Benchmarking Responsible Investment

Transcription

1 Benchmarking Responsible Investment 1 September 2011 Prepared for National Employment Savings Trust Prepared by Tim Currell Aon Hewitt Limited Registered in England & Wales No Registered office: 8 Devonshire Square London EC2M 4PL This report and any enclosures or attachments are prepared on the understanding that it is solely for the benefit of the addressee(s). Unless we provide express prior written consent no part of this report should be reproduced, distributed or communicated to anyone else and, in providing this report, we do not accept or assume any responsibility for any other purpose or to anyone other than the addressee(s) of this report. Copyright 2011 Aon Hewitt Limited. All rights reserved. Authorised and regulated by the Financial Services Authority.

2 Contents 1. Executive Summary 1 2. General Data 2 3. Policies, codes and standards 7 4. Voting and engagement Smaller schemes 20 Appendix A - Questionnaire 21 Appendix B - Participants 32 Disclaimer 33 Aon Hewitt Limited

3 1. Executive summary Introduction Hewitt Associates Limited were appointed by the Personal Accounts Delivery Authority (now National Employment Savings Trust, 'NEST') to carry out a benchmarking exercise in the area of responsible investment (RI). We believe that long term asset owners have both rights and responsibilities. Furthermore, investment decisions made by pension schemes can have an impact, either positive or negative, on the wider world and therefore on the individual members of those pension schemes. NEST has requested that we approach pension schemes around the world in order to find out how they approach responsible investment issues particularly at this stage with respect to voting and engagement for equity investments. We therefore created a questionnaire that would explore a range of responsible investment issues which was sent to pension schemes of varying size around the globe, and this report summarises the 15 responses. This report also draws on our general experience gained over ten years activity in this area. In a few areas we have been able to shed additional light by referring to the data collected by the Secretariat of the UN Principles for Responsible Investment ('PRI') in their 2011 Survey. We are grateful to the Secretariat for their assistance. Our Questionnaire is included in Appendix A to this report. In addition, Appendix B lists the direct respondents. NEST undertook a consultation period, the results of which showed that participants believed that the scheme should be acting in a "world class" manner with regard to investment considerations, including responsible investment and ownership. This paper sets out the steps that NEST may wish to take in order to progress its capabilities in these areas, and includes advice with regard to the more important steps to be taken in the shorter term. At various points through the report, we have identified recommendations or suggestions for NEST to consider. These are indicated by bold and italicised text. Key recommendations from the report include: To consider internal asset management once assets reach a more significant size. A dedicated RI internal team would become increasingly important as assets grow, whether to perform some functions and monitor third parties performing others, or providing the full range of RI operations. The development of RI policy across asset classes, starting with developed market equities (and / or bonds), with other asset classes added as the scheme size grows. The use of a specialist provider to perform functions such as engagement and collaboration, particularly in the earlier stages of NEST's development, is seen as a cost effective solution. Developing a clear idea of how the composition and function of the RI team will develop as NEST assets grow, given a number of differences in the abilities and strength of small versus larger funds. Aon Hewitt Limited 1

4 2. General data Scheme Size (Q2) The chart shows the spread of respondents in terms of value of assets. Spread of Fund size 21% < 10 bn bn 51% bn 21% >50 bn 7% Slightly more than half of our sample were fairly small funds (Iess than 10bn Sterling), but the sample also included some very large funds. DB or DC (Q1) Of the fifteen respondents, there was a fairly even split between different types of scheme. The single respondent in the 'other' category was a multi-scheme, with both DB and DC sections. 1 Number of pension funds that are A pure DC scheme A pure DB scheme A DB scheme with a DC section A reserve fund 3 4 Other continued on next page Aon Hewitt Limited 2

5 DB or DC (continued) Another way of looking at this data is to consider the size of the various different types of asset pools. With this perspective we can see that pure DC schemes were the largest single category of respondents, accounting for almost half of the assets. % of assets which are... 13% 6% A Pure DC Scheme A pure DB scheme 48% A DB scheme with a DC section Reserve Fund 27% Other 6% Internal management of assets (Q3) From our sample, there is a trend for the largest schemes to manage most if not all of their assets internally. This applies equally to the DC schemes (either pure DC, or those with a mix of DB and DC) as the others in our sample. This is not surprising since larger schemes can exploit economies of scale to make internal asset management cost effective. % internally managed Value ( bn, log scale) Note: In this chart (and in similar following charts) the pure red dots represent the pure DC schemes in our sample, whilst the green dots have a mix of DB and DC members. Blue dots are the other scheme structures, including pure DB schemes and reserve funds. A logarithmic scale has been used in some of these charts to better highlight trends based on size, with the plans ranging from c. 150m to c. 275bn. Recommendation We believe it is worthwhile for larger schemes to evaluate the cost / benefit of internal management for at least a portion of their assets once the assets in the scheme reach c. 5bn, if not before. Aon Hewitt Limited 3

