View more on these topics

Why past performance numbers don’t add up

Advisers, investors and managers need to look past the numbers to judge fund performance 

Five-year performance records for investment funds now begin after stockmarkets such as the FTSE 100 and S&P 500 bottomed out in March 2009 following the financial crisis. Negative numbers generated by funds during the financial crisis are dropping out and being replaced by positive numbers that reflect the economic recovery.

Analysis of data provided by FE Analytics shows losing the 2008-2009 numbers means a huge jump in returns for many funds. For example,  In the IMA UK Equity Income sector,  £1,000 invested in Standard Life Investment’s UK Equity Income Unconstrained fund in the five years to 12 August returned £2,181.22, with the fund fourth out of 75 funds in the sector. Seven-year numbers reduce the return to £1,352.82 and the sector ranking to 40 out of 61 funds. On the same basis, Aberdeen Property Share TR returned £1818.61, with the fund fourth in the IMA Property sector out of 35 funds. Seven-year numbers reduce the return to £850.25, with the fund 24th out of 28 funds.

Past performance numbers need to be put into some sort of context for them to make any sense. Without reference to the sector average, benchmark, peer group performance, market conditions at the time and management style, it is impossible to judge the management of the fund as good, bad or indifferent, so the numbers become meaningless.

FE head of research Rob Gleeson says looking at fixed periods of past performance is of little value. “We find it far more helpful to look at past performance in different market conditions and use that as an indicator of how the fund’s strategy will react to different scenarios. It is much more useful to know if a fund does well in a fast-growing market or whether it is more suited to falling markets than how it has done over an arbitrary time period that may cover multiple market conditions,” he says.

The danger with taking five-year figures at face value is poor funds could start to look better even if nothing about the fund or the skill of its manager has changed. 

“It then becomes difficult to see if managers have ridden the wave of recovery or whether they are using their skills to generate performance. Either way, the temptation to use the improved numbers to market funds may be hard to resist. The five-year performance numbers for most funds now look great – or at least positive – and the marketing machines will be kicking into overdrive,” says GAM discretionary fund management investment director Charles Hepworth.

But advisers, investors and multi-managers need to look past the numbers and dig deeper to understand the reasons why funds have performed the way they have.

F&C co-head of multi-manager Gary Potter believes past performance numbers should be taken with a pinch of salt although he concedes they do provide a useful starting point for further analysis. “Judging real manager capability is about how did the manager perform in down markets. Only then can you deliberate whether a manager is good at bull market capability or bear market capability. You rarely find both,” he says.

“If you were to look at July or August last year, five-year numbers looked poor. Average returns from global funds were around the 20 to 30 per cent mark. Move forward a few months and the average returns were 40 to 60 per cent. Market performance was positive so the numbers look better.”

Five-year figures can also create false impressions of funds that were in positive territory during the financial crisis but subseq-uently underperformed because their style or investment process is suited to different market conditions.

Fleming Family & Partners investment analyst Ahmet Feridun stresses the importance of assessing managers over a full market cycle, so you can see how they perform in different market conditions. “The issue with looking at five-year track records is the vast injections of liquidity by global central banks over this period has led to an extended and largely uninterrupted bull run in equity markets. This means it is almost impossible to ascertain how these strategies would react in more normalised or stressed market conditions. “

Feridun sees the strong performance of US and UK small cap funds as evidence of this. “Many of these funds would have been a very bad investment for those that invested in them over the five- year period to the end of 2008, for example,” he says.

Former Charteris Treasury Portfolio Managers investment manager Tony Yousefian says: “It’s crucial to understand what the fund manager is doing and why 
they are doing it. Looking purely at the numbers wouldn’t mean anything. If you look at the last five years, it is only part of the economic cycle – and it hasn’t been a normal five years.”

Newsletter

News and expert analysis straight to your inbox

Sign up

Comments

There are 4 comments at the moment, we would love to hear your opinion too.

  1. A fairly anodyne piece reminding us that the line connecting two prices across market peaks and troughs can quickly pivot. Contributors opine that the last 5 years hasn’t been “normal”. So what is normal, exactly? I am reminded of the yacht captain who proclaims his skills as he speeds along with a following wind, yet blames the weather when becalmed.

    And don’t imagine marketing departments will be including the following caveat in their material “Don’t take these 5 year figures seriously. This wasn’t a normal period. If our style doesn’t suit the cycle, we’ll be crap. You have been warned…”

  2. Very nicely put Graham. Another aspect to consider is survivor bias. A client of mine, and I use that description extremely loosely, contacted me as he had ‘read’ an article on ETF’s and was wondering why he should bother to use my services (and he has my sympathy there!) as these ETF things as he called them can ‘change direction and sell out before a crash and get in at the bottom and essentially make money when the market was moving in any direction’. Wow! there goes my business model I thought, so for some masochistic reason I thought I would look into ETF’s past performance to see how these structures performed through the banking crisis. To my shock, despite there being several hundred recorded on FE Analytics, only a handful commenced before March 2009 (other than commodity ETF’s). Obviously they (Equity based ETF’s) simply didn’t exist before 2009 ! I’d have to conclude that this is some form of major survivor bias – not limited to ETF’s of course – but talk about lies, damn lies and statistics. Ho hum, back to work!

  3. Funnily enough most of those “ETF things” (and there were plenty around before 2009!!) performed pretty much exactly in line with the index they were tracking…..which of course is what they are supposed to do. Survivorship bias would make little or no difference to the average performance of index trackers in developed markets.

  4. Hi Andrew, yes of course they performed in line with the index they track – I was re-stating what my client was misquoting from his article. The survivorship bias I am talking about is the fact that so many of those ETF’s that were trading very happily before 2009 don’t show up on FE Analytics, but hundreds do show up after 2009. Maybe renamed, repackaged, whatever, so the survivor bias here is that those that fell with the indices during the banking crisis (all of them!) simply don’t exist anymore – the bad stuff didn’t happen – because the new versions came after the crash – ta dah! Just like magic. I have no problem with ETF’s at all and our whole investment strategy for the past 9 years has been passive or pseudo passive based. My comments really relate to past performance data in line with the article and the way clients or ‘victims’ can be duped.

Leave a comment