4.6.8 Based on literature and previous studies, how is the national corporate governance code applied in practice?

Code compliance is studied comprehensively and annually in the Netherlands. This already started in 1997 with several compliance studies on the "Recommendations on Corporate Governance in the Netherlands" of the committee Peters (De Jong, De Jong et al. 2005) (Dutch Monitoring Committee 1998) (De Jong and Roosenboom 2002, p. 927). As stated above, these forty recommendations were not followed as often as had been anticipated (De Jong and Roosenboom 2002, p. 927). In the latest study of the 40 recommendations, De Jong and Roosenboom compared the situation between 1997 and 2001 of 40 companies and concluded that at first glance progress had been made. For example, more information was provided on the supervisory board committees, their number of meetings, the remuneration of individual board members and their shareholdings, and profiling and regulations for the supervisory board were in use. However, these results need to be qualified since stagnation or even decline had been measured on some points, such as mentioning the number of supervisory board meetings or the companies that have a separate corporate governance paragraph. On average the companies tried to create the impression that they have taken on the recommendations. However, the ultimate impact of their recommendations was very limited or even absent (De Jong and Roosenboom 2002, p. 927). With respect to the compliance with the Dutch Corporate Governance Code 2003 and the 2008 code, an annual research was performed by the Corporate Governance Code Monitoring Committee1 established on the 6th of December 2004 by the Minister of Finance, the Minister of Justice and the State Secretary of Economic Affairs. Simultaneously other researchers conducted some compliance studies as well. The main results of these studies will be summarised below in a chronological order, for the purpose of presenting some concluding remarks on code compliance in practice in the Netherlands up till now.

Akkermans et al. examined Dutch code compliance during 2004 for 150 companies as commissioned by the Corporate Governance Code Monitoring Committee and concluded that: "The findings indicate a high level ofcom-pliance with the Code. Moreover, the extent of compliance is positively associated with company size. (...) In addition, the nature and content of the explanations provided for non-compliance are remarkably similar across companies, which may indicate symbolic compliance with the Code sbest practice provisions" (Akkerman, Van Ees et al. 2007, p. 1106). Taking into consideration that this concerns the code's first year, the level of compliance is regarded as high: on average 81% of the best practice provisions are complied with directly and 7% are explained2 (Dutch Monitoring Committee 2005, p. 5). An average of almost 5 deviations per company can be seen. A fixed set of provisions are deviated from (only 16 provisions are associated with 80 per cent of the deviations), for example: provisions related to the maximum appointment period of executives (provision II.1.1 explained by 100 companies), the maximum severance pay (provision II.2.7 explained by 81 companies), the contents of the remuneration report (provision II.2.10 explained by 50 companies), the rules regarding executive shareholdings (provisions II.2.6 and III.7.3 explained by 49 companies) and the providing of instruments such as webcasting at gatherings with analysts (provision IV.3.1 explained by 45 companies) (RUG Report Akkermans, Van Ees et al. 2005, p. 4).

With respect to 2005 a higher level of compliance is measured, especially with regard to the small-cap companies (RUG Report Akkermans, Van Ees et al. 2005, p. 68) (Dutch Monitoring Committee 2006, p. 18). The direct level of compliance involves 92% of the best practice provisions and 4% of the provisions are explained. The provisions on the remuneration policy (provisions II.2.9 and II.2.10) and the statement on the internal risk and control systems (provisions II.1.3 and II.1.4) are complied with worst of all (Dutch Monitoring Committee 2006, p. 7). Moreover, the quality of the explanations was researched in 2006. Some degree of subjectivity is inevitable; therefore the committee is careful with regard to the results. 85% of the explanations can be classified as "understandable". The "controllability" of the explanations (using publicly available information) is, in 47% of the cases, labelled as positive or neutral. In 12% of the cases a negative view exists and in the other cases no clear opinion on the controllability of the explanations is possible (Dutch Monitoring Committee 2006, p. 44).

As regards 2006 a certain point of stabilisation seems to have been achieved; the level of compliance compared to 2005 has decreased slightly (from 92% to 90%). Especially the provisions on board remuneration remain difficult to comply with (70%). Furthermore, the Monitoring Committee does not consider the increased standardisation in the explanations in the case of non-compliance a desirable development (Dutch Monitoring Committee 2007, p. 6) (RUG Report Akkermans, Van Ees et al. 2007, p. 2 and 4). The top 5 deviations remain almost the same as in previous years: the maximum board appointment period (II. 1.1), the maximum severance payment (II.2.7), the rules regarding shareholdings (II.2.6 and III.7.3) and the establishment of supervisory board sub-committees (III.5) (RUG Report Akkermans, Van Ees et al. 2007, p. 4). With respect to 2007 the level of compliance has again decreased to 89% and compliance with the best practice provisions on board remunerations (67%) remains relatively low (Dutch Monitoring Committee 2008, p. 5). The provisions on internal risk and control systems are no longer a real issue (RUG Report Akkermans, Van Ees et al. 2008, p. 4). To be comprehensive: the Association of Securities Holders (Vereniging van Effectenbezitters, the VEB) conducted research on the compliance as well -although this was less extensive - and mentions a level of compliance in 2006 and 2007 of 63% (Tomic and de Vries 2006).

