A study of 300 US pension funds by CEM Benchmarking reinforces the importance of asset allocation, highlighting the performance of asset classes, as well as new evidence on correlations between asset classes. Alex Beath, author of the study, discusses the implications for asset allocation with Amanda White.

A CEM Benchmarking study “Asset Allocation and Fund Performance of Defined Benefit Pension Funds in the United States Between 1998-2011,” has revealed the power of asset allocation as a driver of performance, and how asset class correlations add vulnerability to diversification.

Alex Beath, author of the study that looked at the asset allocation, fees and returns of 300 US pension funds with combined assets of $2.8 trillion from 1998-2011, says during that period the defined benefit pension plans included in the study increased their investments in alternative assets – private equity and hedge funds, real estate and other real assets such as commodities and infrastructure – by nearly 400 percent on average.

CEM used its extensive databases to examine how this reallocation to alternatives paid off in terms of gross returns and realized returns net of fees charged by investment managers.

And basically it didn’t.

Large corporate plans were the best performers among the funds analysed and this was due to their higher allocations to long duration as part of a liability matching strategy.

The report concludes that many pension funds could have improved their performance by choosing different asset allocation strategies and optimizing management fees.

CEM found that listed REITs were the best performing asset class over the period, and that hedge funds were the worst performer.

“Hedge funds didn’t just perform poorly but it also revealed that on average hedge funds look like US large cap stocks,” Beath says. “Hedge funds have a lower volatility than US large cap so that is a benefit, but they perform like US large cap stocks, but have much higher fees.”

In addition to the correlation between hedge funds and US large caps being very high, the study revealed a very high correlation between private real estate and REITs, and between private equity and US small cap stocks.

“If you look at raw data it looks different but once you correct the data to account for lag then you see there is a very strong correlation,” he says. “You might think by investing in these asset classes you are increasing diversification but you are not at all. Thinking you are de-risking by adding in uncorrelated asset classes like private real estate is totally an illusion.”

The performance differences in the cohorts of funds analysed in the study are due primarily to their asset allocation differences, the study found.

Large corporate plans were the best performers, and this was due to those investors allocating more to long duration fixed income, which is a direct result of those funds more consciously investing according to liability matching strategies.

“These large corporate funds had a much higher net return compared to public funds due to a shift in asset allocation prior to the financial crisis and a timely increase in long duration fixed income and a decrease in US large cap stocks. Corporate funds are more interested in liability matching than public funds and embrace LDI, which means they are more likely to alter allocations.”

The study found that corporate sector plans reduced their average allocations to US large cap equities and US broad fixed income securities by 19.6 percentage points and 10.3 percentage points, respectively, with the largest declines among the largest plans, those with total assets of more than $10 billion. Over the same period, allocations to long duration US bonds and hedge funds increased 16.9 percentage points and 5.0 percentage points, respectively, and again most dramatically among the largest plans.

Public sector plans also reduced on average their allocations to US large cap equities (by 18.7 percentage points) and to US broad fixed income (by 11.0 percentage points), although the reductions in both cases were somewhat larger for mid‐sized plans.

Unlike corporate sector plans, however, the allocation to long duration US bonds among public plans was notably unchanged over the period as public sector plans did not embrace LDI to the degree of their corporate sector counterparts. Public plans on average increased their allocations primarily to non‐US equities (by 11.0 percentage points), TAA/hedge funds (by 4.6 percentage points) and private equity (by 6.0 percentage points).

Large corporate sector plans were the best performing among all major pension fund cohorts classified by fund type and fund size, achieving an average annualized compound net return of 7.54 percent over the period 1998‐2011, significantly outperforming the 6.61 percent of all funds combined.

Linking the asset allocation decisions to performance and to determine the potential impact of a shift in asset allocations on portfolio net total returns over the period 1998‐2011, the study looked at estimates for each year of the marginal benefit or loss of a one percentage point increase in the portfolio weight for each of the main asset classes, with the average annual impact shown in the bottom row.

The estimates reveal that US long duration fixed income, listed equity REITs, and other real assets (infrastructure, commodities, etc.) would have provided the largest marginal increases to total portfolio annualised average net returns of 4.4 basis points, 3.9 basis points, and 3.8 basis points per year, respectively, for each percentage point increase in allocation over a period that includes the financial crisis.

