Bye, bye Blogger

 From now on, you'll find me on Substack at laurafspira.substack.com. See you there!

Poor research and lazy journalism

Well, it's been a while since I've been sufficiently provoked to let off steam here but today's Guardian headline has made me very cross: "Companies with female leaders outperform those dominated by men, data shows" 

Data doesn't show any such thing. What matters is how data is interpreted. The extensive literature on board gender diversity shows that relevant data can be interpreted in a variety of ways. This means that policy makers and lazy journalists can cherry pick to find support for their arguments.

Anneliese Dodds no doubt has worthy intentions in flagging up an important role for women in post-pandemic economic recovery but her researchers have not helped her cause. The article does at least provide links to the research cited in support of the misleading headline: a report from McKinsey (interesting but not peer-reviewed academic research) and a single article  from a 3 star journal. A competent researcher in the HoC library could have found the work of Alex Edmans, whose academic credentials are rather more impressive than those authors. Alex says it all here. I shall be sending this link to Anneliese (who happens to be my MP) and to Alexandra Topping at the Guardian.

Rant over. Back to worrying about the much more important stuff going on in the world.

What do directors say they do? New insights?

A tweet from Raj Thamotheram alerted me to this Bloomberg headline: "Corporate Directors Say It’s Not Their Job to Monitor CEO: Study".

 I was immediately intrigued as this looked like a challenge to the conventional wisdom about the board’s role - and I was amused by this delightfully acerbic comment in the article:

 

“Usually directors at least pretend to acknowledge their legal obligation to provide oversight of CEOs on behalf of shareholders,” said Nell Minow, who advises institutional investors on corporate governance issues at ValueEdge Advisors. “This acknowledgment that directors see themselves as corporate cheerleaders instead of skeptics whose job is to push back, question, and insist on better is further proof that shareholders will need to support more Engine No. 1-style challenges.”

 

The article helpfully provided a link to the study (so many don’t!) so I headed off to Strategic Management Journal to read more.

 

What a surprise – a paper by US academics using grounded theory! And a paper challenging the assumptions of agency theory! I was quite thrilled but as I read on through the thirty odd pages my disappointment grew. 

 

I was interested in the research method. The predominant approach to corporate governance research in the US is quantitative so it is refreshing to see a qualitative perspective. But how was the research actually undertaken? 50 interviews were transcribed, coded and analysed. There are four authors. The coding process is described in the paper and the appendix provides a lengthy list of interview questions used (is this really grounded theory?) but it is not clear who conducted the interviews. Does this matter? I think so: interviewing is not a straightforward process and I think the paper should at least include some reflection on the variability of the experience. 

 

I read the literature review with interest. Little attention seems to have been paid to the work of UK or European scholars. A 1999 paper by McNulty and Pettigrew is cited but I would have expected to see some reference to the work of Annie Pye, to the work of John Roberts and colleagues on NEDs or even some reference to Bob Tricker’s useful analysis of board roles.

But apart from geographical silos, the literature review also reflects disciplinary silos. I would have expected to see reference to the work of US legal scholars in critiquing board roles, especially Kellie Alces’ paper “Beyond the Board of Directors”.

 

A central finding of the paper seems to be the evidence that the directors interviewed had little enthusiasm for the monitoring aspect of their roles. I don’t think this is a great surprise. Strategy is far more interesting work than compliance. But it is surprising to find directors not only admitting that they don’t monitor but arguing that they can’t and it’s not worth trying. My work on UK audit committees suggests that the monitoring role is challenging for NEDs but it is accepted as part of their role.  Oddly, there is no mention of audit in the paper although the interview questions cover the work of the audit committee. Did any of the directors interviewed interact with external or internal auditors?

 

I have long believed that the agency theory model does not capture the dynamics of corporate governance behaviour and has inhibited the development of corporate governance research so I am always pleased to see it challenged. But it is worth remembering that the need for monitoring predates this theory: financial reporting  and audit exist to support the monitoring process. 

 

The paper cites Professor Boivie’s earlier paper which is an important critique of the ability of boards to monitor effectively and includes a very comprehensive review of relevant literature (but again seems to overlook the work of US legal scholars). On the basis of the copious evidence they marshal, Boivie et al remark that: 

 “… we believe that future research and theorizing needs to focus on boards as advice-giving bodies, or bodies that get involved in punctuated events, and look to other corporate governance mechanisms to secure monitoring." 