6 Active vs. passive (Q4) The results for Question 4 (portion of assets actively and passively managed) do not show a clear pattern. This is to be expected since there are many factors that affect the active/passive decision, including asset class considerations, geographical preferences and the sponsor's or trustees' own philosophical view. % managed actively Value of assets ( bn) Internal ESG / RI teams (Q5) Recommendation Two thirds of the respondents have a dedicated internal ESG / RI team. This proportion is higher for the larger schemes, being almost 85% (5 of 6) of those with assets above 15bn, but even among the smaller schemes the proportion with a dedicated internal team is significant (half of the schemes with assets under 5bn have a dedicated team). The schemes in our sample all have an interest in responsible ownership and therefore it is no surprise to find that so many of them, including many of modest asset size, have a dedicated internal resource. We believe that the quality of understanding of RI issues, and service provided by fund managers and other specialists in this area, have improved somewhat over recent years. It is still the case that schemes with a particular interest in this area are likely to benefit from developing a dedicated internal resource, however this result supports the view that, at least in the early stages, NEST could rely on its external providers for implementing any policies or practices that it chooses to adopt in the area of RI. Once the assets under management have grown NEST may wish to consider building out the internal team and developing the functions that it performs for the scheme. Even in the absence of an internal team providing the full range of RI services, we believe that an internal team dedicated to RI monitoring, in this case of external providers, is important. However, perhaps as important as the use of an internal or external team is the integration of the ESG/RI considerations in the investment process. We believe that the integration of these factors is critical to their successful implementation. Aon Hewitt Limited 4

7 ESG/RI staff (Q6) It is no surprise that the largest respondent to our survey employs the most full time equivalent ("FTE") ESG/RI specialists. However, beyond this there was no clear pattern in the results from this question. The majority of respondents answered that, where a team existed, it was less than two FTEs. More than 5 FTE staff Between 2 and 5 FTE Less than 2 FTE staff No internal resource Value of assets increasing (log scale) Proportion spent on RI activities (Q7) Only one respondent (the smallest scheme in our group, with an asset value equivalent to c. 150m) spends more than 0.5bps of the scheme's value on RI activities each year. All other respondents spend less than this level. Time horizon (Q30) Almost three quarters of the respondents view 'long term' as being more than 5 years (green and blue sections in the graph below). Interestingly there was no significant difference in the pattern of responses between DC and DB schemes. Split of funds in defining long-term investment time horizons 7% 20% 46% Between 1 and 3 years Over 3 years Over 5 years Over 10 years 27% Aon Hewitt Limited 5

8 Monitoring investment activity and impact (Q26 & Q29) Only one of the respondents attempts to measure the impact of RI activities on their portfolio, although the majority do discuss RI with their fund managers in various forums, as indicated in the following pie chart: How is RI/ESG monitored with fund managers? 15% 15% Separate periodic meetings dedicated to RI/ESG Discussion on RI/ESG within the regular investment meetings 20% Ad-hoc phone calls to discuss ESG issues as they arise Regular reporting 35% Other method 10% 5% Not monitored Recommendation The majority of respondents in the 'other' category use an annual review of their managers to monitor RI and ESG issues and activity. The most common response was for those participants that discuss these issues with their fund managers at regular investment meetings, with over half of the responses covering these issues in these meetings or specially arranged meetings. We strongly believe that a key step in the evolution of RI from a 'feel good' investment idea to a core value-adding investment proposition will be the measurement of the impact of RI activities. We recognise that this is a difficult topic but we support those fund managers and providers (such as Hermes) that are attempting to develop metrics in this area. Aon Hewitt Limited 6

9 3. Policies, codes and standards Written Policies (Q9 & Q10) All but one of the respondents has a policy that makes specific reference to RI / ESG issues. However, despite not having a written policy, this respondent was active in RI, having collaborated with fund managers on a variety of issues, and having created their own voting and engagement guidelines, which were executed in-house. Of those respondents to our survey that had policies, there was a no particular pattern some of the smaller schemes had comprehensive policies covering multiple areas; whilst some of the larger schemes had policies focussing entirely on ownership activities (voting and engagement). Number of funds which are 14 Integrating ESG issues into asset allocation process Using managers that integrate ESG into their stock/security selection process Screening (positive or negative) Incorporating ESG issues into ownership activities (voting and engagement) Seeking appropriate disclosure on ESG issues from the entities in which you invest Initiating or supporting legal action Asset classes covered (Q11 & Q31) There was surprising divergence of the asset classes that were covered by the RI Policy. Three respondents have a broad policy covering all asset classes, whilst one focussed exclusively on developed market equities. Between these two extremes there was a considerable mix of different approaches however the broad trend is that the larger schemes have RI policies that cover a larger range of assets. 12 Number of different asset classes ,000 Value of assets ( bn, log scale) continued on next page Aon Hewitt Limited 7

