Keeping it simple - back to the future for asset allocation strategies
20 years of progress
When I joined this industry nearly 20 years ago superannuation was largely a cottage industry. There were very few professionally run funds, investment strategy and performance were typically outsourced to one or a small number of balanced managers and consultants were more focussed on making benefit changes and determining contribution and solvency rates than anything to do with manager selection or investment strategy. There were probably (well, certainly) less than 50 real investment managers in the country and the term “boutique” applied to small clothing stores in Mosman, Double Bay or Toorak Road, South Yarra. The number of people directly involved in the industry on a full time basis numbered in the hundreds.
Roll the clock forward and today we have an industry with assets in the trillions (i.e., about the size of our GDP), employees in the tens of thousands, investment managers in the hundreds in Australian equities alone, our own cabinet minister, superannuation funds advertising on TV and sponsoring football teams and fund performance running for weeks as the top news story. Benefit design and contribution and solvency testing have long since disappeared (save a few SMSF actuarial certificates) and consulting to superannuation funds in Australia is now made up almost entirely of “asset” consultants – investment strategy, manager selection, asset allocation and even running the money themselves.
While an industry that has grown to be a significant force in Australian (and even global) capital markets needs a high degree of sophistication, it is appropriate to step back from time to time and ask “is it worth it”? Has the heightened level of interest, increasing complexity and cost delivered a better outcome than would have been achieved without it? Has the pendulum of progress swung too far and somewhere we’ve missed the wood for the trees?
While the industry can rationalise the poor results of the last 2 years, our clients and “constituents” may not be so forgiving.
The Prime Minister’s latest (25th July 2009) essay states “To preserve their standard of living, we need to encourage savings from an early age, and to improve the efficiency and investment returns of our retirement-incomes system.”
That doesn’t sound like an A+ on the report card to me.
The Australian superannuation industry is generally looked upon as one of the more successful attempts at managing retirement incomes in a first world country – and rightly so in terms of its ability to force a level of self-funding of retirement and the positive impact that it has had on Australian capital markets. However when the PM thinks it’s necessary to tell the industry to “improve its efficiency and investment returns” then quite clearly things are not going as well as we would like.
Where did we go wrong?
Over the years I have become a firm subscriber to the KISS school of investing. This is not some vague reference to a quantitative trading scheme developed by Paul Stanley and Gene Simmons et al, but rather the notion that investment is not a complicated game. The mantra of “Keep it Simple Stupid” applies probably as much too successful investing than any other paid endeavour. Investing is not complicated. That’s not to say it isn’t hard, and doesn’t require skill, but it is not complicated. Unfortunately our industry – and those that surround it - earns a good deal from making it just so.
As our industry has grown and “matured” the level of complexity within the industry has grown enormously. To my mind, the current raging public debate about fees on retail funds versus not for profit is a furphy in the grand scheme of things – ultimately that’s about advice and who pays. If you really want to reduce costs in the industry at a meaningful level then reduce complexity and the number of participants (particularly high fee ones). That will have a far greater impact on the 70% of industry assets that don’t pay for advice versus the 30% who do.
The following chart shows the aggregate costs of investing in different asset classes globally for institutions. You probably don’t need to ask about where has been the greatest “growth” over the last decade….or where boards and investment committees have spent cumulatively most of their time and energy.
Source: CEM Benchmarking Inc, 2006, The World’s Lowest Cost Funds
I would also argue that the costs of hedge funds and private equity in the above are probably somewhat understated. The great drive to complexity in our industry comes at significant expense.
More broadly, prima facia examples of the complexity in our industry include the whole “structured debt” class of securities (take stuff you can buy already, repackage it, charge a fee and sell it), hedge funds and especially hedge fund of funds (charging a lot for what in aggregate hasn’t looked any better than a conservative balanced fund), and private equity (charging you a multiple to leverage what you could already buy – and probably already owned).
Outside of these more obvious examples the complexity theme has worked its way into fundamentally quite simple structures. The whole “segregation and specialisation” of the traditional asset classes and the move away from simple balanced structures ultimately represents a move to greater complexity. Even within asset classes, we would contend that the growth in aggregate portfolio sizes which should have on the whole reduced fees has been more than offset by an increase in the complexity of manager structures.