Other portfolio allocation shifts that would have appreciably increased portfolio net total returns over the period include reduced exposure to US large cap equities (2.9 basis points per year for a one percentage point lower allocation), increased exposure to private equity (2.8 basis points per year for a one percentage point higher allocation), and reduced exposure to TAA/hedge funds (2.1 basis points per year for a one percentage point lower allocation).

In terms of costs, not surprisingly, US fixed income was the lowest at around 17 basis points, with costs increasing up to 10 times that in private equity.

The CEM study also touched on implementation styles with the performance of small public funds highlighting the cost of using fund of funds.

“You can explain why different cohorts of funds do better mostly by looking at asset allocation, which explains about 90 per cent. But with small public funds you can’t do that, their asset allocation is fine and so they should have got average asset performance, but their private equity lagged so much because they use fund of funds – the fund of funds performance lagged even before fees were considered.”

Numerous surveys suggest that Australians are not completely satisfied with superannuation as it exists today.

First, fund members tend to think that they will not have enough to retire and second, that investment plan providers are not necessarily acting in their best interest.

In this context, we asked in a recent study supported by AXA Investment Managers whether the recent rapid development of self-managed superannuation funds (SMSFs) may be related to the level of dissatisfaction with the more mainstream types of pension funds (retail, industry and non-profit) especially amongst the relatively more financially literate and wealthier segment of the population.

The original attractiveness of SMSFs springs from tax incentives available to wealthy individuals and small business owners.

But today it can be argued that if Superannuation funds, especially default option embodied by the MySuper legislation, gave individuals a clearer and more explicit vision of what their pension savings are invested to achieve at the relevant horizon, a number of well-off middle class households would not feel compelled to switch to a “do-it-yourself” version of pension wealth management.

This trend may be seen as symptomatic of the difficulty for an industry to demonstrate the benefits of delegation, leading a group of the more active and engaged members to take matters into their own hands.

 

Developing a reporting standard

 

While more financial education of individual members is necessary, the objective cannot be one of making each individual member a specialist of long-term pension investment.

Moreover, behavioural biases are likely to prevent even the best informed from optimally managing their pension savings. If education can only play a limited role and delegation is necessary, the quality of reporting (i.e. what performance information pension investment solutions providers return to their members) is an essential dimension of the credibility of the pension system as a whole.

Existing reporting requirements leave individuals faced with the problem of computing the contribution of a particular investment option to their pension investment objectives by themselves.

The current debate between industry and regulator about the most useful type of reporting will continue so that innovation and learning can take place, while minimising the burden of compliance costs. Part of this effort can come from superannuation funds themselves and the leadership of major providers is important in this regard.

Regulatory effectiveness and regulatory costs could be optimised by developing a reporting standard of only the most relevant information for fund members. This standard, we argue, can only be developed as part of an iterative dialogue between industry and regulator and pension solutions providers should lead it.

 

Certifying solutions

As well as transparency and the role of reporting, the ability of a pension system to apply sound governance principles is an important criterion against which it may be judged. Australia has long been in line with most OECD guidelines for fund governance. With the exception of self-managed pension funds, super funds are structured around trustee arrangements by which there is complete separation between members and trustees, who have fiduciary obligations and take ultimate responsibility for the fund.

A further industry-led step in the direction of good governance and transparency is the establishment of a governance body to oversee standards for the industry. To support innovation in the pension investment sector in Australia could be the creation of a certification scheme for retirement solutions compliance with generally accepted / best retirement investment and risk management and governance practices.

Such a scheme could be created by the industry to inform the default option selection work of the Fair Work Commission, but also signal service quality to individual investors and employers.

 

Better information and possibility of choice

 

Over the past decade, the super sector has been characterised by both increasing fund sizes, which should lead to scale economies that, in turn, should translate in lower costs for members. At the same time, the presence of a significant number of smaller funds also preserves the possibility of competition.

The premise of recent reforms of superannuation is that with better information and the possibility of choice, competition can play its role to improve the quality of service and the adequacy of pension investment products. However, increasing competition through member choice has not been associated with a dramatic decrease of costs, and the impact of economies of scale has also been muted.

