There is no appreciable difference in the returns of asset owners that do pay performance fees and those that don’t, so what’s the point of these fees?

That’s the question to ask, based on a research paper from Maastricht University and De Nederlasche Bank, in the Netherlands. The researchers looked at 218 Dutch pension funds between 2012 and 2015. They found no difference between the investment performance of those that paid performance fees and those that didn’t.

The paper, Does it pay to pay performance fees? Empirical evidence from Dutch pension funds, claims to be the first that examines the relationship between performance fees and the net investment performance of a pension fund and, in doing so, shows whether it pays for pension funds to pay performance fees at all.

The research looks at total returns, excess return and performance fees for the total portfolio and six major asset classes, and finds that those investors that paid performance fees do not produce higher, or lower, returns than those that don’t.

(There is one minor exception. Pension funds that pay performance fees for hedge funds report a net excess return 3 basis points higher, on average, but that disappears when the authors correct for risk.)

In 2015, the funds, whose combined assets are €985 billion ($1.1 billion), paid €1.5 billion ($1.7 billion) in performance fees, across a variety of asset classes but most commonly in private equity and hedge funds. These fees represented a 36 per cent increase from 2012, when the funds paid €1.1 billion ($1.25 billion) in performance fees.

Paying performance fees could be “economically rational”, the paper states, if they enable pension funds to enhance their overall net performance by recovering these costs with superior returns or higher diversification benefits. The findings show this isn’t necessarily what happens.

The authors observed that “performance fees primarily relate to gross excess returns for equities and hedge funds where investors pay 2.1 basis points and 30.5 basis points, respectively, for every 100 basis points of gross excess return”.

The 218 Dutch funds’ total return over the period studied was 9.72 per cent, with a 0.11 per cent excess return over the benchmark, after costs. For private equity, the total return was 11.54 per cent, with an excess return of 1.4 per cent, and for hedge funds it was 2.64 per cent, with an excess return of 0.12 per cent.

The paper further calls into question the relevance of performance fees by showing that pension fund size and specialisation are “economically more important for net returns than paying performance fees”.

Specialisation is positively correlated with net returns for private equity and hedge funds, the paper demonstrates; pension funds with an allocation to these asset classes 1 percentage point higher than their peers’ reported a 1.31 and 1.37 basis point, respectively, higher net total return.

As for size, the research also supports the notion that bigger is better. It finds empirical evidence that larger and more specialised pension funds pay less in performance fees for a given level of excess return in alternative asset classes, possibly as a result of their better negotiating power. Larger funds also apply performance fees more often, and pay significantly less for the same performance by asset managers.

The authors did recognise that, in theory, performance fees could be a valuable tool to minimise the principal-agent conflict between pension funds and asset managers.

The research paper cites academic studies outlining the benefits of performance fees, including aligning the managers’ incentives with the pension fund’s interests, and linking the manager’s reward to performance. Ideally, this should increase the effort from the manager and translate into higher investment returns.

The drawbacks are that the manager receives the same performance fee whether the performance comes from skill or luck, and the payments create a skewed incentive structure, as the manager benefits from excess returns but does not suffer from losses.

“We find no statistical evidence that paying fees for most asset classes adds or subtracts value,” the authors conclude.

The paper is by Dirk Broeders, Arco van Oord, and David Rijsbergen.

Migros-Pensionskasse (MPK) the CHF21 billion ($21.9 billion) pension fund for Switzerland’s largest retailer, Migros, distinguishes itself from peers with its large domestic real-estate allocation. It also remains an exception in Switzerland’s Pensionskassen landscape by still running a defined benefit scheme that is open to new members.

Like many other Swiss funds, MPK applies a core-satellite approach to managing assets. Typically, this strategy means more than two-thirds of assets lie in a core, primarily passive, portfolio, with the rest in satellite vehicles aiming for either higher returns through active investments, or a lower risk profile, in comparison with the core.

At MPK, assets are split between a 40 per cent allocation to fixed income, a 30 per cent equity allocation and a 30 per cent real-estate allocation, with about 5 per cent of each allocation in satellite investments. Internal portfolio management oversees the core allocation, and external managers oversee the satellite.

“In the satellite, externally managed allocations, we aim to add additional asset classes we don’t have in the core allocation,” says Adrian Ryser, head of asset management and chief investment officer at the Zurich-based fund. “So assets here are different; core and satellite comprise different universes. Satellite allocations could be more detailed, or smaller, allocations that improve the risk-return profile and diversification effect. Examples are small-cap equity, high yield, senior loans, and foreign real estate.

