On the challenge of explaining how boards work

In 1960, Jane Goodall, a trailblazer in the field of primatology, visited the Gombe Stream National Park in Tanzania to study chimpanzees. And so began a 60-year study of chimpanzee social interaction. The study was groundbreaking; revealing new insights about chimpanzee behaviour and interaction. Goodall observed directly, for the first time, human-like behaviours in chimpanzees. These included toolmaking (albeit rudimentary) and armed conflict between competing individuals and groups. Consequently, humanity's understanding, of both chimpanzees and itself, changed.

Some centuries earlier, Copernicus produced insights about the transit of planets; specifically, that the planets have the Sun as a fixed point around which they orbit. This observation undid conventional wisdom, which held that the Earth was the hub around which other bodies orbited. Later, Kepler explained the observations.

These examples illuminate the value of long-term direct observations of dynamic entities, especially groups of entities, to achieve more accurate understandings of not only the entities, but their actions and interactions.

The principle holds in contemporary society. Sociologists and anthropologists, for example, have long seen the importance of observing social groups first hand (long-term ethnographic studies, sometimes involving full participation) to gain insights that might lead to more complete understandings and explanations of group dynamics, and the impacts and consequences of group action and interaction. 

Moving now to consider a subject of great personal interest: boards and governance. How do boards work, and what are the characteristics of an effective board of directors? Can, and if so how, boards influence company performance? And how might one go about finding out?

To date, the predominant approach to tackling these questions has been to apply scientific principles, in search of linkages between attributes of boards and company performance. But this enquiry has raised yet more questions. For example, can a comprehensive understanding of the function, interaction and impact of boards be gained by studying isolated attributes of boards, such as the number of directors, independence, 'diversity', or other static attributes, all from outside the boardroom? Or by applying statistical methods to search for regularities (or differences) in publicly available data? Or by interviewing or surveying directors and/or managers about their perceptions about the conduct or behaviours of directors during board meetings?

Enquiries utilising these approaches have produced thousands of research papers and published articles. They have been helpful in so far as they have provided clues about what may or may not be material to identifying the characteristics of high performing boards and the impact of boards on company performance. But the basis of these studies is not as it first seems. These are not studies of boards in action, they are studies based on representations of specific attributes associated with boards, not actual data about the board going about its work—just as the headline picture looks like a pair of giraffes but they are representations, not giraffes.

A small but burgeoning group of researchers have taken a different approach. Invoking Goodall, they have completed long-term observation studies of boards of directors going about their work (i.e., the researcher in the board room, silently observing the board in session, over an extended period to move beyond the behaviour modification that naturally occurs when someone or something arrives in the environment). To date, fewer than a dozen studies have been published. These studies have produced insights that are somewhat different from those produced by remote studies of isolated attributes of directors and boards. In particular, the importance of certain director capabilities, board activities (tasks) and director behaviours is highlighted. Static attributes, such as board structure and composition, seem to be far less relevant. 

So, two different approaches, and two different sets of conclusions. ​That is perhaps not unexpected. But it does leave a rather awkward question—the same as that faced by Kepler, Copernicus and Goodall, and others who have reached observation-based conclusions that have differed from conventional wisdom. Might the small group be on to something? And, if so, might the majority (in this case, business school academics, regulators, institutions, governance consultants) be prepared to set conventional wisdom aside, to pursue a different understanding of how boards can influence the performance of the companies they are charged with governing?