Jonathan Reynolds from Capital Cranfield, ETFGI's Deborah Fuhr, iShares' Mark Johnson, JLT's Peter Martin and Simon Riviere of PTL discuss the costs of exchange traded funds in the second part of this three-part debate.
Mark Johnson: The reality is it costs more to manage an ETF than it does an index fund for institutional investors. There are costs associated with developing an ETF, with running an ETF on an exchange, with listing fees, indices and so forth. But ETFs have a key difference, which is the ability to trade them via the secondary market.
We encourage investors to look at the total cost of ownership. That is to say, the cost of buying into the ETF, holding it and exiting the ETF.
Once you have taken into account the cost-benefit analysis for the period you are holding, the underlying fund management fees of an ETF and the bid-offer spreads, for less than a year you might be better off with an ETF than a traditional passive product
Peter Martin, JLT
If you simply look at one aspect of that, which is the cost of holding an ETF versus an institutional fund, you are only seeing one part of the picture. What we will be able to do for clients is map those two costs to identify the break-even period, and to be able to say to you, depending on your timeframe, there is an optimal way of accessing that asset class.
It may be that for, say, up to two years, the ETF is more cost efficient and after two years, depending on your holding-period requirements, a pooled fund may make more sense, as the impact of the lower TER starts to have an effect over time.
Jonathan Reynolds: I would say that is highly relevant for an investment manager, but I am not entirely sure it is particularly relevant for a trustee. For a trustee, I hope I am not going to be trading ETFs on a regular basis and making tactical calls.
Johnson: You are not, and I would agree with you. But I think of a large trade we did last year for a UK defined benefit plan – they were between active managers, they were looking to identify a new home for monies that had been earmarked for emerging markets, but they knew it was going to take them some time to identify a new active strategy. I would therefore agree with you that your typical institutional investor is not looking for short-term liquid beta.
Peter Martin: A lot of what we do as consultants for trustees is long-term strategic allocations, so the holding periods will generally be much longer than two years.
If it is your starting point, depending on your governance budget, that may rule out ETFs for the majority of those types of uses. But if you were to look at shorter-term holding periods I would very much agree with Mark. Once you have taken into account the cost-benefit analysis for the period you are holding, the underlying fund management fees of an ETF and the bid-offer spreads, for less than a year you might be better off with an ETF than a traditional passive product, if that was available to you.
But if you are a trustee for most pension funds you do not use short-term tactical holding periods. It may be that some of the larger schemes, if they do wish to make a shortish-term market call, use dynamic strategic allocation, of which you see discussion now.
The ongoing underlying fees are higher in comparison with most traditional passive products. For long-term strategic holdings we therefore normally say no to ETFs for most mainstream asset classes.
So for traditional equities it may be the right thing to do. But it might be more esoteric areas, more niche areas, where ETFs can give you that exposure.
Deborah Fuhr: Depending on the sophistication of the pension fund, one of the things you can do with ETFs that you cannot do with mutual funds is lend the ETF shares. If you lend stock you can lend the shares of an ETF you own. That will quite often mean the annual cost of owning an ETF can actually be less expensive at times than being in a segregated account or some other commingled products.
Simon Riviere: That brings a risk in itself. If you are lending to a third party you then have the counterparty risk that whoever you are lending to potentially cannot then let you have the stocks back.
When you talk about ETFs being expensive, the reality is that active managers can be much more expensive and you are not always getting that better than benchmark performance
Deborah Fuhr, ETFGI
Fuhr: In a typical lending programme you would have someone who is responsible for checking and approving who you lend to. You would also be responsible for taking on collateral. So, typically, you would over-collateralise anything that is leant out. It gets marked to market every night and that would be managed so that if something was to happen to someone who borrowed, you would be able to sell that collateral you own to buy the underlying shares if that was necessary.
Moreover, we have seen that process work during the Lehman situation. So there are ways ETFs can be used, if you do that already with shares, that could make ETFs perform differently and better than some other solutions.
If we look at the fees, the average right now in Europe for ETFs is 35 basis points. So they are pretty low and, as we heard, they provide exposure to many benchmarks you would not find in a commingled set of alternatives.
You say that many investors prefer active management. If you look at studies that measure performance of active managers against their benchmarks, especially over a longer time horizon, you tend to find, on average, six to seven out of 10 managers do not consistently beat the benchmark. So when you talk about ETFs being expensive, the reality is that active managers can be much more expensive and you are not always getting that better than benchmark performance.
We have to be careful. People want active and better performance, but I think that has led to a new type of ETF. I do not particularly like the term ‘smart beta’ because it implies anyone who is an index manager is dumb. But you do find that today the type of index products you can gain exposure to are beyond just market cap strategies.
Whether you want minimum volatility, whether you want equal weighting, whether you are looking for yield – there are six academically supported types of factors that deliver performance typically better than market cap. Many of those are now delivered through ETF wrappers. What is more, many of the larger consulting firms have said for many years that using passive strategies that are not based on market cap would be preferable for many of their consulting clients. This is an area that we have seen ETFs move into over the past couple of years.
