Fund Selection

OldScientist

chiang_mai said:

If you want certainty that you will match the index every year for ever, index trackers are the way to go. But in doing that, you should realise that you will have no choice but to take the ride down when the markets crash and you will have to wait for the markets to recover before your tracker returns to profit. Know also, that approach guarantees that you will never beat the index, not ever. Some of my earlier posts in this thread mention the success I've had this year with managed Artemis funds that have out performed and I've beaten the index for periods of around six months before swapping out into other funds. That requires effort, time and analysis, which not everyone is capable of nor wishes to do. There's a middle ground which involves a greater understanding of the funds you hold and their contenmts et al, which once understood, let;s you sleep better at night.   

Leaving aside tracking error, index funds, will as you say, match the index (less fees) year after year not beating the index and conversely not losing to it.

You mentioned your Artemis funds have beaten the market over a 6 month period. Have any of your selections underperformed the market over the same time? If not, what about the previous 6 months or the next 6 months? From a total return point of view it is only returns of the entire portfolio that are important.

OldScientist

chiang_mai said:

It's time to wrap this up, I think. What I've got out of this thread is that most people appear unaware of the underlying internal details of funds and are happy to buy packages off the shelf, presumably on the basis of their reputation or as the result of product comparisons. This has come as a surprise. If they are aware, they don't appear to pay much attention to the metrics data. This would go some way towards explaining why platforms regularly publish the top 10 most purchased/most profitable funds etc. It also explains the limited availability of accurate fund data on some UK platform sites. Americans may regard things differently, perhaps, Morningstar publishes lots of fund data so there must be a reason and a demand.  

It's unfortunate that some see it as a competition between trackers and managed funds, despite having said repeatedly that it isn't that at all and that they are complementary, there is still something of a purist view that prevails in some quarters.

If I've learmed anything about investing this year, the April slump has reinforced the need for investers to manage the degree to which they invest in different markets. This would suggest that a single global fund that spans all markets based on market sizes, is better off being replaced by individual regional funds that can be tilted according to risk tolerance. This year, for the first time, my US holdings are at 25%, UK is at 19%, Europe is at 18%, Dev Asia is at 8% and EM is around 18% (with half of that allocated to China). By using separate funds per region I can easily see which ones are generating good returns and which are laggards, it also means I can derisk slightly by tilting in and out of regions as I see fit....EM has produced my best returns this year and has often moved inversly to other markets.

Thank you for your comments and the discussion. 

I would heartily disagree with your summary as a distortion of the passive (bogleheads) approach. The underlying details of a cap weighted equity index fund are trivial to understand. They hold securities in proportion to their cap weight according to the selection criteria of the index (region, cap size, etc.). In which case, while there are some slight differences between index providers (e.g., FTSE and MSCI), the sole criterion then becomes one of fees (which, once down to 20bp makes little difference) and the proportions in which to hold major asset classes (equities, bonds, others).

The metrics data is historical data and is not an indication of future performance - because an equity fund has had an annualised return over the previous 5 years of 10% does not mean that it will return 10% over the next 5 years. This goes for individual regions too - because EM is doing well this year does not mean it will do well next year, and so on.

More interesting (to me) is your implied question, is there a rationale for moving away from the global portfolio and constructing you own set of regional weightings? On diversification grounds (i.e., moving away from the so-called magnificent 7 which are now ~20% of the global index) may be there is. However, what that set of weightings should be based on (ad-hoc, guess work, analysis of PE ratios or something else?) is, to me, non-trivial since it would involve a political and economic assessment of future returns on a regional basis not an investigation of past returns (which are not a good predictor of the future). I do note that, to date, underweighting the US 5 years ago would have resulted in lower returns than the global index.

coyrls

chiang_mai said:

It's time to wrap this up, I think. What I've got out of this thread is that most people appear unaware of the underlying internal details of funds and are happy to buy packages off the shelf, presumably on the basis of their reputation or as the result of product comparisons. This has come as a surprise. If they are aware, they don't appear to pay much attention to the metrics data. This would go some way towards explaining why platforms regularly publish the top 10 most purchased/most profitable funds etc. It also explains the limited availability of accurate fund data on some UK platform sites. Americans may regard things differently, perhaps, Morningstar publishes lots of fund data so there must be a reason and a demand.  