No significant changes as regards compliance in 2008: the top 5 deviations remain almost the same again (RUG Report Akkermans, Van Ees et al. 2009, p. 1) and the number of deviations has decreased from 683 in 2007 to 619 in 2008 (Dutch Monitoring Committee 2009, p. 16). It is remarkable that the level ofcompliance receives little attention in the 2009 report but that is perhaps because of the point of stabilisation reached in the level of compliance. In the EU comparative study on code compliance during 2008, all 15 companies researched disclosed comply or explain information and an average of four deviations per company was identified, with a difference between large-cap companies (average of six deviations) and mid-cap companies (average of four deviations). Most of the deviations concern the board of directors (45%), followed by the remuneration (32%), or, more specifically, the maximum severance payment, maximum duration of board appointments, having no more than one non-independent board member on the supervisory board and the performance criteria for stock options (RiskMetrics Group 2009, p. 93). Van Ees and Hooghiemstra compared the changes in the explanations of 89 companies between 2005 and 2008. They noticed a decrease in deviations (27.2%) and that, compared to 2005, 61.2% of the explanations provided remained unchanged in 2008. They concluded that the non-compliance and the reasons provided for non-compliance show little in the way of dynamics and they doubt the efficiency of the comply or explain principle (Van Ees and Hooghiemstra 2010, p. 165).

The Dutch Corporate Governance Code 2008 became applicable as from 1 January 2009, hence the Monitoring Report of 2010 is the first report that reviews the compliance with the new code. An average of 6 deviations per company can be seen (RUG Report Akkermans, Van Ees et al. 2010, p. 3). The compliance percentages on board remuneration remain traditionally low (especially provisions II.2.12 and II.2.13), as well as on the maximum appointment period and maximum severance payment (RUG Report Akkermans, Van Ees et al. 2010, p. 2). However, as it is an assumption, the compliance with the new provisions of the actualised code is, in general, high. In many cases the compliance must be assumed, since hardly any explicit information is provided on, for example, corporate social responsibility and the involvement of the supervisory board whenever a public takeover occurs (RUG Report Akkermans, Van Ees et al. 2010, p. 2). Abma and Olaerts researched the code compliance with four 'hard to comply with' code provisions of 100 companies during 2010 and concluded that a lack of proper motivation still exists and, like Van Ees and Hooghiemstra above, they doubt the efficiency of the comply or explain principle (Abma and Olaerts 2011, p. 109). In the Monitoring Report of 2011 the Corporate Governance Code Monitoring Committee states, in response to the EU Green Paper 2011, that the Dutch system of the comply or explain principle and its monitoring score very high on a European level. Although a few points for improvement can be seen, which the Committee will discuss with the companies concerned (Dutch Monitoring Committee 2011, p. 5). In comparison with the compliance during 2009, the Dutch Monitoring Committee states that a slight increase is visible in the average number of best practice provisions explained per company (5.2 explanations in 2009 and 5.4 in 2010). Moreover, the top 10 best practice provisions explained the most remain - except for a few position changes - the same (Dutch Monitoring Committee 2011, p. 21).3 Three of its main conclusions are of extra interest: (i) respecting the existing remuneration arrangements (not in line with the code) is only acceptable in the case of management board members appointed before 2004, (ii) a reference to the company's own regulation without further information is regarded as non-compliance explanation, and (iii) temporary non-compliance for a period longer than one year requires an explanation as to when the code will be expected to be complied with again. Hence, the Dutch Corporate Governance Code Monitoring Committee rightly focuses on sufficient explanations and criticises ' copying behaviour' (Dutch Monitoring Committee 2011,p.12).

To conclude, the national corporate governance code seems to be complied with properly in the Netherlands, although some comments should be made. After an increase in the level of compliance, in recent years a point of saturation has been reached or the level of compliance has slightly decreased. Possibly, this development is the result of a lack of attention to the code by the companies involved and, as a result of the actualised code, has improved. Moreover, the reasons provided for deviations from code provisions and their quality remain an item to be improved. A fixed set of often reoccurring deviations can still be seen where ' traditional' topics are concerned, such as the appointment period, severance payment and board remuneration. An average of 5 deviations per company seems quite high, given the attention paid to corporate governance in the Netherlands. Possibly, the fact that the Dutch code is quite detailed plays a role and also the fact that the Dutch government does not choose to legislate the unpopular best practice provisions, since it believes in a 'one size does not fit all' mentality.