“Over the last few years, we have increased the portfolio position in the satellite allocation, but not across all asset classes. For example, our small-cap equity allocation is the same; however, because of the negative yield curve in our core allocation to Swiss Government bonds, in the satellite fixed income allocation we now invest more in sub investment-grade credit and high yield.

“We also recently introduced senior loans. This is a long-only allocation. It brings a credit spread at the top of Libor [the London interbank offered rate], it is short duration and it is below investment grade. It is a significantly higher yield than Swiss fixed income investments, even after hedging the currency risk.”

Infrastructure joined the satellite asset mix about three years ago via a new infrastructure platform; this was done in conjunction with five other Swiss funds to create an economy of scale.

“We invest only in infrastructure funds; we don’t do any direct investments in infrastructure. We are still building up the allocation,” Ryser explains. “We add a little every year, and at the moment the allocation accounts for 1.5 per cent of our assets, so it is still small. We prefer open-end funds and have one or two secondary market investments. It’s a big market and we are making careful and slow progress.”

MPK’s large allocation to real estate, of which 25 per cent, or CHF5.2 billion ($5.3 billion), lies in Swiss properties, helps give stable returns. The Swiss allocation is managed in-house each step of the process, from the original acquisition through to management. Investments are made in residential and commercial properties, developments and undeveloped land. Outside Switzerland, MPK invests through funds.

There are six internal portfolio managers. A further internal team of four runs the manager selection, monitoring the mandates for high yield, senior loans and small-cap equity. All these mandates are long-only and have market benchmarks.

“We prefer long-term, stable manager relationships and review results every three years,” Ryser says. “We don’t give up after 12 months. Most of our existing managers have been with us for over five years.”

Challenges and adaptations

MPK has successfully adapted to the challenge of changing demographics and decreases in investment returns and fixed contributions by raising the retirement age. Currently at 64, it has been progressively raised from 62 in 2005.

“Changing the retirement age is an efficient solution to financial stability because it increases the period people pay contributions and decreases the time they are paid pensions,” MPK director Christoph Ryter says.

Such flexibility isn’t as easily applied to investment strategy, where regulation informs Swiss funds’ asset allocations, especially the 15 per cent cap on alternative investments. It leaves some Swiss funds feeling over-regulated, although Ryter says there is room to manoeuvre.

“The regulator set up some limits for investments in different asset classes,” he says. “Investors can either comply with these limits or explain why they want to have a higher allocation. It is, therefore, not a hard limit in most cases. It is more based around the idea of ‘comply or explain’.”

Other investment regulations include a 50 per cent cap on equities and a 30 per cent cap on real estate.

For 2017, the fund has no plans to change its asset allocation. MPK shed hedge funds following in- depth analysis of fees and expenses in the wake of the Swiss Government introducing full transparency on asset management costs and mandatory total-expense-ratio reports for all Pensionskassen investments.

It meant being “convinced” of a high net return to “explain high management fees” and this was no longer “justifiable for hedge funds”, Ryser says. MPK’s mix of asset classes can go through difficult times without hedge funds, he concludes.

It may seem like a hidden truth but asset owners are in competition with one another.

They are in competition for the best alpha ideas, the best manager products and the best research – all with the aim of improving risk-return trade-offs to increase the likelihood of meeting their liabilities.

As a result, many asset owners find it difficult to collaborate, even on initiatives that may prove mutually beneficial.

At the Thinking Ahead Institute’s recent Sydney roundtable, asset owners highlighted the top three barriers to successful collaboration with peers: difficulties being transparent; lack of time and resources; and difficulties aligning interests. At the same time, however, attendees agreed on the value to funds of collaborating on industry structure, regulation, and a universal agenda for alignment of owners’ interests.

The word ‘co-opetition’ was described in Adam Brandenburger and Barry Nalebuff’s 1997 book of that name and refers to the ability of competing businesses to co-operate for mutually beneficial outcomes. The authors took insights directly from game theory, which also applies to the myriad investment decisions pension fund boards need to make in efforts to fulfil the requirements of several potentially misaligned stakeholders.

The pursuit of rational but non-collaborative strategies generally produces poorer outcomes (prisoners’ dilemma) whereas better payoffs can often be produced through effective methods of collaboration or government influence.