10 Asset classes covered (continued) The 2011 UN PRI Survey also explored what proportion of asset owner respondents had policies covering other (non-equity) assets. The other asset classes most commonly 'covered' by active ownership policies were corporate bonds and directly held property, with almost a third of respondents having policies in these asset classes. Percentage of respondents with active ownership policies 35% 30% 25% 20% 15% 10% 5% 0% Sovereigns Credit Property Listed property Infrastructure In the majority of cases, these policies covered all three dimensions of environmental, social and governance issues although if one single area was favoured it tended to be that of governance. In addition, some respondents in our survey provided extra information about the kinds of approaches for different asset classes referred to in their policies. There was a wide variety of responses to this question: One respondent explained that their policy only relates to (developed and emerging market) equities and that the only tool they exploit is voting; Another respondent also focused solely on (developed) equities, but referred to multiple approaches including integrating ESG issues into fund management; monitoring fund managers; voting and engagement; A third respondent stated that their policy focuses only on voting, but across multiple asset classes; Other respondents had policies that exploit different approaches across different asset classes. Examples include the ESG screening of hedge funds and listed property funds; engagement with private equity funds; and an ESG thematic approach in infrastructure and direct property. These responses are summarised in the following table: Asset class Developed equities Emerging market Corporate equities bonds Government bonds Private equity Listed property Non listed property Hedge funds Infrastructure Commodities Investment beliefs & policy 73% 55% 45% 27% 36% 36% 36% 18% 18% 9% Selecting managers 36% 27% 36% 18% 27% 27% 18% 18% 18% 9% Reviewing managers 55% 36% 45% 27% 36% 27% 36% 9% 27% 9% ESG integration 55% 27% 36% 27% 36% 36% 45% 9% 36% 9% Screening (-ve/+ve) 45% 27% 36% 18% 9% 9% 0% 18% 0% 0% Thematic investments 9% 9% 0% 0% 18% 9% 18% 0% 9% 0% Carbon intensity measurement 27% 9% 0% 0% 0% 9% 9% 0% 0% 0% Voting 100% 73% 18% 9% 18% 27% 9% 9% 9% 9% Engagement 73% 45% 27% 0% 18% 18% 9% 0% 9% 0% Collaboration 45% 36% 9% 9% 18% 18% 9% 9% 9% 9% Legal remedies 18% 18% 18% 9% 9% 0% 9% 9% 9% 9% Aon Hewitt Limited 8

11 Asset classes covered (continued) Recommendation In the table above, and others of this format, grey squares indicate that no respondents gave this answer, orange is for between 0% and 30% of respondents, yellow for over 30% and up to 60%, and green for over 60% of respondents giving this answer. From this display we can easily see that the majority of respondents have policies that cover developed market equities, and a significant number include RI issues in their selection and review of managers across multiple asset classes. Developed equities, followed by emerging market equities and corporate bonds are the asset classes covered by the most policies; The tools most regularly used are investment beliefs and policy, reviewing managers and ESG integration. We recommend that NEST initially develops a policy to cover only developed market equities (and/or perhaps sovereign and corporate bonds) but as the assets within NEST grow in value then other asset classes could be added over time. Excluded classes (Q12) Only one respondent indicated any investments for which they feel that RI would never be appropriate. They stated that they feel certain absolute return funds (such as global tactical asset allocation funds) are not suited for responsible investment or ownership. We would agree that RI policies are applicable to the vast majority of asset classes although clearly different policies and approaches will be required for different assets. Indeed even within global tactical asset allocation funds, some investors might choose to 'black list' certain countries, or exclude speculative investment in, for example, agricultural commodities. Policy drivers (Q13) Six respondents indicated that external drivers such as macro risks, portfolio level risks and media coverage had helped to form their RI policy. What drove the themes within your policy? 19% 4% 18% Macro risks Portfolio level risks 19% 18% Bottom up/company specific risks Interests or specific circumstances of beneficiaries Best practice standards Media coverage 4% 18% Reputation continued on next page Aon Hewitt Limited 9

12 Policy drivers (Q13) Recommendation It is striking that all of the respondents to this question identified multiple reasons for developing an RI policy, and that non-investment issues (e.g. adverse media coverage or reputation risk) were cited almost as often as investment issues (macro, portfolio or stock-specific risks). We believe that there are sound investment reasons for schemes to develop an RI policy. However it is still the case that many schemes view RI, at least in part, as a public relations 'insurance policy'. Even so, we continue to encourage interested schemes to create and pursue an RI policy. Codes and Standards (Q14, Q15 & Q16) Only 20% of respondents require particular standards of their investee companies. However, the majority of respondents participate in global or industry initiatives such as the UN PRI (the most well supported) and the Carbon Disclosure Project. 14 In which of the following codes and initiatives do you participate? UN Principles of Responsible Investment (UN PRI) UN Global Compact Carbon Disclosure Project (CDP) Institutional Investors Group on Climate Change (or regional equivalent) Investor Network on Climate Risk (INCR) International Corporate Governance Network (ICGN) Extractive Industries Transparency Initiative (EITI) Global Institutional Governance Network (GIGN) Asian Corporate Governance Association (ACGA) Other international or regional initiatives (please specify) Recommendation One respondent commented that they participate in these initiatives via their specialist engagement overlay provider. In our opinion this would also be a cost effective way for NEST to participate in a wide range of global initiatives, particularly in the early years, before the asset value has built up sufficiently for NEST to have an internal team to manage these functions. As a key global initiative, NEST may also wish to directly participate in the UN PRI, independent of any third parties. Five respondents state that they seek to go beyond existing best practice, in the areas of corporate governance (working with the NAPF or the UNPRI Clearing House) and standards in the mining industry. We were encouraged to see evidence that investors with an interest in RI are prepared to take the initiative and, where they see areas of weakness, attempt to improve best practice. Aon Hewitt Limited 10

