27 September 2013 Victoria Edwards Sent by email to: LGPSReform@communities.gsi.gov.uk Dear Ms Edwards, IMA Response to DCLG Call for Evidence on the Future Structure of the Local Government Pension Scheme The IMA welcomes the opportunity to respond to this call for evidence. 1 We represent firms which manage 4.5 trillion of assets from the UK on behalf of domestic and overseas investors. This includes over 95% of the assets managed on behalf of Local Government Pension Scheme (LGPS) clients. As such we acknowledge our industry s responsibility to assist its clients in meeting their objectives and to deliver the outcomes desired by society. We note the number of initiatives amongst LGPS and elected representatives to collaborate and seek efficiencies in the operation of their schemes. As agents, discretionary asset managers are bound to act in their clients best interests, and to work with the structures in place. We have, therefore, sought to share their experiences and opinions to assist clients and Government alike. It has not always been straightforward to fit some of these viewpoints into the question structure used by the Call for Evidence but we have sought to follow the flow of the Call without overly repeating points across questions. In preparing this response, we have sought the views of senior representatives of our members. While we hope that the response offers an accurate representation of industry views, member firms have not been asked for endorsement, and it is understood that some firms will have chosen to submit individual responses in addition to ours. Question 1: How can the LGPS best achieve a high level of accountability to local taxpayers and other interested parties including through the availability of transparent and comparable data on costs and income while adapting to become more efficient and to promote stronger investment performance? We believe that all of the abovementioned objectives accountability, efficiency and strong investment performance are best attained through an improved focus on good governance. 1 The IMA represents the asset management industry operating in the UK. Our members include independent fund managers, the investment arms of retail banks, life insurers and investment banks, and the in-house managers of occupational pension schemes. 65 Kings way London W C2B 6TD Tel:+44(0)20 7831 0898 Fax:+44(0) 7831 9975 w w w. i n v e s t m e n t u k. o r g Investment Management Association is a company limited by guarantee registered in England and Wales. Registered number 4343737. Registered office as above.
To that end, we welcome the consultation on new governance arrangements to LGPS undertaken by DCLG earlier this year and, as part of our current response, propose an increased focus on the following three areas (each of which is further developed under the headings below): 1. Appropriate targeting of cost-efficiency measures 2. Improved consistency of professional standards 3. Promotion of a long-term view Like others, we consider good governance represents an amalgamation of features across a number of areas, such as competence and efficiency of decision-making, appropriateness of investment options, member communication, and efficiency of cost and performance review. 2 While the efficiency of cost and performance review is an essential component of any appropriately governed scheme arrangement, it must be remembered that low cost is not synonymous with best value or most appropriate, and the striving for cost-efficiency should not take precedence over ensuring the best possible retirement income for scheme members, present and future. No discussion around charges should be undertaken without a simultaneous focus on the value added. Equally, we acknowledge that the argument cannot be used to suggest that cost is not a factor on which there should be client focus and strong competition. Value for money or, in other words, strong performance adjusted for cost should be the objective. Many existing local authority schemes represent good examples of well-functioning, appropriately governed pension arrangements. However, the diversity of skillsets that, by some accounts, characterises the LGPS landscape risks creating inconsistencies which may ultimately work against the attainment of the abovementioned features of good governance. An additional challenge is created by manager turnover and the impact of electoral cycles, which all too often undermine the consistency of decision-making and, ultimately, investment performance. As such, we believe that a greater focus on the targeting of cost-efficiency measures, together with increased consistency of competencies and a promotion of long-termism would serve to enhance the governance structure of LGPS, and better equip them to achieve the mentioned objectives. 1. Appropriate targeting of cost-efficiency measures We recognise the need to focus on cost-efficiency measures, especially in light of the large dispersion of deficits across LGPS funds at present. The benefit from economies of scale, which is evident across the asset management industry, should also be made available to LGPS funds, particularly in light of their sheer number and wide variation in size. 2 See, for example, tpr s Principles for investment governance of work-based DC pension schemes : http://www.thepensionsregulator.gov.uk/about-us/principles-igg-dc.aspx. - 2 -
