Fund managers

darkpool

masonic wrote: »

There was an interesting article in the Telegraph today on the subject of active vs. tracker funds. According to some research by Morningstar, the average managed UK fund has been underperforming the All-Share index, which is no surprise. However, more noteworthy is that the average UK index tracker has been performing even worse than the average UK managed fund...

That's an interesting article, however i would say the research came from morningstar - a website that shows fund performance (maybe a bit biased). It also seems funny to use a 7 year comparison period. It would be intersting to see the comparison in performance for longer time frames.

Have a look at this website. Pure internet gold.

http://www.candidmoney.com/investment/trackers.aspx
Are active managers any better than monkeys?

Suppose we let 1,000 monkeys run their own investment funds, with them picking shares from a hat (i.e. the funds are run randomly), how well would we expect them to perform?
Well, assuming no charges, we'd expect 500 to beat the index and 500 to underperform the index in any given year. We would also expect 500 to beat the index cumulatively (i.e. in total) over five years but just 31 to beat the index every year for five years.
If our monkeys start charging total annual fees of 1.6% (typical for unit trusts), then we'd expect 429 to beat the index in any year, 345 to beat it cumulatively over five years and only 14 to beat it every year for five years (assuming an annual standard deviation (i.e. volatility) of 9%).
How does this compare with what happens in the real world?
Monkeys versus Active Fund ManagersManagers% beating FTSE All Share Index (figures to 05/11/10)Over the last yearOver the last 5 yearsIn each of the last 5 yearsMonkeys42.9%34.5%1.4%Active Managers41.0%36.6%0.5%Monkey figures intended to show probability assuming random stock selection with an annual standard deviation of 9% and total annual charges of 1.6%. Active manager figures based on actively managed funds from the IMA UK All Companies Sector.
So, based on these figures, the only major differences between monkeys and typical fund managers seem to be a fat salary and pinstripe suit! You should take this with a pinch of salt, after all, some fund managers have proven themselves to be exceptionally good over many years. But these figures highlight the difficulty of finding good managers and are a great advert for buying low cost tracking funds, for exposure to the mainstream UK stockmarket at least.

masonic

gadgetmind wrote: »

They go on the say that the average in managed funds masks the *huge* performance between them. (Trackers differ very little)

The *huge* difference in performance is a valid reason for choosing a tracker. I am certainly not trying to suggest otherwise. What I am saying is that trackers do not outperform managed funds on average, which is an argument often made here.

Incidentally, the fact trackers differ very little suggests that on the whole they are underperforming their index by far more than they should be. Apparently, they are not doing what they say on the tin. The underperformance is too large to be accounted for by explicit charges. So what's going on? High turnover? Tracking error? Poor replication of the index?

This is to be expected. Managed funds are a coin tossing contest. Do you feel lucky? Can you pick the winning fund, or might you instead get a motley collection of dogs?

If you pick your funds by 'coin toss', then that is indeed possible. I'm not going to be drawn into any arguments about picking funds. This has been discussed ad nauseum and it is clear that some people believe it is essentially impossible to avoid the dogs, while others disagree. I disagree.

And don't think that searching for a star manager, or using "recentism" and going for the sectors that have done well over the last few years (months!) will help, because statistical analysis shows that it won't.

Sorry.

Some of the "statistical analysis" that has been discussed here recently has been based upon fallacious assumptions about average performances. It is the interpretation of that "data" that I wanted to address.

I've yet to see a set of statistics that holds up to scrutiny, so I'll reserve judgement about whether good funds can be identified if it's all the same to you. What I do know categorically is that over the past 5 years my managed funds have returned more than my trackers.

If you want to take a fund flutter with a small percentage of your portfolio, then so be it: I'll probably do the same. But please use rational and mathematical/statistical reasoning to deduce what to do with the bulk.

While my managed funds are 'up' I will continue to hold them. The moment that performance gain is eroded I will not waste any time in going passive. The worst position I can find myself in is equivalent to having invested passively from day 1 and I'm likely to have given up hope on managed funds long before I reached that level.

gadgetmind

masonic wrote: »

While my managed funds are 'up' I will continue to hold them

By what algorithm do you evaluate this?

masonic

darkpool wrote: »

That's an interesting article, however i would say the research came from morningstar - a website that shows fund performance (maybe a bit biased). It also seems funny to use a 7 year comparison period. It would be intersting to see the comparison in performance for longer time frames.

I had considered the same thing. However, the comparison period shouldn't matter when comparing the trackers against the index. I think the trackers should have performed better than they did. Am I wrong to think so?

Have a look at this website. Pure internet gold.

http://www.candidmoney.com/investment/trackers.aspx
Are active managers any better than monkeys?

