should be mandatory reading for everyone interested in finance

[colin]

mc2fool wrote One was 154% and the other 363%, so for each £1 invested you'd have ended up with £2.54 and £4.63 respectively, and the latter is 82% more than the former, not over three times as much.

Thanks that's a more arithmetically literate way of looking at the figures but it still represents a very big difference between two 10 year periods separated by 11 months, Global markets as a whole were volatile then but not that volatile?

[colin]

Hiriskpaul wrote..
The AIC stats don't work the same way as indices and are subject to survivorship bias, so moving a poor performer out could significantly increase the average performance.
Top

Yes that's what needs to be checked before the stats can be used to compare investment trust sectors with passive indices. If a trust is wound up and the funds rolled into another trust in the same sector would the previous performance of the closed trust be encapsulated in the performance of the now larger 'host' trust? Presumably if trusts are wound up and money returned to shareholders then the performance record of that trust becomes lost to the subsequent years statistics, I know that happened with John Laing Infrastructure in 20018 but how often does a complete winding up of a trust occur within a ten year period.?

[mc2fool]

mc2fool wrote One was 154% and the other 363%, so for each £1 invested you'd have ended up with £2.54 and £4.63 respectively, and the latter is 82% more than the former, not over three times as much.

Thanks that's a more arithmetically literate way of looking at the figures but it still represents a very big difference between two 10 year periods separated by 11 months, Global markets as a whole were volatile then but not that volatile?

Well, the MSCI World index (capital only) closed Dec-07 at 1,588.80 and Dec-17 at 2,103.45, which is +32%, turning each £1 invested into £1.32.

And it closed Nov-08 at 892.93 and Nov-18 at 2,041.36, which is +129%, turning each £1 invested into £2.29 -- which is 73% more than £1.32, and that's just the capital only, not TR. https://uk.investing.com/indices/msci-world-historical-data

So, the 82% more from the AIC global sector TR as a relative difference across those periods looks about right. However, in comparing returns themselves (rather than the difference between periods) the AIC sector appears to have done much better -- 154% vs 32% and 363% vs 129% -- more so than TR vs capital only and/or £ vs $ could reasonably explain, so I'd repeat my earlier comment about the AIC global sector not really being MSCI-World-like, and also HRP's comment re survivorship bias.

If you really want to get into the nitty gritty of that you can download and compare the AIC's Monthly information releases, https://www.theaic.co.uk/aic/statistics/mir, which are .CSV's you can suck into a spreadsheet and go back to the start of 2007.

Or you could ask the AIC. I've done so for a couple of things over the years and they're fairly responsive. https://www.theaic.co.uk/aic/contact-us

P.S. If instead of prefixing copied text with "mc2fool wrote" you select the text and click the "Quote" button at the top right of the post, then the system will automatically produce a quote box with the name of the poster (as above) and notify that poster that you've replied to them (if they have notifications turned on).

[johnhemming]

The underlying problem with looking at 10 year figures when there are discontinuities during those periods is that the period that you select has a substantial impact on the conclusion. It is the eternal problem of trying to avoid buying at the top.

If you look at the eleven years you get both the financial crisis of 2008ish and the impact of the brexit vote and trump (Which I think are the main two causes of the drop in the FTSE100 in the past year or so, mainly brexit, but trump does not help)

[colin]

Thanks for that particularly useful post mc2fool.
On a total return basis the figures for the MSCI index are 63% for Dec 2007-Dec 2017 and 181% for Nov 2008-Nov 2018. Giving a difference of 72% so close to your capital only calculation. That's priced in Dollars, I remember back in 2008 getting 2 dollars for my pound on holiday , but still the Global IT sector returns in sterling are way ahead. When I have some time on a wet day i shall go through the stats looking for trusts which might have dropped out.
Cheers again and Happy New Year.

[Hariseldon58]

I have 30+ downloads of the AIC Stats from 2007 to 2017 unfortunately they do not provide much extra info, as they do not cover the relevant months for the apparent discrepancy, found for 10 year returns on the interactive stats page. There has been quite a few changes in the Global Growth sector over 11 years , trusts have moved sectors etc. Closed or merged..

The earlier downloads contained the comparable indices for the right time periods which made life easy for comparing trusts to their indices.

[colin]

That's OK Hariseldon58
I shall go through the Global sector stats provided in mc2fool's link , draw up a list of trusts not represented every year in a 10 year period, do a bit of research into how their performance records might have been retained and that will be good enough i think to give me a feel for what the extent of survivor ship bias might have been over a ten year period.
Thanks .

[hiriskpaul]

Thanks for that particularly useful post mc2fool.
On a total return basis the figures for the MSCI index are 63% for Dec 2007-Dec 2017 and 181% for Nov 2008-Nov 2018. Giving a difference of 72% so close to your capital only calculation. That's priced in Dollars, I remember back in 2008 getting 2 dollars for my pound on holiday , but still the Global IT sector returns in sterling are way ahead. When I have some time on a wet day i shall go through the stats looking for trusts which might have dropped out.
Cheers again and Happy New Year.

