Oriana Bandiera (LSE) Stephen Hansen (Oxford) Andrea Prat (Columbia) Raffaella Sadun (Harvard)

We measure the behavior of 1,114 CEOs in six countries parsing granular CEO diary data through an unsupervised machine learning algorithm. The algorithm uncovers two distinct behavioral types: “leaders” and “managers”. Leaders focus on multi-function, high-level meetings, while managers focus on one-to-one meetings with core functions. Firms with leader CEOs are on average more productive, and this difference arises only after the CEO is hired. The data is consistent with horizontal differentiation of CEO behavioral types, and firm-CEO matching frictions. We estimate that 17% of sample CEOs are mismatched, and that mismatches are associated with significant productivity losses.

Oriana Bandiera (LSE) Andrea Prat (Columbia) Renata Lemos (World Bank) Raffaella Sadun (Harvard)

We build a comparable and bottom-up measure of CEO labor supply for 1,114 CEOs, and investigate whether family and professional CEOs differ on this dimension of effort. Family CEOs work 9% fewer hours relative to professional CEOs. CEO hours worked are positively correlated with firm performance, and account for 18% of the performance gap between family and professional CEOs. We study the sources of the differences in labor supply across family and professional CEOs by exploiting firm and industry heterogeneity, and variation in meteorological and sport events. The evidence suggests that family CEOs value–or can pursue–leisure activities more than professional CEOs.

Oriana Bandiera, Andrea Prat, Raffaella Sadun and Julie Wulf, February 2012

For both practitioners and researchers, span of control plays an important role in defining and understanding the role of the CEO. In this paper, we combine organizational chart information for a sample of 65 companies with detailed data on how their CEOs allocate their work time, which we define as their span of activity. Span of activity provides a direct measure of the CEOs management style, including the attention devoted to specific subordinates and functions, the time devoted to individual work and outside constituencies, a preference for multilateral or bilateral interaction, the degree of planning, etc. We find that CEOs with a larger number of reports spend more time with subordinates, more time on large meetings, less time on unplanned activities. The presence of a delegate, such as the COO, allows the CEO to reduce the time spent with insiders and to focus on bilateral and unplanned activities. These results suggest that time-use information is helpful in interpreting how span of control determines management style.

Oriana Bandiera, Luigi Guiso, Andrea Prat, Raffaella Sadun, October 2010

This paper utilizes data collected on time use for different work activities by 94 CEOs of top-600 Italian firms. It analyses the correlation between time use, managerial effort, quality of governance and firm performance and interpret the empirical findings under the context of effective corporate governance as well as imperfect governance.


Oriana Bandiera (LSE) Andrea Prat (LSE) Raffaella Sadun (HBS)

Managerial practices are a key determinant of firm productivity, and yet little is known about the managerial capital of the top executives who shape these practices. Using a novel survey instrument, we record with unparalleled detail the activities undertaken by 357 Chief Executive Officers of listed Indian manufacturing firms over a specific, exogenously chosen, window of time. Our preliminary findings reveal substantial heterogeneity in the way CEOs use their time along three dimensions: (i) their focus on firm employees vs. outsiders, (ii) their focus on different functional areas and (iii) whether their schedule is planned in advance. Matching our time use data with the firms’ balance sheets reveals that firms’ productivity is higher when CEOs work longer hours but not all CEOs’ time is equally productive. In particular, the correlation between total hours worked and output is entirely driven by planned time with firm employees, especially those belonging to the functional areas of production, finance, and labour relations.

Michael Blanding, April 2011

A CEO’s schedule is especially important to a firm’s financial success, which raises a few questions: What do they do all day? Can they be more efficient time managers? HBS professor Raffaella Sadun and colleagues set out to find some answers. Key concepts include:

  • On average, some 85 percent of a CEO’s time was spent working with other people, with only 15 percent spent working alone.
  • The time CEOs spent with outsiders had no measurable impact on firm performance. But time spent with other people inside the company was strongly correlated with positive increases in productivity.
  • In companies with stronger governance, CEOs spent more time with insiders and less time with outsiders, and at the same time were more productive.
  • The research could help CEOs learn to be more productive.

Edited by Tito Boeri, Antonio Merlo, and Andrea Prat, October 2010

This book provides an understanding of what drives the formation of a ruling class, and the relationship between politics and business firms. Focusing on Italy, it uses labour economics to analyse the selection of the ruling class, the labour market of politicians, the allocation of managers’ time, and their incentives, remunerations, and career paths. It draws on contributions from two teams of leading scholars and on research undertaken by the Fondazione Rodolfo DeBenedetti.

Oriana Bandiera, Luigi Guiso, Andrea Prat, Raffaella Sadun, May 2008
Presented at the X European Conference, Gaeta

This report utilizes extensive information collected on information on the characteristics of Italian managers and of the firms that employ them. It analyses the incentive structure that managers face, their career profile, and their use of time. Our data indicate that a fraction of firms – especially non-family firms and multinationals – while other firms – especially family firms and firms that operate on the national market only – instead adopt a fidelity model of managerial talent development.