View more on these topics

Allocation, allocation, allocation

Andrew Wilson, head of investment at JS&P Towry Law, explains why asset allocation is a vital force.

Numerous academic studies have shown that asset allocation is the primary explanatory factor behind portfolio returns.

To state to the contrary – “this is only true if you get it right” – with no supporting evidence is not a credible position. If you get it wrong, it can be extremely painful but it is still the asset allocation which has driven the return and the academic evidence holds true.

Mark Dampier’s column in Money Marketing last month used Jupiter’s multi-manager team as case in point.

The Jupiter team have, in the past, done well with their asset-allocation decisions and this explains their heterogeneous returns versus their peer group. However, this is also true during the times when they are less successful. Therefore asset allocation calls do indeed explain much of their returns, for better and for worse.

To suggest that time should spent simply selecting the “best” fund managers (whatever that means) rather than asset-allocating and then to end the article with a theory that bonds and property are “fairly fully valued” and that “a more truly diversified portfolio now should be far more one of cash and equities” seems disingenuous in the extreme.

What is true to say though is that it is incredibly difficult to get asset-allocation decisions consistently right.

David Fuller, a professional investor and commentator, reckons: “No one, and I mean no one, can assimilate, let alone make sense, of all the data available. And guess what – we would be no wiser if we could. We would only be a walking database – a repository for conflicting information and opinions.”

The vast majority of investment managers are better off doing what they are supposedly good at – invest-ing – rather than guessing or effectively betting. The asset allocator may have called Japan right one year but he has to repeat that success the following year. Actually, he needs to get two “bets” right each time. This is because any asset-allocation decision involves two calls, what to sell and what to buy and both have to be timed right to be successful.

Hence, it might not be enough to correctly call an upswing in Japan if emerging markets perform even better and that is where you took monies from.

An alternative to the stressful life of constantly trying to know more about geopolitics and macro-economics than the comb-ined knowledge of thous-ands of market participants is to take a longer strategic asset-allocation view and spend more of your time on where you can add value.

This was identified in the Money Marketing article, albeit in a slightly simplistic and self-evident manner, as the pursuit of the “best” fund managers.

Perhaps even more important is the blending of such managers into a riskcontrolled portfolio, whereby the end-investor has a properly balanced exposure to each asset class and is compensated for every unit of risk being taken, and monitored, on his or her behalf. This is what the best, and most highly resourced portfolio managers can do for you.

However, this is an incredibly tough discipline in which to succeed. Markets are zero-sum games to their participants and the addition of trading costs and fees mean the average investor will underperform the market as a whole. This will compound up to a position whereby they seriously lag the relevant index over time.

The best active managers are notoriously difficult to identify in advance and most seem to outperform only for a limited period of time.

They either find their “edge” difficult to replicate with much bigger sums under management or they may be poached by another group or depart to run hedge fund money. Surveys continually show that a big majority of fund managers have been in situ for less than three years.

This is too short a period to be able to judge their capabilities fully, partly because it is less than a full economic or market cycle. But this does not stop the industry trying to torture their past performance data into revealing a definite answer.

The more you demand of the data the less useful it is and most fund selectors simply end up buying the funds that are top decile over one and or three years, their differing torture techniques ending up with an answer less sophisticated than their process would have you believe.

Most funds at either the top or the bottom of performance league tables are usually those that have taken the most risk and got away with it (or not), or those whose manger’s style has been most suited to recent conditions (or not).

There are clear implications for mean reversion – witness the sorry sight of some portfolio managers switching into the latest hot fund just as it pays a visit to the fridge.

It is probably unfair to choose one example (and apologies to all concerned), but Dave Mitchinson’s JPM Japan fund was an interesting case in 2006. Mitchinson had a track record of strong performance, from his time at Framlington. But this was over a reasonably short time period. Under different market conditions and at a different investment house it became an altogether sorrier story.

Had they called market conditions wrong in Japan, or not understood the fund in the first place? Presumably, it has to be one or the other.

At this point, the asset allocators may be smirking, as, to them, it is less important whether they pick a first or third-quartile fund. Instead, it is more important that they correctly, in the case of 2006, call European equities over bonds.

In this way, if they can get the right funds, then it is icing on the cake and if they can’t, it should not be too much of a drag on their overall success.

So, two very different ways of skinning the investment cat. Asset allocators need to keep defying the odds and can suffer particularly unpleasant returns when they get it wrong.

Those relying on squeezing blood out of the active manager stone are fighting mathematics itself and ignoring the well documented benefits of a properly diversified and optimised portfolio.

That is not to say that consistently selecting outperforming managers is impossible but it does involve a huge amount of time and resources to do properly. Even when done correctly, it usually results in only small gains over indices, even for those at the very top of their game. It is probably best that most do not try. However, the default and benchmark option of pure index exposure is impossible to capture. An ETF (or tracker fund) with a TER of, say, 0.75 per cent, would, in a market rising by 10 per cent a year for 10 years, “underperform” that index by 20 per cent.