 

The interpretation of the data in this subsequent paper certainly appears to support that belief. Now, I hesitate to suggest that eminent scholars might be subject to confirmation bias. I have just been reading a pair of papers by Professor Brendan McSweeney on confirmation bias in qualitative research, so such a suggestion might lead to accusations of my own confirmation bias… But I do think that researchers, especially those claiming to be using grounded theory, need to reflect on their interpretive choices in more detail than is shown in this paper. The appendix provided a little more detail but I really wanted to know how the interviewers responded to some of the comments with follow-up questions.

 

I am left with a strong sense of the wide differences in approaches to corporate governance roles between the US and the UK. Some of this is attributable to differing regulatory frameworks but this study suggests to me that there are also very significant cultural differences. I’d very much like to see such a study replicated in the UK.

 

However we react to the insights of this paper, it does raise a fundamental question about the relationship between corporate governance and management. 

I think the boundary is more blurred than the conventional wisdom suggests (Sir Adrian certainly didn’t agree, we had some interesting chats about it!) and some time ago I started to explore this as reflected in the use of language around governance and management (I've yet to write them up but you can listen to my thoughts here and I'm happy to supply copies of the accompanying slides if you are a real glutton for punishment) Interviewing NEDs on NHS Trust boards provided some interesting insights into how they viewed their roles within the different accountability structure of the public sector: they were able to shift quite easily between providing hands-on support to managers in applying new government instructions and reverting to a monitoring role.

 

I’m looking forward to the imminent publication of ISO 37000 which apparently aims, among other things, at “clarifying the distinction between the governance realm and the management realm.” 

(Some of the links will lead you to full papers, others are not open access but I'm happy to send you a copy.)

 

 

 

Board diversity - there is always more to say!

This week the Financial Reporting Council has published a lengthy (129 pages!) report entitled "Board Diversity and Effectiveness in FTSE 350 Companies". The heatwave had pretty much fried my brain but I was still eager enough to learn its contents to sit in my very warm study staring at my computer screen to read it as soon as it landed. (Pause for short rant: I know I am old-fashioned - I'm certainly old - but I really do prefer to receive hard copies of publications like this. I like to scribble notes on them and flick back and forth between the pages. I'm sure I absorb information from a printed page much more easily than I do from a screen. Yes, I could have printed it out but that would have taken ages and a lot of printer ink and made my study even warmer.)

Was the effort worthwhile? I think it will repay further reading but here are my first impressions.

It's a well written report. I hope this encourages readers to continue beyond the executive summary.

It's very encouraging to see some academic rigour applied to the topic to underpin the recommendations. The research methods are clearly described, in considerable detail. The appendices include a substantial list of references and a dip into the academic literature with a balanced approach that doesn't just cherry-pick studies which support previously determined recommendations, as so often happens in earlier reports. But it's a limited list that does not venture into other disciplines such as law and sociology where scholars have also tackled this topic and offer useful insights.

The researchers are very clear about the problems of inferring causality from correlations (page 9) and provide important caveats about interpreting the statistical data. Other limitations of the research are clearly set out (page 21).

The role of the nominations committee, the least researched of the board committees originally recommended in the Cadbury Committee report, is emphasised.

The importance to boards of diversity of thought is clearly demonstrated and the need for further research into its assumed relationship to demographic diversity is emphasised. This is particularly important as demands on boards to move beyond gender and ethnic diversity continue.

Two aspects that trouble me:

1. The authors assert that the debate should move beyond discussion of why diversity is important to explore how it can be achieved, criticising the business case approach (rightly, in my view) for leading to potentially adverse consequences. However, my reading of the debate over many years is that the advancement of the business case was politically motivated and has led to other arguments for board diversity being insufficiently discussed. The perspective of this report does little to change that. For example: 

"In addition to gender and ethnicity, there are a number of other ways in which directors are not representative of the UK working population" (page 12) 

Why should directors be representative of the working population? Under the current regulatory regime (admittedly a regime that does not adequately reflect twenty-first century business or social environment) directors represent shareholders alone; the arguments for wider representation on boards need to be clearly articulated. Further, the practical problem faced by boards in addressing demands for diversity beyond that of gender is not fully considered. There are also potential issues arising from the transgender debate which may impact boards. 