Studies of the relationship between fees and investment performance suggests strongly that super funds with higher costs do not generate a consistently higher investment return.

The literature also finds an inverse relationship between costs and active performance and concludes that higher expenses cannot be justified with claims of more active management aiming, successfully of not, to achieve superior performance, especially in the case of retail funds.

When institutions allowing market participants to compete without restriction on prices or volumes are in place and the expected benefits of competition do not materialise, market mechanisms can be considered to be failing.

The Australian super sector is large and not dominated by any particular provider. Nevertheless, neither choice nor scale appears to have had a marked effect on prices (costs) or the relationship between prices and service quality and adequacy.

 

Adequate default option

The current problem faced by the average super fund member, as well as the regulator, is one of identification of the most desirable investment solution. The possibility of choice and the potential for competition have not yet been sufficient conditions to reveal what the advanced service providers could offer.

With perfect information about what investment solutions can be created to achieve a set of long-term objectives, competition between fund members would work as expected: individual members exercise choice and decide which solutions are best suited to their needs e.g. a real consumption/wealth target (liability) to be achieved at different points in their lifecycle while respecting a certain risk budget.

The difficulty arises from the absence of information (i.e. reporting) and point of reference (i.e. certification) for fund members, who cannot discriminate between providers. In this setting, efficient plan providers have an incentive to mimic inefficient providers (moral hazard), make no costly effort to design advanced investment solutions and provide the same products as inefficient providers. What drives up costs in this case is not the absence of competition, but the tendency for all providers to “pool” and behave like the inefficient type of service provider.

To avoid this pooling equilibrium, market participants can create “sorting devices” or “revelation mechanisms”. In this context, we note that consultants may play an important role in reducing asymmetrical information and that progress in this respect progress should be encouraged.

In their role as gatekeepers, consultants can for example challenge providers to improve reporting along lines that would be consistent with state-of-the-art principles for retirement management, and to redesign products to add value.

In the end, either individual pension fund members, through the regulator, can request bids for a limited number of pre-defined investment solutions (i.e. the regulated default option) in an auction, or pension plan providers can choose to highlight the different solutions that are available through the kind of reporting standard and certification scheme that we discussed above.

 

Designing high-quality default options

Whether auctions or certification is used to determine the most adequate default option in a defined contribution system, the question of what the optimal default fund should be is extensively discussed in the literature and its conclusions tend to be in contrast with the content of existing investment products.

The question with default options is to determine whether or not they should be standardised and, if necessary, regulated, which amounts to asking if on average the combination of a given set of asset allocation strategies and risk management techniques tends to deliver superior outcomes.

Improving the performance of default options in superannuation plans could begin with maximising the benefits of diversification both within and between asset classes, to design better performance portfolios to allow more ambitious targets at different levels of risk tolerance.

Likewise, better liability-hedging portfolios would reduce the risk of dispersion around the objective at the horizon and allow for a higher allocation to the performance portfolio at a given level of risk tolerance.

Going beyond a static liability-driven approach, dynamic allocation between the performance-seeking portfolio and the liability-hedging portfolio can recognise the time-varying nature of asset and liability risks and advantage of the mean-reversion of asset prices over the investors’ long-term horizon.

Without short-term downside protection included in their investment products, fund members are exposed to drastic losses when risky assets plummet simultaneously, as has been the case in the recent financial crisis.

Such insurance can be complex and costly to manage, yet a pension system aiming to rely on individual DC funds should aim to integrate the third pillar of risk management into its default option.

Indeed, the separate management of long-term liability risks (via hedging) and short-term loss aversion (via insurance) allows the investor to optimise risk taking in a dynamic manner over the investment period.

The latest advances in risk and investment management can be used to design a new generation of retirement products that would not only offer better performance and risk control features, but also more risk customisation.

In this context, the definition of advanced lifecycle benchmarks (representing reasonable partitions of the population of members according to investment horizons, short- and long-term risk aversion, etc.) would support the design of adequate default options would support convergence towards better default options within MySuper.

Lifecycle investment benchmarks would promote performance and innovation in this market. Providing information about reference asset allocation, they would be combined with generally accepted investment benchmarks to generate performance figures that could be presented using the standardised reporting framework discussed above, to serve as references for the industry.