The “third way”, if that is not now a discredited term, is to use strategic long-term asset allocation, which should reduce volatility and add consistency to portfolio returns. The theory here is to diversify into multiple asset classes, which makes intuitive sense, if only from an “eggs and baskets” perspective.

Clearly, all asset classes used must be expected to appreciate over time (and to have a long-term history of doing so) and have differing characteristics.

The next step is to blend these asset classes in a mathematically optimal manner, based on the way they behave relative to one another. This needs some basic and generic assumptions about the future, for example, over any reasonable timeframe one might expect equities to outperform bonds and smaller companies to outperform blue chips.

The result should be an improved risk/reward tradeoff. Benjamin Graham himself was a fan of managing risks as opposed to trying to manage returns (the former should lead to the latter in any case) – and an efficient frontier of statistically optimal portfolios.

In reality, as we have already discussed, the future will always contain surprises, and the best asset-allocation optimisers, indeed the only ones worth using, are sophisticated enough to allow for uncertainty and can run simulations of possible futures, based upon probable risks.

The resulting portfolio will need little tweaking at the asset-allocation level. Each year, one can add another 12 months of correlation data – essentially further information on how asset classes behave relative to one another – and your broad assumptions about the medium-term future can be massaged if short-term market movements are so violent as to demand it.

The next step is to rebalance the portfolio regularly back to these target asset-allocation weights, perhaps annually. This is incredibly important. An annual rebalance allows a portfolio to benefit from momentum in asset classes, which has been shown to last for about a year and also to take advantage of mean reversion which some studies indicate kicks in over three years.

However, perhaps the greatest strength of this discipline is the behavioural angle. It can be hard to sell an investment that is appreciating or to allocate more resources to one that has gone down. But the rebalancing method does result in regularly taking profits and reinvesting in asset classes that are relatively “cheaper”.

In fact, if there was one tip we would give to an amateur investor, it would be to keep rebalancing (a contrarian strategy that keeps you diversified) back to a target asset allocation.

Why? The history of markets is littered with investors who have bought high and sold low. A study by Dalbar, a North American financial research firm, shows that in the 20 years to 2005, the average US investor earned 3.9 per cent, compared with 11.9 per cent from the S&P 500. The inescapable conclusion being that they could not resist buying high and selling low (and probably weren’t diversified enough either).

Investment is a discipline where few are heroes for long, and the price of error can be high or at least higher than the end-investor might be willing to bear.

Therefore it makes sense, if you can, to consider the latest and provable technology and to prefer to have mathematics on your side where possible.

Competing where the odds are against you and where you self-evidently have less information than the market opposition is not necessarily a great idea for most investors, especially those who want to grow their wealth rather than risk it on a hunch.

Funds can be a great way to access various asset classes but active managers seem best left to well resourced teams of professionals, with proven track records.

Sophisticated wealth management is available for investors and the evidence suggests it is very much worth their while outsourcing to the best teams, with attractive and plausible philosophies and strong risk-adjusted returns.

For those that would rather go it alone, perhaps after unhappy experiences, we would suggest that the path of least resistance involves passive exposure to a spread of asset classes, sensibly weighted and, crucially, rebalanced regularly and consistently.


Bee returns fire at pension snipers

Pension guru Steve Bee has lashed out at critics of the pension industry and says he is tired of constant sniping.Scottish Life head of pension strategy Bee, who is currently engaged in a second Battle of the Blogs with pensions minister James Purnell over his claims that people could find up to 40 per cent […]

Broker wins judicial review against FOS

A judicial review has clarified the right for advisers to make time-limited cash offers to claimants as an alternative to taking complaints to the Financial Ombudsman Service.The case originated from a dispute involving an insurance broker who offered a claimant £3,000 – half the amount claimed – if they accepted the offer within a month.The […]

F&C cut to dividend ‘will help develop specialist products’

F&C’s decision to cut dividends to plough millions of pounds into the business is a welcome move for advisers, says the firm’s head of communications and strategy Jason Hollands.Shares in the group fell by 18 per cent as it announced that the board had endorsed a three-year accelerated growth plan which will see the development […]

Rates of change

Early in January, the Bank of England caught the market by surprise by raising interest rates to 5.25 per cent. However, it was always likely that the bank would have to move rates higher fairly soon in response to the pick-up in core inflation over recent months.Until recently, higher input costs had not been substantially […]


Guide: day-to-day tasks ​— can your system manage?

This guide from Johnson Fleming will take you through the required communication and also give ideas for additional actions that will ensure your auto-enrolment project is a success. As well as highlighting what is required from a system to ensure it is up to the tasks, an overview of the following is also provided: data validation; data categorisation; employee communication; opt-in process; opt-out process; produce contribution schedule; contribution reconciliation process; upload of member data to pension provider; upload contribution to pension provider; manage salary sacrifice process; enrolment process; re-enrolment process; and management of increased employee queries.


News and expert analysis straight to your inbox

Sign up


    Leave a comment