Voici un bref document explicatif de Per Lekvall, membre du Swedish Corporate Governance Board, sur le modèle d’application du « comply or explain » à l’échelle suédoise. L’expérience suédoise en la matière est intéressante à plusieurs égards, notamment parce que l’on peut en évaluer les effets sur plusieurs années. Ce document a été transmis par ecoDa – The European Confederation of Directors’ Associations http://www.ecoda.org/ , association à laquelle le Collège des administrateurs de sociétés adhère.

Après une brève introduction, M. Lekvall explique comment le code suédois est appliqué et quels sont les résultats sur une périodes de 6 ans.

The Swedish Corporate Governance Code, based on the comply-or-explain principle, was introduced 1 July 2005 for the about 100 largest companies listed on the Stockholm Stock Exchange. Three years later, 1 July 2008, the requirement to apply the Code was widened to include all companies listed on a regulated market in Sweden, currently around 260 companies. Having initially been considerably questioned, after only a few years the Code became well accepted by the companies and is now generally seen as an integral part of the corporate governance regulatory system in Sweden. It is administered entirely within the Swedish business sector self-regulation framework, which has a long tradition as a complementary regulation to law and other statutory regulation in Sweden.

The system for managing, implementing and monitoring the Code is in short as follows:

The Swedish Corporate Governance Board is responsible for defining the Code and for keeping it up-to-date with regard to new developments in the field in Sweden and internationally. To this end, the Board annually follows up how the companies’ use the Code, but only as a means to analyze its functioning, not to supervise how individual companies apply the Code.

This duty instead rests with the two regulated markets in Sweden, Nasdaq OMX Stockholm and NGM Equity. This is based on the fact that all companies listed on these exchanges are contractually obliged to apply the Code. The exchanges monitor the adequate application of the Code by their member companies on an individual basis according to a certain procedure, with the possibility to report unsatisfactory application, should a company refuse to respond properly to questions about this, to their respective Disciplinary Committees. Theses, in turn, have an arsenal of increasingly severe sanctions at their disposal, none of which, however, knowingly have been used so far.

Still the stock exchanges only monitor that companies apply the Code properly, not whether the corporate governance behavior they report is satisfactory or not from an investment point of view. This is entirely left to the capital market, i.e. the shareholders and their advisors and intermediaries, to decide on and act upon accordingly.

Hence the Swedish system can be described as strict on the requirement to apply the Code but relatively soft on obligations to comply with individual Code rules. The aim of the Swedish Corporate Governance Board is that all listed companies should apply the Code properly, but      not that all companies must comply with all its rules all the time. On the contrary, the Board encourages companies to use the Code with the flexibility intended with the comply-or-explain mechanism and would, in fact, be concerned if all companies would comply with all rules in the Code. Such a situation would indicate that the Code is not ambitious enough.

Nevertheless, Swedish listed companies are quite compliant to the Code, as shown by the latest follow-up numbers (referring to the reporting year 2010):  50% of the companies reported no case of non-compliance and another 39% reported non-compliance to a single Code rule adding up to almost 90% of the companies reporting no or at most one case of non-compliance. These numbers have been more or less the same over the last three years.

The Board considers these results slightly on the high side in terms of compliance. On the other hand they show that the companies find the Code relevant and can apply it without much trouble.

Another crucial issue of code application is the quality of the explanations given in terms of their information value to the capital market. This has been followed up annually since the introduction of the Code through a fixed methodology each year, thus ensuring a reasonable degree of consistency over time. (The methodology was also to some extent “validated” in the RiskMetrics study of code monitoring and enforcement practices in the EU some years ago, which reported results for Sweden very close to those produced by our method.)

According to this methodology all explanations reported each year are classified according to their information value to the market (not whether they are considered satisfactory or not from an investment point of view) into one of the categories Good, Acceptable and Unsatisfactory/Non-existent. The key issue here is the share of Unsatisfactory/Non-existent explanations, which has developed as follows since the introduction of the Code:

          2005          2006            2007          2008            2009            2010

          28%           23%             15%           27%             29%             15%

The interesting thing with this series is that it demonstrates, first, a learning curve of successive improvement 2005– 2007, during which time only the Large-Cap companies were obliged to apply the Code, then a bounce back up when the Code application was broadened to include all listed companies, and finally a second phase of  downwards learning curve leading back to the 15% level.

The significant drop in 2010 no doubt also has to do with an important change of the Code imposed this year, whereby companies were obliged not only to motivate any case of non-compliance but also to describe the solution they had chosen in lieu of what the Code prescribes. This simple measure has significantly improved the information value of the explanations.

Even though this later development is encouraging, the Board is not satisfied with a situation where about 15% of the explanations are non-existent or not deemed informative enough to the capital market. In principle a “zero tolerance vision” should be applied (although it may in practice be difficult to reach this level entirely). Therefore the Board is currently considering further measures to decrease the number of unsatisfactory explanations.