 

Obsolete distinction between “pre” and “post” retirement

Australian retirees are likely to spend several decades of their lives in the “post-retirement” stage and the generation of adequate income to fund their desired level of real consumption over such periods should be integrated in the broader question of individuals’ asset and liability management over their entire life.

In effect, the period of retirement has its own stages, characterised by different consumption needs and risks, which would benefit from a lifecycle approach.

Crucially, the retirement of one’s human capital, the end of labour income, is the largest but also the most predictable asset allocation shock in one’s lifetime. We argue that an integrated, whole-lifecycle approach can thus improve wealth outcomes if this predictable shock is taken into account.

The choice of horizon at which labour income ceases could conceivably be managed as a variable in a dynamic lifecycle environment.

The standardisation of reporting and the provision of planning tools to members could allow the impact of financial, career and lifestyle decisions on retirement preparedness to be simulated, and even accommodate the de-regulation of the pensionable age and members’ increasing freedoms to transition out of the labour force.

Given the high expected inflation of the costs of healthcare and long-term care, savings accumulated until retirement risk being insufficient to cover these predictable expenses for a significant share of members.

Neither would a typical life annuity be indexed on healthcare price inflation. It thus follows that a proportion of individuals’ savings should stay optimally exposed to rewarded risk for a much longer period than the official retirement date currently suggests.

Since Australian pensioners are still exposed to investment, inflation or interest rate risk when they retire, and as they face a multi-decade liability/consumption objective with a likely upward slopping profile (a significant proportion of medical costs are incurred in the last years of life), they can still benefit considerably from applying efficient and targeted asset and risk management techniques to their pension savings after they have retired.

The most adequate investment horizon in the super system could thus be a function of individual member’s life expectancy, not the age from which they cease to receive labour income.

Hence, default options under MySuper could be designed to encompass the entire lifecycle of fund members and use several significant investment horizons including the official retirement date (the end of labour income) but also the stochastic horizon implied by life expectancy and longevity risk.

The investment and risk management technology described above remain the proper way to address this problem: optimise diversification benefits through the design of efficient building blocks, dynamically manage a combination of performance seeking and liability hedging portfolios to target long-term liabilities/wealth objectives taking into account not only age and risk preferences, but also changing market conditions, and control risk budgets through insurance mechanisms.

 

Frederic Blanc-Brude, research director, EDHEC Risk Institute—Asia, and Frederic Ducoulombier, director, EDHEC Risk Institute—Asia

G20 financial ministers and central bank governors welcomed the findings of the G20/OECD roundtable on institutional investors and long-term investment last month, which included clear plans to incentivise institutional investors to undertake more long-term investments.

The roundtable, “From solutions to actions: implementing measures to encourage institutional long-term investment financing”, held in Singapore recognised that long-term investment in infrastructure will contribute to global efforts to return to self-sustaining growth, and that institutional investors can help fill the finance gap. The key issue, the roundtable discussed, is the intermediation of available private capital into infrastructure.

Key messages raised by participants during the roundtable included that to incentivise institutional investors to undertake more long-term investments, performance measurements and compensation should be based on longer-term metrics and should not be penalised for short-term market fluctuations.

It was suggested that inflation-linked debt is an attractive asset class, and that introducing instruments that have a clear and explicit link with inflation would contribute to the better matching of assets and liabilities for pension funds.

While it was recognised that as traditional financing sources such as governments and banks become increasingly constrained and institutional investors can fill the financing gap, participation of non-bank private capital in infrastructure financing is hindered by the different interests of the various stakeholders in the project loan market.

To bridge the different needs and risk appetites, it would be efficient if capital could be “right-sighted”, so that commercial banks can finance projects at the construction stage, while institutional investors take over post-construction projects with stable cash flow.

It was also discussed that accounting standards can play a role in enhancing transparency, an essential part of attracting finance, and helping investors make informed decisions about long-term investment.

Some investors view current regulatory and accounting treatments as favouring mark-to-market accounting and low risk liquid assets instead of long-term investments.