An interesting presentation of a familiar theme. On the whole, active managers don't seem to be very good at their jobs. I don't think there is any disputing that. However, they aren't all bad - I don't think anybody disagrees there either. The point of contention is whether one can top-slice, bottom-slice or otherwise improve ones chances of identifying the better managers, or indeed whether there is a formula or process for identifying them. I hold trackers in several sectors, so clearly I don't think I can do so in those sectors. In other sectors I have more confidence. It'll be at least another 10-15 years before I know if I am right or wrong on my calls, but so far I have made a good start.

masonic

gadgetmind wrote: »

By what algorithm do you evaluate this?

I compare the managed fund's performance with the tracker I would have otherwise held over the period I have been invested in the fund.

gadgetmind

masonic wrote: »

I compare the managed fund's performance with the tracker I would have otherwise held over the period I have been invested in the fund.

And you ditch the fund the moment (month, quarter, year?) that it underperforms?

What do you then buy?

What is your long-term return based on this strategy?

thelawnet

masonic wrote: »

The *huge* difference in performance is a valid reason for choosing a tracker. I am certainly not trying to suggest otherwise. What I am saying is that trackers do not outperform managed funds on average, which is an argument often made here.

An assertion with no supporting evidence, ok.

Incidentally, the fact trackers differ very little suggests that on the whole they are underperforming their index by far more than they should be. Apparently, they are not doing what they say on the tin. The underperformance is too large to be accounted for by explicit charges. So what's going on? High turnover? Tracking error? Poor replication of the index?

They do?

Looking at Ishares, FTSE100:
http://uk.ishares.com/en/rc/products/ISF/performance

Year-by-year performance vs. benchmark:
Y-1: -0.41%
Y-2: -0.45%
Y-3: -0.28%
Y-4: -0.31%
Y-5: -0.50%
Annualised since inception: -0.44%

TER is 0.4%, so the annualised underperformance is a mere 0.04%. Pretty good if you ask me.

For IUKD (UK Dividend+)
Annualised underperformance -0.15% - 0.25% BETTER than the TER of 0.4%

MIDD (FTSE 250)
Annualised undeperformance -0.52% - 0.12% below the TER of 0.4%

IUKP (Property)
-0.32%, slightly better than the 0.4% TER

I've yet to see a set of statistics that holds up to scrutiny, so I'll reserve judgement about whether good funds can be identified if it's all the same to you. What I do know categorically is that over the past 5 years my managed funds have returned more than my trackers.

Well that doesn't tell us much without knowing where the managed funds are invested and likewise the trackers. Equally the coin toss principle dictates that one has to be better than the other. Finally, given that markets have performed poorly over the last 5 years, it's no surprise that managed funds, which dampen down both good and bad performance have performed well.

cepheus

Why does Mutual Fund Performance not Persist?

The Impact and Interaction of Fund Flows and Manager Changes

.....investors should pay close attention to fund flows and the resulting changes in fund size, as well as to the career paths of individual fund managers amongst different funds: our results show that past performance is only an indicator of future performance if the manager is not replaced and fund flows do not eliminate the persistence.

see also the graph on page 34

thelawnet

masonic wrote: »

There was an interesting article in the Telegraph today on the subject of active vs. tracker funds

Well it's typically (for a journalist) imprecise.

'the average tracker fund '

What is the 'average tracker fund'? Overpriced crap sold to captive pension fund holders? Does it include all kinds of trackers or only FTSE All Share trackers?

Personally I think I suspect it's a lot of rubbish:

'Active managers also underperformed the index over five years, with returns of 4.3pc against 7.3pc. Trackers produced 4.1pc on average.'

Eh? Trackers returned 4.1% versus 7.3% for the index? Impossible I say - unless the 'trackers' were not all tracking the same thing.

In which case what's the point of this article? They can choose a different index, the FTSE 350 or something and use it to come to different conclusion.

The only thing that matters is the average tracking error for trackers, since trackers publish their costs, and you have the freedom to choose a cheaper tracker instead of a more expensive one..... So we just add 'average tracking error' to TER, and that's what we need to know. Then we compare the average return of the index, minus TER and tracking error, with the average return of managed funds. Not this nonsense about the 'average tracker', implying they are tracking very badly, which isn't the case.....

masonic

gadgetmind wrote: »

And you ditch the fund the moment (month, quarter, year?) that it underperforms?

What do you then buy?

It depends. I tend to check funds roughly monthly, but it would take something significant to make me abandon a fund just because it underperformed during that month. However, if I see the cumulative gains made by holding the fund eroded away significantly I will switch into the appropriate tracker. I have done this once so far.

What is your long-term return based on this strategy?

That is a tricky question to answer. This is not a strategy that I am using universally. For the parts of my portfolio this strategy does apply to:

Emerging Markets (30% allocated, 2 funds): fund avg down 3%, tracker down 12.5%, Gain 9.5% over ~1 year [Tracker launched only recently, but funds held for ~4-5 years]
Europe (8% allocated): fund up 35%, tracker down 5%, Gain 30% over 5 years
UK small cap (8% allocated): fund up 85%, tracker up 70%, Gain 15% over ~3 years.

The other funds I hold are either trackers (40% of my portfolio) or funds I can't directly compare with a tracker (14% of my portfolio).