: Facilitating Pooled Asset Management for Ontario’s Public-Sector Institutions

Download PDF

Report from the Pension Investment Advisor to the Deputy Premier and Minister of Finance, October 2012


Table of Contents

Message from the Advisor
1. Introduction
1.1. The Idea in Brief
1.2. Mandate
1.3. Process
2. Context
2.1. The Ontario Landscape
2.2. Potential Sources of Cost Savings
2.3. Empirical Evidence of Cost Savings
2.4. Potential Cost Savings for Ontario Public-Sector Entities
2.5. Trends
2.6. The Opportunity
3. Design Considerations
3.1. Establishing a New Entity versus Using Existing Large Pension Plans
3.2. Mandatory versus Voluntary Participation
3.3. Structure and Governance
4. Implementation Considerations
4.1. Implementation Strategy, Timelines and Costs
4.2. Operational Considerations
4.3. Legislative and other Considerations
5. Conclusion

Message from the Advisor

On May 30, 2012, the Deputy Premier of Ontario and Minister of Finance, Dwight Duncan, announced my appointment as Pension Investment Advisor. Minister Duncan asked me to help the government determine the advantages of pooled asset management for Ontario’s public-sector pension funds, and, if appropriate, to recommend a path to implementation.

I am pleased to report there is a significant opportunity for pension funds to realize benefits from the economies of scale and other advantages that a pooling framework would generate. More specifically, implementation of such a framework would reduce duplication and costs, broaden access to additional asset classes and enhance risk management practices. To the extent that these advantages support more diversified portfolios among participating institutions, pooled asset management may also help realize improved investment returns over the long term.

I estimate that, with political will and effort, a pooling framework would achieve potential savings of between $75 million and $100 million annually, once fully implemented. These savings would enhance the sustainability of participating pension and investment funds, to the benefit of members and taxpayers. It’s important to consider the perspective of current and future pensioners as you read this report. Achieving cost savings and improving fund returns would enhance the security of pension benefits, especially as pension plan members hear about the challenges their plans face due to increased longevity and persistently low interest rates. Any savings or improved returns may also reduce the need for increased contribution rates on the part of employers or employees.

Much of what I discovered in my role as Pension Investment Advisor was not unexpected. Ontario’s public sector features a large number of pension funds with diverse approaches to governance, investment and risk management. Most of the organizations I met with are fully engaged in the management of their pension assets and the results they achieve are within the expected norms for their respective sizes and asset allocations.

However, there were some surprising findings. The sheer number of pension plans is greater than expected, with the degree of fragmentation suggesting obvious cost-saving opportunities and, in some cases, a need for greater day-to-day oversight. I heard from leaders who are frustrated by the responsibility of managing pension assets, given the value of these assets and their own lack of investment management expertise. While risk management is an important consideration at Ontario’s public-sector institutions, some organizations have neither the time nor the understanding of how to embark on or sustain appropriate risk management efforts in respect of pension assets.

My mandate to engage the public-sector constituents was rewarded with an interested audience. I met with or spoke to well over 100 individuals and groups, representing labour and management at Ontario’s public-sector institutions; managers as well as current and former leaders of large Canadian pension funds; investment management professionals; interested industry associations; and retirees. There was consensus that an opportunity exists to improve the management of public-sector pension assets, and the overwhelming majority of the participants in my meetings acknowledged the potential advantages of a new pooled asset manager featuring sufficient scale, excellent governance, independence from government and professional management.

In the body of this report, I set out an agenda for the development of a new pooled asset manager that would oversee investments on behalf of Ontario’s public sector pension and other investment funds. I identify the critical success factors for this new institution, including appropriate scale, approach to governance and management, and the need for individual institutions to maintain control over asset allocation decisions. Adherence to these recommendations would better enable participating institutions to embrace the new pooled asset manager.

Based on responses from potential participants, there is sufficient support among Ontario public-sector institutions to envisage a new pooled asset manager overseeing well over $50 billion in assets and thus achieving the benefits of scale outlined above. Successful implementation of this new fund must address a number of challenges that the new institution would face.

As reported in The Economist1, Canada has an enviable record for establishing and managing large, successful pension funds. As the quest for superior returns requires greater capital and worldwide breadth, establishing a new pooled asset manager to oversee investments on behalf of participating public-sector institutions would build on this Canadian success story, and enable us to compete for best-in-class returns with other global players on an equal footing.


William Morneau
Pension Investment Advisor

1. Introduction

Pension plans are facing significant funding challenges, precipitated by persistently low long-term interest rates, financial market volatility and increasing longevity of plan members. Concurrently, public-sector institutions are tasked with addressing operating pressures at a time when the government is making a concerted effort to manage expense growth and eliminate the deficit.

The 2012 Ontario Budget announced the government’s intention to develop a legislative framework facilitating pooled asset management for smaller public-sector pension funds. Public-sector pension funds could, while maintaining ownership of and fiduciary responsibility for their assets, realize benefits from the economies of scale that a pooling framework would generate.

Consistent with government direction, I have consulted with affected stakeholders and developed recommendations for consideration. This report summarizes the potential merits of pooled asset management for public-sector pension funds, and provides a strategy to mitigate or avoid some of the intrinsic challenges related to the implementation of any pooling framework.

Chapter 1 briefly outlines my mandate and the consultation process, and identifies the principles that ultimately informed the recommendations.

Chapter 2 reviews the current pension landscape in Ontario’s public-sector, in an effort to define the appropriate scale and potential scope of any pooling framework. It also describes the potential benefits of a pooling framework, provides comparisons between the investment management costs faced by larger and smaller pension plans and outlines some recent developments in other jurisdictions.

Chapter 3 identifies the critical design questions related to the development of a pooling framework. It outlines policy options including recommendations on the desirability of establishing a new or leveraging an existing investment management entity; to what extent participation should be mandatory or voluntary; and a potential governance structure.

Finally, acknowledging the complexity of this undertaking, Chapter 4 describes a potential implementation and transition strategy. This strategy includes a description of how savings from a pooling framework could be realized over time, and a mechanism by which the responsibility for asset management could be assumed by a new institution while managing investment risks.

1.1. The Idea in Brief

There are currently over 100 public-sector pension funds in Ontario, each of which is responsible for its own investment management and administrative functions, with a wide range of investment results. In many cases, these funds are unable to realize the lowest possible investment management costs or access the range of asset classes desired. Many of these funds are also concerned that they cannot ensure the best risk management in their portfolios.

Implementing a pooling framework would reduce duplication and costs, broaden access to additional asset classes, and enhance risk management practices. To the extent that these advantages support more diversified portfolios among participating institutions, pooled asset management may also help realize improved investment returns over the long term.

This report recommends the creation of a new pooled asset manager to oversee investments on behalf of Ontario’s public-sector pension funds, as well as several non-pension investment funds such as the Ontario Nuclear Funds, which would also benefit from the advantages described above.
Participating institutions would retain ownership of their assets, but invest them with the new pooled asset manager through a family of unitized pooled funds, similar to mutual funds:

  • employees and retirees would remain members of their existing plans;
  • the relationship between pension plans and their members would not change; and
  • institutions would maintain their current plan designs including benefit levels and contribution rates, as well as funding policies and approaches to administration.

There is considerable support for the concept of pooled asset management as an opportunity to achieve economies of scale.  All else being equal, lowering costs is the only certain way to improve fund returns. Lower costs support pension plan sustainability and ultimately benefit taxpayers by reducing pension funding demands. I acknowledge that additional actions may also be required to ensure the sustainability of Ontario’s public-sector pension plans, but they are beyond the scope of this report. Implementing a pooling framework as outlined above would not conflict with any potential pension plan design changes.