Analysis of long term average investment management costs incurred by larger funds around the world show that Australian funds are amongst the most expensive globally in terms of their investment management costs.
Source: CEM Benchmarking Inc, 2006, The World’s Lowest Cost Funds
The prima facia reason for this is explained as the higher proportion of more expensive asset classes – principally hedge funds, private equity etc. – that is, greater complexity.
By way of example let’s take the relatively simple concept of constructing an equity portfolio. We have witnessed several fads and foibles over the years in the most efficient way to “build” a portfolio of equities/managers. A non-exhaustive list would include:
- Style diversification (value vs growth – is that sensible in a market as narrow as Australia?).
- Small is beautiful (small, new boutiques – yep there are some good ones, but being small isn’t what makes them good).
- Core/satellite (blending low tracking error “core” managers with “high alpha” managers – see “What Price Complexity – Rethinking Australian Equity Multi-manager Portfolios”, Schroders March 2009).
- Emulation (purportedly to reduce trading/implementation/tax costs).
- Long/short, 130/30 etc and the associated significantly higher fees.
Many of these have sounded quite simple and rational and hence seen broad support, yet closer examination can often reveal unnecessary complexity and illogical conclusions.
It is probably fair to say that there are many stakeholders who benefit from complexity – with the principle exception of the end beneficiary.
2009 – A turning point?
With the results for the 2008/9 financial year now out it would be hard to conclude otherwise that the last 2 years have not been a good time for superannuation. Market surveys point to the average fund returning circa -13% in 2008/9, following a -7% in 2007/8. As such the emphasis from most funds in reporting to members is likely to be a focus on the long term “value” of superannuation and their long term performance.
But how long is long term? What time horizon do we need to have to be confident that as an industry we’ve done a good job? With 5-7 year average returns less than cash – clearly you need a much longer time frame than what many funds encapsulate in their investment objectives (typically to outperform CPI by 4% to 5% over rolling 5-7 year periods).
As the table below shows, you currently need to go out to periods of 15 years or longer before the average fund has performed even close to its long term objectives – and that is hardly a stellar outcome.
Long Term Superannuation Fund Performance Relative to Objectives
|Period to 30 June 2009||ChantWest Median Return (%p.a.)||CPI (%p.a.)||Margin over CPI (%)|
Source: ChantWest, Schroders
Even more enlightening from a complexity perspective, the 15 year performance of the average Australian pooled fund (ie. old school balanced fund) is, 6.4%p.a. over the period to 30 June 2009 (source: Mercer Pooled Fund Survey) – exactly the same as the average fund today (albeit you could buy the average balanced fund at substantially less than an average fee of 44bps in larger sizes). Similarly, while a few months out of date, the medium term data for implemented consultants vs single manager balanced funds set out below shows little, if any, in the way of difference between the averages.
Gross returns for periods to 31 March 2009
|3 months||1 year||2 years||3 year||5 years||Standard deviation 3 years||Standard deviation 5 years|
|Single manager median||-4.6||-21.2||-13.4||-5.7||3.7||10.5||9.3|
Source: Rainmaker/CPG Implemented Consulting Survey, Multi-year periods in %p.a.
What the above highlights is that on a post fee-post tax basis, the move towards greater complexity in investment manager structures has had little positive impact, if any, in performance terms on aggregate fund results. More importantly for us, it has probably meant that too much attention has been devoted to debating different implementation structures and not nearly enough to the single most critical issue.
The AFR ran a large-type call-out in its wrap up on superannuation fund performance for the year on 23 July 2009:
“Asset managers will find it increasingly difficult to justify their hefty salaries”.
While some of us would argue that you don’t need a year of -13% to make that statement (and in most cases make it quite accurately), the reality is that for all the huff and puff made by the fund management industry, the biggest determinant of returns is completely and utterly outside of most of their control. Over the last 15 years, the principal driver of portfolio returns has been increasingly under the explicit control of the board or the investment committee (in many cases via asset consultants).