There was a large drop in the pound which flatters the figures. Dec 2007-Dec 2017 MSCI World TR in pounds was 155%. Nov 2008-Nov 2018 was 258%.

When comparing past returns with ITs, etc. it is worth considering not just absolute return, but the spread of historic returns. With trackers, the spread tends to be very small. With actively managed funds it is vast. With passives the risk of significant underformance is largely eliminated. With active the risk of significant underperformance is the price you have to pay in order to have any chance of outperformance. I prefer to take risks with an expected positive payoff, but all the evidence shows that the expected payoff from actively managed funds is negative. ITs are a little different from other types of funds though as they can use gearing, but don't charge the outrageous fees of hedge funds. They also often trade at a discount, the effect of which can offset their fees.

[Hariseldon58]

@HiRiskPaul

Well in the passive/active debate I like both.

There are times , the last few years prior to this summer where you only need be along for the ride.

As private investors I can go against the trend if I feel confident and using active ITs it’s more a question of being able to avoid some things that are popular and buy some unpopular things.

I thought that smart beta ETFs would be my Investment if choive in this manner but experience so far has been very mixed. The various UK equity income smart beta investments eg iukd, UKDV and the Vanguard UK Hi Yield fund have not compared favourably to the UK Equity Income ITs, the point about the spread of returns between active funds is valid , a mix of half a dozen ITs for each specialised area is my approach.

Swapping ETFs at nav for ITs at a discount has a reasonable chance of paying off and swapping back in the future, it’s worked before

[OhNoNotimAgain]

:D @HiRiskPaul

Well in the passive/active debate I like both.

There are times , the last few years prior to this summer where you only need be along for the ride.

As private investors I can go against the trend if I feel confident and using active ITs it’s more a question of being able to avoid some things that are popular and buy some unpopular things.

I thought that smart beta ETFs would be my Investment if choive in this manner but experience so far has been very mixed. The various UK equity income smart beta investments eg iukd, UKDV and the Vanguard UK Hi Yield fund have not compared favourably to the UK Equity Income ITs, the point about the spread of returns between active funds is valid , a mix of half a dozen ITs for each specialised area is my approach.

Swapping ETFs at nav for ITs at a discount has a reasonable chance of paying off and swapping back in the future, it’s worked before

I have yet to hear a a good argument in favour of active when a rules-based alternative is available.

[Lootman]

I have yet to hear a a good argument in favour of active when a rules-based alternative is available.

So the very worst rules-based fund always beats the very best active fund?

Wow.

[colin]

I have yet to hear a a good argument in favour of active when a rules-based alternative is available.

Such as?

[OhNoNotimAgain]

I have yet to hear a a good argument in favour of active when a rules-based alternative is available.

So the very worst rules-based fund always beats the very best active fund?

Wow.

Who said that?

There are three different active funds at the top of the UK All Companies performance tables over 1, 3 and 5 years.
Most passive funds are in the the second quartile over those same time horizons.
So the data tells us that the average passive beats the average active fund. Sure, some active funds will beat everything over some periods but the data also tells us that outperformance by any one active fund will not be sustained.

[Lootman]

I have yet to hear a a good argument in favour of active when a rules-based alternative is available.

So the very worst rules-based fund always beats the very best active fund? Wow.

Who said that?

There are three different active funds at the top of the UK All Companies performance tables over 1, 3 and 5 years.
Most passive funds are in the the second quartile over those same time horizons.

So the data tells us that the average passive beats the average active fund. Sure, some active funds will beat everything over some periods but the data also tells us that outperformance by any one active fund will not be sustained.

You've changed your story there somewhat. You started talking about rules-based funds and then about passive funds. I do not see those as identical.

The only purely passive funds are cap-weighted trackers. They literally run themselves for the most part.

The universe of rules-based funds is broader, as they include all the so-called "smart beta" funds. They are "passive" in a sense but need more work and effort, since they have to be re-balanced frequency, unlike cap-weighted funds. Which means that they are more expensive as well.

I can believe that most truly passive funds would be in the second quartile much of the time, relative to their underlying benchmark. But the more active type of passive funds have a much wider distribution, and can fail just as spectacularly as a purely active fund. It just depends how appropriate the chosen rules end up being.

Your comment that "rules-based" alternatives are always better than active funds is an exaggeration. They may or may not be better depending on how successful the chosen rules actually are in practice.

[OhNoNotimAgain]

So the very worst rules-based fund always beats the very best active fund? Wow.

Who said that?

There are three different active funds at the top of the UK All Companies performance tables over 1, 3 and 5 years.
Most passive funds are in the the second quartile over those same time horizons.

So the data tells us that the average passive beats the average active fund. Sure, some active funds will beat everything over some periods but the data also tells us that outperformance by any one active fund will not be sustained.

You've changed your story there somewhat. You started talking about rules-based funds and then about passive funds. I do not see those as identical.