My recommendations include some important caveats. Establishing a new pooled asset manager to oversee investments on behalf of participating public-sector institutions would only be effective if key implementation and operating principles are respected. Any pooled asset manager must:

  • permit participating institutions to retain fiduciary responsibility and control over asset allocation decisions, given variations in the liability profiles across pension plans and investment funds. This would require a family of unitized pooled funds, similar to mutual funds, so that participating pension plans or investment funds could tailor their investment portfolios to meet their particular needs;
  • operate at arm’s-length from government and be responsible to an independent and representative board of directors, to ensure that investment decisions are made solely on the basis of seeking the best returns without political influence;
  • feature world-class governance, professional investment and risk management, competitive compensation and effective oversight, to ensure the confidence of its constituents, including the management teams of participating institutions and pension plan members; and
  • have sufficient assets under management – at least $50 billion – to support investments in a broad range of asset classes at the most competitive costs. All else being equal, no participating institution should face higher investment costs than it does at present.

The creation of a new pooled asset manager would be challenging, especially given the large potential number of participating institutions. To support the transition to the new fund, the government should provide funding, which would be repaid through cost savings realized over time.

1.2. Mandate

The 2012 Ontario Budget announced the government’s intention to “introduce framework legislation in the fall of 2012 that would pool investment management functions of smaller public-sector pension plans in Ontario. Under this framework, management of assets could be transferred to a new entity or to an existing large public-sector fund.”

The pooling framework was just one of a number of measures announced in the 2012 Ontario Budget to help improve the sustainability, affordability and efficiency of Ontario’s public-sector pension plans.  Separate consultations on measures to address the liabilities and funding of jointly-sponsored and single-employer pension plans are ongoing and are beyond of the scope of this report. Further, it was not within the scope of my mandate to recommend:

  • changes to the administration or governance of any pension plan, including mergers;
  • changes to the contribution or benefit provisions of any pension plan; or
  • direct or indirect overrides of plan sponsors’ asset allocation decisions.

Since May 30, 2012, I have undertaken extensive consultations with pension plans, stakeholders and other interested parties. Input from these consultations, a review of applicable research and a scan of related developments in other jurisdictions have all informed various considerations in this report, including:

  • evidence (e.g., potential benefits);
  • scale (e.g., its significance, sufficient size to realize potential benefits);
  • scope (e.g., potential participants, mandatory versus voluntary participation);
  • mechanism (e.g., new entity, leverage one or more existing large pension plans);
  • governance (e.g., corporate structure, board composition);
  • implementation (e.g., corporate functions, appropriate timing, estimated costs); and
  • legislation.

The focus of the 2012 Budget announcement was on smaller public-sector pension funds (i.e., those with assets of less than $1 billion). The total value of these defined benefit and hybrid pension plans is estimated to about $10 billion.

However, my findings suggest that $10 billion is not sufficiently large to fully realize economies of scale. Furthermore, my research suggests the benefits of pooled asset management could extend to some larger pension funds. Several pension funds with significantly more than $1 billion in assets have determined that investing through a new pooled asset manager would be to their benefit, either for all or some asset classes. I also became aware of other non-pension investment funds such as the Workplace Safety & Insurance Board funds and Ontario Nuclear Funds, as well as certain practices such as co-management of pension and endowment fund assets, that necessitated a broadening of my mandate to consider the potential implications of pooled asset management for these other types of funds.

The recommendations that flow from my mandate are intended to maximize the potential benefits of pooled asset management while minimizing potential implementation risks. It is my intention that no participating institution should face higher investment management costs as a result of the design or implementation strategy proposed in this report.

1.3. Process

A discussion document was posted on the Ministry of Finance website, posing eight questions to stakeholders and inviting interested parties to make submissions for consideration. After conducting more than 40 formal consultations, speaking with many other constituents and reviewing numerous written submissions, I have collected and considered input from four groups:

  • representatives of public-sector pension and investment administrators and sponsors;
  • representatives of public-sector labour groups and retirees;
  • current and former leaders of large pension and other investment funds; and
  • representatives of Ontario’s investment management community.

While positions varied both within and across these groups, there was consensus on a number of critical success factors for a new pooled asset manager. These factors, listed below, informed my recommendations:

  • flexibility to accommodate the asset allocation decisions of participating institutions;
  • arm’s-length relationship from government to allow for effective investment and risk management discipline;
  • strong professional governance and management, with effective investor/client relationship management; and
  • sufficient scale to achieve the lowest possible costs as well as retain superior leadership and investment management talent.

It is my view that the absence of any of the above-mentioned factors would seriously undermine the potential success of any pooling framework for Ontario’s public sector.

2. Context

As noted by The Commission on the Reform of Ontario’s Public Services and reiterated in the 2012 Ontario Budget, the pension landscape of the Province’s public sector features a high degree of fragmentation: there are over 100 pension plans, each of which is responsible for its own investment management and administrative functions.

Many of the Province’s school boards, municipalities, hospitals and colleges already participate in relatively large, well-governed, sector-based pension plans, to the benefit of plan members and taxpayers. Empirical evidence suggests that larger plans generally benefit from higher returns at lower costs than their smaller counterparts. Many jurisdictions, both within Canada and internationally, are pursuing pooled asset management to reduce costs and broaden investment opportunities, particularly for smaller funds.

The 2008 Report of the Ontario Expert Commission on Pensions highlighted the importance of fund size in pension policy innovation. The report noted that smaller and medium-sized plans were unable to operate across the full investment spectrum, generally lacking the necessary infrastructure, resources, expertise and/or inclination to undertake maximum portfolio diversification. Furthermore, Commissioner Harry Arthurs emphasized the significantly higher fees of financial intermediaries that smaller and medium-sized plans pay compared to larger plans as a considerable disadvantage. The report recommends that “[p]ension policy and legislation ought to facilitate the growth and operation of large-scale pension plans or to enable and encourage cooperation among small and medium-sized plans.”

2.1. The Ontario Landscape

Ontario’s public-sector institutions administrate more than 100 defined-benefit, defined-contribution, and hybrid pension plans as well as a number of other investment funds, with a majority of these pension plans featuring a defined-benefit or hybrid structure. Aside from some co-management of pension and endowment funds, each of these plans makes investment decisions independently, each with its own infrastructure for doing so.

Excluding the Ontario Teachers’ Pension Plan (Teachers’), the Ontario Municipal Employees Retirement System (OMERS) Pension Plan and the Healthcare of Ontario Pension Plan (HOOPP), Ontario’s public-sector pension plans and investment funds manage more than $100 billion in assets on behalf of their members. Over 75 per cent of public-sector defined-benefit or hybrid pension plans manage assets of less than $1 billion each. 

Like other jurisdictions, the fragmented pension landscape is difficult to address without government intervention. Management at many public-sector pension plans recognize the existing inefficiencies, but has no simple means to improve the situation. Throughout my consultations, concerns over costs, access to asset classes and risk management were frequently cited, lending support to government concerns that the current framework leaves pension plans and their members in a sub-optimal position. The small size of public-sector pension funds presents challenges as the funds attempt to broaden access to asset classes. Increasingly, the competition for assets includes global organizations and sovereign wealth funds with capital, resources and reach significantly greater than those of Ontario’s smaller pension funds.

