EUROPEAN COMMISSION

Brussels, 8.12.2016

COM(2016) 774 final

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT AND THE COUNCIL

on benchmarking of diversity practices under Directive 2013/36/EU

I.INTRODUCTION

It has been widely acknowledged that weaknesses in governance arrangements and in the monitoring of management decisions contributed to less prudent risk taking behaviour from the part of institutions in the run-up to the financial crisis of 2007-2008.

To address this, the Capital Requirements Directive 1 (CRD) aims to strengthen the ability of the management bodies 2 of credit institutions and investment firms (hereafter referred to as ‘institutions’) to exercise effective oversight and decision making through a series of measures, including by ensuring that the composition of management bodies is sufficiently diverse.

Article 161(5) of the CRD requires the Commission to review and report to the European Parliament and the Council on the results achieved under the diversity benchmarking, including the appropriateness of benchmarking diversity practices. This report is prepared in view of this requirement.

II.BACKGROUND

The requirement related to the diversity of management bodies was introduced in order to ensure that a broad set of backgrounds and competences are represented in such bodies. This facilitates independent opinions and constructive discussion, and in doing so counters the ‘groupthink’ phenomenon, which is considered to have led to insufficient oversight exercised by the management body in the past.

As clarified by recital 60 of the CRD, management bodies should be sufficiently diverse in terms of age, gender, geographical provenance and educational and professional background. Employee representation is seen as another positive way of improving diversity, by adding a key perspective and genuine knowledge of how a given institution works.

Under the CRD, Member States or competent authorities must require institutions to take into account a broad set of qualities and competences when recruiting the members of their management bodies, and for this purpose to put in place a policy promoting management body diversity. 3  

In addition, recognising 4 that gender balance ‘is of particular importance to ensure adequate representation of population’, the CRD requires ‘significant’ 5 institutions to establish a nomination committee which must (i) decide on a target for the representation of the underrepresented gender in the management body, and (ii) prepare a policy on how to increase the number of the underrepresented gender in the management body in order to meet that target. 6  

Institutions must publish their diversity policy, its objectives and any relevant targets, as well as the extent to which these have been met, as part of the regular corporate governance disclosures mandated by Article 435(2)(c) of the Capital Requirements Regulation 7 (CRR).

It is worth noting that the CRD requirements concerning diversity policy are similar to those of Directive 2014/95/EU on disclosure of non-financial and diversity information, 8 which requires listed undertakings, including listed credit institutions and investment firms, to report, on a ‘comply or explain’ basis, the ‘diversity policy applied in relation to the undertaking’s administrative, management and supervisory bodies with regard to aspects such as, for instance, age, gender, or educational and professional backgrounds, the objectives of that diversity policy, how it has been implemented and the results in the reporting period’.

The CRD 9 also requires competent authorities to collect the diversity information disclosed by institutions and to use it to benchmark diversity practices. Competent authorities must also provide this information to the European Banking Authority (EBA), who must use it to benchmark diversity practices at EU level.

Article 161(5) of the CRD, on the basis of which this report is presented, requires the Commission to review and report by 31 December 2016 to the European Parliament and the Council on the results achieved under Article 91(11) of the CRD, including the appropriateness of benchmarking diversity practices and taking into account all relevant EU and international developments. Article 161(5) of the CRD also mentions the possibility of submitting the report together with a legislative proposal, if appropriate.

In carrying out this review, the Commission drew on the findings of the first diversity benchmarking report 10 issued by EBA. The report, compiled on the basis of information collected from Member States in 2015, was published on 8 July 2016.

In line with the legal requirement, this report takes stock of relevant international developments (part III), reviews the results of the first diversity benchmarking exercise (part IV), and assesses the appropriateness of the benchmarking process (part V). Based on this, it then draws conclusions on ways of improving the effectiveness of the benchmarking exercise and on its usefulness as a tool for the continuous monitoring of diversity developments in the EU (part VI).

III.INTERNATIONAL CONTEXT

The different initiatives that address the diversity of management bodies concern mainly listed companies from various business sectors, therefore this section does not refer specifically to credit institutions and investment firms.

