading to a long-lasting improvement in the productivity of an existing workforce. 12/15

Thus, incidentally, they might help explain why there are persistent performance differences among seemingly similar enterprises, as widely (and puzzlingly) documented by the literature. 13/15

Matching workers to jobs provides new, cost-effective, policy levers. It does not involve worker firing, hiring, training. Other policy implications concern the selection & training of managers in discovering workers’ skills and their rotation within firms so to max impact. 14/15

There are many other interesting details in the paper that I could not cover in the tweet (including evidence on ruling out other channels, heterogeneity analysis) https://sites.google.com/view/virginiaminni And thank you @UniCredit_PR@twitter.com and @EEANews@twitter.com for the Best JMP Award 2022! 15/15

@virginiaminni Thanks for writing this thread! Very exciting work.