Numerous studies over the last 20 years (and longer) have pointed to the importance of asset allocation as the key determinant of total fund performance. Whether the number is 90%, 95% or over 100% (as some passive players may argue), the key point is that it is big. For all the noise made by active investment managers, given current very specialised investment structures, at best over time they are unlikely to influence the aggregate return of a typical Australian superannuation fund post fees and tax by anything more than a small margin – that’s not to say that it’s not important, but it wont make a major difference relative to the big picture call of asset allocation.
We have written for many years on the importance of asset allocation as a determinant of total portfolio returns and the heavy bias of most funds in Australia to equity and growth type assets irrespective of the risk return environment (for our most recent, see the paper from January 2009, “It’s about risk, not return”). Key points we raised then and before are that fundamentally the risk of a superannuation fund (including their Trustee boards and their investment committees) comes back to two broad principals:
- Can we pay our investors back their money when they want it; and
- Will we have added any “real” value to that money while we had it?
The implications for our industry of one or both of these not being possible for even a short period (a few weeks in the case of the first and say 2-3 years in the case of the second) are quite severe. We are seeing this latter issue in the press today and there are a few funds with real issues on the first.
At the total fund level risk is generally described and reflected in a funds’ Strategic Asset Allocation (SAA). This SAA is often fixed (or largely fixed) for a long period of time. Our industry largely ascribes the riskiness of a fund as represented by its SAA – in particular its allocation to “equity” type growth assets and “bond” like defensive assets. The entire “balanced”, “growth”, “conservative” framework on which many investment options are classified is predicated on fixed exposures to different asset types.
However, in our view, while the SAA of a fund may be static, the risk of that SAA is anything but static. In periods when risk premia change, the required outcome – especially in the case of a portfolio aiming to deliver positive real returns through time – is unlikely to have moved to any great degree. This implies that under a typical SAA approach the likelihood of delivering the real return outcome would shift materially as risk premia change.
On this basis, it no longer sounds intuitively sensible to describe a fund’s risk and return expectations in the context of its SAA. In our view, if a change in risk premia on an asset means that it is significantly more likely to achieve the required return outcome relative to other assets then its weight in the portfolio should change materially.
As such, the starting point for portfolio construction should be significantly more dynamic than that in traditional portfolios, and is particularly sensitive to the risk premia embedded in asset exposures at a particular point in time. As these risk premia change, the asset allocation should change.
Unfortunately, this approach flies in the face of the generally accepted advice most funds have received over the years, and requires a completely different “structure” to the way most funds are currently structured. The notion of broadly fixed asset allocations to range of different “asset classes” managed by underlying asset class specialists doesn’t make sense when we have the potential for a significant change in asset allocation (or it is certainly more difficult to manage). In particular, funds still spend significantly more effort on selecting the specialist to manage an asset class than they do on whether it is the right asset class to own (or keep owning).
Of even more importance (for trustees and investment committees) is how to make the decisions on asset allocation in this framework – the process for which has been gradually taken back from asset managers over the last 20 years and placed firmly in the hands of the trustees and their advisors.
As we’ve noted previously, there have been periods (generally corresponding with a change in the prevailing macro-economic environment) where the strategic return has been substantially negative or very low for a long period of time. So, getting it wrong matters, a lot.
If the last 2 years has taught us anything it should be the overall importance of getting asset allocation right.
The responses to this have generally been that market returns are difficult to predict with any degree of accuracy and consequently a fixed asset allocation allows fund sponsors to concentrate on where they believe they have a greater propensity to make a difference – alpha.
However, for an individual member the consequences of getting asset allocation wrong on the downside are quite significant – probably more so than most would think. If we consider the example of a typical “average” member contributing 9% of salary from age 20, in a fund that returns “on average” 7%p.a. through time, the impact on their final lump sum balance at age 65 of a 15% market fall at different times in their career is set out below:
Impact of a market fall on final balance
|Years to retirement||Change in final benefit|
Source: Schroders, Assumes 9% contribution less 15% tax from age 20 to age 65, salary based on AWOTE with salary inflation of 3% p.a., initial salary 30% less than AWOTE and linear change to age 65 where it is 34% above AWOTE. Returns in all years of 6.5% p.a. except “shock” year of -15%.