However, we believe that the often discussed merger of funds to drive cost and fee reduction could fail to fulfil its desired purpose. This is for the following reasons: a) Ancillary costs. While consolidation at fund level can create downward pressure on fees, the ultimate cost savings are often limited in the shorter term due to the offsetting effect of associated ancillary costs. The transition management costs of the merger itself, including the possibility for trading issues to occur during the transition phase, risk creating considerable challenges. Moreover, the potential remains for increased running costs of the merged fund, e.g. in terms of staffing requirements. That said, much will depend upon the size of merger that is contemplated. The Australian experience provides, according to a KPMG-ACFS monograph, some support for mergers whilst evidencing an inability to reduce costs per member over a 5-year period. 3 b) Diseconomies of scale. The relationship between fund size and performance has attracted a lot of attention on both sides of the argument. On the one hand, scale undoubtedly bestows access to opportunities that can put larger pools of money at an advantage over smaller ones. On the other hand, a number of authors have found no conclusive evidence for the argument that there is a linear relationship between scale and a fund s strategic advantage. 4 Research by AXA Investment Managers (IM), moreover, finds that, beyond a certain point, larger funds may face diseconomies of scale in terms of meeting their target asset allocation, maintaining flexibility of execution or facing additional market risks. 5 Indeed, it has been suggested that certain asset classes would be excluded from large funds due to capacity constraints, with potentially adverse consequences on small and mid-cap investing. These capacity constraints are best assessed on a case-by-case basis and may be better addressed through the existing initiatives, than a one-size-fits-all approach from Central Government. Where we see the potential for improvements arising from consolidation, those would primarily be in terms of greater access to more sophisticated decision-making and governance tools, such as independent professional trustees and external advisers. Indeed, in the case of LGPS funds where consolidation has taken place, the reported benefit has been limited on the cost side, and overwhelmingly in terms of improved governance, which in turn created a positive knock-on effect on long-term investment performance. However, these types of scale economy can be achieved without the requirement of a fullscale merger. In agreement with the conclusions reached by the authors at AXA IM and other firms, we therefore argue in favour of an approach that would enable funds to take advantage of the scale economies accessible to larger funds without having to undergo a formal merger. Existing practice shows that this is possible through increased cooperation in terms of investment collectivisation, e.g. through the use of collectivised investment vehicles (CIVs), and shared resource in other areas. These include, but are not restricted to, the streamlining of back-office functions and greater cooperation through framework agreements with an aim to create efficiencies in the otherwise lengthy and complicated procurement process. 3 KPMG-ACFS (2011) Superannuation trends and implications, pp. 16-17. 4 See, for example, the WM Company s WM UK Local Authority Annual Review 2012/13 or the work of Chen et al (2004) Does Fund Size Erode Mutual Fund Performance? The Role of Liquidity and Organization, American Economic Review 94:5. 5 AXA Investment Managers (Sept 2013) Is Big Better? An investment perspective on the impact of pension fund consolidation, p. 4. - 3 -
2. Improved consistency of professional standards While the quality and timeliness of decision-making both with regard to the asset allocation and manager selection of a fund have always had a great impact on investment performance, since the onset of the crisis they have become virtually indispensable. Consequently, governance of LGPS would be significantly improved through greater consistency of professional standards among local authority schemes. These are, at present, reported to vary widely, which risks creating challenges particularly when a growing number of mandates are becoming either specialist 6 or increasingly more diversified into alternative asset classes. 7 While the trend towards diversification and increased incorporation of alternative asset classes has been shown to generate desirable outcomes in terms of return and volatility, it increases the need for highly specialised skills across the entire LGPS landscape. Unsurprisingly, our members experience and findings from the latest WM Company Local Authority report link internal management to stronger performance. 8 And while it must be acknowledged that internal management requires an initial cost of set-up and scale that not all LGPS may be able to meet, it emphasises the benefit of improved consistency of skillsets across the LGPS landscape. Indeed, among the larger corporate pension schemes, the appointment of an independent trustee and/or a Chief Investment Officer (CIO) is associated typically with increased levels of knowledge and engagement with the schemes investment process. We recognise the differences in the expected pay scales between CIOs and local authority employees, and as such understand that this option may not always be feasible. However, in light of the size of some of the deficits and the impact they have on scheme members and taxpayers alike, we do believe that a promotion of more consistent skillsets across the industry is vital. This also includes a greater focus on risk which, according to some firms, would bring benefit at a number of LGPS. In recognition of the above, asset management firms already provide training to a wide range of stakeholders, including those involved in the governance of local authority schemes, and there is an interest to become more proactively involved in this area going forward. Improved consistency of skillsets across the