The London Pensions Fund Authority, (LPFA), is one of the UK’s leading public sector pension schemes, with 73,000 members and £3.2bn (€4.7bn) in pension fund assets.
It was set up in 1989 as a stand-alone public body to take over the running of the pension fund of the Greater London Council (GLC) following its abolition in 1989. It now runs the pension funds of 220 employing authorities.
For investment purposes the LPFA pension fund is divided into two sub-funds – a £1.8bn fund for conributing employers with a 50:50 mix of pensioners and active members, and a £1.4bn fund for pensioners who no longer have a contributing employer, principally former employees of the GLC and the former Inner London Education Authority The LPFA administers a defined benefit type scheme which delivers statutory, inflation-proofed pensions. The strategic aim is to finance these pensions by holding contributions steady and maximising the fund’s investment returns.
Until 2001, the authority had achieved this aim comfortably. Yet by 2004 the pension fund had a gap of 18% between the valuation of assets and liabilities.
The period between the fund’s triennial valuations in 2001 and 2004 had coincided with a sharp fall in the equity markets. The active fund, which was 85% invested in equities, was particularly hard hit.
Peter Scales, the chief executive of the LPFA, says the effect was to turn an actuarially expected gain in asset values into an unexpected loss. “The actuaries’ expectation was that the fund’s income each year would be about 6.66%, so we had assumed a 20% increase in value over the three years. In fact the net position was a reduction of 10%, so we lost 30% during that period.”
As a result, funding gaps appeared in the two subfunds. The active sub-fund was worst affected. Its funding level fell from 103% in 2001 to 74% in 2004. The funding level of the pensioner sub-fund was less affected, since it was invested primarily in bonds. It fell from 99% to 91%.
Scales and his advisers at Hymans Robertson recognised that returns from the recovering equity markets would not be enough to provide the rewards they needed. “We needed to recoup that 30% loss on top of excess returns going forward,” he says.
Neither could the LPFA expect to close the gap with a substantial increase in employer contributions. The public sector employers that contribute to the LPFA funds were not in a position make a large, one-off contributions payment into the fund, in the way that many private sector companies in the UK have done.
“Many employers have made cash contributions phased over time but they weren’t going to be enough to give us a prospect of restoring the position in a reasonable timescale,” says Scales.
Nor could they expect excess returns from their current equities investment strategy, he says.
“What we found – and this is not particular to our managers – was that the traditional approach to equities management had suffered during this fall in markets and that managers were protecting positions by hugging the benchmark too closely.
“At the same time it became very clear to us that a significant negative return didn’t do very much good for funding and we really needed positive returns.”
The LPFA did not follow the path taken by many other pension funds of changing the asset allocation by switching from equities into bonds. Instead, it decided to adjust the way the existing allocation was structured and managed.
In broad terms this meant a switch from a passive to an active approach, and a search for absolute rather than relative returns. “We still believe in equities but we are using them in a slightly different way,” says Scales.
The new strategy, which came into effect in January this year, signalled a shift away from the traditional bond/equity structure to one that is more related to liabilities and cash flow requirements.
It involved a major transition of £2.5bn of the two fund’s assets. In the active sub-fund the bulk (75%) of the active sub-fund’s £1.8bn portfolio had been invested in global equities. These were managed by three managers: Legal & General Investment Management (LGIM) with an index tracker; Henderson Global Investors, with a target of 1% above a bespoke benchmark; and Goldman Sachs Asset Management, with a quant-driven mandate based on the same bespoke benchmark and target of plus 1%.
The remaining 25% of the portfolio consisted of 10% in global bonds, also managed by Henderson Global Investors, with a mandate of benchmark plus 0.75%, and 15% in alternative investments – private equity, real estate and public/private infrastructure projects.
In the new strategy, LPFA has reduced the direct allocation to equities from 75% to 65%. In the place of the index tracker and the traditional benchmark plus 1% mandates, LPFA introduced two new global equities mandates with more aggressive performance targets of MSCI world index plus 2%, but with no benchmarks or index constraints, to be managed by Newton, a UK global equities stock picker specialising in thematic investing, and MFS Investment Management.
LPFA also added a currency overlay programme to the equities portfolio, to be managed by Record Currency Management. The first task was to hedge currency risk, says Scales. “We had not looked at currency hedging before but it became very clear to us with the scale of equities that we had that we needed to reduce risk by hedging our exposure.
“So the basic mandate is to hedge half the overseas exposure, and for this they put on a fairly passive hedge. It doesn’t duplicate what the individual equity managers are doing but simply acts as an overlay.”
The second task was to earn some extra returns from active currency management, he says. “Having taken out the currency risk, we wanted to see if we could get some added alpha out of the currency itself. There is no particular target, although there is an expectation that the manager could over time achieve plus 0.5%.
“They have the whole of the overseas equity value to work with and we give them a small pool of about 5% in cash, which they can use to buy the derivative overlays and forwards on the currency market. We think that’s a very small bit of risk that we’re taking, and if they can add value over time, that is extra money that we can put in the bank.”
The key change in the active sub-fund, however, is the introduction of a target return mandate divided equally between Merrill Lynch Investment Managers and UBS Global Asset Management, with a target of retail price index (RPI) plus 5%.
Target return mandates, which are not constrained by a benchmark or any particular set of asset classes, are more likely to add value over time than traditional mandates, Scales says.
