Companies with at least one woman on their boards outperform: correlation, or causality?
Credit Suisse Research issued a report last week that shows that companies with at least one woman on the board of directors outperform companies with zero women on their boards. What do they mean by ‘outperform?’ Better stock market performance, higher returns on equity, higher valuations and higher payout ratios (aka bigger dividends.) Sounds great. But various critics have pointed out that correlation does not imply causality. Let’s take a look at that in more detail.
First, are these results meaningful or just random chance? CS looked at the performance of 3,000 companies over an eight year period (2006-2014.) Their sample is more than large enough to produce statistically valid results, and the time period is very good, spanning a market peak, collapse and subsequent recovery that was global in scope. It would be nice to have an even longer time frame, but as you go back in time, there get to be fewer and fewer companies with even one woman on their boards (WoB.) 30 year data would interesting, but almost certainly not meaningful due to the low sample size of WoB companies in 1984. I would conclude that the results are highly likely (99%+) to be valid, and not due to chance.
Maybe it is a sector thing? That is a legitimate question: some industries are much more like to have zero women board member than others. As you can see from Table 4 below, over 40% of companies in the Energy, Material and Technology (shame, shame: tech should be a leader in this trend!) sectors have no women on their boards, while less than 30% of companies in the Financials, Utilities, Healthcare and Consumer Staples sector have no female board members. Point One: the CS Research folks show sector-neutral returns in all their charts, so they are adjusting for that. Point Two: even if there is a flaw in how they are adjusting for sector, we now have an eight year track record, and have been through enough sector rotations that sector is not likely to be a significant factor.
Now that the “straw women” arguments are out of the way, let’s move to the heart of the matter. Credit Suisse Research puts it very well:
“While our statistical findings suggest that diversity does coincide with better corporate financial performance and higher stock market valuations, we acknowledge that we are not able to answer the causality question and this is an important caveat to the observations below in the report. Do better companies hire more women, do women choose to work for more successful companies, or do women themselves help improve companies’ performance? The most likely answer is a combination of the three.”
I don’t have a definitive answer, but I have some reasons to believe that the first two factors are material but not more than 50% of the effect; meaning that the women on boards are – in and of themselves – leading to better decision making and better performance.
The problem with the idea that women are choosing to work for “more successful” companies is that it requires women joining boards to have some sort of ability to figure out which companies will have better stock market performance IN THE FUTURE. Years of robust data show that individuals are not capable of doing that, whether female or male. Women have been joining boards for a variety of reasons, but there is no reason to think they are any better than average at joining because their stocks will go up!
Next, are “better” companies more likely to pick at least one woman to sit on their board? If anything, the reverse is likely to be true. Companies where everything is going well tend to have very stable board memberships. Meanwhile, troubled companies frequently appoint new board members in an attempt to turn things around. Therefore, over the period studied (when we moved from fewer companies having at least one WoB to a period when more companies have female board members) troubled companies should be over-represented in the “at least one woman on board” category due to turnover and recruiting.
Which leads me to suspect that a significant part (perhaps more than 50%) of the effect is due to the additional diversity having a woman on the board provides. There is copious academic research on the dangers of groupthink and the benefits of diverse voices, ranging from Irving L Janis’ pioneering work to a lovely 2011 study from Ilan Yaniv.
Diversity doesn’t end at just having one woman on your board, of course. But it’s a start!
Just as one more note, there are some significant differences among countries. The chart below shows the percentage of women on boards around the world. Those in the UK or Israel might gloat at having around 18% female representation versus Canada at 16% and the USA at 14%. And kudos to you. But we are all WAY behind Belgium at 23%, Denmark at 25%, with France and Sweden at 30%.
Final note: the full report can be downloaded by clicking on the link here: CSRI_CS Gender 3000 – Women in Senior Management_final