Proof in the research

There are numerous academic articles and research projects that prove this assertion. Here are three examples:

In the 2009 article titled “Improving pension management and delivery: an (im)modest and likely (un)popular proposal”, Ron Bird and Jack Gray argue that excessive competition among retirement savings providers has undermined their key objective of maximising net returns to members in three main ways, namely:

  • Inefficient pricing: The race to outperform one another (largely but not exclusively through listed equities), often forces asset managers to rely heavily on momentum and other non-information-based strategies. This causes mispricing away from fundamental values, leading to sub-optimal capital allocation, which lowers long-term returns.
  • Agency costs: The growth of intermediaries and other agents has led to increased complexity and uncertainty, and substantial increases in costs. And given that active management is, in effect, a negative-sum game after fees, aggregate returns are reduced.
  1. Excessive choice: Bird and Gray refer to Joshua Fear and Geraldine Pace’s 2009 article “Australia’s ‘choice of fund’ legislation: success or failure?” in arguing that despite the plethora of investment strategies available, a large portion of Australian institutional retirement savings funds are essentially identical, with little investment choice exercised. Therefore, members bear the direct and indirect costs of competition-induced excessive choice. Additionally, the average fund size was seen to be well below that needed to benefit from economies of scale including lower fees. (There is a trade-off involved here, between economies of scale enjoyed by larger funds, and the ability of smaller funds to express conviction and flexibly alter their positions. Better outcomes would have been achieved with better default design for workers who ‘choose not to choose’.

Bird and Gray suggest that these leakages can be plugged by rationalising the retirement savings industry and its agents and through greater co-operation (such as through joint research efforts) while retaining the genuine benefits of productive competition.

In universal owners’ interest

Willis Towers Watson’s Roger Urwin, in his 2011 paper “Pension funds as universal owners: opportunity beckons and leadership calls”, argues that it’s in the interest of universal owners, which through their portfolios own a slice of the whole economy and the market, to collaborate with other asset owners to ensure the health of the investment ecosystem as a whole. In a nutshell, while universal owners adapt their actions to try to directly enhance the value of their portfolios, they indirectly help the whole economy secure a more prosperous and sustainable future.

Finally, in its survey this year of 15 best-practice asset owners carried out on behalf of the Future Fund, Willis Towers Watson observed the following trends:

  1. Some participants have developed more strategic partnerships and have seen benefits in sharing information in areas such as operations, human resources and technology. All participants agreed that peer collaboration had proved valuable to some extent, but noted that further work needed to be done to crystallise these opportunities.
  2. The participants were cognisant of their external profiles, and greater success was aligned with where their profiles had been deliberately and carefully cultivated, often through proactive and highly visible strategies. Willis Towers Watson noted that there were growing expectations of leading asset owners to exercise positive influences in pursuit of their financial goals in co-operation with others, and to consider environmental, social and governance issues through their ownership interests. Peer relationships and collaborations are particularly helpful in this regard.
  3. Many participants outlined explicit goals to enhance collaboration, whilst some described instances of co-investment success, although most saw this as more limited in reality than they had initially hoped. Several are now looking to be more discerning and targeted in their collaboration activities, aiming to make one or two relationships much richer and deeper. The limits to senior time and bandwidth are clear constraints.

Effective collaboration without sacrificing the genuine benefits of competition requires clearly defined objectives and goals. It also requires a mindset shift among asset owners that recognises these strategic partnerships have the potential to be mutually beneficial.

Marisa Hall is senior investment consultant in the Thinking Ahead Group, an independent research team at Willis Towers Watson and executive to the Thinking Ahead Institute.

Investors should be examining the drivers of active return from managers – in other words monitoring exactly what they are paying for. And now with the help of a study by MSCI, investors can compare the extent to which factors and stock selection are driving active returns, making it easier for them to evaluate managers’ skill.

A new paper by MSCI, Anatomy of active portfolios, looks at the attribution of active performance – that is, above the benchmark or beta performance – and finds that factors contribute more than half of active return.

The authors examined five-year active performance data from 1315 diversified active global and international mutual funds over 13 years from September 2003 to December 2016. They found that for the top-performing funds, 55 per cent of active return was due to factors and 45 per cent was from stock selection.

“In short, we found that, based on absolute contribution fractions, 16 style factors, collectively, had the largest impact on a typical fund’s performance,” authors Leon Roisenberg, Raman Aylur Subramanian and George Bonne, from MSCI, state.

