Is picking funds an easy route to outperformance?

15th Oct '18 by Oliver Cooper

43 comments 8922 reads

Kevfle recently posted an intriguing article called “Can we beat the professionals?’ In this he touched on some interesting questions around the investment industry, which provoked a flurry of responses from our fantastic community here at Stockopedia.

Kevfle seemed to have several questions on his mind, including:

Can we actually ever outperform the pros?
Am I wasting my time?
Is it easier to choose and evaluate a group of funds instead?
Can I achieve similar results leaving it to the pros?

In my previous article, I considered the first two questions regarding a more personalised investment approach. In the second of this two part series, I investigate the alternative option of buying funds and look at some of the difficulties that must be overcome.

Is it easier to choose and evaluate a group of funds?

Purchasing a group of funds (be they active or passive) has many advantages. Investing in a fund is an easy way to gain exposure to a certain market or asset class. They are efficient in gaining access to markets that are more difficult to trade in and and can provide the investor with a more diversified exposure for smaller investment sums.

Buying funds also requires very little time investment and you utilise the combined expertise of the talented members of the fund management team.

I will argue however that it is possibly more difficult to pick and evaluate ‘winners’ in this space.

The usual starting place for determining the level of skill of the fund manager is to look at how they have historically performed. Whilst we are all aware that we ‘cannot buy past performance’ it is still often used as a guide for how talented the fund manager is.

However due to the nature of luck involved in investment returns, past performance is a poor proxy for determining manager skill. Using returns data standalone, you would need a very long time series of return observations to have enough data to deduce a statistically significant outcome about a fund manager’s level of skill.

A popular way to filter the universe of funds is to use the Morningstar star ratings. These ratings are based on past performance and rate each fund between 1 and 5 stars, where 1 is the worst and 5 is the best. Seeing many stars on a fund factsheet may be visually appealing, however in line with…

Unlock with Email

Unlock with Google

Already have an account?
Login here

kevfle

16th Oct '18

Oliver

As I originally posed the question I was very interested to read your two articles, both of which are excellent and very informative.

Like many private investors who subscribe to Stockopedia, I regard the time which I spend on research as a hobby. I am semi-retired so have some spare time which I can devote to this. I enjoy the challenge - except when we have steep sell offs such as last week. I have still not trained my brain to treat this as an opportunity although I am working on it.

However, I started to look at valuing the time which I spend on research and it was that which prompted my question. Based on the size of my portfolio and the research time which I spend, I calculated that I needed to outperform a portfolio of funds (requiring little or no research) by about 300bps a year to recover the value of my time. In other words, if the funds achieved a 5% annual return after costs, I needed to achieve 8%.

I will probably continue this as a hobby in which case the value of my time is irrelevant. However, if you do treat investment as a business, then I think you do have to build in a cost for the value of your time when assessing performance.

pka

Here are some more reading material on the short-term persistence of short-term out-performance of funds:

https://repositorio.ucp.pt/bitstream/10400.14/11863/1/How%20to%20Profit%20From%20Mutual%20Fund%20Performance%20Persistence%20-%20D.Wessling.pdf

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1462408

https://pdfs.semanticscholar.org/e2d8/c98192110333f1cd451a266ead273e5785a0.pdf

http://www.kowal.com/sites/default/files/00U_short_term_alpha_as_a_predictor_michael_hartmann.pdf

https://www.crai.com/sites/default/files/publications/performance-persistence-in-UK-equity-funds-an-empirical-analysis.pdf

My conclusion from these papers is that there is some evidence for momentum in funds, but the momentum effect in funds seems to be weaker than that in individual stocks. So buying a number of funds that have out-performed over the previous year, holding them for a year, and then selling and repeating the process might lead to a modest out-performance of a few percent on average over the relevant market indices. But this would only be a profitable strategy if one can buy and sell funds relatively cheaply. Having to pay a 5% initial charge to buy a fund (which used to be the norm) would kill this strategy. Fortunately nowadays one can buy and sell funds for a small fixed fee without a percentage initial charge through platforms such as Alliance Trust Savings.