13 Scope of collaboration (Q17 & 18) The majority of respondents to this question (75%) collaborate with other investors. This is slightly lower than the proportion in the 2011 UN PRI survey where 90% of asset owner respondents said that they collaborate with other parties. The chart below shows which bodies the respondents in our survey collaborate with, and shows that the most popular partners are other pension funds. Have you collaborated on particular issues? 4% 21% Yes with other pension funds 46% Yes with asset managers Yes with specialist providers Yes - with national professional bodies 29% It is also notable that several respondents collaborated with more than one of the parties named in the survey. For example, the one respondent that collaborated with national professional bodies also collaborated with other pension funds, asset managers and specialist providers it could be said that they collaborated along four different 'dimensions'. The following chart illustrates this. Collaborate in four dimensions Collaborate in three dimensions Collaborate in two dimensions Only collaborate in one dimension continued on next page Aon Hewitt Limited 11

14 Scope of collaboration (continued) It is interesting to note that the respondent that collaborates along four dimensions is not the largest respondent in our survey, indeed it is one of the smaller schemes. As the chart below shows there is little correlation between the 'dimensions of collaboration' and asset size. This suggests that other factors contribute to the decision surrounding collaboration. 4 Dimensions of collaboration Value increasing (log scale) Finally, whilst several respondents collaborate only in their own region or country, almost as many respondents also collaborated internationally. Have you collaborated with others in the same region or internationally? 2 In the same country / region only 7 Both locally and internationally Do not collaborate 6 Aon Hewitt Limited 12

15 Focus of collaboration (Q19) Within our survey, the most common collaborative activity was engagement with individual companies, although higher level engagement was also significant. How do you exert influence? 29% 17% Meeting and engaging with regulators to influence codes and standards Engaging at the company level 54% Responding to local or international consultations Recommendation One respondent commented that they use their specialist engagement overlay provider to carry out these collaborative activities, which is a route we would recommend for NEST, particularly in the early stages. The UN PRI survey showed the perhaps surprising result that more than half of the asset owner respondents collaborated most frequently on issues other than Active Ownership (such as Integration and Promoting the Principles). This is shown in the chart below. Which of the UN PRI Princliples did you collaborate most on? Principle 1 - Integration Principle 2 - Active Ownership Principle 3 - Disclosure Principle 4 - Promote the PRI Principle 6 - Reporting Aon Hewitt Limited 13

16 4. Voting and engagement Voting guidelines (Q20) In our sample it is more common for investors to have created their own voting guidelines. Five respondents rely on the voting policies produced by their fund manager or their specialist provider (or both in the case of one survey participant). Note that none of the respondents have produced different voting guidelines for different markets (although some third party providers do adopt a differentiated approach between markets). Have you established your own guidelines for voting? 1 3 Yes, general guidelines that are applied in all markets No, use fund manager guidelines or best practice 2 9 No, use voting specialist provider s guidelines No, have policy not to vote The chart below looks at the pattern by value of assets. The clear indication from this sample is that larger schemes produce their own guidelines. Note that there is no clear relationship between size of internal team and voting policy. Own guidelines Specialist's guidelines Manager's guidelines Don't vote Value ( bn) continued on next page Aon Hewitt Limited 14

17 Voting guidelines (continued) Recommendation Despite the trend among those investors that take RI seriously to establish their own voting policies, we actually support the use of externally produced guidelines as we question whether even large well-resourced schemes have the skill, knowledge and experience required to establish their own effective voting guidelines. This is also our recommended approach for NEST. We do however encourage investors to inform their agents (fund managers or specialist providers) of any particular issues (such as labour rights or climate change) that they wish to pay particular attention to. In this way, NEST can ensure that their agents take particular care over voting opportunities in those key issues. Engagement guidelines (Q21) The pattern is different (more balanced) with respect to engagement, with a third of respondents using external guidelines and around half using their own guidelines. Have you established your own guidelines for engagement? 4 4 Yes, general guidelines that are applied in all markets Yes, different guidelines for different markets No, use fund manager guidelines or best practice 2 2 No, use engagement specialist provider s guidelines No Guidelines 3 The chart on the following page compares policies in the areas of voting (vertical axis) and engagement (horizontal axis). Note that the larger dots indicate multiple respondents at this point. The chart indicates that, generally, those respondents that put more effort into their voting will also have a more highly developed approach to engagement (that is, a trend from bottom left to top right) with one outlier / exception. continued on next page Aon Hewitt Limited 15

