"From formal body to strategic coach" A conversation with Prof. emer. Herman Daems on the evolution of corporate governance in Belgium

Slagmulder: Within the broader governance framework in Belgium and internationally, how have corporate governance codes historically grown? 

Daems: 
If you look at it historically, you have to go back to the early nineties. This is when the first attempts to codify corporate governance emerged. In Belgium, there were several initiatives, including from the Federation of Belgian Enterprises and the VKW National, but they never fully matured. A real breakthrough came with the development of a code specifically for listed companies. This eventually led to the so-called Code Lippens at the beginning of this century. This formed the first coherent reference framework for corporate governance practices in Belgium. 

Internationally, Belgium was not alone. In the United Kingdom you had the Cadbury Code, in the Netherlands the Tabaksblat Code, and similar initiatives also arose in France and other countries. The common goal of these codes was to make explicit the role and functioning of boards of directors, about which there was often little clarity until then. So the approach was to arrive at a kind of description of what boards of directors do – or should do. 

Slagmulder: What was the reason for the revision that led to the Belgian Corporate Governance Code 2009? 

Daems: 
The financial crisis of 2008 was a turning point. It then became clear that governance practices were inadequate and that there was a need for a renewed framework. At that time, I was asked by the then chairman of the FEB, Thomas Leysen, to take on the chairmanship of the Corporate Governance Committee. 

There was also an institutional incentive. After all, the business community preferred to develop a system of self-regulation, rather than having a framework imposed by the Belgian government. The challenge was therefore to develop a code that was sufficiently robust to be recognized by the government, without getting bogged down in political or other negotiation processes that would slow down or dilute the whole. Eventually, the code was recognized and laid down in a Royal Decree. 

The 2009 Code built on the foundations of the Lippens Code and was inspired by international developments in corporate governance, as well as the expectations of the financial markets in terms of transparency, reporting and governance structure. I would like to emphasise that this was not a personalised code – despite the fact that some have started calling it the "Code Daems". I have always resolutely rejected that designation; It is advisable that a governance code stands on its own, independent of an individual. 

Slagmulder: A recurring area of tension in governance is that between structure and behaviour. How was that discussed when the 2009 code was drawn up? 

Daems: 
Perhaps the most difficult challenge and one of the things I have constantly worried about is that we would not fall into the trap of just setting up structures. Structures are relatively easy to codify: board composition, independence, committee structures, separation between CEO and chairman, and so on. But corporate governance is essentially behavioural. It is about how directors function, how they make decisions, how they deal with information, and so on. That is much more difficult to grasp. 

I have often used the analogy of a traffic code. It determines rules and prohibition and commandment signs, but implicitly assumes that drivers know how to drive a car. The latter is arranged through the driver's license. We did not want to introduce "proof of governance", but it was clear that structural rules alone are insufficient. That is why we have tried to include principles of conduct in the code, such as preparation of meetings, duty of discretion, and dealing with conflicts of interest. 

Slagmulder: How do you see the relationship between legislation and corporate governance codes? 

Daems: 
That brings us to the distinction between hard law and soft law. The law – in particular company law – always takes precedence. The code supplements that law, without lapsing into all kinds of rules and exceptions. The principle of "Comply or Explain" is essential here because it introduces the necessary flexibility in a context of great diversity between companies. You cannot impose one uniform set of rules on all listed companies. Some have a dispersed shareholding, others a dominant reference shareholder, and others are part of a group structure. "Comply or explain" allows companies to deviate from certain recommendations of the code, provided that they justify this transparently. This avoids a rigid, one-size-fits-all approach. For example, the code prescribes that the chairman of the audit committee must be an independent person, or that the chairman should not be an ex-CEO, but in certain circumstances it may still make sense to organize the board in this way. 

Initially, there was resistance to the "comply or explain" principle, from certain political quarters. Some found it unacceptable that companies could decide for themselves not to follow certain rules. But in the end, this principle has proven its effectiveness, precisely because it takes into account the complexity. It has been understood that you are better off with a code that brings together a number of principles and allows for their smooth application. Personally, I think "comply or explain" is a great invention and I would greatly regret it if this principle were to be lost. 

I think that the "comply or explain" principle also has the advantage that if you as a company decide not to follow a certain rule, you have to think about it carefully. If you have to explain why in your annual report, it contributes to a critical evaluation. Afterwards, we did notice that the "explanation" was too non-committal by some companies. That is why we have developed additional guidelines on what exactly a qualitative explanation entails. 

Slagmulder: How has the role of the board of directors evolved since the introduction of the corporate governance codes? 

Daems: 
Thirty years ago, boards of directors were often no more than formal entities that ratified decisions but had little else to say. Today, they are much more involved in strategy and long-term value creation. I distinguish three phases: 

Slagmulder: What is your vision of the future of corporate governance codes? 

Daems: 
There was a period in which corporate governance was reduced to ticking off checklists, from a so-called box-ticking mentality. But I think we are gradually moving beyond that. Boards are starting to understand their own role better and recognize that true governance goes far beyond just formal compliance. 

Despite the great progress, there are still gaps that show that codes, legislation and regulators are not always enough. For example, the tension between majority and minority shareholders remains a fundamental problem, especially in Europe. This is made even more complex by changing shareholder structures, such as the rise of index funds and short-term investors. 

The relative importance of codes may be decreasing. They have largely fulfilled their role of institutionalising good governance practices. I think that people are now sufficiently aware that a well-functioning board of directors involves more than just a code, just as good driving with a car is more than just knowing the traffic regulations. In summary, I would dare to say that the transition from the board of directors as a purely formal body to a coaching body is the most important achievement of the codes. But the work is far from finished. The challenge lies in further improving the quality and diversity of directors, the dynamics within boards of directors, and the interaction with management. 

As far as the impact of new technologies is concerned, I am a bit sceptical about the statement that AI will fundamentally transform boards of directors. There is a lot of talk about the impact of AI on directors, but relatively much less about the impact on management. In my opinion, corporate governance remains highly dependent on implicit knowledge: intuition, experience, contextual insight. That kind of knowledge is often more difficult to codify and therefore more difficult for AI to replicate. Moreover, if all companies rely on the same datasets to generate alternative scenarios, we risk convergence in decision-making. Plenty of food for thought...