Some smaller plans operate and perform well, though some of the smallest pension funds are managed on a part-time or voluntary basis, which may lead to a host of other risks related to governance, investment expertise and succession planning. Furthermore, risk management systems are non-existent at some smaller plans and inadequate in many other cases. While risk management is cited as important to smaller plans, the added cost and complexity of such systems can be difficult to justify without sufficient scale.

The review of the landscape supports the conclusion that there exists significant opportunity for improvement through pooled asset management.

2.2. Potential Sources of Cost Savings

Investment management fees are typically charged based on the value and type of assets under management. Generally, the costs charged by external investment managers to smaller plans will be higher than to larger plans, the latter tending to have greater negotiating power by virtue of their size. Not only are larger plans positioned to pay less for external management, a large enough asset base could also justify bringing expertise in-house. While a costly and complex undertaking, internal investment and risk management is widely seen as key to lowering costs and improving control and performance, provided it is appropriate based on investment objectives, has the ability to recruit and retain expertise at competitive rates, and has sufficient assets under management. Much of the recognized success of the large Canadian public-sector pension funds has come through the implementation of internal management.

Investment management costs for alternative asset classes such as real estate, infrastructure and private equity are generally much higher than for traditional assets such as stocks, bonds and cash. While investors appreciate the advantages of higher potential returns and the opportunity to diversify away from traditional asset classes, pension fund investors may also benefit from stable cash flows and investment horizons that uniquely suit their long-term liabilities.

Direct investments in alternative asset classes require specialized expertise and large amounts of capital, both initially – to find and structure the investments – and on an ongoing basis, often over a long timeframe. Accordingly, such direct investments can only be achieved efficiently by the largest funds with dedicated in-house expertise and enough capital to make sufficient allocations to alternative asset classes given their preferred asset mix. As with more traditional asset classes, the largest of funds have reduced costs by internally managing alternative asset classes. The cost-effectiveness of this approach enables a greater allocation to these classes, resulting in a more diversified investment portfolio. Many smaller funds currently pay the comparatively high costs to external managers for small allocations in alternative asset classes.

A pooling framework would also facilitate a more rigorous, dedicated and expert system of investment and risk management.

2.3. Empirical Evidence of Cost Savings

There is strong evidence to suggest that large pension funds outperform smaller and medium-sized funds, with lower investment costs, better overall returns and improved diversification across asset classes. While there is no single threshold defining a large plan, there is sufficient empirical evidence to support the thesis that smaller funds cannot realize the same level of advantages that are typically enjoyed by much larger funds.

My research considered both Canadian and international studies with samples dating back to about 1990. However, many of these studies were published in the last five years as volatile financial markets and record low interest rates adversely affected plan funding. Some of these studies considered the link between scale and increasing investment in alternative asset investments, due in part to the emergence of larger funds (e.g., sovereign wealth funds and large pension funds) with liabilities that align with the investment attributes of these asset classes.

My research suggests the threshold to achieve economies of scale ranges from about $5 billion to $90 billion in assets under management, depending on the sample. The positive economies of scale associated with these larger plans produce the following advantages:

  • lower fees paid to external investment managers and other service providers;
  • increased and more cost-effective access to alternative investment classes;
  • higher gross and net returns; and
  • improved risk management and investment monitoring.

The cost efficiencies noted above are significant, as demonstrated by the following estimates: 

  • One-quarter of the gains associated with positive economies are a result of reduced costs from favourable negotiations with external managers.2
  • Gains from internal management added 3.6 to 4.1 basis points in net value for each 10 per cent increase in the proportion of assets managed internally.3
  • About one-third to one-half of unit cost gains of larger funds can be attributed to reliance on internal management.4
  • The savings associated with internal management results in better net returns for some asset classes such as foreign equity that would otherwise face higher management costs.5

External investment management costs for alternative assets vary based on investment size, rate structure and performance. The most common rate structure includes a 2 per cent management fee on committed funds and a 20 per cent performance fee above a set return. The resulting investment management costs for these asset classes ranged from 100 to over 500 basis points, with performance of the investments being the single most significant determinant of actual costs. The 100 basis points is at the very low end of the range, and in many cases the fees are not shown as an expense but are a reduction in the realized return. Having the scale to make larger financial commitments to these asset classes enables more effective negotiating power, generating cost-efficient fees from external managers of these assets. Internal management of alternative asset classes can also significantly reduce investment costs.

While research suggests large plans consistently feature lower investment management unit costs, estimates of enhanced returns are more mixed. In many studies, increased investment returns by large plans were associated with greater allocations to alternative asset classes, where scale tends to be a prerequisite for cost-effective participation. Generally, it appeared that the higher costs associated with investing in alternative asset classes were more than offset by improved returns. A number of studies made the link between internal management and cost-effective access to alternative asset classes, indicating that about 40 per cent to 60 per cent of gains were attributable to increased investments in alternatives.6 

It is important to note that some studies show that the very largest plans can underperform small to large plans in some of the traditional asset classes, particularly in down markets.7 This underperformance was due largely to the liquidity challenges of trying to trade large volumes of assets, particularly when there were fewer buyers. In my view, any underperformance with respect to certain asset classes would be more than offset by the advantages of a pooling framework described above. While I appreciate that investment unit costs should not be expected to decline indefinitely as assets under management increase, I expect that in aggregate, economies of scale would be positive, if diminishing, for a fund up to and including the potential scale being contemplated. 

Based on my research review, I conclude:

  • the cost advantages of scale are significant, particularly as it relates to the development of in-house expertise; 
  • more cost-effective access to alternative asset classes made possible through a larger fund can reasonably be expected to result in more diversified investments; and
  • a pooling framework should help facilitate the potential achievement of higher net returns.

2.4. Potential Cost Savings for Ontario Public-Sector Entities

In this section, I combine investment cost and strategy data from the CEM Benchmarking Inc. (CEM) database and de-identified target asset allocation data from the Financial Services Commission of Ontario’s Investment Information Summary (IIS) database to generate rough estimates of the potential annual savings.

The analysis groups pension plans into four categories:

  • small plans with assets of less than $1 billion;
  • medium plans with assets of $1 billion to less than $5 billion;
  • large plans with assets of $5 billion to less than $40 billion; and
  • largest plans with assets of $40 billion or more.

While investment costs and savings estimates are aggregated in this section for the purposes of presentation, the analysis was undertaken at the asset-class level to account for asset-class specific economies of scale. Due to data limitations the analysis excludes potential savings from non-pension funds and six pension funds with total assets of about $40 billion.

The target asset allocations for the three smaller categories of plans are reported below at a high level. These target asset allocations are based on information collected through the IIS, and weighted by plan assets.

Weighted Average Target Asset Allocations (%)

Asset Class

Small

Medium

Large

Equities

59

53

49

Bonds

38

34

35

Cash

2

1

1

Commodities

0

1

1

Alternatives

1

11

14

Total

100

100

100

Estimated baseline investment costs were generated by combining the representative asset mixes developed using the IIS data with the corresponding representative unit costs for plans of a given size as reported by CEM. The results closely approximate the actual investment costs as contained in the IIS database.

Scenario A is intended to estimate the potential investment cost savings for the smaller, medium and large funds if they were able to benefit from the cost structures of the largest funds generated only by the ability to lower fees with external asset managers. The analysis holds constant the proportion of assets in each asset class managed internally as well as the proportion of assets in each asset class managed using passive investment strategies.

One of the principal benefits of pooled asset management would be the ability to employ in-house investment management expertise. Scenario B is intended to estimate the incremental potential investment cost savings associated with a shift toward internal management, consistent with the approach employed by the largest funds, while assuming no change in the proportion of assets managed passively and the asset allocation by plan size.