18 Engagement guidelines (continued) Own guidelines Specialist's guidelines Manager's guidelines Don't vote No engagement guidelines Use fund manager's guidelines Use specialist's guidelines Use own global guidelines Use own regional guidelines Recommendation Voting patterns (Q22) As with voting guidelines, our recommendation is that NEST utilises the engagement guidelines provided by the fund manager or the engagement overlay specialist, perhaps with some additional direction towards specific issues of concern. The question on voting patterns revealed some interesting information. The most important issues (number of respondents, weighted by ranking) were compensation, the remuneration report and director elections. Clearly these issues are of critical importance for the company, and it is appropriate that responsible investors should focus on these. However we also encourage investors to protect their shareholder rights which is why the approval of stock options grants and new shares are also important issues for shareholders to consider. Top 3 resolution types for abstaining or voting against management Electing Directors 8% 17% 6% Remuneration report Compensation related 12% Share placement 17% Share issuances 5% 4% Stock options Amend Article / Bylaw s / Charter 31% Other Recommendation If NEST follows our recommendation with respect to voting policies and relies on the policies of their managers or specialist provider, we would nevertheless encourage NEST to make it clear to their agents in this area (both initially, and in ongoing monitoring) that they consider issues around shareholder rights as important as corporate governance issues such as director remuneration. Aon Hewitt Limited 16

19 Voting execution (Q23) Six of the fifteen respondents execute their own voting policy in house; a majority (even those with their own policy) use their fund managers or overlay specialists to carry out the voting activity, with one using a combination of their own and third parties' and therefore appearing in two chart sections below. How do you execute your votes? 2 In-house 3 6 Own policy executed by specialist provider or fund manager Specialist provider's policy Fund manager's policy Fund manager s or specialist provider s policy, with over-ride option Own policy executed by specialist provider or fund manager, with over-ride option Recommendation We believe that it is more cost effective to use fund managers or overlay specialists to carry out voting activity, and we therefore recommend this approach to NEST. However only a minority of fund managers attempt to monitor or audit the 'voting chain'. We believe this is a critical part of voting activity since a low 'success rate' in actually delivering votes will seriously impede the effectiveness of any voting policy. Stock lending (Q24 & Q25) Stock lending can be an impediment to voting activity and yet the majority of respondents participate in stock lending to some extent. Do you participate in securities lending? We or our managers lend securities in all regions 5 We or our managers lend in certain regions 7 We or our managers lend out only a proportion of shares for each company 1 We do not participate in securities lending 2 continued on next page Aon Hewitt Limited 17

20 Stock lending (Q24 & Q25) Of the respondents who lend stock, all but two recall stock in order to vote to some extent, although only a minority of those that lend stock (2 from 10) recall stock for voting all of the time. These results are broadly similar to those recorded in the UN PRI 2011 Survey. Of the 122 asset owners that responded, less than half (49) said that they had no stock lending programme, although a further quarter did not give any details about their policy with respect to recalling stock. Of the 40 respondents that did provide details, 28% say they recall stock all the time; 28% say they never recall stock and the remainder say they recall stock some of the time. This means that the 'partial' approach of recalling stock occasionally is the most common approach both within our study and also within the group of PRI respondents. Monitoring voting and engagement activity (Q27 & Q28) Just over half of the schemes in our sample monitor whether votes have been delivered to investee companies as per instructions and within the correct timeframe, whilst the remainder did not. There was no pattern to the split of respondents on this question. We were pleased to note that none of our sample schemes monitor engagement activity by simply counting up the number of engagements and whilst around 40% attempt to evaluate the effectiveness of engagement, the remaining 60% do not attempt to track the progress of their managers' engagement activity in any way. This latter group includes several respondents that have an engagement policy (their own, or their fund managers'). How do you track the progress of the engagement activity? 16% Achievement of predefined objectives or milestones 59% 6% 19% Achievement of predefined objectives or milestones within specific time period Level of change implemented by company Completed number of engagements 0% Do not track progress of engagement activity It is also interesting to analyse this result in comparison with other features. Of the eight pure DC or mixed DB/DC schemes, only two attempt to monitor their managers' engagement activity. Coincidentally, one of these two is fully internally managed, whilst the other is fully externally managed. Three of the schemes that do not track the progress of their engagement activity employ internal ESG/RI specialists. This is perhaps a surprising result but probably reflects the different priorities for schemes in different geographies. continued on next page Aon Hewitt Limited 18