2. The approach of the authors to qualitative research concerns me.  The predominant background of the research team appears to be psychology and behavioural research, mostly using a quantitative approach. I suspect that their main use of interviews as a data collection method has been in the context of providing consultancy advice to boards  - the interesting BEP/Q-sort tool suggests this, as do the comments on page 103 about the limitations of interviews. Ethnographers have strategies to overcome such limitations, although the significant constraints on conducting research interviews in the context of a pandemic must be recognised. Research described as "mixed methods" (quantitative and qualitative approaches combined) presents important challenges to underlying assumptions about what counts as knowledge and requires more than adding a series of interviews to bolster statistical analysis (page 26)

The addition of an anthropological perspective to the research team could have provided further insightful analysis, exploring in greater depth the context in which board diversity is addressed and experienced and the rituals and symbols which provide clues to how it is conceptualised. Gillian Tett's recent book "Anthro-vision" beautifully demonstrates the value of an anthropological approach to organisations.

Overall,  I think the report is a very useful addition to the knowledge base with the potential for stimulating further well-informed debate around the issues identified, particularly how boards can meet the challenge of demands for different sorts of diversity and how these can stand as proxies for diversity of thought. I do hope that the authors will produce a paper or two in peer-reviewed journals from their data to confirm the validity of what looks to me like very useful data analysis. 

I'll read it again when the weather changes...

Gender and scholarship

 Started reading this paper but got no further than the abstract.

"In December 2020, Nasdaq asked the Securities and Exchange Commission to approve new diversity rules. The aim is for most Nasdaq-listed firms to have at least one director self-identifying as female and another self-identifying as an underrepresented minority or LGBTQ+. While Nasdaq claims these rules will benefit investors, the empirical evidence provides little support for the claim 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 board diversity can actually lead to lower share prices. Adoption of Nasdaq’s proposed rules would thus generate substantial risks for investors."

It's the parenthesis in that penultimate sentence that brought me up short - specifically, the implicit assumption that the reader might be surprised that female scholars should present evidence that board diversity might not be a good thing. While I'm not a "leading female economist", I do find that a bit insulting. And is the suggestion that less eminent female economists might be expected to be less rigorous and more biased in favour of diversity?

The author of this paper is male but I hadn't noticed that before I started reading. Had the author been female, I might possibly have thought it even more egregious. But does the gender of the author of a scholarly article matter? It's not always obvious, anyway. Does it matter more when the research relates to gender? 

In reading the academic literature on board diversity over many years, I don't think it has ever occurred to me to take note of the gender of authors of studies published in peer reviewed journals. I would certainly take note of their disciplinary area. I might take note of their academic affiliation. But my focus would be on the arguments proposed and the evidence adduced. Author characteristics matter to me much more when reading non-academic material. 

OK, have now finished reading the paper. It's not a peer-reviewed publication, it's more of an opinion piece, critiquing the Nasdaq proposal by examining the academic evidence it cites. The author points out, correctly, that Nasdaq has done a poor job, cherry picking sources and citing with a spin in favour of its proposal. The author cites some well known studies and doesn't pretend to be providing a literature review but I find it a little odd that a law professor makes no mention of the growing stream of literature from US legal scholars such as Lisa Fairfax (ooh, a female law professor!) on the topic - an example is here 

It's refreshing to see scholars (of whatever sexual orientation!) challenging the "business case" argument which I have never found convincing. As evidence to support this challenge grows, we also need to see a better articulation of other arguments for board level diversity - moral, social, political - which could also open up a broader debate about corporate power, governance and accountability. The G in ESG discourse is in danger of being neglected.

Gamestop and the gap between companies and shareholders

The Gamestop story is fascinating for a variety of reasons but one issue that it illustrates is the disconnect between investors and companies. The underlying story of Gamestop is almost entirely irrelevant to the trading in its shares. None of the frantic market activity has anything to do with company resources.

When our current corporate reporting framework was established in the nineteenth century, it had a clear purpose: it allowed creditors and those buying company shares to see what had happened to the resources they had provided and to hold those managing those resources to account.

The economic and social environment of the twenty-first century is very different. The development of institutional shareholders and the complex array of investment intermediaries has fragmented that original direct relationship and obscured accountability relationships. Corporate reporting has not kept up with this. Periodically the flaws in the reporting model become too obvious to ignore but the solutions have been pragmatic and politically expedient, rather than based on a reconsideration of the fundamental assumptions on which the model is based.