LGPS landscape is a prerequisite for any desired improvement in the level of client-manager interaction and, ultimately, a more joined-up decision-making process. We wish to be clear. We would be disappointed if our response was read as a criticism of individuals or of our members clients in general. A well-resourced, qualified and specialist cadre of investment-savvy individuals at LGPS will only engender an environment in which there is greater challenge of ideas and expectation of service. This is something which all asset managers welcome, for they will be more likely to be chosen and rewarded for the right reasons strong performance within a given risk and cost budget. To that end, we would welcome a discussion on the possibility of greater collaboration in the area of professional standards, including the potential for a cross-industry initiative to arise 6 IMA Asset Management in the UK 2012/13 (see Annex 1). 7 WM Company WM UK Local Authority Annual Review 2012/13, p. 3. 8 WM Company WM UK Local Authority Annual Review 2012/13, p. 7. - 4 -
in conjunction with consultants, advisors and the local authorities themselves. This could include an expansion of existing educational and training initiatives as well as the potential for an establishment of industry best practice, so as to enable comparison, self-evaluation and tools for improvement among the schemes themselves. 3. Promotion of a long-term view The LGPS landscape also continues to be impacted by electoral cycles and increased manager turnover. This serves to shift the focus away from long-term investment returns, and greatly increases the risk of adverse outcomes from continuous tinkering. While concerns over policy instability have been present across the entire UK pension landscape, the additional challenge for LGPS arises through political timetables and performance monitoring, which are not in line with a focus on long-term investment. Whilst hard to measure, given the asset classes to which LGPS are exposed, it appears axiomatic that short-term judgements will commonly be misaligned to the best interests of the LGPS. Just as the appointment of independent trustees and professional investment officers has been linked to improved governance and investment performance, it is widely agreed that this is undermined by high turnover and a lack of continuity in client-manager relations, thus ultimately resulting in sub-optimal outcomes for scheme members. We also believe the OECD s view as to why long-term investment is important provides indicators of what might need to be considered: 9 a) 'Patient' capital allows investors to access illiquidity premia, lowers turnover, encourages less pro-cyclical investment strategies and therefore higher net returns and greater financial stability; b) 'Engaged capital encourages active voting policies, leading to better corporate governance; c) 'Productive' capital provides support for infrastructure development, green growth initiatives, financing of small and medium-sized enterprises etc., thus leading to sustainable growth. In this way, further work could be designed to assist with encouraging the right environment in which it would be rational to re-direct capital into the abovementioned areas. Consistent with the findings of the Kay Review 10 and the WM Company, 11 we therefore call for a greater promotion of the principles of long-termism in the governance and investment strategy of LGPS, and would welcome the opportunity for greater engagement with councillors on this subject. 9 OECD (2011) Promoting Longer-Term Investment by Institutional Investors: Selected Issues and Policies. 10 See The Kay Review of UK Equity Markets and Long-term Decision Making. 11 WM Company WM UK Local Authority Annual Review 2012/13, p. 7. - 5 -
Question 2: Are the high level objectives listed above those we should be focusing on and why? If not, what objectives should be the focus of reform and why? How should success against these objectives be measured? While welcoming the Department s high-level focus on dealing with deficits and improving investment returns, one must acknowledge that both are consequences of an underlying issue, and as such cannot be addressed without a prior focus on improvements in governance standards. As we elaborate in our response to Question 1, we propose a primary, high-level focus on good governance, and therein specific areas which we consider particularly relevant across the LGPS landscape, consistent with our belief that they are a prerequisite for the creation of a more robust LGPS governance framework. Without first tackling the issue of cost-efficiency targeting, consistency of skillsets, and a greater promotion of a long-term focus on governance and investment, the existing differences in local authority governance are unlikely to diminish, with a potentially detrimental impact on scheme members and taxpayers alike. In this regard, and as further elaborated on in our answer to Question 4, alteration of the investment regulations is long overdue. Question 3: What options for reform would best meet the high level objectives and why? The wide range of local authority schemes, and the diverse forms of collaboration already underway across a number of LGPS, argues against the introduction of a single reform option. A one-size-fits-all approach is rarely appropriate beyond the formulation of highlevel principles, and across the diverse LGPS landscape risks introducing overly simplistic or excessively restrictive reforms. We believe, therefore, that the most optimal form of cooperation is best decided at a regional level, where local authorities in conjunction with industry professionals are allowed sufficient space and flexibility to properly address any local specificity in need