20 April 2021

Overstating the business case for diversity understates the change we need to make

There is irrefutable evidence that diverse organisations perform better.

This statement, or its equivalents, is now so commonplace that we cease to question it. So it’s a shock to the system when someone does, as Jesse Fried of Harvard Law School did in a recently published article. Fried was responding to proposals from NASDAQ, which aim to make it a listing requirement for firms to “diversify or explain”. Specifically, companies on NASDAQ would, under the proposals, have to have at least one director self-identifying as female and another self-identifying as an underrepresented minority or LGBTQ+, or explain why not.

NASDAQ claims to have “…reviewed dozens of empirical studies and found that an extensive body of academic research demonstrates that diverse boards are positive associated with improved corporate governance and financial performance.”

An inconvenient truth

Unfortunately, as Fried’s work shows, NASDAQ’s conclusion is incorrect. He states that: “…the empirical evidence provides little support for the notion that gender or ethnic diversity in the boardroom increases shareholder value. In fact, rigorous scholarship – much of it by leading female economists – suggests that increasing diversity can actually lead to lower share prices.”

What’s interesting is that Fried’s statement wouldn’t cause anyone familiar with the academic research on diversity to bat an eyelid. It appears that the non-relationship between board diversity and performance is well established. 

In a very readable review of the academic evidence on gender diversity, Katherine Klein, Edward H. Bowman Professor of Management at Wharton and the Vice Dean of the Wharton Social Impact Initiative, stated that:  “Rigorous, peer-reviewed studies suggest that companies do not perform better when they have women on the board. Nor do they perform worse. Depending on which meta-analysis you read, board gender diversity either has a very weak relationship with board performance or no relationship at all.” Klein goes on to explore some of the reasons why this may be the case.

In a recent Harvard Business Review article, revealingly titled “Getting Serious About Diversity: Enough Already with the Business Case”, academics Robin J. Ely and David A. Thomas of Harvard Business School said: “Leaders may mean well when they tout the economic payoffs of hiring more women and people of color, but there is no research support for the notion that diversifying the workforce automatically improves a company’s performance.”

In his evidence to a UK Parliamentary Committee Inquiry into Corporate Governance, Alex Edmans of London Business School gives an overview of the evidence, cautioning with great politeness that “the evidence base for diversity improving performance is much weaker than stated [by the Committee]”.

All of these authors highlight that the strongest support for linking diversity to performance comes from what Jesse Fried describes as “…reports – prepared by consulting and financial firms for marketing purposes – that claim to find a correlation between the two. But these reports are not academic studies; rather, they are aimed at attracting clients.” In other words, we shouldn’t expect consultancies to let the rigorous facts get in the way of a good story. And as Klein and Edmans point out, even the claims of correlation are not particulary strong.

A high profile series of reports that is often cited in this regard comes from McKinsey, the latest iteration of which is titled Diversity Wins. However, this is not rigorous peer reviewed academic research. Indeed, Edmans has provided an analysis of the methodology adopted for studies in the McKinsey series, showing that what could be concluded from the study was very limited indeed.

So here is a range of distinguished academics, many of them prominent advocates for greater diversity and inclusiveness in business, agreed on one thing: the evidence that more diverse organisations perform better is threadbare, despite practitioner rhetoric to the contrary.

This really matters, for at least four reasons.

• Repeatedly making untrue claims will over time harm the credibility of the people who make them, undermining the case for the change we wish to see, and brushing valuable insight under the carpet.

• Time and energy will be focussed on issues that don’t seem to make a difference, like board diversity, as opposed to those that do, and may even give false comfort as to the progress being made.

• Claims of a mechanistic relationship between diversity and performance will lead to inadequate focus on what has to change for diverse organisations to be successful.

• Focus on the business case distracts attention from the much stronger moral, social, and economic cases, which is what we should be drawing on to motivate change.

Harming credibility

Truth matters. However well-intentioned we are as diversity advocates, we’re on a slippery slope if we start being selective with the evidence. And the more we do this the more we provide fuel for a potential counter-reaction. In this case the ends do not justify the means.  

As if by way of justification it’s been said to me that: people won’t change without a business case. Well, I’m afraid that we’re just going to have to work to come up with better arguments. And if we need a business case to treat people fairly then we’re really in a bad place.

Presenting a business case immediately makes the issue contestable, subject to empirical study. And at the moment, according to this narrow framing, the data isn’t on the side of change. Making a business case that turns out to be illusory will simply undermine the case for change. 

When we’re passionate about an outcome, then win-win thinking is seductive. It can help build the broadest constituency for change. But Ely and Thomas warn: “…when diversity initiatives promise financial gains but fail to deliver, people are likely to withdraw their support of them.” They go on to say that: “research suggests that when company diversity statements emphasize the economic payoffs, people from underrepresented groups start questioning whether the organisation is a place where they really belong, which reduces their interest in joining it.” 