It's unfortunate that some see it as a competition between trackers and managed funds, despite having said repeatedly that it isn't that at all and that they are complementary, there is still something of a purist view that prevails in some quarters.

If I've learmed anything about investing this year, the April slump has reinforced the need for investers to manage the degree to which they invest in different markets. This would suggest that a single global fund that spans all markets based on market sizes, is better off being replaced by individual regional funds that can be tilted according to risk tolerance. This year, for the first time, my US holdings are at 25%, UK is at 19%, Europe is at 18%, Dev Asia is at 8% and EM is around 18% (with half of that allocated to China). By using separate funds per region I can easily see which ones are generating good returns and which are laggards, it also means I can derisk slightly by tilting in and out of regions as I see fit....EM has produced my best returns this year and has often moved inversly to other markets.

Thank you for your comments and the discussion. 

I am not sure that your conclusion is what you got out of this thread, I suspect that was your opinion before you started the thread.  I wouldn't have drawn the same conclusion as you from this thread.  Where did anyone say they bought funds on the basis of reputation or product comparisons?  That is,as you say, your presumption and I doubt it was a surprise, as I say, I suspect it was already your opinion before you started the thread.

The conclusion that I would draw is that some contributors are suspicious of using backward facing metrics and ratings to predict future performance and don't think that you can know when to sell before performance drops off whereas you are in favour of such an approach.

thunderroad88

“it would involve a political and economic assessment of future returns on a regional basis”

Isn’t that what fund managers do, so if a manager has demonstrated consistently over a respectable period of time that they can make those assessments successfully, isn’t it reasonable to think they have as good a chance as an index of producing average or even better results?

“Where did anyone say they bought funds on the basis of reputation or product comparisons?”

I think many do (they’d be silly not to consider comparative performance)… many put fund cost first too, even when the % difference in performance is significantly more than the % difference in cost

Dead_keen

chiang_mai said:

most people appear unaware of the underlying internal details of funds 

I suspect you are right.  I used to design the structure of funds (when to use LP/Ltds/SCSp/LLPs/etc for what roles and in what jurisdictions, etc) but it's quite niche.  I must admit I've never designed an ETF but that's my loss.

chiang_mai said:

If they are aware, they don't appear to pay much attention to the metrics data

That's me.  I am aware but pay almost no attention to fund metrics. I'm not really sure which ones are going to be relevant and so don't worry about it.  I do look at the real performance of my portfolio every now and then but only to sense check my consumption.

chiang_mai said:

If I've learmed anything about investing this year,

It's good that you are still learning stuff.  I wonder what you'll learn next year that will change what you thought you'd learned this year?

chiang_mai said:

This would suggest that a single global fund that spans all markets based on market sizes, is better off being replaced by individual regional funds that can be tilted according to risk tolerance.

Sorry, but this reminds me of the first rule of fight Dunning-Kruger club.

Eyeful

thunderroad88 said:

“it would involve a political and economic assessment of future returns on a regional basis”

Isn’t that what fund managers do, so if a manager has demonstrated consistently over a respectable period of time that they can make those assessments successfully, isn’t it reasonable to think they have as good a chance as an index of producing average or even better results?

“Where did anyone say they bought funds on the basis of reputation or product comparisons?”