21 Monitoring voting and engagement activity (Q27 & Q28) (continued) Recommendation We accept that engagement activity is more onerous and less easy to monitor than voting activity. However, we believe that it is also a much more effective means of communicating with investee companies. Furthermore we note that engagement is applicable across other asset classes, not only equities (e.g. bonds, property, infrastructure and private equity). Our preferred approach to engagement is to establish clear objectives and milestones before embarking on engagement activity, and then monitoring progress against these milestones. We recommend that NEST appoints an engagement overlay specialist, or appoints fund managers with a genuine approach to engagement, and monitors the progress of their engagement activity using milestones. Engagement activities (Q32 & Q33) Some respondents were able to give additional information with respect to engagement activity. Firstly, how frequently engagement with investee companies occurs, split by asset class: Total = 13 Frequently (e.g. monthly) Occasionally (e.g. quarterly) Rarely (e.g. yearly) Planning to in 2010 Not at all Listed equity (developed markets) 31% 31% 15% 0% 23% Listed equity (emerging markets) 8% 15% 15% 0% 31% Fixed income (corporate issuers) 0% 15% 8% 0% 23% Fixed income (government and other non-corporate issuers) 0% 0% 0% 0% 0% Private equity 8% 23% 15% 0% 23% Listed real estate or property 8% 8% 8% 0% 31% Non-listed real estate or property 0% 8% 0% 0% 31% Hedge funds 0% 0% 0% 0% 0% Infrastructure 8% 0% 0% 0% 0% Secondly, how responsibility for certain engagement-related activities were split between different participants. Total = 13 Delegated to specialist provider Delegated to investment manager(s) In-house Did not employ Decisions on particular votes 69% 62% 15% 8% Execution of voting decisions 46% 54% 15% 8% Instigating or supporting specific shareholder resolutions 69% 38% 8% 31% Requesting an Extraordinary General Meeting 15% 8% 0% 54% Identifying issues for engagement 77% 38% 15% 8% Engagement activity 62% 31% 15% 23% Initiating class actions or other legal remedies 23% 8% 0% 46% Participating in class actions 23% 23% 0% 31% Monitoring fund managers or specialist 46% 8% 0% 23% Making public statements 23% 15% 0% 31% Recommendation These results show that even those investors that are thoroughly engaged with RI issues do not necessarily embrace the more 'activist' approaches such as requesting EGMs or participating in class actions. However, at the higher level, it can be seen that the majority of respondents do not delegate engagement activities to investment managers, with the majority of activities carried out by more participants in-house than delegated to a specialist. As mentioned earlier, we believe that the best action for NEST would be to delegate the majority of these activities to a specialist provider from the start, with a view to potentially taking a number of these functions in-house upon reaching greater size. Aon Hewitt Limited 19

22 5. Smaller schemes Respondents Five of our sample answered the additional questions relating specifically to smaller schemes. These questions were designed to explore what constraints, if any, were experienced by those smaller schemes or when the larger schemes were in the earlier stages. The respondents to these questions had assets between 150m and 22bn. Liaison with third-party managers Recommendation Four of the respondents said that they took RI policies and practices into account to some extent when selecting their external investment managers. These respondents also said that they sought to influence the RI policies of their third-party managers. Our broader experience supports this finding. Those schemes that have an interest in RI issues will typically seek to discuss those issues with their third-party managers both during the appointment process and also subsequently as part of the ongoing relationship. Indeed we see evidence that the number of schemes engaging in this way with their fund managers on RI issues is gradually increasing. We encourage NEST to adopt this approach to the appointment and ongoing relationship with its external managers. Evolution of RI practices Each of the five respondents had different views as to what RI practices they would introduce next as their assets grow. We have weighted their responses (to allow for ranking) which produced the following ranked order: Introduce RI activities to other asset classes More rigorous measurement activity Improve the degree of reporting Take on more activity 'in house' and delegate less Become more active in forums and initiatives We are not surprised that the five respondents to this part of the survey had differing views of their next priorities since these depend very much on the current situation and also on the views of the individuals involved. In our view, the most valuable developmental steps from the list would be to improve monitoring and measurement; and to extend RI activities to other asset classes. Constraints Recommendation None of the respondents pointed to a 'lack of support' as a constraining factor on their development of RI practices. However, all the other factors (lack of resource, small asset size, cost, passive / pooled vehicles) were identified as being barriers to progress. In our wider experience, we have found that passive / pooled vehicles present the biggest barrier to schemes that wish to develop their activity in the area of RI. However, even within a pooled or passive investment framework it is still possible to find ways to develop and implement certain RI policies and practices. We recommend that NEST take these issues into consideration in the design phase of their investment framework. Aon Hewitt Limited 20

23 Appendix A - Questionnaire Prepared for: Personal Accounts Delivery Authority Prepared by: Tim Currell Responsible Ownership Benchmarking Project Background 1. Is your scheme a pure DC scheme, a pure DB scheme, a DB scheme with a DC section, a reserve fund, or another (please specify )? 2. What is the approximate value of assets (please state currency)? 3. Please provide an approximate percentage breakdown of assets managed: internally % externally % 4. Please provide an approximate percentage breakdown of assets managed: actively % on a passive (index-tracking) basis % 5. Do you have an in-house team dedicated to RI / ESG? Yes No 6. If you have an in-house team dedicated to RI / ESG, please specify the number of "full-time equivalent" members of staff in that team Less than 2 Between 2 and 5 More than 5 Continued on next page Aon Hewitt Limited 21