For example, financial scandals in the 1960s prompted the accountancy profession to establish accounting standards. At the time, Professor Will Baxter outlined the possible consequences of this. He was especially prescient with regard to the impact of standards on education and the potential stifling of original thought about the role of accounting in society.

This avoidance of revisiting basic assumptions is very apparent in the recent discussion paper issued by the FRC which seems to be little more than an attempt to reconfigure the presentation of corporate information without providing any fundamental reflection on the role of corporate reports. It is worth comparing this attempt with the 1975 discussion paper The Corporate Report and the 1988 publication Making Corporate Reports Valuable, both of which tried to assess the underlying purpose of reporting before offering innovative ways to achieve this. Sadly, accounting history does not feature prominently in accounting education. 

The FRC discussion paper claims to offer a "stakeholder neutral" approach to the content of corporate reports. Is this logical? Surely the audience for corporate reports is corporate stakeholders, however they may be defined? "Neutral" may be intended to indicate that no particular stakeholder group will be given prominence but it is surely impossible to decide on the content of reports without considering the information requirements of the audience and the very act of considering these must involve some kind of ordering. 

A narrow shareholder focus for reporting is presumably no longer deemed appropriate. Certainly the reporting framework seems to assume that shareholders are a homogenous group with common interests which is clearly not the case now, if it ever was. The Gamestop story again illustrates that. But shareholders in the original joint stock companies were more likely to be providing resources directly to the company than they are today. 

A more radical approach to corporate reporting could begin by identifying those who provide the resources which enable companies to operate. Assessing their information needs could redefine the context and communication of corporate reporting within a clear framework of accountability and assurance which could underpin a reassessment of the role of audit and the necessary corporate governance mechanisms.

Conventional wisdom

I am increasingly irritated by those who respond to every business issue by pointing out that the board involved is dysfunctional because it is not diverse.

Aston Martin is in the news today. There are no women on its board. More interesting to me is that there is a preponderance of chartered accountants and the issue that has arisen relates to accounting. The company web site is not very informative about the current board but, although it has gone through a major shake up recently, two of the board members - described as shareholder representatives (aren't all board members shareholder representatives?) - have been directors within the group for a while and are CPAs. Asleep at the wheel?

J K Galbraith defined conventional wisdom in "The Affluent Society":

It will be convenient to have a name for the ideas which are esteemed at any time for their acceptability, and it should be a term that emphasises this predictability. I shall refer to these ideas henceforth as the conventional wisdom.

The idea that diversity is a remedy for the poor outcomes of groupthink has become conventional wisdom in the board diversity discourse: one could even argue that it is the result of groupthink among those promoting it.

But it is worth reminding ourselves that no board or committee can operate effectively without consensus: finding a way to reconcile challenge and arrive at consensus is the job of an effective chairman but board meetings take place behind closed doors and minutes provide little insight into how consensus is arrived at. Even personal reports of meetings can vary among attendees. And, of course, there is no guarantee that consensus will lead to good business decisions. How can we distinguish between groupthink and negotiated consensus?

Tracing the history of UK corporate governance regulation suggests that there seems to be an implicit acceptance that NEDs provide insufficient challenge: it was first assumed that the reason for this lay in their lack of independence and this underpinned the approach of Cadbury Code. As a result, the balance of board membership between executive and non-executive changed and boards became more measurably independent from management. There is little evidence that this has improved corporate governance, however you choose to measure such potential improvement, a challenge in itself. And it is telling that the annual Grant Thornton corporate governance analyses show repeatedly that the area of least code compliance is that relating to NED numbers. Research into this issue would be useful but I'm not aware of any. And a moment's thought might lead one to reflect that the change from a situation when the entire board sat round the same table to a situation where contact between NEDs and the full group of EDs can only take place under different circumstances might have had unforeseen consequences with regard to effective communication. Again, this could be a useful research area.

But now it seems that the reason for imputed lack of challenge in the boardroom has more to do with lack of diversity. Both independence and diversity are measurable substitutes for that which cannot be measured - the independent and diverse thought and behaviour which are seen as desirable in board members. But without sound evidence that the prescription of board composition improves corporate governance, boards are likely to be further hamstrung by imposed requirements, especially if all aspects of diversity are to be addressed.

The debate about the purpose of the corporation is welcome. When corporate purpose is substantively and transparently addressed, accountability processes can then be designed appropriately. Do we need boards? Read this and ponder.