First, there is no persist' />

Chasing fund and asset class returns can hurt, writes Christopher Worthington, senior economist at Gareth Morgan Investments.

The "stylised facts" of funds management, according to a research paper that landed on my desk this week, are twofold.

First, there is no persistence in a fund's ability to outperform their peers. No persistence, in this sense, means that the top-performing fund managers from last year are no more likely to have good performance this year than a manager selected at random. Secondly, despite fact one, investors actively chase the best returner of yesterday.

I couldn't help but chuckle as the authors, American academics Jonathan Reuter and Eric Zitzewitz, concluded: "The traditional interpretations of these facts are that fund managers are unskilled and fund investors are unsophisticated."

Whether this is true or not, it certainly reveals the gulf between finance as understood by academics and finance as conceived by the average investor.

The fact that most managers don't outperform the benchmark is repeated enough to assume that it is starting to filter into the general investor consciousness.

But the fact that there is almost no persistence in manager returns is perhaps less well known, despite it being equally well documented.

And even if it were well known, it is hard to imagine investors accepting it. Human beings have evolved to recognise patterns and build explanatory narratives - think for example about how we explain sports results. We discount the role of luck, even when the evidence points to luck having a large impact. In the case of fund out-performance, luck appears to be the main factor.

Interestingly, the Reuter and Zitzewitz paper was actually concerned with one objection to these stylised facts. An earlier argument made by Jonathan Berk is that the same observed facts are equally consistent with investors being excellent at identifying (truly) skilled managers.

His argument was that as investors shift their funds to more skilled managers, that skill is effectively diluted by those extra funds - to the point where in equilibrium the returns to all investors are again the same.

However, the end result is still that more skilled managers end up with larger pots of money to manage, as we would hope for in an efficient economy.

The evidence seems to reject Berk's theory. The problem of funds inflow reducing the manager's ability to earn returns does seem to have a small negative impact. But it is an order of magnitude too small to explain the fact that managers' outperformance does not persist from year to year.

Ordinarily, the impact of investors being "unsophisticated" would be small - after all, being in a fund with average rather than above-average performance doesn't really matter, unless investors are being charged for above-average performance (or, heaven forbid, relying on it for their retirement).

Where the costs do add up is if investors continue to chase last year's winner, year in and out. Since shifting between funds usually involves crossing a spread between the entry and exit price, this churn does result in more of the investors' returns ending up in the hands of the managers. For the funds industry as a whole, the misconception that luck is skill, and the churn it creates is very lucrative, as long as everyone gets a turn at the top of the charts from time to time.

Investors are not just chasers of funds performance, but also of asset class returns. Globally, investors have responded to the financial crisis of the past three years by exiting sharemarkets and loading up on bonds, responding to the crappy returns on the former and the excellent returns on the latter.

That same, seemingly reasonable belief - that there is persistence in returns - is even more flawed in this case. Over the long run, periods of unusual performance tend to reverse rather than continue. Asset classes that have performed poorly over a long period (shares in this case) tend to be cheaply valued relative to their earnings, raising their expected returns in the future. Bond yields, on the other hand, are extremely low, practically guaranteeing that future bond returns will also be low.

In other words, investors looking to improve their long-run returns should be moving from bonds to equities at the moment. But they're doing the opposite.

Some investors, no doubt, have a legitimate reason for portfolio switching. It's hard to estimate your risk tolerance, and the wild ride in recent years has no doubt brought home how risky equity investments can be. But most are simply chasing (or running from) last year's result.

Is this behaviour costly too? Research by top academic Ilia Dichev suggests that market timing costs the average sharemarket investor 1.3 per cent a year compared to the buy and hold market return, purely because investors hold less money in equities during periods with subsequently high returns, and have more invested in periods with subsequently low returns.

So chasing fund performance and chasing asset class returns can both potentially hurt investor returns. The tragicomic aspect of this behaviour is that in general, the investors making these bad decisions are the ones with higher levels of financial literacy, or at least the interest and patience to monitor market developments.

It's just that the seemingly reasonable assumption they use - that returns are persistent - is utterly wrong. Or as Mark Twain once said: "It ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so."