Furthermore, the idea that diversity directly creates performance improvements becomes ever less credible the smaller the minority. So should we only worry about large minorities? Do we just focus on women? What’s the threshold?

Focussing on the visible at the expense of the important

All the evidence seems to suggest that board diversity doesn’t have a positive impact on performance. But it’s an easy area to focus on. The data’s available. Shareholders can vote on it. Corporate governance codes can apply to it. It meets the test that “something must be done”. So we focus on it regardless, and wish the inconvenient truth away. 

Andrew MacKenzie, now Chair of Shell but previously CEO of BHP, set a target for 50% female representation in the global miner (of all industries!) by 2025. He was deadly serious, and devoted significant management time to the task, seeing it as a key part of a broader culture change. In a recent podcast interview with my colleague Randall Peterson he asserted that a narrow focus on board representation encourages a tick box mentality, and also betrays a misunderstanding of governance: it’s the executive team that runs the company not the board.

Indeed the idea that executives rather than the board matter is also supported by recent research by Karl V. Lins, Lukas Roth, Henri Servaes and Ane Tamayo. They look at the share price reaction to the emergence of the Weinstein scandal and find that firms that had women among their five highest paid executives experienced excess share price returns and positive analyst earnings revisions as #MeToo broke. This suggests that investors viewed the presence of female executive leadership as a positive indicator of a non-sexist corporate culture. This was borne out by the fact that firms with women in the Top 5 also had more gender diversity in the lower ranks of senior management. But female non-executive board membership had no impact, suggesting that simply changing the composition of the board doesn’t affect perceptions of corporate culture. Note that the authors can’t determine whether women in leadership led to the non-sexist culture or the non-sexist culture led to more women in leadership. But this appears consistent with what Klein refers to as the “…small but dependably positive associations of female representation in CEO positions and [Top Management Teams] with long-term value creation…”

This all suggests that serious efforts on diversity will start within the management of the organisation not the board. Many organisations now recruit at junior levels from a diverse pool. But at a certain level in the organisation the pipeline dries up. The critical focus for long-term change is probably the levels above and below where this happens. This will rarely be the board.

Perhaps it doesn’t matter if we focus on the wrong issue. It’s still a signal that change has to happen. Perhaps we can view diverse boards as a beachhead.

I’m not aware of any evidence that there’s a trickle down from board diversity to wider organisational change. Indeed, the opposite may be true. Oriane Georgeac of Yale and Aneeta Rattan of London Business School found that communicating success in one dimension of diversity, for example board diversity, could make people feel less concerned about wider aspects of diversity that persisted (for example pay gaps, representation gaps across the workforce). If there’s no longer a tip, people think there’s no iceberg.

Could meeting diversity targets on boards simply be a way of appearing to do something so everyone feels better while really achieving nothing at all?

Inadequate focus on what has to change

Perhaps the most damaging aspect of the narrative that diversity in and of itself leads to better performance is the lack of focus on the deep and fundamental changes needed for diversity to yield progress. As Ely and Thomas state: “Taking an ‘add diversity and stir’ approach, while busines continues as usual, will not spur leaps in your firm’s effectiveness or financial performance.” 

We shouldn’t be surprised if putting women, ethnic or LGBTQ+ minorities into a system created and run by and for heterosexual white men doesn’t immediately lead to success. The organisation where I had my own professional career, PwC, is one that over the last decade or more has in my view been genuinely committed to improving diversity, inclusion, and equality in the work place. But my experience in being part of that effort taught me how much hidden bias exists in how work is allocated, coaching provided, networks formed, performance evaluated, promotion decisions made. Female partner candidates could be criticised for “lacking ambition”, “being emotional, lacking resilience”, or “being mothering”. Yet on successful promotion these characteristics turned out to be much more valuable in the long term than sharp elbows, lack of emotional intelligence, and inability to be caring or nurturing. 

Identifying, surfacing, and changing these inbuilt biases is a project that requires fundamental work. It may require more than just changing processes to reimagining how work itself is delivered. For example, as Claudia Goldin has shown, many high paying jobs are organised in a way that is inherently more attractive to men than women, at least within today’s cultural norms.

This is also borne out anecdotally by MacKenzie. He describes the wholesale change to culture, performance evaluation, hiring and promotion, and very practical aspects of workplace design that was required to support the push towards gender parity in a way that was consistent with performance. This demanded a substantial portion of his personal attention as CEO.

It’s long been recognised in the academic research literature that diversity has the potential to both benefit and harm team performance – the so-called “double-edged sword” perspective. The widely quoted benefit is that diversity of views brings more information into the discussion enabling better decisions to be made. The less publicised downside is that diversity can weaken the social integration of teams, making it more difficult to make use of shared knowledge and creating implicit or explicit friction within the team that makes collaboration harder. Whether the benefit of improving information outweighs the risks to social integration depends on the nature of the task and starting point of the team. 