Finally, I also observe that a number of the largest funds invest a greater proportion of their assets  internally and using passive strategies at extremely low costs – 2 to 5 basis points – which compares favourably against the externally-managed active strategies currently employed by many small funds in Ontario’s broader public sector.  Consistent with that approach, Scenario C is intended to estimate the incremental potential investment cost savings by assuming a shift toward passive management, consistent with the approach employed by the largest funds. As with the other scenarios above, Scenario C does not assume any change in asset allocation.

Estimated Average Aggregate Investment Costs by Plan Size (%)

Scenarios

Small

Medium

Larger

Baseline Cost Structure

0.0033

0.0040

0.0036

Scenario A Cost Structure

0.0027

0.0037

0.0035

     Incremental Savings

0.0006

0.0003

0.0001

Scenario B Cost Structure

0.0018

0.0026

0.0027

     Incremental Savings

0.0009

0.0011

0.0008

Scenario C Cost Structure

0.0013

0.0024

0.0024

     Incremental Savings

0.0005

0.0002

0.0003

Total Potential Savings

0.0020

0.0016

0.0012

Estimated Average Aggregate Potential Savings by Plan Size
($ Millions)

Scenario

Small

Medium

Larger

Total Assets (All Scenarios)

$9,500

$15,800

$31,900

Scenario A Incremental Savings

$6

$3

$3

Scenario B Incremental Savings

$9

$17

$26

Scenario C Incremental Savings

$5

$3

$10

Total Potential Annual Savings

$20

$23

$39

Based on the analysis above, I expect that the pooling framework proposed could generate cost savings of up to $82 million annually. As mentioned earlier, this savings estimate does not include potential savings from excluded non-pension investment funds and a handful of pension plans for which IIS data was unavailable. Assuming that these funds have similar target asset mixes and cost structures as comparably-sized pension funds, the savings estimate increases to about $130 million annually.

In addition to using the CEM data on investment cost and management styles, I conducted a similar analysis using data from a sample fund with assets putting it in the largest category. This alternative methodology produces total savings of $147 million annually, suggesting my savings estimates may represent a conservative estimate of the potential financial benefits that would be generated by a pooling framework.

Over time, I expect that improved risk management processes, larger and more cost-effective allocations to alternative asset classes and the resulting portfolio diversification could reasonably be expected to result in higher investment returns, although I have not attempted to quantify those impacts here. Using the above estimates as guidelines, I estimate a pooling framework would achieve potential savings of between $75 million and $100 million annually, once fully implemented.

2.5. Trends

Canada is emerging as a world-wide leader in successfully adapting the advantages of large funds to the public sector. Several federal and provincial entities have been established as arm’s-length investment management entities with sufficient scale, independent boards and internal investment management, remunerated at rates competitive with the private sector. The following organizations are examples of Canadian pension plans that have developed the internal capacity to manage their funds over the last 25 years:

  • The Ontario Teachers’ Pension Plan: $117.1 billion8;
  • The Canada Pension Plan Investment Board: $165.8 billion9;
  • The Public Sector Pension Investment Board: $64.5 billion10;
  • The Caisse de dépôt et placement du Québec: $159 billion11;
  • The Ontario Municipal Employees Retirement System: $55 billion12;
  • The British Columbia Investment Management Corporation (bcIMC): $92.1 billion13; and
  • The Alberta Investment Management Corporation (AIMCo): $69.7 billion14.

Notably, unlike the other examples listed above, the Caisse de dépôt et placement du Québec, bcIMC and AIMCo were established with a structure designed to manage pooled investment portfolios, allowing for client-controlled asset allocation for multiple public-sector pension plans and investment funds. Through pooled asset management, these entities achieve sufficient scale to produce significant cost savings through internal investment management and access to alternative asset classes.

British Columbia Investment Management Corporation administers pooled assets from several public-sector pension plans, public bodies, publicly-administered trust funds and government operating funds. Its investments help finance insurance and benefit funds that cover more than 2.3 million workers, from diverse public and broader public sectors, including B.C. public-sector workers, university and college instructors and staff, municipal employees, healthcare workers, firefighters, police officers, teachers and employees of WorkSafeBC, the Insurance Corporation of British Columbia and BC Hydro.

Alberta Investment Management Corporation manages pooled assets from public-sector pension plans, provincial endowment funds (e.g., the Alberta Heritage Savings Trust Fund, formerly managed within the Alberta Ministry of Finance) and government-mandated funds (e.g., Government of Alberta bank accounts) with diverse client groups, including municipal employees, public servants, police officers and provincial judges.

The success of Canadian entities such as the Canada Pension Plan, Teachers’, OMERS and HOOPP has garnered world-wide attention. The following are several international examples of pooled asset managers in operation or in development.

In the United Kingdom, the London Pensions Fund Authority is reviewing with the Treasury the potential for creating a new investment entity, which would manage £30 billion in assets pooled from the approximately 35 independent pension funds. Many of the independent London funds have less than £1 billion in assets. Local government has proposed the initiative based on the great potential for savings through the streamlining of costs for actuaries, investment advisors and fund managers.

In Sweden, the national pension funds are invested and managed separately in four independent funds, each managing assets of approximately $30 billion. The separation of assets was originally conceived to reduce the impact of the funds on the domestic market, diversify management risk, enhance performance through competition and mitigate the risk of political interference, as well as to diversify strategic risks. Since then, all of these reasons for establishing disaggregated funds have been widely questioned and mostly discredited. Therefore, Sweden is currently reconsidering this disaggregated investment model and favouring the implementation of one large fund, citing that the current multi-fund structure creates unnecessary costs and lacks sufficient economies of scale.

In New York City in 2011, the comptroller proposed consolidating five of the city’s pension plans into one pooled fund, professionally managed by a new investment management entity. The plan, supported by the mayor and several unions, would merge the pension plans’ five boards, currently totalling 58 directors, into one far smaller board that would oversee the plans’ investments of $120 billion. The plans cover 237,000 retirees and more than 300,000 current city and city-affiliated employees, including teachers, firefighters, police officers, sanitation workers and corrections officers. The savings through consolidation were estimated to be at least $1 billion per year. Currently, implementation is on hold as the city is having trouble rallying support from all affected unions.

In Australia, a broad review of the country’s superannuation system was released in 2010: Review into the Governance, Efficiency, Structure and Operation of Australia’s Superannuation System. One of the review’s key recommendations was that superannuation funds should “actively examine and conclude whether, on an annual basis, its [fund] has sufficient scale on its own (with respect to both assets and number of members) to continue providing optimal benefits to members.” The report clearly recognized the benefits of scale and that the consolidation of pension schemes should be supported by “removing barriers…to consolidation, so that scale can be more easily achieved.” Australian legislation removing tax barriers to consolidation and creating consolidated investment entities has been and continues to be introduced.

Around the world, governments are recognizing that public-sector pension funds can benefit from economies of scale. Ontario and Canada have shown world-wide leadership in this area, successfully demonstrating the advantages of large and, in some cases, pooled funds. Ontario has the opportunity to continue this tradition by implementing a new pooled asset manager for Ontario’s smaller public-sector pension plans.

