Tough economic times demand greater return on investment

In early October, Labour MP Frank Field wrote to Pensions Week warning that pension funds were running “alarming risks” by allowing their securities to be lent to third parties. Field’s letter drew a blistering response from International Securities Lending Association (ISLA) chief executive David Rule, who itemised all the points he thought Field had got wrong. However, politicians shrug off that sort of thing readily enough and Field’s subsequent responses showed him determined to uphold the position on this sometimes controversial practice.

But a lack of knowledge about securities lending is by no means unusual. It is an arcane topic well off to one side of the usual final salary ­ versus ­ defined contributions debates, and there probably are trustees in large schemes who have very active securities lending programmes that would find a briefing from their pensions advisors on this theme useful.

A basic tutorial would very briefly explain what securities lending is all about. A pension fund has a portfolio of assets, including government bonds and equities which it trades to achieve the fund’s objectives as laid down by the trustees. If it is a buy-and-hold fund, which pension funds tend to be, since they are trying to achieve a match of assets to very long-dated liabilities, once it buys the assets they sit on its books and gain (or lose) value under their own momentum. What securities lending does is to enable the fund to earn a handful of basis points from a borrower by assigning them a security for a defined period, in exchange for sufficient collateral.

One of the very good things to come out of the fuss Field kicked up was a collaboration between Pensions Week and Data Explorers, an organisation that gathers data about securities lending around the world. This collaboration resulted in a survey of the top 50 UK pension funds, conducted in the summer, which found that 68% were participating in a securities lending programme. All respondents said they were participating by choice ­ putting paid to one of Field’s major concerns that trustees of pension funds did not know their agent lenders (the banks who provide custodial services to the fund investment management team) were lending out their securities. Around 88% of the sample said they lent through their custodian, while 11% engaged a specialist third-party agent to manage their securities lending programme.

On whose authority
ISLA’s Rule made the point that there is no way a reputable custodian would indulge in a lending programme without the explicit authorisation of the beneficial owner of the securities. Who would it pay the proceeds to? Imagine the conversation:
Bank: “Excuse me, Mr/Ms Trustee, here’s £1m.
Trustee: “Eh? What’s this for?
Bank: Well, we’ve been lending your securities on your behalf, we thought it would be a good thing to do.”

A second major fear Field expressed was that, while securities lending is done on a fully collateralised basis ­ I lend you £1m of my portfolio of securities, you put up £1.05m worth of government gilts to cover the trade ­ there is still some risk. The risk he focused on is trading volatility, suggesting that markets sometimes lose 10% in a day and so the industry standard practice of accepting 105% collateral is, to his mind, risky. But this point missed the fact that it is standard practice, when taking equities as collateral, to mark-to-market on a daily basis, which means whoever is managing the lending programme would be automatically pulling more security from the borrower to bring the collateral up to the level specified by the lender.

Of course, if the borrower could not afford to meet the additional security call, then there is an insolvency situation where the agent acting for the lender would begin liquidating the borrower’s collateral to replenish the lender’s securities. Does this not still involve the risk of loss to the lender? No, not unless the custodian bank also fails: as the Data Explorers survey discovered, more than 82% of lending agents indemnify the lender against counterparty default.

It only takes a moment’s thought to see that if there is a market event of such cataclysmic proportions as to wipe out both a borrower and the custodian bank itself, sending both into default, then a glitch in its securities lending programme is likely to be the least of any pension fund’s worries. The global financial network would be in meltdown.

The point of all this is that there is some counterparty risk to securities lending, but nothing to get anyone excited. Two rather more significant and related obstacles standing in the way of pension fund trustees getting involved in securities lending is short-selling and the reputational risk to the fund, which boils down to thinking short-selling is simply a bad thing, or the concern that your pension fund is in any way associated with that bad thing. It is a case of education.