Merlotman

Thanks Oliver - another thoughtful article.
On fees - As many readers will also be aware brokers often levy a management charge which has differential rates for funds versus shares (this latter category includes IT and ETFs). For instance HL charges a % sliding fee directly related to the value of the funds you hold with no cap whilst the fee on shares is capped at £45 per holding.
Therefore for a say £1m portfolio of funds the annual fee is £3k whilst a portfolio of say 20 shares would cost £900. Obviously for smaller portfolios the economics are less differentiated. On the other hand on the face of it funds don't charge a bid offer spread but I personally don't like the lack of price transparency where you don't know how much you will pay per unit until end of day.
I agree that for someone armed with a bit of investment knowledge picking shares is more straightforward than picking funds because the elements of quality, value and momentum are easier to identify in a business.

Rusty2

Yes that is true however ETFs are classed under shares, that's why I tend to buy those, also you instant trading unlike funds which take a day or 2 to buy and sell. Just sold my Vietnamese ETF. Funds can give much more access to shares than just buying individual shares. As for picking funds I would look at the managers record, some have done well over a long period. In some ways it's easier than a companies management which will often chop and change. As I said in the other post about funds, I'm going back more towards them, they are good if you don't have the time to manage a lot of shares. If you are making over 20% a year and have the time then carry with shares.

JMLDutch

Good comment as being an HL customer, these broker management fees have put me off from investing in funds!

And it is actually even better - the £45 annual fee for shares is for the entire share portfolio (and includes ETFs). I pay £45 annually for an HL ISA consisting of multiple shares and ETFs.

timarr

The research shows, time and again, that passive index funds outperform active management, especially when you take fees into account. For instance, Europe:

https://www.fnlondon.com/artic...

Morningstar’s research, which looked at 9,400 active and passive mutual funds in Europe, in which around €2.9tn is invested, also found that fewer than a quarter of active managers beat the index over 10 years, in more than half of the categories surveyed.

Or in the US:

https://www.ft.com/content/e13...

The suitability of investing with asset managers that try to beat the market has been thrown into question by figures that show almost all US, global and emerging market funds have failed to outperform since 2006.

So basically, if you're buying active funds at random you've a very high probability of losing money compared to an index tracker. And if it's not done at random what's the method?

That is interesting, I had some European ones that did fairly well. I can’t read those links because I’m not a subscriber. But what about the uk? Also more international funds such as Fundsmith which I think makes about 20% pa. If you compare against the FTSE100 I should think quite few outperform it.

GrahamF

I’m never quite sure why when comparing active vs passive funds the comparison is against the average active fund. We’re trying to find the best managers here, no?

Bestinvest Tilney have their MRI (manager record index), which is a statistical analysis of the manager’s performance and is a starting point. Also, We should select a diverse range of managers for our portfolio (similar to having a diverse portfolio of shares in the NAPS portfolio), not all will outperform but hopefully overall they will.

I also tend to believe that there are a handful of star managers that overall tend to perform well. The likes of Terry Smith, Nick Train, James Anderson, and so on are not too difficult to spot.

Those figures are not for US funds that only invest in the US or European funds that only invest in Europe. They cover all active funds based in those countries - e.g. a US fund that invests in SE Asia, or commodities, or whatever.

As for the UK:

https://www.cass.city.ac.uk/fa...

UK investors would be 1.44% p.a. better off switching to a passive tracker fund, say researchers at Cass Business School. Their study claims that funds should split at critical size to protect investor returns. Almost all active fund managers fail to outperform the market once fees are extracted from returns, according to this research.

And:

Yet active management remains the dominant investment strategy and there are recognised 'star' fund managers. Based on the findings just one per cent of fund managers actually prove to be 'stars', being able to generate superior performance in excess of operating and trading costs. However, they extract all of this for themselves via fees, leaving nothing for investors.

