ETF providers say custodians need to step up stock lending programmes to serve the European market. Our ETF operations panel discusses this obstacle to industry development, and highlights the problem of multi-market listings. Edited by Nick Fitzpatrick.
Andrew Jamieson, head of client execution sales, Emea, BlackRock
Michael John Lytle, chief development officer, Source
Fiona Moore, head of ETF fund administration, Emea, Northern Trust
Ivan Nicora, director, investment fund product management, Euroclear
Gillian Walmsley, head of ETFs, London Stock Exchange
Funds Europe: What challenges face ETF providers as they initiate a fund launch in Europe and how does this contrast with a traditional index fund?
Andrew Jamieson, BlackRock: The big difference is ETFs trade on the secondary market and so issuers have to engage with the market-making community and relevant broker/dealers.
There is significantly more secondary market participation by issuers because we want to monitor whether the ETF is priced appropriately and ensure there’s adequate liquidity. And also, from the start, providers need a seed partner. With the challenges that a lot of the banks are experiencing in being able to allocate balance sheet and capital to seeding today, there tends to be more proactive input and discussion required by the ETF issuer.
Gillian Walmsley, London Stock Exchange: One of the areas that is often talked about with fund providers coming to market is which market to select in terms of liquidity and how to address fragmentation in the European market.
Jamieson: Funds have to be registered in different markets, and ensuring we are launching in markets that would warrant the demand in the first place adds further complexity. Adding the market fragmentation issue also draws some of the centralised liquidity away, which is a problem in itself.
Michael John Lytle, Source ETF: There is an issue around transparency of transaction data. Disclosure of OTC [over-the-counter] trades is important. The LSE requires this; some other exchanges don’t. An important step in Europe would be a move towards capturing OTC data across the board. One of the biggest challenges investors face is just getting their heads around how liquid the products are. If there are 35 different Euro Stoxx 50 ETFs, can they all be equally liquid? In theory they are, but practically speaking there is this difference between implied liquidity and realised liquidity.
Walmsley: There is often a misconception that ETFs lack liquidity because not all of the volume is published due to the large OTC market – this was the key driver for our introduction of exchange rules to require printing of off-book trades so that investors see a fuller picture of true ETF liquidity. And we see this as part of a more general trend – as European regulation moves towards greater transparency and with Mifid II, where it is expected that there will be a requirement to print all ETF trading, this will be helpful in terms of showing the full picture of true ETF liquidity.
Lytle: Large market makers would prefer to have more data disclosed. That transparency would make it clear the size of the businesses that they’re building, what a good job they’re doing and which venue is the right place for trading a security. People who are core to the market benefit from transparency.
Fiona Moore, Northern Trust: Market fragmentation and the impact this has on settlement of the ETF and distribution is one of the key challenges faced. From an asset servicing perspective, making sure the right infrastructure is in place is a key requirement to help overcome these challenges.
The workflow for an ETF is more involved than a traditional mutual fund; there are a lot more linkages arising from the fact that ETFs are traded on multiple exchanges. It’s crucial that you have the right infrastructure in place in terms of monitoring this workflow and also to facilitate prompt reporting.
Infrastructure from an asset servicing perspective needs to be scalable and adaptable to ensure an ETF provider can capitalise on innovation.
Lytle: Scalability is key; in Europe this has not been challenged but hopefully will be. We worked with Bank of Ireland for nine months, before it was purchased by Northern Trust, building the infrastructure to support Source. This might have been surprising to most market observers because Bank of Ireland at the time was already supporting iShares. You’d have thought their solution would have worked right out of the box. But the reality is that different funds have different requirements. If we hadn’t worked to build an infrastructure upfront that could handle scale, we could have been faced with a significant problem in the future with no time to fix it.
Ivan Nicora, Euroclear: We’ve been dealing with ETFs for quite a long time and it’s only recently that we have started to hear from our clients about the hurdles they encounter when processing ETF trades across trading venues.