2.6. The Opportunity

Over the course of my consultation process, there was some dispute over the scale necessary to fully realize economies of scale and related benefits. Based on available research, I conclude that assets under management should be in excess of $50 billion to ensure access to positive economies of scale across a broad range of asset classes including alternatives. With these figures in mind, I considered the following categories of funds:

Categories

Defined Benefit / Hybrid Plans and Investment Funds

Assets
($ Billions)

Government of Ontario

  • Public Service Pension Plan

$17

Investment Funds

  • Nuclear Funds
  • Workplace Safety and Insurance Board Consolidated Fund
  • Agricorp Production Insurance Fund


$28

University Sector

  • University of Toronto
  • York University
  • Queen’s University
  • University of Ottawa
  • 22 Additional Plans with Known Assets of less than $1 Billion

 

$12

Miscellaneous Broader Public Sector

  • Workplace Safety and Insurance Board Pension Plan
  • Hospital for Sick Children
  • St. Michael’s Hospital
  • St. Joseph’s Health Centre
  • Grand-River Hospital
  • 7 Additional Plans with Known Assets of less than $200 Million

 

 

$5

Municipal Sector

  • Various  City of Toronto Plans
  • Various City of Hamilton Plans
  • City of Ottawa
  • 2 Additional Plans with Known Assets of less than $200 Million

 

$2

Electricity Sector

  • Ontario Power Generation
  • Hydro One
  • Independent Electricity System Operator
  • Electrical Safety Authority

 

$14

Smaller Jointly-Sponsored Pension Plans

  • OPSEU Pension Plan
  • Colleges of Applied Arts and Technology Pension Plan
  • Toronto Transit Commission Pension Fund Society


$23

The assets of the plans and funds included in the table above total about $100 billion. I see this list as the starting point for determining how a pooling framework might be constituted. Not included above are the three largest sector-based, jointly sponsored pension plans: Teachers’, OMERS and HOOPP.  I have determined that these plans are sufficiently large and well established that they should be considered outside the scope of the pooling framework, except where they might be in a position to offer services to some of the smaller plans (see Section 3.1).  Also not included above are defined-contribution pension plans, supplementary pension plans or endowment funds, all of which are potential voluntary participants in a pooling framework.

I received advice on the appropriate scale required to invest cost-effectively in the broad range of asset classes. My starting point in considering the advice was to examine the current situation. Some key facts emerged during the consultation process:

  • A significant majority of the funds under consideration believe that they should have a material allocation of assets to alternative asset classes.
  • In most cases, funds are not satisfied with their current ability to access these asset classes at an attractive cost and are under-allocating to these asset classes.
  • The overwhelming majority of the assets in the smaller funds are actively managed by external fund managers.
  • A much larger allocation of assets in the largest funds was internally managed, with a significant allocation of traditional asset classes managed passively.

Cost-effective external fund management is possible at a reasonably modest scale in publicly-traded equities and fixed-income instruments. However, it is only through significant scale and the development of internal expertise that cost-effective access to alternative asset classes can be achieved.

Almost all of the smaller and medium-sized funds I spoke with did not have the capacity to invest directly in these alternative asset classes, and, in most cases, do not have access to the best funds. The cost to find, negotiate and manage alternative assets is prohibitively high for smaller funds, and the cost to invest in external alternative asset managers is significantly higher than the cost to invest in traditional asset classes. I received advice that $3 billion to $5 billion of assets in each of these alternative asset classes would be sufficient to build and maintain internal management teams for these asset classes. Recognizing that these asset classes would not be chosen by all institutions, and would likely represent only a modest percentage of overall assets of any individual institution, the guidance above suggests an overall fund needs to be of significant size to enable the development of internal expertise (to manage and negotiate with external managers) and internal management teams.

Furthermore, sufficient scale would enable the investment management entity to attract world-class talent to internal management teams and leadership positions, which is key to the success of the pooling framework. Sufficient scale would be a magnet for best-in-class investment management talent and appeal to the kind of experienced leaders the investment management entity would need to retain. Therefore, I estimate the requisite threshold size to be roughly $50 billion.

Recommendation 2-1:  Pooled asset management should only be undertaken if it achieves sufficient scale to support the development of cost-effective internal investment management teams and to attract and retain world-class leadership. I estimate this scale threshold to be at least $50 billion.

3. Design Considerations

The recommendations that follow are intended to maximize the benefits of a pooling framework for participating public-sector institutions and to ensure effective, responsive governance of the new pooled asset manager. The design options presented are considered in light of their relative merits and support the success of a pooling framework. Any pooled asset manager must:

  • permit participating institutions to retain fiduciary responsibility and control over asset allocation decisions, given variations in the liability profiles across pension plans and investment funds. This would require a family of unitized pooled funds, similar to mutual funds;
  • operate at arm’s-length from government and be responsible to an independent and representative board of directors, to ensure that investment decisions are made solely on the basis of seeking the best returns without political influence;
  • feature world-class governance, professional management, competitive compensation and effective oversight, to ensure the confidence of its constituents, including the management teams of participating institutions, and pension plan members; and
  • have sufficient assets under management – at least $50 billion – to support investments in a broad range of asset classes at the most competitive costs. All else being equal, no participating institution should face higher investment costs than it does at present.

3.1. Establishing a New Entity versus Using Existing Large Pension Plans

As suggested in the 2012 Ontario Budget, a pooling framework could rely on either a new or an existing entity to support pooled asset management.

I heard from a number of stakeholders who raised the possibility of establishing multiple, sectoral investment management entities. While smaller public-sector pension funds may at times benefit from pooling with one of the existing large pension funds, I do not view the creation of multiple new sector-based investment management entities as the best solution. For example, the university and electricity sectors each have assets of approximately $15 billion. As discussed in the previous chapter, pooled funds of this size would not be sufficiently large to achieve the lowest possible costs or cost-effective access to a broader range of investment opportunities. This arrangement would also lead to duplication of start-up and operational costs.

Teachers’, OMERS and HOOPP are each of sufficient scale to achieve cost savings for the sectors they serve as well as to act as potential platforms to support investment pooling in Ontario. In addition, plans wishing to have their investments managed by these funds have a history and track record to assess their performance. While both Teachers’ and OMERS currently have the authority to offer investment management services, only the latter has a structure in place to manage third-party assets. Significant investments and organizational changes would be required before either the Teachers’ Plan or HOOPP would be in a position to manage pooled funds.

Furthermore, the total value of third-party assets that these larger plans may be asked to manage could rival the value of assets they already manage on behalf of their own members. While these plans are well-governed at present, the appropriateness of their governance structures and accountabilities to potential participating institutions would need to be re-examined before these larger plans could assume responsibility for the large-scale management of third-party assets.

Throughout my consultations, I heard strong views on the appropriateness of transferring assets from smaller institutions into one of these plans. Many potential participants expressed a preference for the government to establish a new pooled asset manager rather than asking an existing large pension plan to manage additional assets. Key reasons cited for this preference included the following:

  • the opportunity to establish a governance structure that met the expectations and requirements of potential participants;
  • the ability to establish a family of pooled funds that could accommodate the asset allocation decisions of participating institutions of all sizes; and
  • the ability of a new entity to reduce transition risks by assuming the assets of participating institutions in-kind.

Recommendation 3-1: The Province should introduce legislation to establish a new pooled asset manager to facilitate pooled asset management for Ontario’s smaller public-sector institutions, hereafter referred to as the Ontario Investment Management Corporation (“the Corporation”).

3.2. Mandatory Versus Voluntary Participation

Broadly speaking, participation within a pooling framework could be mandatory or voluntary for some or all of Ontario’s public-sector pension and investment funds. While recognizing that participating institutions could retain responsibility for asset allocation decisions, I also considered whether participation as part of a pooling framework should be mandatory or voluntary at the asset class versus institutional levels. For example, an organization could invest with the Corporation, either on a mandatory or voluntary basis, within some or all asset classes.

