3. Discussion and analysis of jurisdictions studied. 3.1.4 The methodology of assessment

«The concept of governance is abstract and latent rather than concrete and observable, and we are not sure how to proxy for this vague concept using observable measures.»

There are a number of ways that effectiveness of CG and a CG regime may be assessed. The discussions of the jurisdictions studied in the Appendices refer to many assessments that have been made of CG performance, or CG improvements in a given market over time. Many of these are undertaken on an annual basis, for example, as typically seen by the UK FRC and the SEHK in relation to the UK CG Code and HK CG Code, respectively.

Broadly speaking, the assessments tend to fall into two categories - those that measure the percentage of CG rules (such as the CG Code) that are complied with (a «Framework Approach»), and those that seek to measure effectiveness by canvassing the views of stakeholders (an «Empirical Approach»). Each of these approaches is based on an observable construct that is used as a proxy to measure the underlying concept of governance. However, as suggested by the above quotation, this requires an assessment of how well the construct in fact measures CG.

These measures are frequently used to inform the policy making process as regards developing standards, as well as assessing the uptake of CG standards and the effectiveness of measures, including but not limited to enforcement, that work to promote improved governance. Accordingly, it is important to understand the information provided by and the limitations of each of these different approaches to measurement.

The Framework Approach measures how many rule-based requirements have been complied with, where each rule is accorded an equal value. For example, this has been done in relation to the HK CG Code as well as the UK CG Code. There are two primary problems with this approach.

First, it glosses over the problem of box-tick compliance. This tends to make a betterВ¬looking story than what is the reality. The UK Institute of Directors (IoD) has suggested that such a measure of CG is "naive", because focusing solely on how companies report compliance with a framework does not consider underlying behaviour. The FRC has also observed the problem of companies paying lip service to CG and so have started to take an active interest in understanding corporate culture and the components relevant for boards to consider as mentioned in Section 3.1.3 «The role of corporate governance culture».

Second, it fails to take account of the relative importance of different CG factors to stakeholders. A high score achieved by complying with a large number of less-significant contributors to good CG may mask non-compliance with requirements that have an important bearing on governance.

The upshot is that the robustness of measures of improvements in CG standards based on an existing framework of rules is subject to significant qualifications. Accordingly, more should not be read into them than are presented, i.e., a measure that indicates more companies are complying with a specific requirement should not be read as implying that this results in the company achieving the desired outcome of the requirement, no matter how propitious that may appear.

It does of course make some basic sense to assess the relative importance of different, measurable CG factors to stakeholders. Particularly so given the premise that good CG is for the benefit of stakeholders. In other words, the question is to what extent has compliance with a rule been effectively translated into practices that benefit stakeholders.

The Empirical Approach is based on stakeholder feedback and involves a more complex process of extracting from responses the relevant factors and assigning a weighting to each factor. An example of this is the approach taken by the IoD, as discussed in see Appendix II.1.1.

While it is tempting to suggest that an approach founded in an assessment of stakeholder views yields a better measure of CG performance or, when assessed over time, improvements in CG standards, a confounding issue is that a factor could attract a lower score not because it is unimportant per se but because it is already sufficiently well implemented in the culture and practices of the market.

To this one must add the problem that the results of an Empirical Approach will be significantly dependent on the profile of the stakeholder population that participates in the study, and this will typically lean toward institutions and away from the retail viewpoint. This bias will be material where retail investors are either significant participants in a market (whether by trading volume or share ownership) or are in need of greater protection.

Underlying this discussion therefore is a more fundamental one: to what extent should CG standards be developed based on a Framework Approach or Empirical Approach, or some combination of the two? In Hong Kong, public consultation, as well as consultation by proxy with different stakeholder groups, tends to be the normal approach to the development of regulations, including in relation to CG. The content of consultations are frequently though not exclusively developed by reference to developments and discussions in other jurisdictions. Standards are often introduced because they appear to make sense in principle. It is trite to point out that a fundamental flaw of an approach based on right-minded concepts being laid down by right-minded individuals, is that it can fail to deal with the reality of persons who do not share the same views. While the drawbacks of the Empirical Approach have already been noted above, it does make some sense to subject rules to empirical ex post verification as to how well companies are performing on them in the eyes of stakeholders - i.e. not just measure whether the company appears to be complying because it has complied with a rule. To fail to look beyond mere rule-based compliance is to mistake the rules for the objectives they set out to achieve.

It is suggested that obtaining weightings through an assessment of stakeholder views are of informational value in relation to regulatory development and enforcement priorities since the weightings can be used as pointers to areas where stakeholders feel more needs to be done.

In Hong Kong, the regulators do not issue any specific assessments of CG performance or developments, although they may do so on an ad hoc basis. Material published by the SFC (such as its Annual Report or other research material published on an ad hoc basis) or the SEHK may include information on CG, such as rates of compliance with HK CG Code.

How often should measures of CG standards be undertaken, and on what basis should they be undertaken to serve as useful guidance for further policy development? Should a body be specifically charged with undertaking such a task and, if so, should it be independent of the apparatus of CG policy development? As regards the latter, the success of the market-practitioner approach to regulatory development in Hong Kong cannot be ignored, whether or not one considers this to be the best «theoretical» approach. Certainly, a body that is independent of the rule makers, one that was established with a view to providing information to the rule makers, may give the review greater weight in the market by avoiding potential conflicts of interest that arise where a rule maker is reviewing the success of its own rules.

An issue that is fundamentally tied up in this question is the issue of corporate culture, as already discussed in Section 3.1.3 «The role of Corporate Governance culture» above. While giving lip service, or taking a box-tick approach, to compliance with CG rules is recognized as a problem, there is little or nothing in the way of formal guidance on CG culture in Hong Kong, it remaining to regulators a somewhat elusive topic that remains embedded in now well-worn catch-phrases such as «tone from the top», more recently to «tone from the middle», or «tone from every layer». CG Culture nevertheless is identifiable and, importantly for regulators, it is increasingly being recognized that wrong CG culture is able to be punished if the correct ground conditions for good CG are clearly laid out. For example, the increasing attention on hard-to-justify executive remuneration packages coupled with greater rights being given to shareholders has resulted in the curtailing of some corporate excesses, as discussed in the remainder of this Section 3 and the Appendices.