Clearly the impact on members at different points in their career from a significant market fall can be quite significant – even for those with a long way to go to retirement.
Consequently, managing asset allocation from the point of view of avoiding large negative drawdowns is in our view as important as managing purely for long term returns. In an industry where we have members constantly entering and leaving the system the path dependency of returns is as important as the “long term” average return.
How predictable are returns?
As noted above, one of the principal defences in the debate about fixed asset allocation is that asset class returns are not predictable. Like any asset (and let’s remember an “asset class” is just a collection of assets with supposedly similar characteristics) the predictability of returns is partly a function of time horizon and partly a function of the price paid for the asset. To think that somehow future return is independent of price paid (which is effectively what most Strategic Asset Allocation processes assume) sounds strange to me.
If we take Australian equities as a good example, the chart below shows, from data over the last 40 years, the range of returns over the subsequent 10 years (and the average return) based on the starting P/E ratios.
Source: MSCI, Schroders
While there is some variability in the return outcomes (as you would expect) there is also a pretty strong correlation between the price you pay and the subsequent returns – sounds obvious, but under most current investment structures not something typically taken into account.
The pattern in international equities points to broadly the same outcome (again unsurprisingly).
Source: MSCI, Schroders
Clearly the above points to the fact that over sufficiently long time horizons, return forecasting, even in a volatile asset class such as equities, is not a random exercise. To be fair, as a famous economist once pointed out - we don’t live in the long term - so the skill is in translating more predictable 10+ year returns into progressively less predictable shorter term time-frames.
In sovereign bond and credit markets the exercise is even easier – the market is telling you a large part of what the return is going to be for a given time horizon – e.g. look at our current (29 July 2009) expected returns on credit vs high yield.
The above suggests to me that the current flurry of activity in high yield is too late – always a danger when you implement specialist structures with slow decision making. Consequently, if we think about allocating to asset classes in the same way we think about allocating to individual stocks or bonds – that essentially we are buying a stream of cash flows – then the only things that matter are:
- What price we pay for those cashflows
- The variability of those cashflows and what drives that variability
- How the above relate to our performance objective
Given the consequences (and potential benefits) of getting the asset allocation decision wrong (or right) it is hard to understand why we would want to put in place structures that make decision making on asset allocation so difficult (except for the desire for greater complexity).
Quite frankly, the more asset classes are segregated out into sub-components and handed out to specialist management the more difficult asset allocation becomes – both the making the decision and the implementation of the decision.
We have specifically observed in the last year or so the desire to carve fixed income mandates into ever increasing levels of specialisation. In a market where the risk premium is explicitly observable but changing rapidly, where indices are completely counter-intuitive (larger weights to those players with more debt) and massive changes likely in the construct of those indices in coming years, it strikes me that absolutely the wrong way to implement a fixed income exposure is through a very sub-asset class specific specialised structure.
Understanding that fixed income is in the portfolio as a diversifier vis a vis the massive growth/equity bias in the portfolio, the focus should be in ensuring that its role as a diversifier is met. That was the mistake of 2007/8, carving up fixed income portfolios too finely risks repeating it all over again.
In terms of developing solutions, we need a mindset change. It is clear from many funds objectives (and certainly the perception of our end beneficiaries which is what really matters) that the role of most funds is to return an appropriate margin over inflation over time. That is quite clearly the point that is being made by the PM and most journalists over the last 12 months.
The “market” for our services will ultimately not care what the asset allocation is, they care rightly about the result achieved. Shouldn’t we manage at least part of our portfolios accordingly? If payments in the industry reflected real value add we would be in a much stronger position to justify those payments.
Consequently, our role as investors should be to allocate assets “where there is suitable compensation for risk”. This means we have to:
- Broaden our view across asset classes not just within asset classes – recognise that our asset allocation should change, possibly materially, through time.
- Set time horizons for the delivery of objectives which are appropriate to market realities (this is not a 1 to 2 year game, it is a 10+ year game made difficult by the need to focus on 3-5 year time frames).
- Understand the real risks of what assets we are buying and compare those assets on a like-for-like (preferably ungeared) basis to our required investment objective.