The absence of any government mandate would make it very unlikely that the Corporation would have sufficient assets to cost-effectively develop the capacity to serve the participating institutions. I estimate that strictly voluntary participation would likely lead to a pooled fund with assets of $30 billion to $40 billion in the near term, a level I deem insufficient to fully realize the potential benefits of pooled asset management. A strictly voluntary approach would almost certainly leave intact the fragmented pension landscape for the foreseeable future. In light of this, I view some measure of legislated participation as a requirement to ensure sufficient scale and effective risk management within a reasonable time horizon.

Recommendation 3-2: The government should legislate the participation of public-sector pension funds that are expected to realize appreciable benefits from pooled asset management.

In the interest of facilitating a faster transition to a pooling framework, legislation could be used to compel the participation of institutions or funds that are expected to realize appreciable benefits from pooled asset management. The legislation would indemnify current fiduciaries from any fiduciary liability arising from the transfer of investment management responsibility to the Corporation.

Recommendation 3-3: The government should include provisions in legislation that would indemnify current fiduciaries from any fiduciary liability arising from the legislated transfer of investment management responsibility to the Corporation.

As previously discussed, I view the potential benefits of pooled asset management as material for public-sector entities, in aggregate. The scale required to achieve the potential benefits necessitates the inclusion of smaller funds that would realize significant economies of scale, as well as at least some of the medium-sized and larger funds, that would realize economies of scale in some cases, and access to a broader range of alternative assets at cost-effective rates. In the absence of evidence to suggest material risks of aggregate diseconomies of scale for a fund up to and including the potential scale being contemplated, my view is that all broader public-sector pension and investment funds with assets of less than $40 billion should be viewed as candidates for potential participation as part of a pooling framework.

Recommendation 3-4: All public-sector institutions with pension funds of less than $40 billion in assets under management should be compelled to pool their assets with the Corporation, subject to limited exceptions.

Aside from pension funds, there are a number of other investment funds that feature separate investment management teams that could be candidates for pooled asset management. These funds include the Ontario Nuclear Funds, the Workplace Safety and Insurance Board investment funds and the Agricorp production insurance fund. Using the Corporation would keep investment management decisions separate from government and ensure that investment objectives are independent from government, while obtaining the best possible returns through cost efficiencies, access to broader asset classes and superior risk management.

Recommendation 3-5: The Ontario Nuclear Funds, the Workplace Safety and Insurance Board investment funds, and the Agricorp Production Insurance fund should be compelled to participate in the pooling framework.

During my consultations, I met with several public-sector entities and other constituents who questioned the merits of asset pooling for their institution, based on specific situations. Some important considerations for government in considering potential exceptions to the overall recommendation for mandatory participation are as follows:

  • The Ontario Public Service Employees Union Pension Plan, the Colleges of Applied Arts and Technology Pension Plan, and the Toronto Transit Commission Pension Fund Society feature jointly-sponsored governance, where decisions on benefit levels and contribution rates, as well as responsibility for funding any shortfalls, are shared between representatives of employers and employees. The government would need to override the current governance structure in respect of investment management decisions in order to implement the recommendation as contained in this report. In each situation, the pension plan is large enough to already achieve significant economies of scale in traditional asset classes.
  • The recommendation to include the assets of several municipal sector pension plans that are not currently managed by OMERS reaches past the direct oversight of the provincial government and into municipal jurisdiction. These plans could also consider the potential amalgamation of their assets into OMERS.
  • The proposed participants from the electricity sector advised me of ongoing efforts to create a sectoral jointly-sponsored pension plan. While achievement of such a sectoral plan would not achieve the full benefits anticipated in this report, significant economies could be realized, particularly in traditional asset classes. A sectoral plan is plausible, given the relatively small number of pension plans under consideration, as well as the similarity in benefit levels across the plans in the sector.
  • While the recommendation is to include the assets of the pension plans of the above-noted entities in the Corporation, I recognize that the size threshold recommended in this report could be achieved with the exclusion of one or more of these entities.

Should the government decide to create an exception to the broad mandate being proposed, there would still exist potential for an exempted entity to realize economies of scale in alternative asset classes. Even the medium and larger funds have difficulty accessing alternative asset classes at reasonable costs.

Recommendation 3-6: All public-sector institutions whose assets are not mandated to be managed by the Corporation should be permitted to voluntarily access the services and individual asset classes available through the Corporation, subject to reasonable terms and conditions, and on a “cost-recovery” pricing basis (i.e., voluntary participants would experience pricing on the same basis as mandated participants).

Some public-sector entities, such as universities and hospitals, also manage endowment funds. These funds are often co-managed with other assets and as a result benefit from lower external management fees. If university pension plans are compelled to have their assets invested with the Corporation, orphaned endowment funds or supplemental employee retirement plans would likely face higher incremental investment management costs. Permitting these other funds to invest their assets with the Corporation on a voluntary basis would reduce this risk. I appreciate that accommodating endowment funds may lead to additional implementation challenges (e.g., differential treatment of endowment funds versus pension funds for tax purposes in some jurisdictions), but view this accommodation as necessary to eliminate any undue burden on participating institutions.

Recommendation 3-7: Broader public-sector institutions should be permitted to voluntarily access the services available through the Corporation for endowment funds and supplemental employee retirement plan funds, subject to reasonable terms and conditions, and on a “cost-recovery” pricing basis. The Corporation should be equipped to accommodate these types of funds immediately upon its establishment.

Some smaller pension funds may wish to be aligned with existing large asset management entities (e.g., closed municipal plans with OMERS, or hospitals with HOOPP). Some of these plans have examined or are examining opportunities to transition to one of these existing plans.

Recommendation 3-8: Any public-sector pension plan that can negotiate an agreement-in-principle to transition its assets to an existing large Ontario asset management entity (i.e., Teachers’, OMERS, or HOOPP) with a signed memorandum of understanding prior to the establishment of the Corporation should not be compelled to pool its assets with the Corporation.

While not the focus of my mandate, I have nonetheless considered the treatment of defined-contribution pension plans. The administrative complexities of transferring defined-contribution plans into a pooled fund would be significant, as investments in these accounts are generally member-directed. Hybrid plans, featuring elements of both defined-contribution and defined-benefit pension plans, are not generally member-directed and would not face similar implementation issues.

Recommendation 3-9: The Corporation should be structured to facilitate the management of defined-contribution assets. Defined-contribution funds would be permitted to pool assets with the Corporation on a voluntary basis, but only at such time as the capacity of the Corporation permits.

Many stakeholders highlighted the importance of a robust relationship management function (i.e., the need to keep an open and transparent relationship with participants to ensure that each organization and its constituents understand the allocation of their assets and the performance of their plan), particularly given the potential diversity of participating institutions with respect to fund characteristics and investment needs. A number of administrators also noted their reliance on investment advisory services. In my view, the ability of the Corporation to provide these types of services would be particularly important as it seeks to offer a broader range of investment opportunities to participating institutions.

Recommendation 3-10: The Corporation should develop the capacity to offer cost-effective advice on asset allocation decisions to participating institutions.

In my consultations, I heard from some stakeholders who thought that the current mandate, focused solely on pooled asset management, could be expanded. They made the case that consolidation of plan administration (e.g., plan recordkeeping and benefit calculations) was at least as important as investment management and that the government could further improve the affordability and sustainability of public-sector pension plans. While I have not included this as a recommendation, I acknowledge that there should be no obstacle to voluntary consolidation of administrative functions by public-sector pension plan sponsors.