Most of the diversity-focused initiatives concern gender diversity. A significant number of studies suggest that women’s presence on boards is beneficial in terms of improving companies’ performance 11 and a growing number of policy initiatives aim to promote increased gender diversity in companies’ top decision-making bodies.

For instance, in 2013 the Organisation for Economic Cooperation and Development (OECD) sent a strong signal by recommending that member countries increase the representation of women on the boards and senior management of listed companies. Countries are using various policy measures to achieve the goal of increased women representation on boards, ranging from legislative quotas and voluntary targets to diversity disclosure requirements and government initiatives for state-owned companies.

Some countries have opted for mandatory minimum female representation percentages and reach relatively high representation levels (e.g. Iceland and Norway require 40 % female representation on boards; France requires 40 % female representation among non-executive directors in large listed and non-listed companies by 2017; Italy required 33 % female representation among executive and non-executive directors in listed companies and state-owned companies by 2015). A number of other countries have shown promising, albeit lower results related to gender diversity without having mandatory quotas. In Australia, for instance, women represent nearly 22 % of board members of listed companies, 12 while in the US women represent 19 % of all directors at S&P 500 companies. 13 At the other end of the spectrum there are countries, for example in the Asia Pacific region, where the proportion of women on the boards of listed companies remains very low, at around 3 %. 14

Other diversity dimensions captured in the EU 15 by the requirements under the CRD (age, geographical provenance, educational and professional background) were less identifiable in international policies and practices relating to the diversity of management bodies.

IV.MAIN RESULTS OF EBA’S FIRST CRD DIVERSITY BENCHMARKING

Earlier this year, EBA analysed the diversity practices of a representative sample 16 of CRD institutions, for which data were collected by national competent authorities. The benchmarking covered two aspects: (i) the diversity policies that the CRD requires institutions to have in place (including possible targets set as part of those policies), and (ii) the current diversity practices of CRD institutions. 17

Diversity policies

Based on data collected in 2015, EBA found that only around 35 % of the institutions in the examined sample had adopted a diversity policy. Denmark was the only Member State where all sampled institutions had a diversity policy in place. This percentage was 93.3 % in Sweden and was above 60 % in only three other Member States: Spain, Ireland and Latvia. 18  

In terms of diversity dimensions addressed, three quarters of diversity policies in place cover professional background, and around two thirds cover gender diversity and educational background. Only half cover age and 40 % cover geographical provenance (with the caveat that the geographical dimension is relevant mostly for institutions with international activities).

Some of the sampled institutions had set specific targets for the diversity of their management bodies with regard to one or more of the diversity dimensions. 19 The benchmarking revealed that more than a third of sampled institutions had set a target for gender representation. In around a third of these, however, 20 this target was set at zero per cent, thereby making the objective irrelevant. In most cases, gender diversity targets have not yet been met, 21 and most institutions did not indicate a timeline within which they intend to achieve their target. 22  

With regard to age and geographical provenance, diversity targets are currently only met in less than a third of the cases in which they have been set. For educational and professional background, the targets set are met in around 42 % and 52 % of cases, respectively. These better results could be linked to the suitability requirements introduced by the CRD, which states that members of management bodies must possess, both individually and collectively, adequate knowledge, skills and experience to be able to understand the institution’s activities, including the main risks. 23  

All in all, the benchmarking results show that most institutions still need to take action to comply with the CRD requirement of having a diversity policy in place, and in order to meet the targets set with respect to one or more of the diversity dimensions of their management bodies. This may in part be attributed to the requirement’s relatively recent entry into force, in 2014.

Diversity practices

The benchmarking of current diversity within the management bodies of the sampled institutions showed a mixed picture and revealed that there is room for further action, to different degrees, with regard to all of the diversity dimensions.

Data revealed that management bodies have insufficient gender diversity, with only 13.63 % of executives in the sampled institutions being female. Moreover, more than two thirds of the sampled institutions have no female executive directors at all, and only 11 % have a female chief executive officer. The trend seems to be positive though: the percentage of new female executive directors recruited went up to 19.39 % in 2014, from 15.17 % over 2010-2013.

With respect to the supervisory function, the percentage of female non-executive directors is 18.90 %, 24 and 39.18 % of the sampled institutions have no female non-executive directors at all. There is a positive trend in the percentage of new non-executive female recruits, which went up to 22.15 % in 2014, from 18.52 % over 2010-2013.