- Change the structure of the governance and implementation models to cope with the required decision making around asset allocation.
- Introduce better fee alignment between service providers and the end beneficiaries.
One of the major issues for plan sponsors in Australia is that the current structure of most plans where assets are managed on an individual asset class basis in highly specialised mandates makes cross asset class analysis difficult, and implementation of cross asset class decisions cumbersome. There are a number of potential ways this can be resolved:
- The use of overlay managers
- Internalise the asset allocation decision
- Outsource the asset allocation decision
- Some combination of 2 and 3 above
This is a bit of a “been there, done that, didn’t work.” Fundamentally the issue with using overlay managers through time to make these sorts of decisions are in my view:
- The time frame for many of the decisions can be a multi-year one, as such using derivatives to implement a multi-year decision is unlikely to be the most cost efficient.
- The basis risk between the available futures and the nature of some of the decisions you want to take can be quite large – e.g. not owning LPT’s for a considerable period, or wanting to be out of emerging market debt.
- The benchmark for the overlay is the underlying strategic asset allocation (SAA) – the problem is that the SAA is the problem, you want to own assets with the right risk reward trade-off relative to your fundamental investment objective – this is not a benchmark relative decision it is an objective relative decision.
- As many have witnessed in 2008 with currency hedges, the mismatch between derivative positions and the need to fund those positions can have unintended consequences on funds ability to maintain the exposures they would like to obtain.
While overlays may perform a valid function (I’m not a buyer if you haven’t guessed) I don’t think they are the answer to the medium to long term asset allocation decision making process.
Internalise – the rise of the strategist?
As larger funds in particular build their internal investment capabilities then the reality is that some may be in a position to buy the appropriate skills. The key issues here are:
- The requisite skills need to be bought – this is no different to buying a capability in any other asset class (albeit the skills are different but the issues and principles are the same) – and the funds’ board and management need to be happy that that is the most cost efficient way to manage their exposures. For some it will be, for most not so (and their might be easier things to internalise before you get to asset allocation given the potential impact and the lack of market talent).
- The governance structure needs to be in place to support the decisions made by the internal team – in particular the greatest issue likely to have to be dealt with by funds willing to take larger asset allocation positions relative to market is peer group risk. Management of peer group risk is an extremely important issue – there is no point being right if you have none of your clients money left to be right with.
- There is still a requirement for the necessary investment mandate structure to be in place to ensure that managers can be appointed or terminated as necessary asset allocation decisions are taken or large cash flows properly handled (and the resulting impact this may have on fees where mandates decrease rapidly in size). All of this needs to take place on a timely basis to ensure that emerging opportunities are not missed (e.g. high yield credit earlier in the year) or not retained too long (probably high yield credit now!).
Like all decisions to outsource the primary issue we see is that sufficient assets are outsourced to make the exercise worthwhile at the total fund level at appropriate cost. Consequently, the issues around identifying appropriately skilled managers and how many managers to appoint remain (albeit you won’t want too many managers else you run the risk of cancelling out the skill bet).
In particular for individual funds one of the decisions that will need to be taken when outsourcing is where to fund the money from? Is it a slice down the portfolio (e.g. say 20% of the total portfolio) or is it funded from an “objective based” allocation? At the total performance level it ultimately doesn’t make a lot of difference, but from a governance and reporting perspective it may.
There is the potential in this option for some of the assets to be outsourced to consultants as well as money managers. While clearly representing a biased opinion, I believe this is an investment management function and skill set not an “advice” one. Consultants perform a valid role but it requires them to have the ability to provide objective independent advice to the board/investment committee as would be expected. You cannot do this if you are responsible for the decision that impacts 90% or more of the outcome. Notwithstanding that, it is fair to say that the track record of the broader consulting community on asset allocation globally is not good (with a few exceptions as always).
One key advantage of outsourcing objective based mandates is the ability to much better align fees with long term performance. “Managers” should or could get paid a completely transparent fee based on results achieved relative to the objective not based on the complexity of the structure or asset class they invest in. Not only is this likely to result in lower fees in aggregate, managers will be the ones required to decide if they want to allocate funds to that expensive hedge fund or private equity capacity versus much cheaper alternatives – I suspect in that framework the level of investment in expensive alternatives would fall rather sharply.