Participants included delegates from G20/OECD Taskforce, IIWG delegates, B20 sherpa and representatives, IOs and senior representatives (CEOs/CIO) from institutional investors including the largest SWFs, pension funds and insurers, but also asset managers and banks. Leaders of the  G20 asked their finance ministers at their meeting in St Petersburg in September 2013 to identify approaches to the implementation of the G20/OECD high-level principles on long-term investment financing by the next leaders meeting, which will be in November 2014 in Brisbane, Australia.

 

The Fiduciary Investors Symposium, to be held on campus at Harvard University on October 26-28, will address the barriers investors need to overcome to invest more in long-term investments. Speakers at this session include:

Jane Ambachtsheer, global head of responsible investment, Mercer

Sharan Burrow, general secretary, International Trade Union Confederation (ITUC)

Raffaele Della Croce, lead manager, long-term investment project, OECD

Conor Kehoe, senior partner, McKinsey

Jameela Pedicini, vice president, sustainable investing at Harvard Management Company

Fiona Reynolds, managing director, PRI

Ethiopis Tafara, vice president and general counsel of International Finance Corporation, a member of the World Bank Group

Chair: David Wood, director of the Initiative for Responsible Investment at Harvard University

 

Long-horizon investors can withstand macro-economic volatility and so should tilt towards strategies that are exposed to that, including value, small cap and momentum. Oleg Ruban, vice president in the applied research team at MSCI says this validates factor-investing and factor-based asset allocation for these investors.

 

Appropriate asset allocation requires explicit attention be paid to the different problems, risks, horizons and constraints of different investors. It’s part of the evolution of modern asset allocation which has moved from equal weighting, risk weighting and risk parity through mean-variance and Black-Litterman reverse optimisation through to a more modern dynamic approach that includes looking at tolerance for macro-economic risk.

Unlike a more traditional approach which says most investors should invest in a similar way, outcome-oriented investing says the way to build a portfolio is to blend the components.

“You start with a market portfolio, and tilt away from the market depending on the different horizons and risks of the investors. The tilts can come from factors, pure alpha or skill and from risk hedging,” Ruban, who focuses on portfolio management and risk related research for asset owners and managers, says.

The commonality among institutional investors is they have a long horizon relative to the average investor which alters their risk tolerance.

Ruban expands that to explore the notion of risk versus uncertainty, and differentiates risk as being inherent in a current opportunity set and uncertainty being how that evolves over time.

“This uncertainty relates to macro economic risk and gives us the idea that long-run horizon investors are more tolerant of macro economic shocks because their time horizon means they have an ability to withstand these.”

From an asset allocation point of view, Ruban says this leads to whether there are strategies naturally more exposed to macro-economic fluctuations versus those that hedge those fluctuations.

“Institutional investors should tilt to those more exposed to fluctuations. A lot of traditional quant factors like value and small cap, and some behavioural strategies like momentum have greater sensitivity to fluctuations in economic growth versus the broad market. This leads to the idea long-horizon investors would want to tilt more to these risk premia strategies than short-horizon investors.”

This in turn, validates the idea of factor investing and factor-based asset allocation.

At the end of last year MSCI conducted a global asset owner survey which asked investors about their asset allocation practices and factor investing.

“The majority of participants believe factor investing can capture alpha,” Ruban says. “Ideologically there was an appreciation for factor-based asset allocation but people were still struggling with implementation.”

Ruban says there have been large institutional investors such as the Norwegian Sovereign Wealth Fund, CalPERS and Alaska Permanent Fund adopting factor-based asset allocation and he sees the trend continuing.

“More investors will follow. While there are some operational difficulties in implementation, the trend will continue,” he says.

Identifying the factors to use in asset allocation is part art and part science, he says, noting that organisational considerations play a role.

For example Alaska has identified “companies” as one of its factors and includes in this equities, private equity and corporate debt.

“There is some logic in that there is commonality but there is also some evidence that corporate debt behaves differently to equities at certain times. There is a statistical aspect to factor investing but it is also driven by certain organisational beliefs.”

Ruban says one of the main obstacles to more investors adopting factor-based asset allocation is the need to access the right technology in order to decompose the portfolio and view it through a factor lens.

“The main advantage of doing this is it gives you more detailed insight into what’s driving your portfolio,” he says. “In times of uncertainty factor-based behaviour can be more intuitive than asset class behaviour.”

The MSCI asset owner survey revealed an average asset allocation to listed equities of 38 per cent, but a risk contribution from equities of 92 per cent.