All the above figures are overall values for all the countries covered; at Member State level the representation of women in both the executive and the supervisory function can differ quite significantly (it ranges from 2.63 % to 40.35 % for the executive function, and from 6.38 % to 28.57 % for the non-executive function).

Based on the data presented, EBA concluded that it is possible to increase the representation of women in management bodies, while noting that this is not only the responsibility of institutions, but also of Member States to implement a social framework that supports equal opportunities for women and men.

As regards the age dimension, the ‘under 40’ age category is relatively poorly represented among executive directors, and especially among non-executive directors. Almost a third of non-executive directors are over the age of 60, and almost a quarter of the sampled institutions had non-executive directors belonging to only one or two age categories. There are however variations between Member States and between different sizes and types of institution (with smaller credit institutions and investment firms having a higher number of directors in the lower age categories than larger credit institutions). While recognising that the higher representation of the higher age brackets is related to lengthier professional experience, EBA concluded that institutions should aim to diversify the age composition of their boards in order to ensure the necessary plurality of perspectives and professional experiences.

As regards geographical provenance, EBA noted that institutions that are internationally active should have geographically diverse management bodies in order to ensure sufficient knowledge about the culture, market specificities and legal frameworks of the areas in which the institution is active. Nevertheless, the benchmarking reveals that the composition of the management bodies of most of the larger credit institutions and investment firms analysed does not represent all the geographical areas 25 in which the institution conducts business. For the smaller institutions analysed, over half are reported to cover all their business areas, but in a number of cases this might be due to the fact that the institution is active in only one geographical area.

As regards professional background, EBA looked at the number of years of experience and the field in which this experience was gained. Unsuprisingly, most executive directors in the insitutions included in the sample had experience in the financial sector and in management and most non-executive directors had experience in the financial sector and in academia. Financial sector experience tends to be lengthy (more than 20 years); this was true for a significant number of executive and non-executive directors.

As regards educational background, the prevalent specialisation is in business administration and economics. In most institutions, only one or two educational backgrounds are represented among executive directors, whereas the supervisory function is generally more diversified.

The above results need to be interpreted in light of considerations related to proportionality and to the difference in size between the executive and the supervisory function. Thus, as is stated in the EBA report, smaller companies (and in particular certain types of investment firms) with a low number of management body members will find it more difficult to reach a high level of diversity than medium-sized and large firms will. Moreover, diversity in all its aspects can be more easily achieved among non-executive directors, given the usually wider composition of the management body in its supervisory function.

The results of the first diversity benchmarking exercise should also be seen in light of the recent entry into force of the CRD diversity rules and the fact that changes in the composition of management bodies typically take time to materialise. As more time needs to be allowed in order to observe the effect of these relatively new diversity requirements, the Commission will continue to closely monitor developments in this area through regular EBA benchmarking exercises.

V.APPROPRIATENESS OF THE BENCHMARKING PROCESS

A number of observations and recommendations can be made regarding various aspects of the process of benchmarking diversity practices on the basis of the first exercise of this type carried out, with a view to the periodic benchmarking activities to be undertaken in the future.

Scope of the benchmarking

As regards CRD institutions included in the benchmarking process, EBA sought to cover a representative sample of entities: at least 10 % of institutions in each Member State were covered by the data collection, and the sample was composed of entities of different sizes and types. The sample used appears to be appropriate in terms of size and composition. It reflects the balance that needs to be found between, on the one hand, the need to ensure that the sample is sufficiently large and representative to make it possible to draw meaningful conclusions and, on the other hand, the concern not to cause excessive burden on the parties involved (institutions, competent authorities and EBA).

With respect to the diversity dimensions analysed, the benchmarking process covered the main dimensions of a diverse management body as specified by the CRD: gender, age, geographical provenance, educational and professional background. However, one aspect listed by the CRD as a positive way of improving a management body’s diversity and variety of perspectives — employee representation — was not covered by this first benchmarking exercise. In order to fully mirror all aspects of diversity as outlined in the CRD, future stock-takings should also include data on the extent to which institutions’ management bodies benefit from employee representation, in accordance with the applicable national legislation.