Post-trade processes in the US are centralised. But in Europe, there are multiple processing infrastructures serving different trading venues without any post-trade harmonisation.
We started to realise the disconnect between the way ETFs had been historically structured and settled – domestically – while over time they were increasingly traded cross-border. This disconnect has become an important issue.
Lytle: Companies are naturally identified with one country. ETFs are not. BP is a UK company, Total is a French company. Everyone expects these companies to be listed on their domestic exchanges. ETFs have a greater need for a supranational solution. However, as a relatively small industry in relation to the overall equity market, people are not necessarily looking to build infrastructures in order to support ETFs.
Walmsley: Do you not think that there is a supranational element already? There are certain centres of liquidity in Europe where there are international participants in those markets. There is a large amount of international participation in London. There is already an international element.
And as we have mentioned, we are some time away from finding a solution to some of the fragmentation associated by individual domestic markets because there are so many other listing aspects that need to be taken into account, such as local tax and regulation.
Funds Europe: The ETF market is a new business segment for asset servicing. What gaps in the service were identified early on and how have asset servicers adapted to serve this new and growing business?
Moore: At a very early stage in the servicing of ETFs we recognised that there was a gap in the middle-office servicing for authorised participants who are an integral part of the overall ETF model. In response to this gap we put in place a dedicated team to ensure that all points in the trading life cycle were covered and supported.
It’s important to leverage core infrastructures and services, such as accounting, custody and transfer agency, but then overlay that with the particular nuances that are required from an ETF perspective.
Nicora: ETFs are not new to the asset servicing industry. What is new is the rate of their growth and the drive for interoperability across markets to support liquidity. This comes back to the point about people thinking that ETFs were little different from shares and that we should be able to replicate equity post-trade processes.
But what we as an industry missed was that, by their nature, ETFs were evolving very quickly into pan-European, cross-border products. Just using the available domestic settlement infrastructure was placing an element of constraint at the core of the market.
Jamieson: The big challenge custodians have not yet addressed is creating an efficient ETF securities lending market in Europe. There are about $400 billion worth of assets in existence. The vast majority of those will be sitting in custody, yet if you want to borrow a Ucits ETF, it’s nigh on impossible.
When we talk about ETF lending we mean lending of the ETF units. The underlying assets are within the BlackRock lending pool and that’s one means of revenue enhancement that gets fed back into the performance of the fund. But there’s a secondary means of revenue for the beneficial owner in lending the ETF, and that’s what we need to address. In the US, it’s a very efficient market as about 20% of the daily trade volume in the securities lending market is ETFs.
There are clearly infrastructure issues that we need to address in Europe and this has numerous knock-on effects to the detriment of the industry in as much as often market makers cannot borrow at any level. From an economic perspective it means they cannot get the assets they need to make markets efficiently. Hence on occasions, the bid-offer spreads are wider than they ought to be.
The ETFs, therefore, become less investable and the people that would otherwise be advocates for buying these assets are questioning whether they are affordable because there’s no means of offsetting the TER compared to underlying assets that they can lend out freely in a normal securities lending pool.
Equally, there are potential end-borrowers, whether it be hedge funds or proprietary traders, that are interested in looking at ETF solutions to express macro views in conjunction with, or as an alternative to, futures, but they cannot borrow efficiently at present.
Therefore, securities lending of ETFs is something that we’re working on to encourage the entire asset servicing community to do, but it’s still very much a work in progress. There are many things we have to overcome, like antiquated systems and overcoming preconceived ideas, such as there’s just no interest in borrowing or lending ETFs.
Lytle: There is a mechanism for creating securities to cover short positions and it’s called ‘create-to-lend’. It has been used extensively in the US and Source has used it effectively in Europe.
If somebody needs to borrow an ETF, it is an open-ended fund; you can create as many units as you need. If a market maker needs securities, theoretically, they can go as an authorised participant of the fund and get the units that they need. When they receive them back, they can redeem them.