of consideration. Believing that the most properly targeted measures are those driven by client demand, we strongly advocate the consideration of the solutions already underway. The collaboration between Cambridgeshire and Northamptonshire, the work underway among the Triboroughs, and initiatives under consideration in Wales and London all demonstrate the willingness to explore approaches from the bottom up. These cover a wide range of areas: a) Collectivised investment. Pooled investment and the use of CIVs are already in existence among a number of multi-employer schemes. The IMA invested significant resource in supporting HM Treasury s development of tax-transparent authorised contractual schemes. We consider their robust design and ease of formation could be easily adapted to the needs of LGPS. The IMA will publish in the next two months a model form of constitution to reduce the cost of development further. These types of organisational arrangement can achieve many of the benefits accessible through increased scale without necessitating a merger or cross-subsidisation between the individual component employers liability profiles. As mentioned in our response to Question 1, we consider this type of collaboration preferable to the option of a merger, - 6 -
both from a cost and a scale economy perspective, whilst delivering any benefit a merger may provide in the investment process. b) Streamlining of other business functions. Cost-efficiency initiatives currently underway reportedly focus mostly on back-office functions such as administration or legal services. c) Framework agreements. We see an increasing use of framework agreements across some parts of the procurement process. Whilst these have been used for administration and other support functions, they are generally not used for investment management services, and we believe that any further review should carefully consider in whose interest these agreements primarily operate, and how much added value they would provide. We are not in a position to comment on whether framework agreements better serve clients or merely avoid the strictures of the procedures required under the Official Journal process. That said, the official tender procurement process itself is not without challenges. The requirement to set the fees out at the beginning of the tender process, together with the fact that these cannot be varied by negotiation, does introduce a different dynamic. In the corporate area, even once an investment manager has been shortlisted, there can be fee negotiations and, by their nature, these will be securing fee reductions for the client as opposed to fee increases for the manager. The experience of managers being negotiated down in the LGPS space varies; further work could be done to assess how to improve the negotiating position of LGPS. Nevertheless the nature of the tender process demands keen pricing from the outset, which often includes significant discounts and/or the incorporation of performance fees, which are seen by a number of schemes as an effective means of aligning reward to return more directly. As such, the tender process itself provides a wide trawl of possible providers, thus giving some opportunity to new entrants. Question 4: To what extent would the options you have proposed under question 3 meet any or all of the secondary objectives? Are there any other secondary objectives that should be included and why? An improved focus on good governance together with the exploration of existing bottom-up approaches to achieving cost-efficiency on a regional and/or local level significantly contributes to the fulfilment of the secondary objectives set out in the document. 1. Reduction of overall cost As mentioned earlier in our response, we believe that a collectivisation of investment without the simultaneous requirement of a full merger would help minimise the ancillary costs of transition management whilst enabling access to the scale economies typically taken advantage of by larger funds. Liability profiles of different LGPS should remain distinct, but their assets could be collectively invested. Bearing in mind the findings by AXA IM and others, however, the relationship between fund size and economies of scale is not necessarily linear and one must therefore take account of the diseconomies of scale thus potentially arising. - 7 -
Clients may of course explore different avenues to drive cost reduction, including a greater use of pooled vehicles, however, the appropriateness of any such decision still needs to be determined by the client. Moreover, in light of the diverse nature of the services offered as well as the increasingly specialist and diversified nature of LGPS mandates, particularly into alternative asset classes, the potential for fee reductions would, by the very nature of those mandates, be limited, especially on capacity limited products. 2. Improvement in the flexibility of investment strategies For some years now we have asked that the investment regulations be updated to allow for investment decisions to be made on a prudent person basis rather than being directed by a schedule of allowable weightings between different asset classes. There is also some variation between schemes as to what they consider to be legally allowable. This remains a wholly unnecessary drag on performance which all parts of the LGPS landscape, including the LGPS themselves, agree needs to be changed but which is still being debated by Government. 