Data collection process

According to EBA feedback following the first CRD diversity benchmarking, a number of aspects could be improved in the context of future exercises in order to ensure better data quality. EBA has identified the following areas which could be improved on the basis of lessons learnt from this first exercise:

Certain concepts could be clarified and the phrasing of some questions improved in order to facilitate a harmonised understanding of the issues, for example for the notion of geographical provenance.

Data collection could be made easier through the use of IT tools.

Additional data quality checks could be applied at all data collection levels in order to ensure that the information compiled is complete and consistent.

The Commission encourages EBA to work on the basis of these first lessons learnt in order to streamline future exercises, and invites it to continue to take note of the obstacles and possible areas for improvement in future diversity benchmarkings.

With a view to advancing the harmonisation of data collection for diversity benchmarking purposes, EBA may want to consider whether developing own-initiative guidelines could further facilitate the data collection process.

Presentation of the findings

While the data and information collected by EBA provide a good picture of the status quo, it nevertheless seems that certain presentational aspects could be revised in future benchmarking exercises in order to render the data even more informative.

One recommendation is that EBA reflect more specifically the findings of the benchmarking exercise for the subset of ‘significant’ institutions. In particular, it would be valuable if future benchmarking reports could assess to what extent ‘significant’ institutions comply with the requirement to set a gender representation target and whether institutions that have set an explicit target for gender representation achieve better results than those that did not have a specific target.

With regard to the data on the geographical provenance of institutions’ directors, it is suggested to also include information on the number of geographical areas in which the analysed institutions conduct activities. In the absence of such information, it is difficult to assess for instance how many of the institutions covering all geographical areas do so only because they are active in only one area.

The presentation of the findings could also benefit from a clearer differentiation between institutions that have diversity policies, diversity targets or both. In all comparable sections, the data should be presented under the same categories, including total EU averages, and all the indicators and notions used should be clearly defined.

Timing of the benchmarking

The first benchmarking report was compiled by EBA based on information collected in the course of 2015. However, the CRD requirements on the diversity of management bodies of institutions were applicable only as of 2014, which means that only a relatively short time had elapsed since they entered into force.

Consequently, according to EBA, some institutions provided information on their planned, not yet adopted, diversity policies, which presents the risk that the policies effectively adopted and implemented will differ from what has been reported at this stage. Another consequence of this short timespan since the rules were introduced is that it might be too early to observe the effects of the CRD diversity provisions.

Article 91(11) of the CRD does not explicitly mention with what frequency the benchmarking should be performed. In order to continue monitoring institutions’ compliance with the diversity requirements and to identify trends in their diversity practices, it is nevertheless important that the benchmarking exercise be performed regularly. Such periodical benchmarking, the results of which are made public, can also play a role in incentivising credit institutions and investment firms to improve their diversity practices, by having a certain peer pressure effect.

In light of the usual length of mandates of management body members, the Commission deems it unnecessary to conduct the benchmarking exercise every year. It nevertheless considers that the time between future benchmarking exercises should not be longer than three years.

VI.CONCLUSIONS

Following the obligation set out in Article 161(5) of the CRD, the Commission has reviewed the results achieved under Article 91(11) of the CRD and has assessed the appropriateness of the benchmarking of diversity practices.

The review of the benchmarking results showed that there is still considerable room for improvement both with regard to having diversity policies in place and achieving greater diversity of the management bodies of institutions. As regards compliance with CRD provisions, the benchmarking revealed that most of the sampled institutions are not at the moment compliant with the requirement of putting in place a policy promoting the diversity of their management bodies. The degree of ‘significant’ institutions’ compliance with the CRD requirement to set a gender representation target could not be assessed; however, the benchmarking revealed that most of the institutions that had set a gender diversity target, be it on a voluntary or on a compulsory basis, have not yet met it and/or have not set a timeline for doing so. Although these results can be at least partly attributed to the rules being recent, they highlight the need for both institutions and supervisors to make further efforts to ensure that the required diversity policies are properly put in place.

The review of the appropriateness of the benchmarking process revealed some aspects that could be improved with a view to future exercises. Most notably, the bechmarking and presentation of the findings should facilitate understanding on the level of ‘significant’ institutions’ compliance with the requirement to set a female representation target; the benchmarking could be extended so as to also cover the employee representation dimension; the benchmarking should be conducted at regular intervals and at least every three years.