3.3. Structure and Governance

Having recommended that the pooling framework be achieved through the establishment of a new pooled asset manager, I now consider the structure and governance of the Corporation. My recommendations are based on a review of the structure and governance models employed by some of Canada’s most successful pooled investment funds and large pension plans.

I heard diverse opinions during my consultations with respect to the structure and governance of the Corporation. However, there was complete consensus on one key criterion: the Corporation must have an arm’s-length relationship with government, meaning that the government would have limited control over and restricted ability to influence the Corporation’s governance and operations.

The government’s control, influence and ability to intervene should be clearly defined and restricted by legislation, but should enable oversight where necessary, including adequate reporting requirements to ensure accountability and transparency. Supporting that goal, the Corporation’s board should have a prescribed mandate requiring it to serve and act in the best interests of its clients, thus ensuring that the board’s duties and responsibilities are to its clients and not to the government.

Recommendation 3-11: The legislation establishing the Corporation should clearly define the relationship between the Government and the Corporation, limiting control and influence to specific areas, including accountability and transparency through reporting requirements. The legislation should include a mandate clarifying that the Board has a duty to serve and act in the best interests of its clients.

I view excellence in governance, including investment, risk management, legal, accounting, human resources and strategic planning expertise, as an essential element of an effective and efficient investment management entity. I also recognize the benefits of active stakeholder involvement in strategic decision making, as well as the constructive role representative directors would play in managing stakeholder relationships. The latter would be particularly important given the potential number of participating institutions.

Therefore, I envision that the Corporation be led by a board of directors, all of whom meet minimum standards of proven financial ability and/or relevant work experience, but with a set number of positions reserved for appointments by client groups and plan members. I propose that the composition of the board be 11 directors, each with one vote:

  • three of the directors would be nominated and appointed by the clients (i.e., plan sponsors and investment funds) of the Corporation, but would not necessarily be in the management of such client nor be union executives of such client;
  • two of the directors would be appointed by the plan members as represented by labour officials; and
  • the remaining six directors, including the chair, would be selected solely on the basis of their professional qualifications from a pool of candidates generated by a nominating committee and appointed by the board.

Each client organization and each labour group would be able to recommend a sufficiently qualified representative to the nominating committee of the board. The nominating committee would have the responsibility to select the three client representatives and two plan member representatives.

It is important to note that while five directors would be nominated and appointed by their respective client groups and unions, all directors, once appointed as fiduciaries of the Corporation, would have a duty and responsibility to the Corporation and to the plans invested with the Corporation. All directors would be required to act in the best interests of the Corporation as a whole, regardless of how they came to be appointed to the board.

Recommendation 3-12: The board of directors should have 11 representatives including a chair. Three directors should be nominated by client groups and two directors should be nominated by plan members. Six, including the chair, should be selected solely based on professional qualifications, and appointed by the board.

The nominating committee would consist of five members from the board of directors; the chair, two independent directors, one director representing clients and one director representing plan members. The nominating committee would verify that board candidates nominated and appointed by their client groups meet the minimum standards of proven financial ability and/or relevant work experience that are required of all directors of the board.

Recommendation 3-13: All candidates for the board would be screened by the nominating committee to ensure they meet minimum qualification standards for eligibility before appointment, and selected based on the decision of the nominating committee as to fit for the Corporation board.

Upon establishment of the Corporation, the Minister of Finance would appoint the initial chair of the board, whose first priority would be to establish the nominating committee and generate the board. The Minister of Finance would appoint directors only until such time that quorum was reached and the board could regenerate itself.

Thereafter, to ensure some level of external oversight, since the Minister would have no influence over the composition of the board, the government should have the legislated right and responsibility to audit the governance of the Corporation. While I expect it to be unlikely, if determined necessary through a transparent audit process, the government should have the authority to require a director to resign for cause, and in extreme cases, have the authority to dismiss the entire board.

4. Implementation Considerations

Almost every stakeholder group highlighted the complexity inherent in developing a pooling framework, including the importance of a transition and implementation plan, risk mitigation strategies, and an appreciation of the lead times required to fully realize the benefits of pooled asset management. I take this opportunity to outline, at a high level, how responsibility for asset management could be assumed by the Corporation, without the need for liquidation of assets or other unnecessary market disruption.

While I believe that some savings could be realized in the near term, I recognize that establishing a mature investment management corporation, and the realization of all of the associated benefits, would take years. I also acknowledge that the establishment of the Corporation would result in some implementation costs, but I view these costs to be both manageable and recoverable.

4.1. Implementation Strategy, Timelines and Costs

In consultations, there was consensus among potential participants that transition to the new Corporation would be complex and time consuming. The complexity of the undertaking would be exacerbated by the following:

  • the large number of potential participating institutions;
  • the inclusion of non-pension funds; and
  • diversity of investment approaches, external managers and assets under management.

A number of stakeholders raised concerns about the costs of implementation and suggested that participants, particularly those compelled by the government to participate in a pooling framework, should not be obliged to pay for these implementation costs.

I propose that the government finance the start-up costs of the new organization, which could subsequently be recovered from savings realized through negotiation with, and consolidation of, external investment managers. By holding investment costs paid by participating institutions constant, I expect that the government could fully recover its loan over time. At that point, the investment management costs paid by participating institutions would be permitted to change, reflecting each client’s asset allocation decisions. I estimate the maximum amount of funding required to finance the operations of the Corporation, and thus the size of the required government loan, to be up to $50 million.

Recommendation 4-1: The government should commit to ensuring that in the first three years of the Corporation’s operation, participating institutions would not face increased total investment management costs, except in cases where an institution changes its asset allocation. Beyond this three-year period, investment costs should be charged to participating institutions on the basis of asset-allocation decisions and the direct costs of investment.

If legislation is proclaimed, an immediate goal would be to secure a chair of the board, the nominating committee to generate additional board members, and a chief executive officer (CEO). The chair would need to be chosen as soon as possible following the proclamation of legislation, with directors and a CEO being selected by the time of passage of legislation, or shortly thereafter. I estimate that this could be achieved on a budget of less than $3 million.

Once the Corporation is in operation, I anticipate that assets could be transferred in-kind from participating institutions, with the Corporation continuing to use existing external investment managers in the near-term. While respecting the existing investment management decisions of participating institutions, this would also serve to reduce market disruption and limit the need to wind-down pre-existing positions.

I recognize that while the vast majority of assets in the public sector are managed externally, there are a few sophisticated investment teams that internally manage assets on behalf of institutions that may be compelled to participate in a new pooling framework. To limit uncertainty and mitigate the risk of professional flight, transitional agreements could be signed with these institutions, retaining these teams to manage assets as might be required before establishing the Corporation, and eventually transferring these teams to the Corporation to support implementation. I expect that transitional arrangements could be negotiated with the existing internal teams.

Recommendation 4-2: Once the Corporation is established, anticipated within approximately six months following the passage of legislation, it should have arrangements in place to transfer the internal investment teams from participating institutions to support the early stages of implementation.

I illustrate at a high-level the anticipated evolution of the operations of the Corporation from the preliminary planning phase to full operation, on the assumption that the Corporation is established on January 1, 2014.

Phase 1: Planning (July 1, 2013, to December 31, 2013)

Phase1

Under the status quo, organizations A, B and C make asset allocation decisions and individually determine which external investment managers to employ at unit costs that vary depending upon the value and performance of assets under management.


Phase 2: Negotiation and Assumption of Assets (January 1, 2014 to June 30, 2014)

Phase2

Investment management agreements are negotiated and the assets of organizations A, B and C are gradually inherited by the new entity. The entity then uses the collective purchasing power of its clients to negotiate lower external management fees.