Given this finding, that more than 50 per cent of active return is due to factors, rather than skill, investors could reconsider what they are paying for and how much it’s worth.

The study found that most funds had a significant exposure to factor groups, which it defines as including country, industry, style and currency. And among these factor groups, style had the largest impact on active performance, with 34 per cent of factor returns, on average, attributed to style.

Among the individual styles, size and growth had the biggest impact, with 63 per cent of funds benefiting from a positive contribution from both of those. This was followed by momentum (which gave 60 per cent of funds a positive contribution), quality (53 per cent) and value (46 per cent).

MSCI’s in-depth analysis of the contribution of style factors to active performance also raises questions about the ability of managers to be true to label. For example, the study found that, for value managers, the active return contributions from volatility, price momentum and profitability factors each accounted for more than the value factor, which contributed 15 per cent of the total style contribution, compared with 19 per cent, 18 per cent and 17 per cent, respectively, from the other factors.

The paper reports that while active funds had a significant exposure to their target, or stated factors, such as value, most funds also had a significant exposure to other factors. This suggests the need for a benchmark that is consistent with a fund’s stated objectives, which would help highlight the drivers of performance.

The paper shows that over the entire period measured, the average trailing five-year active performance was 73 basis points a year, before transaction costs and fees. It doesn’t consider the returns after costs and fees.

Outperformance tougher than thought

Another recent academic study, Why indexing works, shows that outperforming the benchmark is more challenging for active managers than previously thought, regardless of fees.

This paper’s authors – from Chicago Booth School of Business, Oxford University, and law firm, Bartlit Beck Herman Palenchar & Scott – state that active equity managers often find it difficult to keep up with the benchmark, let alone outperform it.

They say previous research has focused primarily on fees as the cost of active management. Instead, they present a model that suggests the much higher cost of active management may be the inherently high chance of underperformance that comes with attempts to select stocks.

“One reason indexing works so well,” the paper states, “is that, on average it seems, active management faces a higher hurdle than previously recognised. Missing (or underweighting) the securities that significantly outperform other securities is a strong headwind for an active manager to overcome. This view of the active-passive problem helps us understand the mystery of how so many smart people, with enormous financial and informational resources, systematically do such a poor job investing money.

“To the extent that those allocating assets have assumed that the only cost of active investing above indexing is the cost of the active manager in fees, that assumption should be revisited. Active managers do not start out [on] an even playing field with passive investing…Put another way, passive investing may have a larger head start on active investing than previously believed. When creating a portfolio combining passive and active strategies, independently of past performance, return estimation should be adjusted for the inherent statistical disadvantage of the active manager combined with the higher fees.”

 

Investors should expect to continue to earn an equity risk premium comparable to a 3.8 per cent historical long-term average, based on new analysis that quantifies the effect of share buybacks – and the decline of dividends – as a major driver of equity returns.

An article by the head of capital markets and asset allocation at Morningstar Investment Management, Philip Straehl, and Professor in the Practice Emeritus of Finance at the Yale School of Management, Roger Ibbotson, highlights the importance of including share buybacks in calculating a “payout yield” for equities.

In the article, “The Long-run Drivers of Stock Returns: Total payouts and the real economy”, published in the Financial Analysts Journal, Straehl and Ibbotson argue the importance of share buybacks is often overlooked but their impact on equity returns can no longer be ignored.

In the July 2017 edition of The Ambachtsheer Letter, KPA Advisory Services’ Keith Ambachtsheer says the good news for investors is that based on what he believes is a reasonable set of assumptions underpinning equity return forecasts by Straehl and Ibbotson, “forward-looking equity risk premium today looks a lot like its historical counterpart”.

Danger of underestimating future returns

Getting an accurate and meaningful bead on the equity risk premium remains of prime concern for long-term investors seeking to allocate capital efficiently across different asset classes and varying periods of time. Opinions remain divided on what the figure is, and even on a suitable definition.

The traditional and widespread approach to forecasting returns using dividend yield and expected growth in dividends, while ignoring buybacks, will cause future returns to be underestimated, Ambachtsheer says.

As the importance of share buybacks becomes better understood and is factored into calculations, the effect of potential underestimation becomes clearer.