3. Greater investment in infrastructure While an interest in infrastructure investment in general has been evident both from the IMA Asset Management Survey 12 and the Smith Institute s report on local authority investment, 13 there are a number of issues that require careful consideration before a wholesale promotion of LGPS investment into these types of asset class. These revolve largely around: Conflicts of interest Rates of return Investment horizons Liquidity issues These reservations cannot be lightly dismissed, and as explained in our response to the Kay Review and other contributions to debates about fiduciary duties, discretionary managers cannot be asked to make decisions otherwise than on their merits as investments. To this end, the use of triennial valuation and accounting methodologies may need to be reviewed as they reportedly skew investment approaches towards the shorter-term, as well as having technical but important impacts such as inadvertently causing listed infrastructure funds to be treated less favourably than direct investment. 12 IMA Asset Management in the UK 2012/13, pp. 28-30. 13 The Smith Institute (2012) Local Authority Pension Funds: Investing for Growth. - 8 -
Question 5: What data is required in order to better assess the current position of the Local Government Pension Scheme, the individual Scheme fund authorities and the options proposed under this call for evidence? How could such data be best produced, collated and analysed? We recognise the need to ensure that data is available and comparable so that clients are able to determine the likely true cost of any service. We would be willing to consider where information can be harmonised across the industry to improve comparisons. However, while we recognise the emphasis on data collection for the purposes of fee and cost comparison, it must be acknowledged that any attempt to show the potential improvement in investment fees would have to take account of the diverse nature of scheme strategies which would, in turn, create inevitable challenges in terms of comparability. This is further complicated through the existence of different calculation methodologies, which often make published statistics of questionable value. For example, accurate performance measures corrected for costs are more informative than just looking at cost per member or headline performance. A LGPS that appears amongst the most expensive on a cost per member basis may be amongst the better performers after costs. We therefore believe that any successful attempt at the harmonisation of data across the highly diverse LGPS landscape must be preceded by a discussion over a consistent use of methodology, including its associated definitions and calculations. That will enable greater consistency and better comparability across LGPS data, which have so far been severely challenged. We would be happy to meet with you to discuss any of the above in further detail, or facilitate a meeting with the asset management side of the industry to assist your further considerations about this important and challenging subject. Yours faithfully Guy Sears Director - Risk, Compliance & Legal - 9 -
Annex 1: Results from the IMA Survey Data from the most recent IMA Asset Management Survey, recognised as the authoritative source of information on the UK asset management industry, sampled firms managing almost 150bn of local authority assets. As part of the annual survey process, the IMA has received data from 38 firms who, as at the end of December 2012, managed a total of 145bn in local government pension scheme (LGPS) mandates. This represents the vast majority of the 150bn estimated to be under management in total. 14 As at the end of 2012, the top 10 respondents by the size of their LGPS mandates managed around 82% of the total mandate size. The total is split relatively evenly between segregated and pooled mandates, with 42% accounted for by pooled mandates and the remaining 58% represented by segregated mandates (see Chart 1). The majority of mandates (64%) are reportedly managed on an active basis (see Chart 2). Chart 1: Breakdown of mandates into segregated and pooled Pooled 42% Segregated 58% Chart 2: Breakdown of mandates into active and passive Passive 36% Active 64% As shown in Chart 3 overleaf, the predominant LGPS mandate type is single-asset or specialist, representing 73% of the total. The rest is accounted for by LDI (18%) and multiasset mandates (9%), although controlling for difference in reporting methodology would likely increase the share of multi-asset mandates to over 10%. 14 NAPF (May 2013) Local Government Pension Scheme 2013: Investing in a changing world, p. 4. - 10 -
Taking a closer look at specialist mandates, one-half (49%) is composed of equity mandates, with the remainder including bond (15%), property (4%), cash (1%) and other asset class mandates (4%). Chart 3: Breakdown by mandate type LDI 18% Multi-asset 9% Single-asset 73% Equities 49% Bonds 15% Cash 1% Property 4% Other 4% Among specialist equity mandates (see Chart 4), the largest proportion has a global focus (35%). This is followed by UK equity mandates with 29% and North American and European equity mandates (13% and 11%, respectively). The remaining 12% is represented by mandates focused on Japan, the emerging markets, Asia-Pacific and other regions. Chart 4: Breakdown of specialist equity mandates by geographic focus Other 1% Global Fixed income mandates: 35% UK 29% Emerging Market 4% Japan 3% Asia-Pacific 4% North America 13% Europe 11% The largest proportion within specialist fixed income mandates is represented by Sterling corporate and index-linked gilt mandates with 26% and 23%, respectively. These are followed by global fixed income mandates (19%). Gilt and Corporate and gilt mandates account for 10% of total specialist fixed income mandates each. - 11 -
Chart 5: Breakdown of specialist fixed income mandates by geography and type Global 19% Other 12% Indexlinked gilt 23% Sterling Corporate 26% Sterling Corporate and gilt 10% Gilt 10% - 12 -