The Investment Management Association has been ridiculed by both industry professionals and the media because of its decision to classify funds that manage their asset allocation across conventional classes as managed A, managed B or managed C. Managed D awaits for funds with a lower risk profile that might currently sit in either cautious managed or absolute return.
In its paper of May 26, the IMA advanced a purist view: “The names of the managed or mixed-asset fund categories are intended deliberately to provide no other information about the sector, thereby encouraging users of the sector to do more due diligence to understand the nature of the funds that fall into the underlying sectors.”
The generous interpretation of its superficially farcical decision is that it was intended to provoke a debate it thinks too many people are not yet engaging in. If this is what it meant to do, we whole-heartedly support it.
Without greater under-standing of the uncertainty inherent in describing and differentiating investment strategies, a number of problems will be perpetuated:
- Investors, both lay and professional, will go on being surprised, regretful or even angry when things do not turn out as expected.
- Regulators will go on laying down principles or even rules for risk assessment and applying them after the event.
- Consumer activists will go on berating the industry because it refuses to say on the tin what its products do.
These are all big problems for consumers, agents, regulators and journalists but they will not be solved without an accommodation with the truth, however inconvenient.
The problem of forming clear expectations about risk and return is much greater in multi-asset class funds than in single-market, single-currency funds, even when they rely on conventional assets such as equities, bonds and cash.
Diversification was never intended to be a way of managing absolute risk
This is a corollary of the 90 per cent rule, in which asset class and market exposure explain most of return variance and long-term outcomes precisely because the differences are typically large and there are fewer of them. Poor predictability arises from the combination of uncertainty about funds’ own risk measures, which are not stable, and their common movement or co-variance, which are highly unstable.
This is no different from the problem of selecting individual discretionary or advisory portfolio managers in a market dominated by semi-customised balanced portfolios. The commonplace solution of positioning a few different versions of balanced management on some notional risk spectrum defined by volatility, often with names that match a similar categorisation of investor needs or preferences, relies on the same certainty about location.
Even if all investors’ utility or welfare was best defined in terms of volatility, locating portfolios on a spectrum defined by volatility is not possible with anything like the accuracy the approach assumes.
Most private wealth is, in any case, defined more by long-term purchasing power outcomes than by volatility. Volatility you can live with but outcomes are what the benefits of wealth depend on.
On a long-term, real outcome risk spectrum, the same portfolios occupy very different positions.
The balanced approach takes its authority from a theory about diversification of risks, aimed at more efficient allocation of risk capital.
Diversification was never intended to be a way of managing absolute risk.
Differentiating mixed funds by the mix, or particular components of the mix, is another way of asking too much of diversification. The IMA’s alphabet soup at least avoids this.
Some may welcome this bizarre development precisely because they believe balanced management has failed as a means of risk control but very few people are as yet confronting the failure.
A far better way to find the right risk spectrum and the right position on it is to start by defining the risk-free asset that best or even perfectly matches the desired outcome at the desired time.
Risk management is then about splitting all asset holdings between that asset and a set of risky assets that also match the desired outcome but with uncertainty.
The second no longer needs a vast array of products to support it – geographically diversified equity exposure is sufficient. This approach is far easier to describe and quantify than combining lots of different assets in an artificial construct (a portfolio or product) based on some expected combination of risk, return and co-movement.
In an important sector of finance, the management of defined-benefit pension funds, balanced management is rapidly being replaced by portfolio separation. In so-called liability-driven investment, assets are split between hedges and a risky or return-seeking portfolio. Setting up a hedge portfolio, which often takes the form of duration-matched index-linked gilt holdings, is typically very cheap compared with the portfolio seeking returns. And this brings us to a major conflict of interest in retail finance.
With the business models of most retail agents based on the value of assets under management or advice, and ad valorem fees set at levels that absorb much or all of any real risk-free return, agents would rather take in all or most of the assets available and find or build the construct that uses all the assets to locate itself on the notional risk spectrum.
They do not want customers splitting their hand between riskless assets that do not need managing and risky assets that do because it reduces their revenue.
I question the common assumption that the difference between the three IMA sectors being renamed from July is best explained in terms of their equity exposure.
Based on historical evidence, however, the asset that most differentiates their location on the risk spectrum, however defined, is conventional fixed income, because its low (often negative) correlation with all equity related risk sources is fairly reliable.
This is also therefore true of the Association of Private Client Investment Managers benchmark indices, which mainly rely on the fixed-income exposure to differentiate the riskiness.
This is remarkable. It means the main form of risk control in such portfolios or products is varying the exposure to an asset that reduces the impact of equity risk, only to be substituting it with inflation risk.
This is not a trade-off worth making consciously and it should not be made by default.
With these flaws in conventional balanced management, it is no wonder we welcome the IMA’s provocation.
Stuart Fowler is founder and director of investment at Fowler Drew