Randall Peterson from London Business School contends that “the more diverse the team the more diverse the outcome”. In a fascinating analysis Fabrice Cavarretta of ESSEC Business School agrees. Extremes of diversity – high and low – lead to extremes of team performance, good and bad. A very non-diverse team might either be exceptionally well task-matched to its objective or could completely miss the point. A highly diverse team may successfully integrate varied information in solving complex problems or could fall apart with no ability to integrate that information into sensible outcomes. Weak (or strong) teams normally have poor (or good) social integration, which is essential to capturing the benefits of diversity. It’s therefore more likely that increasing diversity will help strong teams get better than it will help weak teams improve.  This substructure around implications of diversity could explain why aggregate relationships between diversity and performance are so hard to find.

So without the commitment to change, or to attending to the factors that make diverse teams more likely to be successful, a numbers approach to diversity will not make things better and could even make things worse. It could lead to female and minority candidates failing, burnt out with frustration, or simply throwing in the towel. Or it could lead to diversity being blamed for problems arising from other sources. 

In developing the case for a dynamic model of diversity that goes beyond the double-edged sword hypothesis, Srikanth Kannan, Sarah Harvey and Randall Peterson show how surface level diversity (e.g. demographic and ethnic factors) can be blamed for problems in teams that are actually caused by deeper level cognitive diversity. In the short term, surface level diversity can increase awareness of different perspectives within the team. But over the longer term, when team members who think about the world in very different ways face co-ordination problems, they wrongly attribute this to the surface level differences within the team. The authors point out that much remains to be learned about the processes through which diversity impacts team performance, and how these evolve over time. They replace the “double edged sword” perspective of diversity by a “tropical depression” metaphor. Diverse teams can degenerate into a hurricane or can diffuse into a rainstorm that leads to sunshine. Figuring out the right interventions to manage these dynamics remains a work in progress.

The prescription for change set out by Ely and Thomas shows it shouldn’t be underestimated and is not for the faint hearted. They don’t sugar the pill:

“…firms may have to make financial investments that they won’t recoup, at least in the short turn, and more will be required of top leaders, managers, and rank-and-file employees alike. Everyone will have to learn how to actively listen to others’ perspectives, have difficult conversations, refrain from blame and judgment, and solicit feedback…developing those capacities is no small feat in any context; it is even more challenge for people working across cultural identity differences”.

The moral, social, and economic case

Should we even care whether there’s a business case for diversity? To my mind not. There are three higher order cases for creating more inclusive organisations.

First, we shouldn’t need a business case to enable everyone to fulfil their potential regardless of characteristics or background. This is simply what it means to live in an open, inclusive, and tolerant society in which everyone has the best opportunity to express themselves and their talents. It’s a moral case not a business case. This case is in my view uncontestable.

Second, you’d have to have been living under a rock over the last few years not to recognise that we’ve got a way to go to create social harmony around issues of gender, race, gender identity, and sexuality. That’s not to say things haven’t got a lot better over time. Just that there’s still a long way to go. Creating fairness within organisations – truly open and inclusive cultures – plays an important role in addressing societal injustice. It’s not hard to predict that a society more at peace with itself will be happier and better one.

Third, there is an economic case, even if there’s not a business case. A country where everyone’s talent can find its best expression without barrier or prejudice should see improved economic performance, as we’ve seen with increased female participation in the workforce. But this economic case is subtly different from the business case. Saying that an economy open to all the talents will be stronger is not the same as saying that every organisation that improves diversity will perform better. The economy and society that reaches that state has by definition transformed itself. Whether diversity improves a given company’s performance depends on whether that company has undergone the required change.

Adding it all up

It’s striking that a number of academics, who are passionate advocates of greater diversity, argue against relying on a business case to promote it. Yet it has almost become a condition for acceptance into polite society to support the notion that more diversity necessarily improves performance. Confirmation bias causes every scrap of evidence that supports the notion to be seized upon, no matter how flimsy. 

This actually undermines the case for a more diverse business world. And distracts attention from the significant work needed to make more diverse business more successful business, and to truly embed equality of opportunity. 

The business world has an important role to play and can help to lead and reinforce the social change we need. But a simple numbers game won’t cut it. And the extent of focus on board diversity, although it’s doable and scratches the itch of needing to do something, looks like a distraction from more important changes. 

We should strive for a world where everyone has a fair opportunity to flourish and is able to contribute their best. There’s good reason to believe this better world will lead to more diverse organisations. There’s less evidence that more diverse organisations will of themselves lead to this better world. Deeper change is needed.

This doesn’t mean we should suspend efforts to promote diversity. Far from it. We need to double down. But as we strive for greater inclusivity and the improved diversity this brings, some businesses will do better and some will do worse. Some may quickly capture the benefits of a more diverse organisation while successfully mitigating the risks. Others may not. It will take time to learn and implement the approaches that consistently harness the benefits of diversity for business performance. We need to be open, curious, and reflective in the face of the inevitable difficulties that may arise. There will be significant bumps along the way. 

It is the moral case, not the business case, that will get us over them.