This is a process that Source built to support the trading of European sector ETFs and they have become some of the most actively traded ETFs in Europe. But then the question is: can it be generalised to other issuer’s funds? It requires a level of co-ordination and organisation across issuers and prime brokers, which has yet to happen.
Jamieson: There is definitely a place for create-to-lend, but from our philosophical perspective, since there’s $400 billion in existence and only roughly two $2 billion on loan, in conclusion there must be almost $398 billion available to be loaned at any cost, above free. Consequently, it is always going to be cheaper than the model because you will have already paid for the creation cost when you bought the asset initially. Therefore, it’s sitting as a non-performing asset in terms of potential revenue if you do nothing with it.
On the other hand, if you go down the create-to-lend route, what you’re doing is adding extra cost; you’re adding the cost to create the asset in the first place that you don’t need, so the entity creating, most likely a bank, would need to cover that cost, by passing it on to the end borrower.
Yet there are certainly instances where the create-to-lend model could be beneficial, if the category of holders of the ETF are the type of entity that would traditionally lend.
From a generic perspective, for someone to borrow something vanilla like our FTSE 100 ETF, there’s £4 billion in existence so there doesn’t need to be someone else creating more of it, in contrast something more specialised like a sector fund, there may be a requirement currently.
What we would like to do is encourage more borrowing of our funds. Today, they just don’t show up on custodian’s radars, unfortunately.
Lytle: There’s no question that create-to-lend works in markets where the creation process is cheap and efficient. In developed markets it works very well; in emerging markets transaction costs create challenges.
However, the good thing about create-to-lend is it can very rapidly collapse the price of borrowing securities. If there is no one offering to lend a security then it will trade on ‘special’. A EuroSTOXX 50 ETF on ‘special’ could easily cost EONIA +2-4% to borrow. Whereas, if you wanted to borrow the components of the EuroSTOXX 50 you might pay EONIA minus. Create-to-lend offers a way of breaking the logjam.
Nicora: The current fragmentation of ETF asset pools is, again, a problem here. A lender may be happy to lend to a broker in the UK to support trading on the LSE or OTC, but the lender’s stock of ETFs could be held in Italy or Germany. As we well know, moving ETFs from one place to another may take days, sometimes weeks.
Jamieson: It certainly adds to the confusion because quite often people agree to lend the FTSE 100 iShare and when they come to deliver it, they find out it’s sitting, for example, in Milan. Conversely, the short position that person might want to borrow against is probably in CREST and, therefore, they would have to go through a process of doing an internal flip, which takes time and has cost implications. Else, it fails.
In future, more centralised settlement of ETFs would offset much of that issue.
Walmsley: And there are moves afoot to address these issues: the new pan-European settlement offering that we are launching on the London Stock Exchange, for example. We are making available new segments at the end of this month and to allow a pan-European settlement solution through Euroclear. This will offer more flexibility for issuers and market participants.
Lytle: When we set up Source just over four years ago, market makers actually asked us to not list on multiple exchanges. They told us that multiple listings fragment and effectively reduce liquidity. Though some French investors might express a preference to buy an ETF on their domestic exchange, ultimately institutional investors generally have access to a range of exchanges. You have to have the courage of your own convictions, pick a market, and then stick to it. Ultimately, it will make products more liquid and reduce settlement fails.
Nicora: Some clients told us that a few years ago ETFs represented 5% of their activity, but 30% of their fails because of the inefficiencies of dealing with different domestic settlement systems.
Lytle: It’s important to note though that most fails don’t age; they are resolved in a few days. From a risk perspective they’re not that meaningful. The key is that only a small percentage of all transactions end in aged fails.
Jamieson: Also to put this in perspective, the noise around ETF fails is mostly a US issue. There’s a different regime for classifying fails for underlying securities versus ETFs in the US. So ETFs are getting caught up as fails when they’re not actually fails under the respective ETF regime. So, again, it’s very misleading to suggest that ETFs are failing as much as people suggest.