I think many do (they’d be silly not to consider comparative performance)… many put fund cost first too, even when the % difference in performance is significantly more than the % difference in cost

1. What do you define as a respectable period of time 10, 15, 20,30 years?

2. Academic research shows that out performance which was once attributed to "Star Managers" is  primarily due to factors such as luck, market efficiency, and statistical reversion rather than consistent skill or unique insights

 3. As academic research repeatedly shows that, after fees and charges are taken into account,
most fund managers cannot match a simple major global index fund.

4. Many put fund costs first, because over long periods of time say 30 or 40 years, they have a significant effect on the amount of money that ends up in your pocket and not in that of the fund manager.
Also cost is something you do have control over. You cannot foretell future fund performance    

5. The active fund industry did for a long time, fool the average investor into thinking that the costs they where paying where small and worth while. Now we know better.
   See your own T=Rex score: https://larrybates.ca/t-rex-score/

6. Can you tell me which active fund manager has for 20 years consistently outperformed lets say the ACWI ?

Linton

thunderroad88 said:

1) “it would involve a political and economic assessment of future returns on a regional basis”

Isn’t that what fund managers do, so if a manager has demonstrated consistently over a respectable period of time that they can make those assessments successfully, isn’t it reasonable to think they have as good a chance as an index of producing average or even better results?

2) “Where did anyone say they bought funds on the basis of reputation or product comparisons?”

I think many do (they’d be silly not to consider comparative performance)… many put fund cost first too, even when the % difference in performance is significantly more than the % difference in cost

1) I strongly believe that the future is unknowable and investing on the basis of anyone's opinions on what the future will bring is as likely to be counter-productive as it is to be beneficial. It is a risk I would not want to take.

What I want from an active equity fund manager is a steady long term investment strategy within their remit and in some markets, filtering of the dross. Then the minimum set of necessary funds can be chosen on the basis of how well they provide the overall portfolio with a highly diversified breadth of coverage of the various categories into which equity can be divided.  In particular: country, sector, size of underlying companies, value vs growth, largest underlying investments and anything else which can be readily determined from the fund portfolio data given by Morningstar.

2) As seen from (1) reputation is irrelevent, everything depends on the underlying investments. Historic performance is very much a secondary factor though consideration of major differences between apparently similar funds can give useful insight.

So in some respects this is similar to the OPs approach though in others very different.  However our objectives are different.  It would seem the OP is after maximum return in the long term for family reasons whilst I am after sufficient greater than inflation returns to ensure our well being in the medium term. Sadly the long term is rapidly disappearing.

leosayer

chiang_mai said:

If I've learmed anything about investing this year, the April slump has reinforced the need for investers to manage the degree to which they invest in different markets. This would suggest that a single global fund that spans all markets based on market sizes, is better off being replaced by individual regional funds that can be tilted according to risk tolerance. This year, for the first time, my US holdings are at 25%, UK is at 19%, Europe is at 18%, Dev Asia is at 8% and EM is around 18% (with half of that allocated to China). By using separate funds per region I can easily see which ones are generating good returns and which are laggards, it also means I can derisk slightly by tilting in and out of regions as I see fit....EM has produced my best returns this year and has often moved inversly to other markets.

The above begs a number of questions such as:

How do you assess and measure your risk tolerance? 

How do you apply your risk tolerance to the metrics you use to decide on an eventual fund/region allocation?

Do you have a system that makes these calculations and if so how often do you run the calcs and how often do you change it?

Have you compared the performance and volatility of your overall portfolio to a global equity index?

I commend you for what you have done but the approach you outline above seems well beyond the capabilities of  most investors, even reasonably sophisticated ones. 

If this approach helps you to sleep at night then great, but after 30+ years of investing I find anything beyond a 2 fund gilt / global equity allocation causes me too much stress and inevitably, a worse outcome.

Linton

Eyeful said:

thunderroad88 said:

“it would involve a political and economic assessment of future returns on a regional basis”

Isn’t that what fund managers do, so if a manager has demonstrated consistently over a respectable period of time that they can make those assessments successfully, isn’t it reasonable to think they have as good a chance as an index of producing average or even better results?

