What Do Boards Really Do?

Boards were more likely to receive updates than to make decisions [and] ... they were rarely presented with alternatives.

In What Do Boards Really Do? Evidence from Minutes of Board Meetings (NBER Working Paper No. 17509), Miriam Schwartz-Ziv and Michael Weisbach find that boards spend most of their time monitoring management rather than making business decisions. However, on occasion they do make managerial decisions.

The researchers analyzed a year's worth of minutes of board meetings from the period 2007-9 for each of nine companies for which the Israeli government had a controlling equity stake. The companies ranged in size from a few dozen employees to more than 10,000 workers. In all, there were 155 board meetings and 247 board-committee meetings, and 2,459 decisions were made or updates given.

The board minutes contain a complete record of everything said at the meetings and were not filtered for sensitive information. The authors constructed a database containing: the topics discussed at the meeting; whether a decision was made, and if it followed the CEO's recommendation; whether the board took an initiative to modify or more broadly define the actions to be taken, or requested further information or an update; whether the board was presented with at least two proposals to consider; and whether there was any dissent around a vote.

These data suggest that, most of the time, boards play supervisory rather than managerial roles. In particular, Schwartz-Ziv and Weisbach find that the boards were more likely to receive updates than to make decisions; that they were rarely presented with alternatives; and that they almost always voted in line with the CEO. Nevertheless, the boards examined were active: in 63 percent of the meetings, the boards took some type of action. On about 8 percent of the issues discussed, the boards requested further information. On roughly as many issues, they took initiatives on their own. Taken together, these findings suggest that boards could be characterized as "active monitors" - their activity chiefly involves supervising management rather than dictating the specifics of how the company is run.

Schwartz-Ziv and Weisbach emphasize that they are studying data from a single year from a small sample of companies in a single country. These companies are government-controlled rather than privately held, which raises a potential concern about the extent to which the findings can be generalized. Government company directors, the authors note, are appointed rather than elected, and their monetary incentives are typically smaller than in privately held companies. Notwithstanding these limitations, the unique data examined here allows them to observe real dynamics within boards and between boards and their CEOs.

--Matt Nesvisky