The only purely passive funds are cap-weighted trackers. They literally run themselves for the most part.

The universe of rules-based funds is broader, as they include all the so-called "smart beta" funds. They are "passive" in a sense but need more work and effort, since they have to be re-balanced frequency, unlike cap-weighted funds. Which means that they are more expensive as well.

I can believe that most truly passive funds would be in the second quartile much of the time, relative to their underlying benchmark. But the more active type of passive funds have a much wider distribution, and can fail just as spectacularly as a purely active fund. It just depends how appropriate the chosen rules end up being.

Your comment that "rules-based" alternatives are always better than active funds is an exaggeration. They may or may not be better depending on how successful the chosen rules actually are in practice.

I use passive as short-hand for rules based. No fund is truly passive.
Many passive funds include derivatives. These are not part of the index so an active decision must be made on when and how to use them. Equally, few, if any, passive funds hold all the the stocks in the index it tracks, again an active decision. Takeovers and other corporate actions all involve someone taking a decision to "do" something. However, if the rules are clear then investors know what they are buying into in a way that is totally impossible with an active fund manager.

[Lootman]

I use passive as short-hand for rules based. No fund is truly passive.

Many passive funds include derivatives. These are not part of the index so an active decision must be made on when and how to use them. Equally, few, if any, passive funds hold all the the stocks in the index it tracks, again an active decision. Takeovers and other corporate actions all involve someone taking a decision to "do" something. However, if the rules are clear then investors know what they are buying into in a way that is totally impossible with an active fund manager.

There is a good choice of trackers that follow the major market indices, that physically hold every share in that index, and require trivial amounts of trading i.e. only for index changes and corporate actions.

A "smart" rules-based alternative that uses a proprietary index will involve all that work as well PLUS frequent re-balancing. As a class of funds I have seen no evidence that smart beta funds beat market cap trackers. Smart beta funds are actively-managed funds masquerading as "passive". And they are very capable of doing very badly, especially given their higher expense ratios.

[OhNoNotimAgain]

I use passive as short-hand for rules based. No fund is truly passive.

Many passive funds include derivatives. These are not part of the index so an active decision must be made on when and how to use them. Equally, few, if any, passive funds hold all the the stocks in the index it tracks, again an active decision. Takeovers and other corporate actions all involve someone taking a decision to "do" something. However, if the rules are clear then investors know what they are buying into in a way that is totally impossible with an active fund manager.

There is a good choice of trackers that follow the major market indices, that physically hold every share in that index, and require trivial amounts of trading i.e. only for index changes and corporate actions.

A "smart" rules-based alternative that uses a proprietary index will involve all that work as well PLUS frequent re-balancing. As a class of funds I have seen no evidence that smart beta funds beat market cap trackers. Smart beta funds are actively-managed funds masquerading as "passive". And they are very capable of doing very badly, especially given their higher expense ratios.

look at this investment objective for the Vanguard UK All Share Index fund

The Fund seeks to track the performance of the FTSE All-Share Index.

"seeks to track" is the crucial weasel clause. Moreover it only holds 573 out of the 637 in the index and has a turnover rate of 7%, which is higher than some other rule-based funds.

In terms of the performance evidence for alternative weighted rule-based funds I cannot help you, but the data is out there; except on Citywire.

Analysing costs can be tricky because some funds exclude stamp duty, by charging a seperate dilution levy, others swallow it and include it in the OCF.

[colin]

OhNoNotimAgain wrote:

"seeks to track" is the crucial weasel clause.

Why use the adjective 'weasel' as though Vanguard is trying to hide something? The information necessary to compare the return on an index with the theoretical return of the relevant index is plainly shown for all passive cap weighted funds that I have ever come across.

Moreover it only holds 573 out of the 637 in the index

So what? the missing small cap shares would be too expensive to hold due to their poor liquidity, and given their low value relative to the all share index they could never contribute much in the way of return one way or the other.

[Lootman]

In terms of the performance evidence for alternative weighted rule-based funds I cannot help you, but the data is out there; except on Citywire.

So you admit that you cannot show us any evidence for your claim?

The last time I looked at your fund it was a third quartile performer with an expense ratio over 1% annually. Investors would have been better off just buying a Vanguard FTSE ETF. Why? Partly because of the expenses I cited earlier. And partly because whatever your rules are, you are over-weighting some names and sectors, and under-weighting others, and you can easily get that dead wrong.

costs can be tricky because some funds exclude stamp duty, by charging a seperate dilution levy, others swallow it and include it in the OCF.

The most accurate way to measure the effect of expenses is the tracking error, which is minimal for the major market ETFs from iShares, Vanguard, State Street etc. No matter how costs and expenses are reported, the tracking error cannot be faked. Most smart beta funds under-perform their cap-weighted benchmark for the same reason most active funds do - the costs are a constant source of friction that decreases returns even if the bets work out OK. You are running into a constant headwind.

[OhNoNotimAgain]

This is a discussion I cannot take any further.