There are also a number of tricks used by fund managers to hide bad performance, including merging and closing underperforming funds and only letting the successful ones continue - a form of survivorship bias. Vanguard did some research into this a few years ago. The conclusion is that the closure of failing funds leads to a relative boost to the overall active fund performance of the continuing funds - although the remaining funds still can't beat the indexes on average.

https://personal.vanguard.com/...

You've also got to account for active funds which are actually shadow index trackers with a bit of gearing around the edges. In good markets, such as the ones we've been in for a decade, these funds will "outperform" the index without really doing anything worth paying for.

There are other tricks as well to artificially boost performance, but you get the general idea.

timarr

Bestinvest Tilney have their MRI (manager record index), which is a statistical analysis of the manager’s performance and is a starting point. Also, We should select a diverse range of managers for our portfolio (similar to having a diverse portfolio of shares in the NAPS portfolio), not all will outperform but hopefully overall they will.

Fund manager performance tends to mean revert - which is what you'd expect if they're adding no real value. There are demonstrable momentum effects in funds which tend to fade along with the momentum in the underlying asset class and the problem the fund manager has of successfully investing new cash attracted by the success.

And there's lots of research about behavioural effects in fund managers. For instance, if your favourite (male) fund manager gets divorced they're likely to start taking a much more risky approach to stock selection. Female fund managers tend to achieve the same performance as male fund managers but to do so with much less volatility. And so on.

As for the obvious handful of star managers - well, I'd guess you'd have included Neil Woodford in there a couple of years ago. Even the star funds will suffer a period of underperformance at some point, the question will then be whether their investors' timeframes match their managers.

“As for the obvious handful of star managers - well, I'd guess you'd have included Neil Woodford in there a couple of years ago. Even the star funds will suffer a period of underperformance at some point, the question will then be whether their investors' timeframes match their managers.”

That’s why we should pick a portfolio of managers rather than a single star like Woodford. The argument that Woodford’s poor recent performance proves fund managers cannot outperform is like saying a single underperforming high Stockrank share underperforming means Stockrank doesn’t work.

The argument that Woodford’s poor recent performance proves fund managers cannot outperform is like saying a single underperforming high Stockrank share underperforming means Stockrank doesn’t work.

That would be an argument so mindbogglingly dense and unredeemingly stupid that I'd be embarrassed ever to post here again ... but I didn't say that - I was merely pointing out that your observation that the star managers "are not too difficult to spot" is open to debate and giving an example of why that's the case.

But the idea of picking a portfolio of funds - or even of fund managers - is hardly new. If it works the people who've figured out the magic formula aren't telling.

"For instance HL charges a % sliding fee directly related to the value of the funds you hold with no cap whilst the fee on shares is capped at £45 per holding."

That's why I don't use Hargreaves Lansdowne as a platform to hold funds. I use Alliance Trust Savings instead, which charges a fixed fee per year regardless of the value of the funds held together with a small transaction charge each time a fund or individual share is bought or sold.

I never picked Woodford either. You could argue that as I read over 90% of PIs don't beat the market then everyone should just buy trackers?
That study was over 20 years, I think that is too long but anyway the FT100 has only gone up about 27% which is about 1% pa, you would have been better to invest in fixed rate savings. I think most UK tracker are either FT100 and Allshare. I would agree there are a lot of funds that don't perform very well but most of those are fairly easy to avoid. Looking at the funds I had 4 years ago, the best went up 168% and the worst +18%, 6 shares outperformed the best fund and 16 underperformed the worst fund. Overall funds were up 54% compared with overall portfolio of +38%. Now this was before I joined Stockopedia.

Hi herbie

Basically you're saying that you have the skill to pick funds that will outperform the indexes. I can't disagree with you, I don't know you. But if you do have that skill you're in the minority of investors overall most of whom would be better off buying trackers.

Note though that there are hundreds of studies showing that index trackers outperform the active fund averages over all timescales. It's perfectly repeatable, so the fund picker has got to add some skill to the process, even if that skill is the ability to avoid obviously bad funds. Mostly people pick the best performing funds from the last few years - which, as I noted earlier, tend to suffer from mean reversion effects.