In our view, short of funds being able to allocate a significant portion of their assets to a small number of managers to make a real difference (probably the best option for smaller funds but practically unrealistic for larger funds), the most likely structure is some combination of the internal and external approaches.
Provided a suitable (small) number of objective based managers can be identified who have the necessary skill-set then there is likely to be some information advantage in understanding when there is mutual agreement amongst the external managers on the calls made. Internal teams should then be in a position (albeit the governance structures still have to be in place) to decide if they want to “leverage” these critical decisions across the broader portfolio.
In particular it has the distinct advantage that the decisions will still be reflected whether the internal team agrees or has the time to implement any additional changes.
In addition to decisions at the aggregate portfolio level where individual asset class portfolios are still used (for example fixed income) I would argue quite strongly that the nature of the market almost demands that the asset allocation decision be left highly flexible and not subdivided into rigid structures.
Strategic Asset Allocation has been the bedrock on which the superannuation/pension/investment industry has established itself over the last 20 years. The SAA typically represents the embodiment of a fund’s risk and return profile.
However, as we have witnessed recently, risk premia in markets change – and substantially so through time, whereas SAA’s typically remain quite static – while being reviewed episodically. As investors, our job is to manage and weigh risks in portfolios, a fixed SAA framework does not allow for the appropriate weighing of risks. This is by far the most important investment decision trustees/investment committees will make.
The objective based nature of superannuation – to achieve CPI+ returns over time – in our view makes it imperative that funds allocate all or at least a meaningful part of their assets in a way that is clearly aligned with those objectives – either via internal management, external management or some combination of the two.
What is required is a shift in thinking from plan sponsors that the process of a fixed strategic allocation to a number of largely equity like asset classes is the most appropriate way to achieve their medium to long term performance objective.
The way forward
Asset allocation has been – and will continue to be - the principal driver of investment returns. However, the sharp movements in risk premia over the last 2 years highlight the need for funds to more explicitly manage their exposures to different assets far more actively than has been considered in the past. This is not a situation that is likely to change in the coming years in our view (cf. the 30’s and 40’s, 60’s and 70’s).
Consequently, recognition of the importance of this decision and the impact negative returns from poor asset allocation decisions can have on member’s end benefits – even those with a long time to retirement – is critical. On a positive note, over the right time horizons, it needs to be noted that the price of an asset not unsurprisingly does impact the expected return quite significantly – and this is somewhat predictable and observable.
However, ensuring that funds have the ability to make and implement these decisions requires a change in structure, new skills and potentially a greater investment in governance – if only in dealing with the implications of the peer group risk that is being taken (meaning that for funds with “closed” membership groups, adopting some of the above is comparatively easier).
Once again, at the end of the day we need to recognise that an “asset class” is no different to any other collection of securities with some cashflow. What matters is:
- What price we pay for those cashflows
- The variability of those cashflows and what drives that variability
- How these relate to the performance objective
If we can recognise that, we are half-way to achieving the PM’s objectives and, more importantly, that of our key constituents.
Opinions, estimates and projections in this article constitute the current judgement of the author as of the date of this article. They do not necessarily reflect the opinions of Schroder Investment Management Australia Limited, ABN 22 000 443 274, AFS Licence 226473 ("Schroders") or any member of the Schroders Group and are subject to change without notice. In preparing this document, we have relied upon and assumed, without independent verification, the accuracy and completeness of all information available from public sources or which was otherwise reviewed by us. Schroders does not give any warranty as to the accuracy, reliability or completeness of information which is contained in this article. Except insofar as liability under any statute cannot be excluded, Schroders and its directors, employees, consultants or any company in the Schroders Group do not accept any liability (whether arising in contract, in tort or negligence or otherwise) for any error or omission in this article or for any resulting loss or damage (whether direct, indirect, consequential or otherwise) suffered by the recipient of this article or any other person. This document does not contain, and should not be relied on as containing any investment, accounting, legal or tax advice. Schroders may record and monitor telephone calls for security, training and compliance purposes.