Anthony Hilton: Alliance Trust takes a strategy change risk

The board of Alliance Trust was under pressure to adapt

Rob Stothard/Getty Images

18 January 2017

One of the core skills of investment consultants is that they make their mistakes in private.

Up and down the land, they sit alongside pension administrators and trustees, and advise them on how best the schemes might invest the money in their charge in order to have enough to pay out the pensions when that time comes. And up and down the land, pension funds are in trouble.

It is not always cause and effect. But because the advice delivered is confidential between the consultant and the pension fund, and the reporting of pension schemes to their members does not go into the detail of why investment decisions were made, there is no way from the outside to see what value the consultants have added.

Similarly, although one hears anecdotally from disgruntled trustees about times when following consultants’ advice made a bad situation worse, such comments are rarely made for the public record.

This week, however, publicly quoted investment fund Alliance Trust announced a complete revamp of its strategy for equity investment whereby it turns its back on employing in-house fund managers and instead parcels up the money among a variety of external managers in the hope they will deliver a significantly better performance.

Alliance has been advised by Towers Watson and, while comment since the event has rightly focused on how Alliance’s reputation will be made or broken by the success or otherwise of this strategy, one might well say the same for its advisers.

If Alliance is transformed, Towers Watson will no doubt feel entitled to praise; if it does not work, it will have broken the golden rule and be seen to have failed in public.

The strategy is certainly ambitious. Alliance has sliced up its equity money between eight investment managers, some of which are well-known but others less so, scattered between Toronto, London, Fort Lauderdale, New York, Los Angeles and Stamford (Connecticut I think, not Lincolnshire). These managers, according to chairman Lord Smith of Kelvin, are “best in class”.

That does not, of course, mean they are any good — only that they are deemed to be better than the competition.

It is surely hard to be definitive, given that past performance is no guide to the future.

It turns out, however, that it is all about the process the fund manager adopts, how stocks are screened and selected, whether it is consistent over time and whether the success or failure of the fund can attributed to that process and various other things in the same vein.

The core idea is that the best-in-class fund manager is one where the process is so robust the individual has virtually no influence over it.

That’s surreal, perhaps, but not totally devoid of logic. There are very few brilliant fund managers and, as Nobel Prize for economics winner Daniel Kahneman says, people have psychological flaws that make them very bad investors.

They fall in love with stocks; they fail to deal with mistakes; they think they are special when they are not; they switch when they shouldn’t and so on.

Kahneman says the best approach is to expect to win a few and lose a few. Unfortunately from the industry’s perspective, while that might be an accurate description of reality, it is not much of a sales pitch. Identifying a unique investment process sounds a lot more convincing.

But should we believe in it? Years ago when I was a pension trustee, I can recall a research paper by Watson Wyatt, the predecessor firm of Towers Watson, which argued that in assessing fund managers at least 15 years of outperformance is needed before one can be reasonably sure any success is down to skill rather than luck.

It is worth noting that the then head of equities at one of the biggest insurance companies poured scorn on this — in his view it took at least 25 years.

This is intriguing because the new fund managers hired by Alliance will be expected to exceed their relevant benchmarks by 2% against the 1% target the earlier lot struggled to achieve.

Without wishing to be sceptical, I also recall a US research paper that said it would take even longer — the academics suggested 200 years — to ascertain whether this degree of outperformance was skill or luck.

Even putting that to one side, it would surely be a miracle if, at a time when the vast majority of equity-fund managers fail to beat the index, Alliance/Towers Watson has suddenly identified no fewer than eight exceptional talents, all of whom — despite being best in class — are sufficiently undiscovered to have room to take on new clients.

The other interesting aspect is that each fund manager will be expected to invest only in their “best ideas” — the few stocks they are most convinced will perform.

Each will select about 20 — except for the emerging markets specialist, who will be allowed to be “convinced” by 40. But, assuming the fund managers all have different ideas, that means there will be about 200 stocks across the portfolio — and that surely is a formula for average performance rather than outperformance.

Alliance risks suffering like all those pension schemes which hired a dozen hedge-fund managers, only to find they cancelled each other out.

However, you have to feel a degree of sympathy for the Alliance board, which was under pressure to “do something”.

There is no way of knowing in advance whether fund managers will deliver or not. But humans cannot cope with total uncertainty; we demand forecasts and processes as comfort blankets to allow us to frame what we convince ourselves are rational decisions.

It is comforting to believe investment is a science that will deliver predictable results if the right rules are followed.

And in these litigious days, it is far more convincing to be able to say in court, if you ever get sued, that you were following a process, not taking a wild punt.

MORE ABOUT

Pension

trust

Investment

Anthony Hilton: Alliance Trust takes a strategy change risk

Read More