“The use of target returns mandates matches our desire to achieve an investment return that reflected our index-linked liabilities and also provided a counterweight to more traditional equity mandates which move in line with equity markets.”
Some features of the active sub-fund remain unaltered in the new structure. The existing Goldman Sachs Asset Management’s global equities portfolio quant approach to investing global equities has been retained as a core, lower risk portfolio and increased by 50%. “Because it is run out of New York, the Goldman Sachs mandate has a different approach, a different perspective on the market, and we quite like the way that worked,” says Scales.
The allocation to alternative investment also remains unchanged for the time being. LPFA made its 15% allocation to alternatives in 2002 and implementation has been slow, says Scales. “We don’t have big cash flows so we needed to sell equities to be able to invest. But in 2002 the equity markets were very low and it wasn’t the right time to sell equities.”
On top of that, the process of investing in private equity has been inevitably protracted, he says.
“Moving money into private equity is quite slow and we’re still feeding that money in. We’ve placed some of the cash in property unit trusts and private equity trusts. In that way we can park the money and at least get the same type of returns we would expect from private equity.”
Other alternative asset classes are always under consideration, he says. “We are continuing to look at that side of the portfolio for assets that have an equity sort of substance to them and are likely to give us both growth and the income that we need. So we might look at commodities and hedge funds.”
The most radical changes to the LPFA pension fund have been made to the pensioner sub-fund, however. This fund had been heavily weighted towards the passive management of fixed income.
A small part of the fund’s portfolio, 15%, was invested in a passive global equities mandate managed by LGIM. A further 20% was allocated to a global bonds mandate with an expected return over a bespoke benchmark of 0.75%. The bulk of the portfolio (65%) was invested in index-linked gilts and cash.
LPFA has retained the passive global equities index tracker to diversify risk but reduced its size slightly to 12.5% of the portfolio. Yet in a bold move it has swept the remaining 87.5% of the portfolio out of index-linked and global bonds into cash flow matching bonds where the aim is to match liability cash flows plus 1.5%.
Scales says the switch was necessary if the pensioner sub-fund was to meet its commitments. “The characteristic of the pensioner sub-fund is that it doesn’t have any contributions coming in and there’s a large cash flow going out in pensioner payments.
“We knew the index-linked gilts that we were holding were not matching those liabilities any more because of the way the markets were going. Plus we couldn’t get any cover at the long end. This was before the issue of 50-year gilts.
“We wanted something that was going to give us those cash flows and a little bit more to pick up the deficit.”
The cash flow matching bonds mandate has been split into three, one managed by EGgM and the other two managed by Barclays Global Investors and Insight, the investment management arm of the Halifax and Bank of Scotland group.
The managers have been chosen for their differing investment approaches, explains Scales. “ECM will use European derivatives, a series of 100 or so interest rate swaps, structured in the Swiss market, to achieve their absolute returns. The other two use different approaches, either pooled products for particular derivative styles or derivative overlays, structuring derivatives over and above a series of bonds.”
The choice of managers was dictated by the need to reduce risk, he says. “We started the process without a clear view of exactly how many managers we would have and what types. It was only when we’d seen their submissions that we could choose different types of products.
“It was a balancing process. We drew up a chart with various mixes of different managers with their different styles and investment approaches, and assessed how they might add or reduce risk against what we were doing.
“The decision in the end was driven partly by diversification of risk. We took three different styles to diversify the risk of the particular products that they were using. ECM is using European derivatives, so it is entering a different market from the other managers and therefore reducing risk. Their portfolios are also more suited to nominal bond yields rather than index linked, and that is useful to us, too.”
The managers have been given considerable freedom to outperform, says Scales “They are very unconstrained, and deliberately so, because we want them to take the decisions.”
LPFA will monitor their progress closely, he says. “The scheme’s liabilities will obviously change over time and the difficult bit will be monitoring the cash flows and re-gauging them. So we will be constantly reviewing the mandates and re-cycling the instruments they are holding.”
Along with its new investment strategy, LPFA has introduced a new way of rewarding its managers – performance-related fees and targets set net of fees.
“We were cautious about performance-related fees because there’s still a lot of argument about whether they incentivise a manager,” says Scales.
“The approach we have taken is certainly to require performance fees on active mandates like global equities. For target returns, where the ability to perform is less certain because they are absolute returns, we gave them an option of performance fees or ad valorem fees.
“For the cash flow matching bond mandates we use performance fees because the nature of the instruments particularly demand skills and performance.”
The LPFA has structured its performance fees to encourage outperformance while not penalising under-performance too heavily, says Scales. “We are using the baseline fee the managers would have been offered in the old style of payment and then reducing it at the point at which our target is set.
“So if they don’t quite hit our target they will just about earn the base fee they would have had otherwise. If they go above our target then they will start to earn even more.
“In effect, we are expecting outperformance before we pay what would have been the baseline fee in the old style of payment.”
The nature of the LPFA scheme also enables it to retain the shape of its new investment strategy while replacing managers, should this become necessary, he says.
“We maintain ownership of our assets in most cases so we can recover our positions by taking them back in-house. This means that if a particular manager isn’t performing we can keep that style of investment but simply get somebody else to do it.”
Yet it is far too early to talk of manager change, says Scales. The LPFA has made a major change in investment strategy and needs time to decide whether this change has been successful.
“The expectation is that within 10 to 15 years we will have hopefully recovered much of the capital loss that we have suffered. But it is a long-term strategy and it is going to take time to see that through.”