How many indices !

[Lootman]

Risk managers are well aware of domino effects! I also still maintain that exposure to an AAA rated exchange, with all positions continuously margined is still much lower risk than exposure against a single counterparty such as a bank. To argue otherwise is to argue against the entire basis of diversification, insurance and financial exchanges.

If you think your risk is with an exchange you are looking in the wrong place.

Ps central bank monitoring of banks has worked very well in the past hasn't it?

We both see this because we are neutral market participants. Ohno is not - he manages a very small "smart beta" passive fund that is an open-ended fund, in a market where nearly all such funds are ETFs. The rise of ETFs has quite simply eaten his lunch and so it is hardly surprising that he conjures up spurious arguments to put down ETFs.

Now I am biased as well since I worked on some of the early ETFs for a well known ETF provider. But then that is a part of how and why I know they are much safer than Ohno claims, and probably safer than his fund whether derivatives are used or not.

ETFs are also much cheaper than his fund which, the last time I looked, had expenses of over 1% per annum, not to mention third quartile performance. Meanwhile Fidelity now has zero cost ETFs in the US.

[OhNoNotimAgain]

Risk managers are well aware of domino effects! I also still maintain that exposure to an AAA rated exchange, with all positions continuously margined is still much lower risk than exposure against a single counterparty such as a bank. To argue otherwise is to argue against the entire basis of diversification, insurance and financial exchanges.

If you think your risk is with an exchange you are looking in the wrong place.

Ps central bank monitoring of banks has worked very well in the past hasn't it?

We both see this because we are neutral market participants.

Now I am biased as well since I worked on some of the early ETFs for a well known ETF provider.

ETFs are also much cheaper

The argument about risk will never be concluded until it is tested in real life.

However, the propoganda about ETFs really needs to be exposed for the blarney it really is.

Morningstar quotes these figures this morning for the 3 year returns from 20/11/2015

SPDR FTSE All Share ETF 121.64
HSBC FTSE All Share Index OEIC 123.35

ETFs are more expensive than OEICs because they have to feed more mouths, namely the broker. They also have to held via a platform which adds about 40bps to the cost. It is possible to buy an OEIC direct from the provider this skipping the platform fee giving a higher net return to the investor.

The industry loves ETFs because it keeps them busy but gives the appearance of reducing costs. The reality is otherwise, but no intermediary will tell you that.

[hiriskpaul]

Well SPDR has a much higher management fee which does not help, but single examples don't prove anything. Comparing 2 S&P 500 funds from the same manager with similar charges gives much closer results. 5 year annualised GBP returns from Morningstar:

iShares Core S&P 500 UCITS ETF USD (Acc) 15.23%
iShares US Equity Index Fund (UK) D Acc 15.24%

A difference of 1bps per year! Well within tracking error. These results are for the cheapest retail share class for the OEIC and the accumulating version of the ETF.

I have found errors in the Morningstar data in the past, so would want to see verification of the data before jumping to any conclusions about ETFs vs funds. Having said that, it would not surprise me to find that for long term holdings, OEICs had an edge over ETFs, provided everything else was the same. i.e. same management and platform charges, same returns from stock lending, same teams managing the funds. I prefer to hold OEICs when I can. The difficulty with OEICs is the platform fee charged by many brokers. For example, my SIPP is with Hargreaves Lansdown. Much as I would like to hold OEICs in my SIPP, there is no way I am going to pay an extra 0.45% to do so, compared to a capped fee of £200 per year to hold ETFs and all the other non-OEIC securities I have in my SIPP.

[Lootman]

The difficulty with OEICs is the platform fee charged by many brokers. For example, my SIPP is with Hargreaves Lansdown. Much as I would like to hold OEICs in my SIPP, there is no way I am going to pay an extra 0.45% to do so, compared to a capped fee of £200 per year to hold ETFs and all the other non-OEIC securities I have in my SIPP.

I haven't held an OEIC for at least 15 years. I have individual shares, ITs and ETFs, all of which give me something a little different. I really do not see an OEIC giving me anything extra, and their expense ratios are typically higher. The one exception I might make is for the Smith fund, but then I hold most of what that fund holds as individual shares.

As you note, OEICs are just peskier to hold. You either have to pay a fee to a platform, or else hold directly with the fund manager which means more work for taxes.

I also used to work for an active fund manager and have a few stories about what goes on behind the scenes, which also makes me not trust them as much. The passive fund manager I worked more was much more ethical, and anyway had less room to deviate from the straight and narrow.

