Does size matter? According to the latest research, it certainly does when it comes to boards.

A team from London, Rotterdam and Hong Kong looked at German supervisory boards, which face mandatory requirements for size, depending on the number of their employees, and found was that as boards grew, performance appeared to drop off. “Our results suggest that forcing firms to have larger boards is costly,” the academics say.

In fact, the researchers, Dirk Jenter of the London School of Economics, Thomas Schmid of Hong Kong University and Daniel Urban of Erasmus University, Rotterdam, found that larger boards may reduce profit margins and lower returns on acquisitions.

The results came about after looking at more than 300 German companies and their performance. German law compels companies of between 2,000 and 10,000 employees to have a supervisory board of 12 members, and companies with more than 10,000 workers to have boards of 16.

What the researchers found was that as soon companies reached 10,000 or more staff and increased their board size, performance fell. The trio say return on investment dropped by 2-3 percentage points at the 10,000 threshold, under one study method. Using another, the drop may be on average as much as 5-6 percentage points.

However, there was a concern that the figures could be manipulated by managers, who could simply down board size by restraining recruitment. So another experiment was undertaken, looking at firms before and after 1976 when the board size laws were introduced.

The figures revealed a similar drop in performance as boardroom members increased.

Efforts to explain why larger boards may be related to lower performance are still in the early stages. The researchers found that it was a “steady state” effect, not just the result of an adjustment as boards transitioned to a bigger membership.

The team notes that directors appointed after a company grows in headcount tend to be less experienced and fewer have doctorates, a common qualification on German boards. They found no evidence that bigger boards increase executive or employee pay, choose less testing pay incentives or cut CEO turnover.

So a mystery remains. Nevertheless, the authors remain convinced of unearthing some vital insight, even if their sample size was small. “Independent of the exact mechanism,” they write, “our findings are a warning that ill-designed board regulations can be costly.”