Every so often an idea comes along which defines the spirit of the age. If the first decade of the millennium is remembered for the explosion of interest in environmental conservationism and the climate change agenda, the second will surely will go down as the one when the diversity agenda moved into the spotlight, certainly in the UK and in the USA. In the case of the UK, the stage was very much set in the Special Duties Regulations (2011), following on from the Equality Act (2010). These set out a requirement on all public bodies or organisations carrying out public duties to provide regular “information to demonstrate their compliance with the public sector equality duty.” This has meant in practice a burden of proof to explain discrepancies in the level of recruitment and remuneration across different genders, races, etc. The spotlight switched ineluctably from processes to outcomes as the focus of scrutiny.
As details of the “gender pay gap” and other inequities started to spill out and the heads of the associated organisations struggled to find convincing explanations of the statistical anomalies, the clamour for change to address the implied injustice grew to the point where it became irresistible. Now, one might argue that it is an impossible task to explain why outcomes are not equal in society. There is little evidence to suggest that equality ever arises naturally in any area, with a myriad reasons why differences arise and persist. Essentially equality is an unstable minimum entropy state from which any change will be such as to effect an increase: so it is equality rather than its absence which ostensibly requires explanation. No matter, the data was out there, those deemed responsible were unable to come up with convincing explanations, and so instead they have had to take upon themselves the enactment of processes designed to address the imbalance.
Although this agenda has played out in the first instance in government-sponsored organisations and educational institutions, in recent years we have seen it encroach increasingly into private corporations. This has not been without its problems because, whereas organisations which receive public funding, by their nature or by the fact of being in receipt, have a public service duty which adds a moral dimension to their mission, private corporations have a primary responsibility to ensure profitability. This means that the moral force of any arguments for imposing a diversity policy has to be balanced against the envisaged impact on the effectiveness of running the business.
What we have therefore seen is a narrative along the lines that teams and organisations which are more “diverse” perform better, resulting in more profitable outcomes, so a diversity policy promoting more equal representation and remuneration across different identified groups is by definition in the interests of the company so should not, and indeed must not, be questioned. A number of stylised facts are typically adduced as evidence of this proposition, a common one being that the most profitable large corporations tend to be those which are able to demonstrate the most progress with the diversity agenda. As this argument is quite central to the issue, it merits examination in some more detail.
The first thing to say is that care needs to be taken with any argument of causality based on statistical evidence. To show that A has caused B, we really need a parallel control experiment in which not-A is seen to result in not-B or less of B. We see the problem when we frame the issue within the wider context of asking what are the factors which make a business successful. We might find statistical “evidence” for many things like breakout rooms with pool and table football, allowing employee sabbaticals, outsourcing IT support and HR, away days in exotic locations for executives, etc. Such things are unlikely to be found in smaller, less successful companies. Does that mean that introducing them would help transform them into bigger, more successful companies? Clearly not. These things have associated costs which make them prohibitive for most companies. The causation, such as there is, is better understood as being in the other direction. Indeed, when one considers how little diversity there tends to be in smaller, less profitable companies in the face of the purportedly compelling evidence of the benefits diversity brings, one might legitimately ask why more of them haven’t cottoned on and got with the diversity program? Alternatively one might trust their judgement that a commitment to diversity is a luxury they cannot afford at their level of development and profitability.
Of course, that is not to say that diversity does not bring any benefits. Such a proposition would find it equally hard to find evidence in support. The truth is almost certainly somewhere in the middle and, I would suggest, requires a more nuanced understanding to get at it. What do I mean? Dr. Matthew Syed, table-tennis champion and current affairs commentator, in his recent book Rebel Ideas: The Power of Diverse Thinking (John Murray, 2019) sets out the case for harnessing what he calls “cognitive diversity” to solve the most complex problems. He draws on examples from psychology, economics, anthropology and genetics, drawing lessons from a dazzling range of case-studies. In essence his point is that we must avoid the dangers of group-think, by encouraging constructive dissent and including on our team people who bring diverse perspectives on problem-solving.
Although the book and many of the examples are new, the idea is not. It can be traced back at least as far as the seminal research of Meredith Belbin in the 1970s. Over the course of a decade or more he developed the Team Role Inventory which consisted of eight (later nine) roles which he found needed to be fulfilled in a high-performing team. This led to the founding of the eponymous company Belbin Associates. Many corporations have invested significant resources into implementing the Belbin program to help ensure that there is sufficient diversity of perspective in teams and to assist team members in understanding the value of roles other than the ones (typically three per person) they most naturally slot into. I participated in a number of such programs in the early 2000s.
It is not my interest here, however, to get into the details of what makes for a high-performing team. The point I would look to make is simply this: insofar as “diversity” as pursued under the banner of a company’s diversity policy encompasses cognitive or team-role diversity as argued for by Syed and Belbin, there is good reason to expect favourable results in terms of team (and therefore company) performance. But when you survey the Web sites of top companies and look at what they are saying about “diversity” you will struggle to find much mention of cognitive or team-role diversity. Rather the assumption appears to be that elevating the degree of representation and remuneration of certain currently under-represented groups in teams and in the company as a whole will in and of itself enhance performance.
Is this a valid assumption? Well, it is certainly a plausible-sounding hypothesis. But whereas with the Belbin approach we start out with a fairly objective set of criteria which can be used to determine which deficiencies a team has in terms of given team roles in the inventory being over- or under-represented, how that can be addressed and what benefits might be expected to accrue, it is less clear in what way a team which encapsulates “diversity” in terms of equal representation of previously under-represented groups will necessarily perform better. A number of issues remain unresolved, indeed unaddressed.
In particular, assuming parity has been achieved in terms of group representation, we might ask how optimal the resulting team is in terms of role balance and diversity of viewpoint. How do we know we haven’t replaced one kind of group-think with another? While it is reasonable to suggest that, in attaining group diversity, viewpoint diversity might at the same be achieved, it is by no means guaranteed. And what assurances are there that a group which satisfies the criterion of group diversity will not over time slide into one or other form of group-think, as has often happened to high-performing teams, particularly in the transition from small- or medium-size company to a large corporation? Based on measurement solely of the level of group representation and remuneration, there is no way of deciding the matter one way or another.
In short, you get what you measure and if all you measure is levels of representation and remuneration, that is all you will be able to claim you are securing. While it can be argued (and frequently is) that equal representation and remuneration at the group level are a good in themselves, it cannot at the same time be claimed that performance has been improved without further investigation and evidence. There is also, of course, the question of whether a degree of unfairness or bias will have to be introduced into company processes in order to secure and maintain such equal representation, which effect might stir up dissatisfaction and cause problems of its own. But that is another matter which would require separate investigation. Hopefully as the discussion about diversity and optimisation of team performance develops, the context in which it takes place will widen and these issues will come to be addressed, as they perhaps were in the recent past, in a more substantive way than they are currently.