Walmsley: It’s something that we’ve looked at closely. There have been changes in terms of the central counterparty and how it enforces settlement discipline. There has been a marked improvement in terms of ETF settlements over the past year or so.
Funds Europe: ETFs are low-cost products. Operational costs can be driven up, potentially, by a low level of funding in some products, and by the complexities associated with multiple listings. How can asset service providers ensure that their own ETF processing businesses remain viable?
Nicora: Building out on what we’ve covered so far, I think the issue that we are trying to tackle has two elements.
One is the existing situation where the industry is telling us that the current post-trade infrastructure fragmentation and the individual market set-ups of ETFs is creating a lot of costs. We are talking about fails, which then become penalties or fines. The second element is cost, which includes a lot of hidden costs. It is difficult to attribute these issues only to the infrastructure, but clearly the current ETF post-trade infrastructure is a very constraining element.
When we were investigating the cumbersome post-trade issues together with BlackRock, the obvious starting point was to look at the US because it seems to be working relatively well there.
Lytle: Maybe that’s because they have one currency, one exchange, fewer issuers and fewer products.
Jamieson: It can be misleading to compare the US versus Europe but, certainly, the US ETF market has led the way in terms of growth and is a much more mature and larger market.
But if you just look at our listings in certain European jurisdictions there can be marked differences. Our FTSE 100 product in Milan has much less liquidity than the product in London and therefore the bid-offer spread is much tighter on the LSE than on the Milan Exchange. That’s something that we would like to address and consequently try to compress the fragmentation of liquidity that currently exists in different CSDs, so we’re very supportive of the Euroclear initiative [in June iShares announced plans to launch the first cross-listed ETF that would settle in one place, Euroclear Bank].
We’re very excited about the opportunities this initiative will bring because it will continue to give investors choice about where they trade to fulfil their regulatory obligations. It will also bring efficiencies, cost savings and benefit from one central liquidity pool.
But to be clear it’s not something we’re looking to do in isolation. For this to be truly successful, it has to be successful for the industry as a whole and we want to see other issuers embrace this concept.
Nicora: Our aim is to find a way to reconcile the need to support pan-European trading on multiple domestic and pan-European platforms, with the extraction of value from centralised settlement and pools of ETF assets.
Lytle: What you just said could have been said in Brussels.
We want to achieve systemic efficiencies while maintaining our independent cultural identities. It’s really the story of Europe.
We have national exchanges because they are important for the implementation of regulatory regimes for retail individuals. They’re relatively unimportant to large institutional investors. Most sizable institutions exist in multiple countries and don’t need the levels of protection required for retail investors.
We’ve seen exchanges buy one another, such as the LSE buying the Portuguese and Italian bourses, but then continue to maintain them as separate functioning entities with some back-end efficiencies. There is significant momentum that keeps the focus on national exchanges.
Walmsley: This really ties into the importance of domestic exchanges for retail investors. Borsa Italiana, part of London Stock Exchange Group, has a large retail segment. It’s really important to underline those aspects. We can offer efficiencies in terms of pan-European settlement, but there are other domestic elements that still need to be taken into account.
Lytle: So how about a two-tiered solution which leaves national exchanges and regulation in place, but introduces a supranational exchange over the top of it? The supranational exchange would be the venue where brokers make markets. An ongoing challenge is that market makers have to maintain IT infrastructure specific to each individual exchange and allocate trading capital to each line that they quote. If you consolidate market making then national exchanges can source liquidity from the supranational exchange instead of having the market maker go directly to each national exchange. This would address the issue in a structural way and allow national exchanges to maintain their role in protecting the local retail investor.
Jamieson: The reality is I don’t think you need individual exchanges to roll up into some super exchange because, as you rightly said, large institutions have got the flexibility and freedom to trade on any of the pan-European bourses that they choose. The settlement infrastructure benefits of having this centralised CSD is what drives down costs and increases efficiencies.