[OhNoNotimAgain]

Well SPDR has a much higher management fee which does not help, but single examples don't prove anything. Comparing 2 S&P 500 funds from the same manager with similar charges gives much closer results. 5 year annualised GBP returns from Morningstar:

iShares Core S&P 500 UCITS ETF USD (Acc) 15.23%
iShares US Equity Index Fund (UK) D Acc 15.24%

A difference of 1bps per year! Well within tracking error. These results are for the cheapest retail share class for the OEIC and the accumulating version of the ETF.

I have found errors in the Morningstar data in the past, so would want to see verification of the data before jumping to any conclusions about ETFs vs funds. Having said that, it would not surprise me to find that for long term holdings, OEICs had an edge over ETFs, provided everything else was the same. i.e. same management and platform charges, same returns from stock lending, same teams managing the funds. I prefer to hold OEICs when I can. The difficulty with OEICs is the platform fee charged by many brokers. For example, my SIPP is with Hargreaves Lansdown. Much as I would like to hold OEICs in my SIPP, there is no way I am going to pay an extra 0.45% to do so, compared to a capped fee of £200 per year to hold ETFs and all the other non-OEIC securities I have in my SIPP.

How many markets and time frames did you look at before finding an example that suited your argument?

And how strange that an intermediary has a pricing structure that favours ETFs over OEICs.

[hiriskpaul]

How many markets and time frames did you look at before finding an example that suited your argument?

And how strange that an intermediary has a pricing structure that favours ETFs over OEICs.

I could ask you exact the same question! My argument was not that you were wrong by the way, I don't know whether you are or not, just that an example of 1 is insufficient evidence.

As it happens the iShares S&P 500 example was the first one that came in to my head to look at. Similar charges, the same manager and both having a 5 year history. My selection also turned out to be flawed as the iShares OEIC does not follow the S&P 500! Instead it follows a FTSE USA index, which seems to be wider than the S&P 500.

It is an interest question though as to which structure has the edge for long term investment. There are now loads of S&P 500 trackers with 5+ years history, so I will take a look. Very few FTSE allshare ETFS, but quite a few FTSE 100, so I will try to look at those as well.

[Lootman]

It is an interesting question though as to which structure has the edge for long term investment.

Take costs into account and the matter becomes clearer. There are now zero cost ETFs. And on the institutional side even negative cost ETFs when the stock lending revenues are split between fund manager and client.

Look at a long-term graph between SPY and the Vanguard S&P 500 OE fund. The difference is negligible.

I have some issues with ETNs and the use of OTC swaps. If you owned a Lehman ETN you might have been in trouble. But not their ETFs. And ETFs in illiquid markets can have issues - a few emerging market exchanges have closed for a while (Kuala Lumpur, Cairo, Lima) and those ETFs went to big discounts. There again an open-ended or closed-ended fund would also have been in trouble as well.

I do worry about ETFs in junk bonds and muni bonds in extreme conditions but, again, how would an open-ended fund handle redemptions or a closed-end fund avoid a discount?

There are now trillions of dollars in ETFs. If they fail, everything is failing, and Ohno's tiddler fund will go to the wall as well, with no big institution to bail it out

[hiriskpaul]

It is an interesting question though as to which structure has the edge for long term investment.

Take costs into account and the matter becomes clearer. There are now zero cost ETFs. And on the institutional side even negative cost ETFs when the stock lending revenues are split between fund manager and client.

Look at a long-term graph between SPY and the Vanguard S&P 500 OE fund. The difference is negligible.

I have some issues with ETNs and the use of OTC swaps. If you owned a Lehman ETN you might have been in trouble. But not their ETFs. And ETFs in illiquid markets can have issues - a few emerging market exchanges have closed for a while (Kuala Lumpur, Cairo, Lima) and those ETFs went to big discounts. There again an open-ended or closed-ended fund would also have been in trouble as well.

I do worry about ETFs in junk bonds and muni bonds in extreme conditions but, again, how would an open-ended fund handle redemptions or a closed-end fund avoid a discount?

There are now trillions of dollars in ETFs. If they fail, everything is failing, and Ohno's tiddler fund will go to the wall as well, with no big institution to bail it out

With ETFS of illiquid securities, the mechanism to arbitrage away the discount/premium can break down in difficult conditions, but investors should still be able to trade in the secondary market. Such break downs has even happened in liquid markets for brief periods. With OEICS though, there is a risk that the fund managers will stop redemptions. This has happened in the past, e.g. with property funds, so even if someone wants to get out, they will not be able to.

[torata]

And how strange that an intermediary has a pricing structure that favours ETFs over OEICs.