One point though - I'm pretty sure that FTSE 100 figure you quoted doesn't include dividends, and certainly doesn't include the compounding effects of reinvesting dividends which would have given you about an 8% p.a return over the last 25 years. Not exactly stunning, you'd think - but it would have turned £1,000 into £6,500 over that period.

That's the baseline for a private investor - with or without using funds.

But most PIs don't outperform the market so I don't see your arguement. Some reports say 90% of investors lose money, one report says this figure is even higher. No the FT100 figure is just capital growth over 20 years. Fundsmith has fallen back in the last week but still is over 18% pa over 5 years which is a lot better than the FT100 which is still only about 1% pa, ex. dividends. Compounding is all well and good if you do it over a long period.

snowdrop

17th Oct '18

For those seeking to balance time spent against likely return, I would recommend doing what I have done with my Funds based pension. I. E, Scan Investment Week's listings of funds using Morning Star data once a quarter. Under each category there are usually 10 or fewer funds which stand out year after year. The performance of my pension has improved radically since I did this. it's VERY simple. My annual pension return is over 9% despite being in tracker funds for years before switching to this. It doesn't help with sector allocation of course.

Since Fundsmith has significant exposure to the USA (currently 60%), it isn't really fair to compare Fundsmith to the FTSE 100. The S&P 500 has had a much better run over the last 5 and 20 years (especially when expressed in pounds due to the decline in GBP). Also, the 4% yield on the FTSE 100 would improve the returns of the FTSE 100 trackers quite a bit (although of course nowhere near to Fundsmith's performance!).

For comparison -
S%P 500 trackers returned 13.51% annualized over the last 5 years.
FTSE 100 trackers returned 6.9% annualized over the last 5 years

Not to take anything away from Fundsmith, the performance of Fundsmith has been extraordinary.

peterthegreat

Hello Oliver and thanks very much for the thought-provoking article and to Kevfle for the earlier article.
If I could perhaps act as devil's adovocate I would suggest that it is irrelevant whether investors choose to invest in funds of directly in shares, the important point is that you (the individual investor) or the fund manager you use, has identified the correct approach to investing, in which case you will get good returns either way.
In common with well informed contributors to this thread, the fund managers I have identified for their excellent long term performance and (just as important) their clearly defined methodology are Terry Smith, Nick Train and Keith Ashworth-Lord.

I have used these fund managers as my mentors for most of my investing activities and have also found that they are willing to answer the occasional questions I pose to them (either directly or through their assistants) which helps me further understand how they invest. I would suggest that it does not matter whether you invest in their funds or directly in the shares (or type of shares) they invest in as you will get good returns either way (but take care if it is an investment trust at a significant premium to nav!). I think I also know why Terry Smith's approach does not work in emerging markets but that is another story.

I use these fund managers to identify which shares to buy both with respect to UK and some overseas companies. Particularly in the case of Terry Smith and Nick Train, I also use their approach to buy shares in smaller companies outside their universe of investing. I tried to persuade Terry Smith to start a fund for smaller companies a couple of years ago as my experience is that his methods also work for smaller companies (glad he has at last taken notice!!).

Of course there is some advantage in buying their funds rather than investing directly in their type of companies because I can not identify what they have bought until some time time after they have announced their purchase and the price may move when they announce the purchase. However, in some cases I can tell fairly quickly what Terry Smith is buying from regulatory filings for his US based fund. For some foreign companies I am also not able to claim back portions of witholding tax which these funds will be able to do for most of their foreign holdings, although I can do this for US and Canadian companies through my broker.

For reasons stated above, you can understand why I hold some of the funds managed by these managers but most of my holdings are directly in companies which they hold or which I have identified using their methods. This way. I avoid their fees for managing the funds. Perhaps I could describe this as a hybrid approach between investing in funds and investing directly in shares and it seems to work for me. However, the disadvantage is the time it has taken me to validate this approach and to execute it and I would agree that most people would not have the interest or background to do this effectively so I would admit that investing in the funds of these managers is probably best for most people.