“Where did anyone say they bought funds on the basis of reputation or product comparisons?”

I think many do (they’d be silly not to consider comparative performance)… many put fund cost first too, even when the % difference in performance is significantly more than the % difference in cost

1. What do you define as a respectable period of time 10, 15, 20,30 years?

2. Academic research shows that out performance which was once attributed to "Star Managers" is  primarily due to factors such as luck, market efficiency, and statistical reversion rather than consistent skill or unique insights

 3. As academic research repeatedly shows that, after fees and charges are taken into account,
most fund managers cannot match a simple major global index fund.

4. Many put fund costs first, because over long periods of time say 30 or 40 years, they have a significant effect on the amount of money that ends up in your pocket and not in that of the fund manager.
Also cost is something you do have control over. You cannot foretell future fund performance    

5. The active fund industry did for a long time, fool the average investor into thinking that the costs they where paying where small and worth while. Now we know better.
   See your own T=Rex score: https://larrybates.ca/t-rex-score/

6. Can you tell me which active fund manager has for 20 years consistently outperformed lets say the ACWI ?

You and the OP are focussing on maximum long term return. I agree that comsistently beating the index over the long term is impossible and is probably due to luck in the short term. For some people, particularly those dependent on their investments to support their long term well-being, maximum return is less important than stable sufficient return. This reduces the importance of low cost.   Different requirements may give rise to different strategies. 

Cus

Eyeful said:

thunderroad88 said:

“it would involve a political and economic assessment of future returns on a regional basis”

Isn’t that what fund managers do, so if a manager has demonstrated consistently over a respectable period of time that they can make those assessments successfully, isn’t it reasonable to think they have as good a chance as an index of producing average or even better results?

“Where did anyone say they bought funds on the basis of reputation or product comparisons?”

I think many do (they’d be silly not to consider comparative performance)… many put fund cost first too, even when the % difference in performance is significantly more than the % difference in cost

1. What do you define as a respectable period of time 10, 15, 20,30 years?

2. Academic research shows that out performance which was once attributed to "Star Managers" is  primarily due to factors such as luck, market efficiency, and statistical reversion rather than consistent skill or unique insights

 3. As academic research repeatedly shows that, after fees and charges are taken into account,
most fund managers cannot match a simple major global index fund.

4. Many put fund costs first, because over long periods of time say 30 or 40 years, they have a significant effect on the amount of money that ends up in your pocket and not in that of the fund manager.
Also cost is something you do have control over. You cannot foretell future fund performance    

5. The active fund industry did for a long time, fool the average investor into thinking that the costs they where paying where small and worth while. Now we know better.
   See your own T=Rex score: https://larrybates.ca/t-rex-score/

6. Can you tell me which active fund manager has for 20 years consistently outperformed lets say the ACWI ?

The OP is discussing how to select funds (and therefore fund managers), and analysis on how to move from one fund to another etc, not how to choose a fund for 20 years.

For point 2/3, this depends on the funds aims and remit, not a relevant comparison.  I have read a lot of SPIVA research, always a bit uneasy as they are an index data company relying on selling their data, so makes me worry they are not unbiased.  It also highlights how in non US large cap fund areas the data is less convincing.
I can't find anything that shows this info for periods before we had such a global concentration in a few mega companies that to me obviously distorts results.

I'm convinced the active fund industry was ripping off the average investor and no doubt the average investor should just go for the global index based passive investing.  I'm also sure that a lot of other elements of the market (etf producers, index data providers, fund of funds providers) are very happy to take the fees and continue to promote the change of perception, they are not here for our benefit either. But we are in a better place visibility wise now and that's good for all.

Just because I couldn't understand the details to a very advanced level of the fund markets won't make me jump onto a different belief that it's 100% random and pointless and I must accept that no-one can beat markets.  I don't need that comfort, and can accept that others might be able to do it.