“Equities is more volatile and correlated so the risk contribution is quite high. But this doesn’t tell you much about the underlying structure of these risks, or factors you are exposed to and whether you believe in them or not.”

Factor-based asset allocation allows you to act at a more granular level.

Ruban says there have been some secular trends since the GFC, including investors in South East Asia changing the structure of their portfolios to be more international and investors in Japan adding more risk.

“Acess to factor-exposure analysis informs these decisions.”

Keith Ambachtsheer’s lead article in the Fall 2014 edition of the Rotman International Journal of Pension Management, takes readers through an historical and logical journey that supports the case for long-termism. Importantly he validates this with four high-profile investor case studies which demonstrate that a long-term view benefits society but also the investors, willing to practice it, in the form of higher returns.

In his article, The Case for Long-Termism, Ambachtsheer examines the historical journey of how mankind has shifted from subsistence societies to wealthier, more stable ones, arguing we are not there yet – there is still a higher plateau of civilisation to aspire to.

“Without long-termism we would still be living in the same subsistence societies our forebears did, continually on the edge of starvation,” he says.

“Surely it was our discovery of the wealth-creating logic of shifting from a mindset of day-to-day survival to one that stretched out to next week, next month, next year.. .and eventually out decades and even centuries. It was this shift towards being able to think and invest in ever longer time frames that made possible the eventual transformation of the subsistence societies of long ago to today’s far wealthier, more stable ones.”

The article goes on to address the fact that today’s societies are far more complex than the societies of long ago and because of this complexity, agents dominate in politics, corporations and finance. And in this world long-termism is not the dominant investment paradigm.

The principal/agent asymmetric information problem is not new in finance, with Ambachtsheer pointing out that Adam Smith addressed its existence in the “first opus on capitalism The Wealth of Nations”. But nor is it going away, and as such needs to be addressed.

Recent academic and practitioner writings on the problem, Ambachtsheer says, conclude that agents too often make business and investment decisions that favour their own short-term interests today, and that principals, or the fiduciaries acting on their behalf, must become more pro-active in fostering decisions that focus on longer-term value creation in their investments.

Part of the solution may lie in insisting that the representatives of asset owners become true fiduciaries, legally required to act in the sole best interest of the people to whom they owe a fiduciary duty.

“The resulting message to the governing boards of pension and other long-horizon organisations.. .is that they must stretch out the time horizon in which they frame their duties, as well as recognising the interconnected impact of their decisions on multiple constituents to whom they owe loyalty.”

Ambachtsheer’s call to action is that the “logical implication of these developments is that the individual and collective actions of the world’s leading pension funds are our best hope to transform investing into more functional, wealth-creating processes”.

The second part of his article gives four different and equally persuasive case studies – John Maynard Keynes managing the Kings College endowment, Warren Buffett’s Berkshire Hathaway, MFS Investment Management, and Ontario Teachers Pension Plan.

Each of the case studies is worth a read, so I won’t ruin it by trying to summarise, however the common threads through each are worth highlighting. For one, they see being out of step with the short-term mainstream as not only acceptable but a competitive advantage.

With MFS and OTPP, Ambachtsheer identifies three further common threads:

  1. Autonomy to act: the organisation does not have to compromise its long-term strategies to serve multiple master with short-term mindsets
  2. Governance and management quality: the organisation’s board can ask the right, hard questions, and its senior executives have good answers to them; both groups are committed to creating long-term value for their beneficiaries/clients.
  3. Human capital: attracting and retaining people committed to executing long-term investment strategies is the organisations number one success driver. This means thinking hard about selection processes, using long-term incentive structures, and creating a collaborative culture and working environment.

For those that see the benefits of long-termism, and need a path to fruition, that says it all.

 

For the full article click here

 

 

 

Generic strategies designed to harvest a certain factor premium regularly conflict with other factor premiums. We find that the premiums associated with these strategies tend to shrink, sometimes even to zero, in these periods of factor disagreement. Enhanced factor strategies, on the contrary, are explicitly designed to avoid stocks that are unattractive on other established factors. As a result, we find not only that their added value is higher on average, but also that they continue to deliver when generic factor strategies struggle. Read more about this white paper.