All in all, the Commission review showed that the diversity benchmarking is a valuable source of information on the degree of institutions’ compliance with the CRD diversity requirements. The exercise is also a useful stock-taking tool, providing a snapshot of the diversity situation and policies of CRD institutions at different points in time.

As such, the benchmarking of diversity practices constitutes a useful tool for assessing the impact and effectiveness of the CRD diversity requirements over time. Regular benchmarking exercises will make it possible to continuously monitor compliance with the relevant provisions and to see future trends in the field of diversity, as more experience is gained with these relatively recent requirements. Furthermore, by enhancing transparency on institutions’ diversity practices, the benchmarking might also incentivise better industry practices by having a peer pressure effect.

Consequently, the Commission is satisfied with the overall added value of benchmarking diversity practices and considers the tool — with the minor adaptations outlined in this report — important in order to identify in future years the effects of the CRD diversity provisions. Therefore, at this time the Commission does not consider it useful to envisage putting forward a legislative proposal to amend these provisions.

(1)

Directive 2013/36/EU of the European Parliament and of the Council of 26 June 2013 on access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms.

(2)

As per Article 3(1)(7) CRD, ‘management body’ means an institution’s body or bodies, which are appointed in accordance with national law, which are empowered to set the institution’s strategy, objectives and overall direction, and which oversee and monitor management decision-making, and include the persons who effectively direct the business of the institution.

(3)

Article 91(10) CRD.

(4)

Recital 60 CRD.

(5)

Article 88(2) CRD refers to ‘significance’ in terms of institutions’ size, internal organisation and the nature, scope and complexity of their activities.

(6)

Article 88(2)(a) CRD.

(7)

Regulation (EU) No 575/2013 of the European Parliament and of the Council of 26 June 2013 on prudential requirements for credit institutions and investment firms.

(8)

The transposition deadline for this directive is 6 December 2016.

(9)

Article 91(11) CRD.

(10)

Report on the benchmarking of diversity practices at European Union level, available at http://www.eba.europa.eu/-/eba-publishes-a-report-on-the-benchmarking-of-diversity-practices-at-european-union-level .

(11)

For instance, a recent IMF working paper (‘Gender Diversity in Senior Positions and Firm Performance: Evidence from Europe’, March 2016) found that replacing one man with one woman in senior management or on the corporate board is associated with between 8 and 13 basis points higher return on assets.

(12)

Korn Ferry Diversity Scorecard 2016 — Building Diversity in Asia Pacific Boardrooms.

(13)

Edward Kamonjoh, ISS, ‘Boardroom Refreshment. A Review of Trends at U.S. Firms’, 2015.

(14)

Korn Ferry Diversity Scorecard 2016 — Building Diversity in Asia Pacific Boardrooms.

(15)

Moreover, on an additional strand of diversity policies, the EU and almost all (27) of its Member States are party to the UN Convention on the Rights of Persons with Disabilities, which in its article 31 on Statistics and data collection requires the collection of information, including statistical and research data, to enable the formulation and implementation of policies.

(16)

Covering around 14 % of the total number of CRD institutions in 27 EU Member States and two EEA countries (Iceland and Norway) that participated in the data collection exercise.

(17)

864 institutions reported information on their diversity policies and 873 institutions on their actual diversity situation.

(18)

Full results per Member State are available on p. 6-7 of the EBA report on the benchmarking of diversity practices.

(19)

Of these diversity dimensions, the CRD only requires targets for the representation of the under-represented gender, and this only in the case of institutions that are significant in terms of their size, internal organisation and the nature, scope and complexity of their activities.

(20)

92 of the surveyed institutions.

(21)

Only 90 surveyed institutions reported that they already meet their gender diversity targets.

(22)

Only 83 surveyed institutions had set a timeline for meeting their gender diversity targets.

(23)

Articles 91(1) and 91(7) CRD.

(24)

However, this total number includes outliers Iceland and Norway, which as per EBA data have a female representation of 52.38 % and 35 %, respectively.

(25)

For the purposes of the EBA benchmarking, the following geographical areas were defined: EU/EEA, Europe non EU/EEA, Africa, America, Asia, Australia.