That said, once securities lending has been explained, there is a good chance many trustees would opt not to pursue it given the relatively paltry rewards on offer. Is an average return of say, five, six or seven additional basis points worth any kind of effort, or worth running even the most modest of risks for? After all, trustees are supposed to be prudent.

Ed Oliver, a director at Data Explorers, points out that the average return to beneficial lenders amounts to around £1m. Take a pension fund grappling with a shortage and ask the trustee board if it would like an extra one million pounds for doing nothing, and the board should leap at the chance if the risk profile is right.

No compromises
Sonja Spinner, senior associate with pensions administrator Mercer, points out the National Association of Pension Funds view on securities lending. It says there is no evidence making assets available for securities lending in any way compromises a pension fund’s goals. That is worth emphasising since one of the arguments against trustees giving the go-ahead to securities lending for their scheme is that securities lending is a major assistance to short-selling – and short-selling drives down the value of assets in the fund’s portfolio.

Data Explorer’s Oliver argues his survey shows that supporting directional short-selling – whatever the merits of short-selling might be – amounts to only a very small percentage of securities lending demand. “Some 50% of securities lending is about lending fixed-income securities and not about equities at all, which means that short-selling doesn’t come into it in any shape or form. Of the remaining 50% of equities lending, Oliver adds that the vast bulk of this is directed at activities such as covering arbitrage opportunities or market failure, leaving only around 4% of the total volume being used to support directional short-selling. The Credit Suisse Tremont Index, a comprehensive index of hedge fund strategies, shows just 0.7% of hedge fund strategies are about pure directional short-selling.

Nevertheless, Data Explorers’ charts show a marked decline in securities lending volumes since the fall of Lehman Brothers, undoubtedly due to some major funds stepping back from securities lending. Mercer’s Spinner points out that in the UK, Hermes, looking after the BT pension fund, was the most high-profile fund to withdraw totally from securities lending.

“Our view is, provided a pension fund’s securities lending programme is properly collateralised, with the kind of securities that are taken as collateral being clearly in line with the fund’s risk profile, then securities lending can be a useful additional source of revenue for a fund,” says Spinner.

What she means by the collateral “being in line with the fund’s risk profile” takes us into another terrain, namely what constitutes sufficient collateral. A conservative approach, which Spinner suggests should logically be the approach adopted by trustees new to securities lending, would be to accept only G7 nation government securities as collateral. “We tell funds that accepting 102% government bonds as collateral (102% by reference to total value of the bundle of securities being lent) is sufficient if there is no currency risk involved, and 105% is sufficient where there is a currency risk. This is the industry standard,” she says.

One form of collateral that pension funds accept is cash, though it is something less common among European or UK funds. It might be thought that cash would be the safest form of collateral for pension fund trustees to accept. However, as it turns out, it is one of the riskiest. When the borrower gives the lender a large sum of money, standard industry practice is for the lender to agree to pay the borrower a set amount of interest at the end of the lend. This puts an investment risk on the lender, since the lender has to invest the money, make a set return to pay the borrower and then an additional return to make the whole deal worthwhile.

Moreover, while custodian banks will indemnify lenders, the beneficial owners of assets, against counterparty loss on trades collateralised with fixed income or equity securities, they won’t indemnify the lender for losses resulting from investment failures around cash. So if you take cash, you’d better be able to make it work for you. This is why the vast majority of UK securities lending programmes use securities as collateral and not cash.

Mark Tidy, head of international business development for global securities lending at Bank of New York Mellon, points out that Lehman’s was a major borrower of securities in Europe as well as in the US and when it crashed, that default stress-tested the securities lending system. “What the failure of Lehman’s showed was that securities lending worked, the collateral system worked and lenders were protected,” he says.

Tidy adds that it is well worth funds having a debate about whether to lend or not to lend, because, ultimately, that helps trustees to be clearer about the whole concept. Essentially, he points out, it is about making idle assets work that bit harder.

For a complete archive of Decisions supplements, go to www.financialdirector.co.uk/decisions