Tim Hale - Smarter Investing

guymo

dunstonh wrote: »

Cash and an index tracker would likely be bad advice for most people. Single sector investing is not considered appropriate nowadays unless justified as to why multi-asset was not suitable.

What about cash and a collection of trackers? Say, something comparable to the Vanguard LifeStrategy 100% equity? Does that cover a range of asset classes that you would be happy with, or is more needed and if so, what?

Glen_Clark

dunstonh wrote: »

Cash and an index tracker would likely be bad advice for most people. Single sector investing is not considered appropriate nowadays

But I never suggested single sector investing, did I?
VWRL invests in over 3,000 companies worldwide

dunstonh

What about cash and a collection of trackers?

That is fine. However, you start building a portfolio and that is less simple and Glen is talking about simple options.

Say, something comparable to the Vanguard LifeStrategy 100% equity? Does that cover a range of asset classes that you would be happy with, or is more needed and if so, what?

VLS is fine. Not sure about 100% version as despite investing millions into the VLS funds, I have yet to have anyone in the 100% version. However, the theory is fine. VLS 80% or lower is multi asset. It also has a element of management (how do they decide the allocations for example). For simple, you cant get much more simple than VLS.

pip895

guymo wrote: »

What about cash and a collection of trackers? Say, something comparable to the Vanguard LifeStrategy 100% equity? Does that cover a range of asset classes that you would be happy with, or is more needed and if so, what?

Smaller Companies

Marazan

guymo wrote: »

If you could guarantee a tiny tracking error then no further study would really be needed, but the Which? report gives some very worrying information about the variation in performance of passive funds, and I think this deserves a closer look.

I would be astounded if the dominant component behind an index fund's tracking error was not their AMC/TER.

EDIT: Which is not to say I think it's the only factor but all other things being equal the bigger the charge the bigger the tracking error must be.

Rollinghome

guymo wrote: »

...but the Which? report gives some very worrying information about the variation in performance of passive funds, and I think this deserves a closer look.

Doesn't need an especially close look. Just read the rest of the paragraph you quoted.

You quoted "we also found striking differences in the performance of tracker funds aiming to match the index. The best tracker, M&G UK Index, which has an ongoing charge of 0.46%, returned 160%. The worst performer, Halifax UK FTSE All Share Index, achieved only 136%".

The full paragraph was "However, we also found striking differences in the performance of tracker funds aiming to match the index. The best tracker, M&G UK Index, which has an ongoing charge of 0.46%, returned 160%. The worst performer, Halifax UK FTSE All Share Index, achieved only 136%. Until October last year, the Halifax fund charged 1.5%, which explains its poor record. It has now reduced the charge to 1%."

Tracker funds will by and large return the index they track minus the charges. There is absolutely no sense in buying a tracker fund such as those offered by Halifax or Virgin that charge more than the 0.25% or so charged by competitors. For HBOS to charge 1.5% for a tracker fund is cynicism bordering on dishonesty.

On the whole it's a fairly rubbish article. You can't make useful comparisons over 10 years unless comparing like with like and without making adjustments for changes in charges - and that's fairly complicated. Annual charges for trackers have gone down in recent years (HSBC dropped the AMC of their trackers from around 1% to 0.25% in 2009) while annual charges for managed funds have gone up. However while most people would have paid an initial charge of 5% on managed funds that would need to be taken into account a few years ago that can usually be avoided now. A further complication is that come April the cheapest trackers will have the cost of a platform fee that will need to be taken into account.

If you want to estimate future returns you need to know current and future costs. Looking at past returns without taking into account changes in costs isn't especially useful.

guymo

Marazan wrote: »

I would be astounded if the dominant component behind an index fund's tracking error was not their AMC/TER.

EDIT: Which is not to say I think it's the only factor but all other things being equal the bigger the charge the bigger the tracking error must be.

Okay, let's take a look. (Thanks for the nudge --- previously I had computed rough approximations to the below in my head and gone a bit wrong.)

If my calculations are correct, using the figures from the Which? article because I have them to hand:

Market returned 171% which is 5.5% annualised.
M&G's tracker returned 160% which is 4.8% annualised, so a tracking error of 0.7% of which 0.46% is the OCF
Halifax's tracker returned 136% which is 3.1% annualised, making a tracking error of 2.4% of which 1.5% is the OCF.

So the M&G tracker looks pretty solid, while the Halifax one looks both expensive and inaccurate.

I still think the question of how many active funds outperform decent trackers could be interesting; and more importantly, how they are distributed. Also interesting would be whether an accurate tracker is likely to stay accurate --- probably it is, since I imagine there is a good deal of automation, isn't there?

guymo

Rollinghome wrote: »

Doesn't need an especially close look. Just read the rest of the paragraph you quoted.

...

The full paragraph was "However, we also found striking differences in the performance of tracker funds aiming to match the index. The best tracker, M&G UK Index, which has an ongoing charge of 0.46%, returned 160%. The worst performer, Halifax UK FTSE All Share Index, achieved only 136%. Until October last year, the Halifax fund charged 1.5%, which explains its poor record. It has now reduced the charge to 1%."

Tracker funds will by and large return the index they track minus the charges.

Thanks for this. I don't think the performance of the Halifax fund is adequately explained by its high costs in fact --- its error is 2.4% of which 1.5% is the charge, so there's a lot missing in between. (Or I've made a mistake, which is very possible.)

But I agree that the article is missing a lot of information and hasn't properly cleaned up the data.

This is a useful discussion, at least for me, so thanks to everyone for participating.

bigadaj

guymo wrote: »

Thanks for this. I don't think the performance of the Halifax fund is adequately explained by its high costs in fact --- its error is 2.4% of which 1.5% is the charge, so there's a lot missing in between. (Or I've made a mistake, which is very possible.)

But I agree that the article is missing a lot of information and hasn't properly cleaned up the data.

This is a useful discussion, at least for me, so thanks to everyone for participating.

But won't there be the some compounding effect on the annual charges over a longer time period? This would explain the increasing tracking error for larger charges over longer time periods.

Of course tracking error is caused by other factors such as how the fund manages itself, things such as futures contracts, synthetic etf components, how frequently and when it changes it's constituents will all have an effect. The are cases where some trackers outperform their benchmark for certain time periods for these reasons. Indeed ther can be valid reasons for this, such as minimising dealing a nd other associated costs associated with churn.

guymo

bigadaj wrote: »

But won't there be the some compounding effect on the annual charges over a longer time period? This would explain the increasing tracking error for larger charges over longer time periods..

Of course errors compound. The calculations I made allow for that --- I calculated the annualised return of the market (5.5%) and the tracker fund (3.1%) by computing the tenth root of the quoted 10-year returns. The gap of 2.4% each year (on average) is partly down to the 1.5% charge and partly to… something else