As an aside, I would admit that if the approach of the fund managers became more popular, the share prices of their holdings could go up so high that the method would no longer work for new investments. Terry Smith has of course discussed this at some length and I would agree with him that this is not currently a problem. It seems to me that most investors still worry about charts, buy and sell limitsand news driven investment decisions so I am confident that the methods used by these fund managers will not become so popular that they will no longer work, but I do monitor this.

I do not think it is any coincidence that contributors to this board (many of whom have also done

I admire Terry Smith, Nick Train and Keith Ashworth-Lord and try to read everything I can by them, but I don’t think investing in all three of their funds is diversification other than in the most limited sense. To jump ahead to my conclusion, it’s likely that if one of those funds does well then all of them will, and if one of them does badly then they all will. And, at some unspecified point in future, they all will.

Diversification can protect you against two forms of risk – systemic and unsystemic. To take an example, if I invest in Rio Tinto (LON:RIO) and BHP Billiton (LON:BLT) I protect myself against the unsystemic risk of Rio’s South American assets being appropriated or BHP’s senior management team being executed by an environmentalist hit team. However, it doesn’t protect me against the systemic risk of a global economic downturn or Donald Trump nuking the Fed or any other event that will cause all commodity stocks to fall in sync.

All three of the aforementioned fund managers take very similar approaches to stock selection such that although the stocks they select tend to be different, their underlying characteristics tend not to be. Given that all three funds have done very well in the last five years it’s tautological to point out that recent market behaviour has favoured the sorts of stocks they like to hold. But perhaps more pertinently, markets have tended to discount the highly cyclical stocks that they shun.

At some point in the future that will change, and the performance of these funds will falter – and when they do, they’ll likely all do so together. Given that they’re all highly rated a net outflow of funds will either lead to a reduction/widening of the premium/discount to net asset value for the quoted funds or the need to sell temporarily discounted shares to fund redemptions for the open ended funds: essentially a double whammy for investors.

You don't have to take my word for it, though. Lindsell Train Investment Trust (LON:LTI) currently trades on a boggling 65% premium to net asset value – some of which can be accounted for by the value of the unquoted LTL management entity – but by no means all. As their Chairman stated in their last Annual Report:

I have asserted before that the Board view the premium as unwarranted and do so again today with particular reference to my comments above about succession. Consistent with that I continue to caution new shareholders about buying shares at a premium as any reversal in markets or in LTL’s performance could cause it to evaporate, leading to significant losses.

https://www.lindselltrain.com/...

And remember, LTL makes its money from the performance of LT's unquoted funds which invest in similar stocks to Lindsell Train Investment Trust (LON:LTI) so that if the latter fall then so do the former and so does the value of LTL which further impacts the value of Lindsell Train Investment Trust (LON:LTI) ... got that? (Basically, if you want to invest in LT invest in the open ended funds).

None of that detracts from the quality of the managements or their long term success – but it’s a given that a narrowly drawn investment approach like this will suffer setbacks at some point. Investing in all three funds doesn't protect against much unsystemic risk other than the possibility of one of the managers going mad, selling everything and investing in cryptos, and adds some correlated systemic risk at the expense of additional fees.

I vaguely get the idea of diversifying across fund managers when you don’t have much insight into their skills, attitudes or abilities but I don’t get it at all at this level. And this is simply a specific example of the general problem of buying into the fund managers who've done well in the recent past: they tend to be highly rated and highly correlated, and the recent past is only a predictor of the recent future until it isn't.

Is picking funds an easy route to outperformance?

Is it easier to choose and evaluate a group of funds?

Unlock the rest of this article with a 14 day trial

About the Author

Oliver Cooper

43 comments

Unlock the metrics for all these stocks and more...

Unlock the metrics for all these stocks and more...