But I agree absolutely with the point around retail. We only have to look at the US, where retail has driven growth in ETFs, and what we hope, in a post-RDR Europe, is that we will see that here.
Lytle: If liquidity were centralised then there would be no difference in the bid-offer spread or market depth between exchanges. It may seem idealistic but we are going to need to make some bold leaps if we are going to transform the existing market and catch up with the levels of trading activity seen in the US.
Jamieson: But won’t competition drive that to a natural conclusion anyway? If the CSD is centralised and there is different pricing on LSE versus Deutsche Bourse, it then comes down to the local bourses to incentivise market makers to make and maintain liquidity pools on their exchange. And presumably that’s where natural competition comes into its own?
Lytle: If you are convinced that the growth of the European ETF market will ultimately be driven by retail then you need to ensure that national exchanges see increased liquidity. You could achieve a very efficient market for institutional investors with an international CSD, but that won’t necessarily reach European retail investors.
Moore: We have had a solution in place for the past five years which overcomes the issue of moving ETFs between CSDs by treating the shares like an equity trade and settling the trades ‘delivery versus payment’ via a recognised clearing system.
Funds Europe: How can ETF providers economise their operational costs levied by asset servicing partners? Is cross-selling of products by service providers to clients one method and – where possible – stock lending?
Jamieson: Securities lending of ETFs is a definite potential revenue driver; we’re very supportive of that.
The key hurdles to it are many. There are perception issues. Some of the messages we hear from custody agents is there’s no benefit to securities lending for ETFs, that there’s no demand. The people that want to borrow are told there’s no supply. The people that have supply are told there’s no demand.
To break that, we have to get out in front of beneficial owners and tell them there is a demand to borrow their ETFs and get custodians to ensure their infrastructure is up to speed because a lot of it seems to be relatively antiquated.
At present, there’s a mechanism for sifting out equities and putting them in the equity pool and there’s a means of segregating fixed income and putting it in the fixed income repo pool. But then there seems to be quite often only one other pool, which might include futures, warrants, funds and ETFs all just sitting there unclassified or invisible. Those assets need to be visible.
It’s about ensuring there’s connectivity between the securities lending system and the custody holdings within an asset servicer and that inventory is made known to the prime brokers and borrowing community.
There is an issue, too, with making sure that there’s collateral acceptability for ETFs. Again, from a risk management perspective there’s a lot of ignorance around what an ETF is and its suitability. An ETF can bring diversification to collateral holdings.
Moore: It’s a new and developing area and in terms of the market evolving in line with the demand, the infrastructure will follow. If there is a demand, if there is a market, I don’t really think the infrastructure will be a constraint.
Funds Europe: In broad terms, how mature is the ETF industry in Europe and how is it likely to evolve in the coming two to three years?
Walmsley: In 12 years the number of products on London Stock Exchange has grown to well over a thousand. In a short time the market has developed significantly, but I think there are still huge opportunities in the European market.
Moore: Product innovation is going to be key in the coming years. The market is just short of 2,000 products and for ETF providers, particularly in terms of securing market share, innovative strategies will be an important differentiator.
Nicora: The international ETF rollout that we are bringing to the market with BlackRock is based on proven infrastructure that exists for other asset classes; we are just reusing existing tools for another asset class. I believe there will be widespread adaptations over the next few years as to how ETFs are structured and settled in Europe.
Jamieson: The industry is at a very exciting crossroads and we’ve come a long way to get to where we are now. Innovation is now becoming the driver. A number of clients have yet to really do more than dabble with ETFs but I think that as we move forward we could easily see this becoming a trillion dollar business in Europe over the next five years.
Lytle: The most salient issue to observe is how well the market is reflecting the latest opportunities to investors. That’s the place where the European market often outstrips the US. It’s flexible, very adaptive and quick to deliver new innovations to investors.
©2013 funds europe