Phase 3: Consolidation (July 1, 2014, to December 31, 2015)

Phase3

The new entity continues to negotiate lower external management fees while assessing investment performance and winding down underperforming positions. The number of external managers is reduced, allowing the new entity to negotiate additional savings.


Phase 4: Internalization (January 1, 2016, to December 31, 2017)

Phase4

The new entity begins to build the capacity to internally manage certain assets, offering its services on a not-for-profit basis. This broadens access to alternative asset classes, facilitates diversified investment strategies, and passes cost savings on to clients.


Phase 5: Full Operation

During this period, internal management would be used where appropriate, as would the full family of pooled funds. All investment management costs would be attributed to participating institutions on the basis of asset allocation.

4.2. Operational Considerations

While the strategic decisions and day-to-day operations should be left to the board and management team, I have a number of general recommendations to support the efficient operation of the organization.

One of the principal longer-term opportunities presented by the creation of a new Corporation would be the ability to offer high-quality investment management services on a cost-recovery basis. To achieve this and offer services on par with those offered by the private sector, the Corporation would need to offer competitive compensation packages, including both base salary as well as incentive pay distributed on the basis of performance against established benchmarks or objectives. This approach to compensation has been used effectively by the Canada Pension Plan Investment Board, Teachers’, OMERS and HOOPP, and has resulted in the development of highly professional investment and leadership teams. Failure to attract and retain qualified leadership and investment management talent could severely compromise the ability of the Corporation to achieve its intended objectives.

Recommendation 4-3: Like existing large pension plans, the Corporation should not be subject to the compensation bands of traditional public-sector entities. Compensation for directors should be comparable to other like pension funds and compensation for management should be competitive to external benchmarks. Furthermore, management compensation arrangements and perquisites should be overseen by a board compensation committee.

Potential participating institutions expressed a strong desire to continue to determine asset allocation based on their distinct investment needs and risk tolerance. This desire could be accommodated by the Corporation offering a suite of investment vehicles each with different investment objectives that could be mixed and matched to accommodate its clients’ needs. I expect that this family of investment pools offered would be based on client demands.

The proposed family of investment pools could also be tailored to the needs of the client organizations through unitization, the process of consolidating the value of all investments that are held in the pool.  Participating client organizations would own individual units of the unitized investment pool.  The mechanics of this form of pooled investing are similar to that of a mutual fund and would allow all client organizations to invest as they see fit in the various pools of assets.
A unitized investment pool operates like a mutual fund on a larger scale.  The units within the investment pool could be clearly segregated to identify each client organization’s share in the total pool. Unitization is frequently used for pools of invested endowment funds.

Recommendation 4-4: The Corporation would need to employ a unitized fund structure, providing the flexibility to accommodate the distinct asset-allocation decisions of each participating institution.

One of the early challenges for the new investment management entity would be to earn the confidence of participating institutions. This could be accomplished by communicating a comprehensive transition plan detailing how and when responsibility for investment management would be transferred from participating institutions. It would also need to establish a corporate and governance policy, outlining at minimum its commitment to ethical governance and management; an arm’s-length relationship with government; and its obligations to participating institutions (e.g., low-cost service provision, rigorous risk management and transparency).

A robust risk-management framework should also be a cornerstone of the Corporation. It is further expected that an internal procurement policy should be established by the Corporation, as part of its corporate governance regime.

While I make some recommendations around how the transition plan should work, I reserve specific recommendations on overall investment philosophy, risk management, procurement policy and ethical governance and management, as these are fundamental priorities for the board and management. That said, it is clearly envisioned that these items would be developed based on industry best practices, consistent with the largest Canadian asset managers. 

Recommendation 4-5: The Corporation, immediately after establishment, would develop and adhere to a clearly stated investment philosophy, robust risk management framework, transparent procurement policy and ethical governance and management guidelines.

4.3. Legislative and other Considerations

To allow the Corporation sufficient time to wind-down underperforming assets, develop a fund family and reduce costs, institutions that were compelled to pool their assets with the Corporation should remain captive clients for a cooling-off period. After that period, to fulfil their responsibility to their plan members, institutions should be able to move all or part of their assets away from the Corporation. The decision-making authority on asset allocation and the selection of individual asset managers would be maintained by the institution. I anticipate the appropriate start-up and operating period prior to allowing participating institutions the ability to withdraw from the Corporation to be seven years.

Recommendation 4-6: After a cooling-off period, participating institutions should be free to withdraw from the pooling framework, as directed by their trustees or governors. This cooling-off period would give the Corporation time to negotiate lower investment management costs, rationalize external investment managers, and develop internal investment expertise. It should also allow for a significant period of full operation and a more accurate assessment of the Corporation’s performance.

5. Conclusion

Around the world, governments are recognizing that public-sector pension funds can benefit from economies of scale. Ontario and Canada have shown world-wide leadership in this area, successfully demonstrating the advantages of large, and in some cases, pooled funds.

Ontario has the opportunity to continue this tradition by implementing a new pooled asset manager for Ontario’s broader public sector pension plans and investment funds. More specifically, implementing pooled asset management would reduce duplication and costs, broaden access to asset classes and enhance risk management practices. To the extent that these advantages support more diversified portfolios among participating institutions, pooled asset management may also help realize improved investment returns over the long-term.

I estimate a pooling framework would achieve potential savings of between $75 million and $100 million annually, once fully implemented. These savings would enhance the sustainability of participating pension and investment funds, to the benefit of members and taxpayers. Any savings or improved returns may also reduce the need for increased contribution rates by employers or employees.

The recommendations contained in this report are intended to maximize the benefits of a pooling framework for participating public-sector institutions and ensure effective, responsive governance of the proposed new pooled asset manager.  Adherence to these recommendations would better enable participating institutions to embrace the new pooled asset manager and help ensure the success of the initiative.

1“Maple Revolutionaries: Canada’s public pension funds are changing the deal-making landscape”; The Economist; March 3, 2012.

2 “Working Paper: Effect of Fund Size on the Performance of Australian Superannuation Funds”, Dr. James Richard Cummings, Australian Prudential Authority (March 2012).

3 “How Large Pension Funds Organize Themselves: Findings from a Unique 19-Fund Survey”, Jody MacIntosh and Tom W. Scheibelhut, Rotman International Journal of Pension Management (Spring 2012); and, “Internal Management Does Better After Costs”, Terrie Miller and Chris Flynn, CEM Benchmarking Inc. (October 2010).

4 “Is Bigger Better? Size and Performance in Pension Plan Management”, Alexander Dyck and Lukasz Pomorski, Rotman School of Management, University of Toronto (November 2011).

5 “Internal Management Does Better After Costs”, Terrie Miller and Chris Flynn, CEM Benchmarking Inc. (October 2010).

6 “Working Paper: Effect of Fund Size on the Performance of Australian Superannuation Funds”, Dr. James Richard Cummings, Australian Prudential Authority (March 2012); and, “Is Bigger Better? Size and Performance in Pension Plan Management”, Alexander Dyck and Lukasz Pomorski, Rotman School of Management, University of Toronto (November 2011).

7 “Does Scale Matter for Public Sector Defined Benefit Plans? Evidence of the Relationship Among Size, Investment Return and Plan Expense”, National Education Association (April 2009).

8 as of December 31, 2011

9 as at June 30, 2012

10 as at March 31, 2012

11 as at December 31, 2011

12 as at December 31, 2011

13 as at March 31, 2012

14 as at March 31, 2012