The Straehl and Ibbotson FAJ article states “the cash flows that corporations supply are the ultimate drivers of stock returns”, and since the early 1980s, buybacks have become an increasingly important part of that supply, exceeding dividends in eight of the last 10 calendar years.

What matters, the article states, is the combination of dividends and buybacks, and buybacks have “a fundamentally different impact on the return generation process than dividends”.

Dividends result in cash in the investor’s pocket, whereas buybacks affect growth by increasing the equity price return for buy-and-hold investors, whose stake in a company increases. Not taking this price growth into account can be a factor in underestimating future returns.

“The advent of share buybacks as the dominant form of payout has created a need for a new set of return models that are independent of the payout method,” the FAJ article states.

Buybacks’ growing importance

In analysing the return from US stocks for the period 1871 to 2014, Straehl and Ibbotson based their calculations on a dividend yield of 4.5 per cent a year, buyback growth (BBgrowth) of 0.8 per cent a year and dividend growth of 1.5 per cent a year – giving a real return of 6.8 per cent a year. However, looking ahead, they base their estimates on a current dividend yield of just 1.9 per cent and buyback growth of 1.7 per cent, the average of the last 10 years.

Ambachtsheer says: “A key message in the [Straehl and Ibbotson] article is that, based on the assumed continuation of the average share buyback experience of the last 10 years, the BBGrowth factor is almost as important as the current dividend yield in calculating the expected long-term return of a broadly based equity portfolio today.”

The Ambachtsheer Letter states that the historical 6.8 per cent a year long-term real return from US equities and the 3 per cent real return from US Treasury bonds implies a historical equity risk premium of 3.8 per cent. However, today’s real bond return is about 0.8 per cent, implying an equity risk premium of 4.4 per cent – based on the average of Straehl and Ibbotson’s forecast of a real return from US equities of 5.1 per cent and what Ambachtsheer says is the projected 5.3 per cent from the S&P 500 Index.

Ambachtsheer says Straehl and Ibbotson’s payout yield calculations support his own analysis that with bond yields at their current levels “the prospects for earning a positive equity risk premium…continue to be good at today’s stock price levels”.

He says bond yields could rise to 1.4 per cent before they begin to affect the 3.8 per cent historical equity risk premium. But the factors depressing bond yields are secular in nature and “unlikely to be reversed any time soon”, he argues, and investors will continue to earn an equity risk premium at or near historical levels.

In 2011, the Research Foundation of the CFA Institute published a collection of papers, Rethinking the Equity Risk Premium, which contained the results of 19 separate studies calculating the equity risk premium in a range between zero and 7 per cent.

“The papers collected in this volume share a general emphasis on supply factors and models for the historical excess return as well as the forward-looking equity risk premium,” the CFA Institute publication states. “After 10 years of low and highly volatile equity returns, there is little consensus about the stability of the [equity risk premium] over changing regimes and time horizons. Interestingly, the group appears to be in agreement more on the actual size of the ERP over the next few years (most agree that it is in the 4 per cent range) than on its stability.”

The causal link between good governance and investment performance has been an elusive domain for financial services academics. Now, in Switzerland, some progress.

A study of 139 Swiss occupational pension plans shows, empirically, governance is positively related to excess returns, benchmark outperformance and Sharpe ratios.

The paper, Is Governance Related to Investment Performance and Asset Allocation? Empirical evidence from Swiss pension funds, investigates the relationship between governance, investment performance and asset allocation at pension funds in Switzerland.

Study authors Manuel Ammann and Christian Ehmann, from the University of St. Gallen, find that fund governance is positively related to investment performance, but only marginally related to funds’ asset choices.

The paper doesn’t give any indication of the direction of causality, but it does show that good governance pertaining to target-setting, defining investment strategy, and risk-management design is positively related to both excess and risk-adjusted net returns.

The academics developed a metric comprising six different governance areas: attributes of organisational design, management incentives, target-setting and investment strategy, investment processes, risk management, and managerial transparency.

The study finds that pension funds in the top governance quartile outperform those in the bottom quartile by about 1 per cent, related to average excess returns and benchmark deviation. It also shows that a clear, written statement specifying organisational goals and strategic targets is positively related to passive benchmark outperformance.

Asset allocation decisions are not related to governance, the study finds, but rather to institutional factors such as size, legal form and the ratio of active managers to pensioners.

The full report can be accessed here:

Is governance related to investment performance and asset allocation? Empirical evidence from Swiss pension funds