24 Background (continued) 7. Approximately what proportion of the scheme's value do you spend on responsible investment activities each year? Less than 0.5bps of the scheme's value Between 0.5bps and 5bps of the scheme's value More than 5bps of the scheme's value 8. Would you be prepared to answer follow-up questions by phone? Yes. (If so, please provide contact name and phone number: ) No Policies 9. Do you have a policy that makes specific reference to RI/ESG issues? Yes No 10. If you have a policy that makes specific reference to RI/ESG issues, which of the following areas does it address? Integrating ESG issues into asset allocation process Integrating ESG issues into manager selection process ensuring that managers integrate ESG into its stock/security selection process Screening (positive or negative) Incorporating ESG issues into ownership activities (voting and engagement) Seeking appropriate disclosure on ESG issues from the entities in which you invest Legal action 11. Please indicate which of the following asset classes your RI/ESG policy refers to? Developed equities Emerging Market equities Corporate bonds Government bonds Private equity Listed property Non-listed property Hedge funds Infrastructure Commodities Other (please specify ) continued on next page Aon Hewitt Limited 22

25 Policies (continued) 12. Are you invested in certain asset classes or investment strategies for which you feel that responsible investment will never be appropriate? If so, please state which they are. 13. Which of the following helped you decide upon your ESG themes? Macro risks Portfolio level risks Bottom up/company specific risks Interests or specific circumstances of beneficiaries Best practice standards Media coverage Other please specify Codes and standards 14. Do you require any particular standards of your investee companies (e.g. participants in the UN Global Compact; reporting under the Global Reporting Initiative; etc.)? Yes No 15. In which of the following codes and initiatives do you participate and/or require your companies to meet? UN Principles of Responsible Investment (UN PRI) UN Global Compact Carbon Disclosure Project (CDP) Institutional Investors Group on Climate Change (or regional equivalent) Investor Network on Climate Risk (INCR) International Corporate Governance Network (ICGN) Extractive Industries Transparency Initiative (EITI) Global Institutional Governance Network (GIGN) Asian Corporate Governance Association (ACGA) Other international or regional initiatives please specify 16. Do you make any attempt to go beyond existing best practice - for example by applying your home-country standards across a global portfolio or working with other investors to raise standards in a particular region or market? Yes, please specify No continued on next page Aon Hewitt Limited 23

26 Codes and standards (continued) 17. Have you collaborated with pension funds, asset managers or other bodies on particular issues? Yes with other pension funds Yes with asset managers Yes with specialist providers Yes - with other bodies (please specify ) 18. Have you collaborated with others in the same region or internationally? In the same country / region Internationally 19. Which of the following have you participated in as a way of exerting influence? (If you have participated in more than one then please rank in descending order of importance 1, 2 and 3) Meeting and engaging with regulators to influence codes and standards Engaging at the company level Responding to local or international consultations Executing responsible ownership 20. Have you established your own guidelines for voting? Yes, general guidelines that are applied in all markets Yes, different guidelines for different markets No - use fund manager guidelines or best practice No, use voting specialist provider s guidelines No, have policy not to vote 21. Have you established your own guidelines for engagement? Yes, general guidelines that are applied in all markets Yes, different guidelines for different markets No - use fund manager guidelines or best practice No, use voting specialist provider s guidelines No, have policy not to vote continued on next page Aon Hewitt Limited 24

27 Executing responsible ownership (continued) 22. When voting against or abstaining can you please provide the top 3 types of resolutions (ranked in descending order of frequency 1, 2, 3) where such a decision was made? Electing Directors Approval of remuneration report Compensation related Approval of share placement Approval of share issuances without pre-emptive rights Approval of stock option grants Amend Article / Bylaws / Charter Approval of executive loans Other reasons, please specify 23. How do you execute your votes? In-house Own policy being executed via voting specialist provider or fund manager Using voting specialist provider's voting policy Using fund manager's voting policy Using fund manager s or specialist provider s voting policy, but with freedom to over-ride individual voting decisions Own policy being executed via specialist provider or fund manager, but with freedom to over-ride individual voting decisions 24. Do you participate in securities lending? We or our managers lend securities in all regions We or our managers lend in certain regions (please specify which ones ) We or our managers lend out only a proportion of shares for each company We only lend a percentage of assets under management. (Please specify what this is ) We do not participate in securities lending 25. Do you or your fund managers recall stock in order to vote? Yes, all of the time It depends on the shares that are out on loan No Aon Hewitt Limited 25

28 Evaluation 26. Do you attempt to measure the impact of responsible ownership activities on the performance of a portfolio, or the Scheme? Yes No 27. Do you monitor whether votes have been successfully delivered to companies as per your instructions within the correct time frames? Yes No 28. How do you track the progress of the engagement activity? Achievement of predefined objectives or milestones Achievement of predefined objectives or milestones within specific time period Level of change implemented by company Completed number of engagements Do not track progress of engagement activity 29. How is RI/ESG monitored with fund managers? Separate periodic meetings dedicated to RI/ESG Discussion on RI/ESG within the regular investment meetings Ad-hoc phone calls to discuss ESG issues as they arise Other method Time horizon 30. What do you consider as being a long-term investment time horizon? Between 1 and 3 years Over 3 years Over 5 years Over 10 years Aon Hewitt Limited 26