AJ Bell / YouInvest also has a increased pricing for holding OEICs (0.5%, from memory), and it's often mentioned as a consideration when the discussion of ETFs vs OEICs comes up on this and other boards.

torata

[Lootman]

And how strange that an intermediary has a pricing structure that favours ETFs over OEICs.

AJ Bell / YouInvest also has a increased pricing for holding OEICs (0.5%, from memory), and it's often mentioned as a consideration when the discussion of ETFs vs OEICs comes up on this and other boards.

I know nothing about such platforms but would imagine that the costs and risks involved with settling fund transactions with a fund manager are much greater than exchange-traded securities.

[OhNoNotimAgain]

And how strange that an intermediary has a pricing structure that favours ETFs over OEICs.

AJ Bell / YouInvest also has a increased pricing for holding OEICs (0.5%, from memory), and it's often mentioned as a consideration when the discussion of ETFs vs OEICs comes up on this and other boards.

I know nothing about such platforms but would imagine that the costs and risks involved with settling fund transactions with a fund manager are much greater than exchange-traded securities.

All settlements are handled by the custodian bank under the eyes of the Trustee so I see no reason why that should be more complex or expensive.

[Lootman]

AJ Bell / YouInvest also has a increased pricing for holding OEICs (0.5%, from memory), and it's often mentioned as a consideration when the discussion of ETFs vs OEICs comes up on this and other boards.

I know nothing about such platforms but would imagine that the costs and risks involved with settling fund transactions with a fund manager are much greater than exchange-traded securities.

All settlements are handled by the custodian bank under the eyes of the Trustee so I see no reason why that should be more complex or expensive.

To use your argument against you, its adds an extra layer that I see no reason for.

What does an OE fund give me that I can't get from ITs and ETFs? Other than higher expenses.

[OhNoNotimAgain]

I know nothing about such platforms but would imagine that the costs and risks involved with settling fund transactions with a fund manager are much greater than exchange-traded securities.

All settlements are handled by the custodian bank under the eyes of the Trustee so I see no reason why that should be more complex or expensive.

To use your argument against you, its adds an extra layer that I see no reason for.

What does an OE fund give me that I can't get from ITs and ETFs? Other than higher expenses.

Simplicity

[Lootman]

All settlements are handled by the custodian bank under the eyes of the Trustee so I see no reason why that should be more complex or expensive.

To use your argument against you, its adds an extra layer that I see no reason for.

What does an OE fund give me that I can't get from ITs and ETFs? Other than higher expenses.

Simplicity

Well, sure, OE funds are marketed to the least sophisticated investors, I will concede that.

[OhNoNotimAgain]

To use your argument against you, its adds an extra layer that I see no reason for.

What does an OE fund give me that I can't get from ITs and ETFs? Other than higher expenses.

Simplicity

Well, sure, OE funds are marketed to the least sophisticated investors, I will concede that.

Yep, that's me. All I understand is dividends and compound interest.

[hiriskpaul]

I had a look at the 5 year returns of S&P 500 OEICs and ETFs. No point going back further than that as trail commission was included in OEICs before 2013 and there are not many funds with a longer trail history anyway. Returns are annualised, in GBP and provided by Morningstar. I could only find 3 OEICs, which is disappointing, as other trackers follow different US indices. Here are the raw results:





So on the face of it, ETFs look like they do better than OEICs!

However, the unpaired t-test for the above comes out at 1.27, implying a 23% probability the result could be just random chance. Furthermore, the Pictet OEIC has quite a high TER at 0.3%, which drags down the OEIC results. Eliminating that OEIC improves the OEIC mean to 14.97 and the probability of random chance of the ETFs beating the OEICs goes up to 80%. Essentially, the best that can be said on whether ETFs or OEICs give better long term returns is "Don't know".

[hiriskpaul]

I get much the same story when comparing FTSE 100 OEICs and ETFs. I could only find 3 OEICs with 5 year histories that did not have obscene charges:



7 ETFs with charges at or below 0.20%:



So, in this case OEICs beat ETFs by 0.02% per year. The t value came out as 0.27, giving a probability that this was random of 80%.

If there is any difference in long term returns between ETFs and OEICs it looks it though it is very small and lost in the tracker error and differences between funds.

[OhNoNotimAgain]

Good piece of work Paul, is it pushing things to ask you to do the same for the FT All Share?

[hiriskpaul]

I will take a look. The difficulty is the lack of low charging ETFS that cover the all share. I might be able to correct for that though.

[OhNoNotimAgain]

I will take a look. The difficulty is the lack of low charging ETFS that cover the all share. I might be able to correct for that though.

Thanks, but I would be interested to see the whole market.