01 May 2005 Features

 

How far can automation go?

Lending of securities has changed from an informal back-room activity to a trillion dollar industry, but can automation replace the relationships at its heart, asks Georgina Stanley

Securities lending is big business. With increasing numbers of hedge funds and general growth in capital markets, securities lending is being used for more than simply covering short trading.

The last five years have seen the sector transformed into a trillion dollar industry with the lending of both equities and fixed income securities being used to cover shorts, prevent fails as well as more complex trading strategies such as risk and tax arbitrage.

Few figures are available but conservative estimates place the total volume of lendable securities at more than $4 trillion, with securities from more than 20 countries around the world available for loan.

With the increasing volumes have come technology platforms to try and ensure best price is found as quickly as possible and to reduce the risk to the lenders by increasing transparency of the trades. But, in a sector as complex as the securities lending market, how far can things ever be automated?

“General collateral transactions are more easily automated but if there’s something very hot, or a liquidity event, or someone’s shorting a stock, this is where the front office people are on the phone, trying to get as much as they can,” says Brian Traquair, president of SunGard Securities Finance. “That’s less likely to be systemised. You can make or lose a lot of money on these transactions and it’s still a relationship business particularly when it comes to hot stock. You tend to go to the person who did a good job last time.

“What you want is to have automation underneath that does the transactions machines can decide and provides an information base for traders for those that they can’t.”

Over the last four years a number of independent firms have cropped up with technology to automate securities lending.

In 2001 a group of 10 financial institutions, including Barclays Global Investors, State Street, Merrill Lynch and JPMorgan, set about developing a global platform to automate securities lending transactions. The result was Equilend, which has a number of solutions in place to increase efficiency by automating front and back office processes while delivering access to liquidity and reducing risk.

Dirk Pruis, outgoing president and chairman of Equilend, explains: “Securities lending was historically very much a voice business. It’s moved quite rapidly into automation in the last three years or so. As this market has matured spreads have declined over time and therefore to continue to be successful you need to be able to scale and do more with less. Technology is an enabler. I expect the growth to continue for the foreseeable future until the market is virtually completely automated.”

Chris Fay, chief executive of UK securities lending platform, SecFinex, agrees with Pruis, saying: “People for a long time have been telling us that stock lending is different. That it can’t be put on a screen and couldn’t go electronic because of the relationships between borrowers and lenders. We’ve demonstrated that that’s not necessarily the case. It’s become much more standardised and people need to become much more efficient. There aren’t the spreads any more to survive on operational inefficiencies.”

Equilend has been live since June 2002 and in that time has lent more than $2.1 trillion of transactions with an average trading day of between $6 billion and $10 billion. The platform is message-based, allowing borrowers looking for specific securities to contact lenders directly to try and fulfil their trades. If the first lender contacted does not have the stock, the request will go automatically to the next lender on the list until it is closed. For high volume, simple transactions the deal can be completed with no human contact.

SecFinex has a similar offering. The technology of both companies is better able to automate lending of highly liquid, readily available, general collateral (GC) stock, on a best price basis than complex trades. The advantage of using these platforms is that they can be integrated into the back office systems of the borrowers and lenders making the booking process more efficient and reducing the risk of re-keying errors.

For more complex transactions it may well be more efficient to carry out the negotiations over the phone as too many messages between the borrower and the lender reduces the usefulness of an automated platform.

“The borrower can conduct simultaneous one-to-one negotiations back and forth,” says Pruis. “It’s an electronic way to carry out a negotiation that otherwise would be done over the phone though typically most traders don’t like to have a huge number of back and forths. If it takes that many it’s generally going to be more efficient to get on the phone.”

In addition to a borrower-lender link similar to Equilend, SecFinex also offers an order market service which is pre-trade anonymous. The service allows trading purely on the basis of price making it possible for lenders to see what their relationship is worth.

Despite the presence of SecFinex in the UK, when it comes to automation the US market is much more advanced than the European. Fay believes that the company is the leader in the UK market but it still only has, according to his estimates, around five to 10 per cent of the market.

“There’s still a lot of phoning, five per cent of the UK market isn’t the answer. It’s still very much a relationship thing. I’m not trying to suggest that this can move from being a relationship business but there are big stp savings through automation. Some of our clients have said it saves their trader more than half a day just covering positions that are GC positions.”

One of the biggest challenges to automating the European market is that the transactions are much more complicated with fewer transactions using cash as collateral.

“The US tends to take the lead in this market,” says Fay. “It’s much bigger than each of the individual European markets and probably similar to them combined.”

Mike McAuley, chief product officer in State Street’s securities finance division, agrees that the European market is more complex than the US, not least because of the different tax regimes across the EU. But he believes that this complexity increases the possible advantages from automation.

“Some of that stuff is harder to automate but on the flip side I would say that the European and especially the UK fixed income market is something that’s much more available for automation.”

Graham Sida, head of business technology and strategic business implementation, State Street securities finance, adds: “Another reason we can execute as many transactions as we’re able to is because there’s such a large demand in the US. You’re talking about thousands of transactions a day. Outside the US that kind of demand, while still being out there, is perhaps nowhere near that volume.”

State Street is one of the biggest securities lenders in the world. It lends on an agent and a third party basis in more than 35 different markets with traders in around 10 locations globally. The bank often has in excess of 100,000 discrete loans out at any time meaning it’s frequently talking to hundreds of thousands of clients.

It offers technology that lets lenders establish their risk/return expectations and view their performance (SL Performance Analyzer) and gives online access to information about their securities lending programmes (SL Performance Reporter).

Given its position Sida claims that State Street’s biggest technology challenge for securities lending is being able to deal with the volume of transactions.

“It’s very much about volume and throughput. Even if you’re not actually executing thousands of loans in one day, and sometimes we do, you’re also having to take any of those open loans and mark them to market. Throughput or capacity is one thing we have to make sure we’re always comfortable with and more importantly that our friends the regulators, are comfortable with.”

As volumes grow and demand to automate increases another challenge according to Sida is ensuring that there is a robust rules structure behind the technology platform — allowing it to take into account different tax, regulatory and compliance requirements around the world. It is also committed to increasing the use of technology in the back office.

Demands from US regulators to increase transparency and ensure lenders and borrowers are correctly recognising and recording their credit exposures to underlying clients will also have an impact on systems.

As companies like SunGard, Equilend, ESecLending and SecFinex all expand their product offerings in the securities lending space, it seems that automation in the market is here to stay. But despite this it seems unlikely that it will ever be fully automated.

“Automation provides so many efficiencies that both borrowers and lenders are eager to subscribe,” says Michelle Phillips, global head of trading at JPMorgan Investor Services. She thinks more could be done to increase automation from front to back office and to increase automation in the Asian markets, specifically Japan and Australia. “But, when you’re dealing with trades that might be convertible bond related, or dividend related or which might be the result of a complex corporate action, you can’t really automate that process of negotiating because each type of trade warrants a different type of fee structure.

“I think the complexity is such that you could not automate it fully. There’s room to automate more but I don’t believe, in this particular business and current environment, you could automate trading 100 per cent.”