29 Supplementary questions The following questions may require a little additional time but we would be very grateful if you were able to answer them as fully as possible: 31. If you have a policy that makes specific reference to RI/ESG issues, which of the following asset classes does it cover using which tools? Please tick accordingly: Asset class Developed equities Emerging Mkt equities Corporate bonds Govt bonds Private equity Listed property Non listed property Hedge funds Infrastructure Commodities Tools Investment beliefs & policy Selecting managers Reviewing managers ESG integration Screening (-ve/+ve) Thematic investments Carbon intensity measurement Voting Engagement Collaboration Legal remedies continued on next page Aon Hewitt Limited 27

30 Supplementary questions (continued) 32. Please fill in the following table to indicate the extent to which you, your investment manager(s) or your external engagement specialist had dialogue with your investee companies during 2009: Listed equity (developed markets) Listed equity (emerging markets) Fixed income (corporate issuers) Frequently (e.g. monthly) Occasionally (e.g. quarterly) Rarely (e.g. yearly) Planning to in 2010 Not at all Fixed income (government and other non-corporate issuers) Private equity Listed real estate or property Non-listed real estate or property Hedge funds Infrastructure 33. Please fill in the following table to indicate which of the named parties carried out the named activities during 2009: Decisions on particular votes Execution of voting decisions Instigating or supporting specific shareholder resolutions Requesting an Extraordinary General Meeting Identifying issues for engagement Engagement activity Initiating class actions or other legal remedies Participating in class actions Monitoring fund managers or specialist Making public statements In-house Delegated to specialist provider Delegated to investment manager(s) Did not employ Aon Hewitt Limited 28

31 Additional questions for smaller schemes or those schemes that invest largely through pooled and/or passive funds 34. To what extent were responsible ownership policies and practices taken into consideration when selecting the external investment manager(s) in whose pooled and/or passive funds you invest? A great deal Somewhat Not at all 35. Have you attempted to influence in any way the existing responsible ownership policies of those investment manager(s) in whose pooled and/or passive funds you invest? Yes No 36. Please indicate the order in which you would expect the following elements of RI / ESG to be introduced (if at all) as the Scheme's assets grow: Introduce responsible ownership activities to other asset classes Take on more activity 'in house' and delegate less to other agents Become more active in national and international forums and initiatives Improve the degree of reporting to beneficiaries and/or the public More rigorous measurement activity 37. What is the biggest constraint to operating your R /ESG policy to its fullest potential? Passive vehicles Pooled vehicles Lack of resource Lack of support from sponsor Small asset size Cost 38. Are you able to vote according to your own policy within a pooled and passive environment? Yes in all markets In some markets (please specify which ones) We do not have our own voting policy 39. Please, if possible, provide a copy of your policies or alternatively provide a link to policies publicly available via website. Aon Hewitt Limited 29

32 Glossary Term Active management Asset allocation Asset class Corporate bonds Corporate engagement Corporate governance Defined benefit pension scheme (DB) Defined contribution pension scheme (DC) Environmental, Social and Governance (ESG) Fiduciary duty Governance Hedge fund Definition An investment approach that pursues a return on investment in excess of a specified benchmark. The proportions in which the holdings of a portfolio are spread across different types of financial instruments and asset classes. A specific category of assets or investments, such as equities, bonds, cash or property. Assets within the same class generally exhibit similar characteristics, such as similar risks and returns. Corporate bonds are debt issued by companies. They are usually tradable in a market. This is where the institutional shareholder builds a working relationship with the board of directors of the company in which it invests to ensure the directors discharge their duties for the benefit of shareholders. Involves the role institutional shareholders play in appointing the board of directors and auditors of the company in which it invests through exercising voting rights. This is to ensure appropriate governance, director remuneration, public reporting and accountability structures. Corporate governance can also include corporate engagement. For this type of scheme, the retirement benefit is specified in advance and the level of contributions (from the employer or the employee) is used as a steering levy to attain the targeted retirement benefit. For this type of scheme the level of contributions (from the employer or the worker) is specified in advance. The retirement benefit or retirement income depends on the investment return that contributions have produced. These types of schemes are classified as money purchase schemes, as the cash proceeds of the fund at retirement are usually used to purchase a lifetime annuity. A broad extra financial concept that that encompasses a wide range of environmental, social and governance issues such as climate change and human rights to executive compensation. Sets out the broad parameters within which trustees must exercise the discretionary powers conferred by the terms of the trust. The system and process concerned with ensuring the overall direction, effectiveness, supervision and accountability of an organisation. A fund which invests money in a wide range of different assets and is allowed to use investment strategies unavailable to standard funds. This can include short selling of securities and use of leveraged positions. Infrastructure funds Institutional investor Institutional shareholder Integration A fund that invests in one or more infrastructure projects, such as bridges, roads and ports. An intermediary that pools and manages funds collectively, usually on behalf of small investors. Institutional investors include pension funds, insurance companies, unit trusts and investment trusts. An institutional investor in its role as a holder of shares in listed companies. Institutional investor and institutional shareholder are often used interchangeably. A type of portfolio management that takes into account environmental, social and governance issues alongside financial issues to determine the potential performance of a prospective investment Aon Hewitt Limited 30