Fennell Betson talks to a champion of the in-house approach
Roy Peters is the man in charge of keeping BG’s asset levels in the UK gas company’s £13bn (E21bn) pension scheme at the right pressure to meet the needs of the 135,000 membership.
But as only 15,000 of these are currently active, the inflow outflow balance is constantly altering over time. “The significant change over the last decade is that the majority of the liabilities are now accounted for by pensioners and deferreds, reflecting the reduced workforce,” says former stockbroker Peters, chief executive of BG Pensions Funds Management in London, which manages over 90% of the two main funds’ assets in-house.
The BG Scheme has followed the fortunes of the gas industry in the UK, and until very recently, that has been the history of British Gas, firstly in its state-owned and later its privatised form. But in his view, the real change came with the greater competition in gas following deregulation and then the demerger of the former British Gas in 1997 into BG, the gas pipeline company and Centrica, the distributor.
“What happened was that most of the assets stayed with the BG schemes along with the vast majority of the pensioners and deferreds. About 10% of the assets were transferred to Centrica’s fund, but with a largely active membership, it is a less mature scheme, and that just accentuated the maturity of the BG schemes.”
The three asset liability studies carried out over the past decade also mark the maturing of the scheme. “The study in the early 1990s just reinforced the existing asset allocation of the time and view that this was not that much different from any other pension scheme, so the WM50 peer group benchmark was appropriate,” says Peters. The 1996 study resulted in a shift of about £1.9bn in assets into index-linked gilts and other bonds. Then after the pension fund split with Centrica, the study done in 1998 and implemented in 1999, a further £1.25bn was switched out of equities into gilts. “So by UK standards the fund has a high bond and index linked content of around 35%, compared with typically 15%, and has, at the same time, moved from the WM50 to a customised benchmark.”
The AL studies yield an optimum asset allocation, around which the tactical asset allocation (TAA) operates in bands. “The bands are ± 3%, which is narrower than previously,” he concedes. “The reasoning behind that decision was as the scheme became more mature, the whole point of the asset liability having a minimum level in bonds, which clearly caps how high we can go in equities.”
The limits could well be a sign of increasing risk aversion on the part of the scheme’s trustees, who wish the scheme to be run in a conservative fashion. In Peters’ view: “The trustees are more concerned about the risks of under performing rather than having too aggressive a strategy to out-perform.” The target is to outperform the benchmark by 0.5% per annum over a rolling three years and the fund must not under-perform by less than 1% in any 12 month-period.
“Our investment style reflects the expectations of the trustees, where we look for small incremental returns in asset allocation and stock selection, rather than a few big bets. Risk control is an important part of our fund management process. While risk analysis won’t help you outperform, it does ensure construction of the portfolio is in line with trustees’ expectations. In our view, controlling investment risk as an essential part of the process, so we look at total portfolio risk broken down between investment policy risk and stock selection risk.”
And the results have confirmed the approach, as within the WM50 universe of the biggest schemes, the BG scheme is among the lowest in terms of risk, but producing above average performance, he says, adding with justifiable satisfaction: “Obviously a nice combination.”
How do they do it? “Internally, we use an asset allocation optimiser to check regularly the investment policy risk or tracking error of the asset allocation against the benchmark and we have access to a number of Barra-type risk services to review the construction of the individual portfolios.” While the funds are actively managed, given the fact £5bn is in UK equities, he reckons “it is not going to have a very high tracking error – it would be about 1%. For an index fund of that size, it could by 0.1%. So it is a low risk equity portfolio.” WM Company analysis of the portfolios shows that the BG managers have been good at stock selection within sectors, but not so good on sectors. “Consequently, we have tended to have a sector neutral policy, with just a few sector tilts, and an emphasis on picking the best stocks within the sectors.” The UK portfolio is relatively large comprising 300, partly because of the smaller companies portfolio run within it, accounting for 120 stocks. Overseas, the number is much smaller, around 100 to 150.
“We still breakdown portfolios geographically, North America accounting for 4% of the equities portfolio, Europe for 7%, the Far East, including Japan and other Pacific markets, for 7%. At the moment Europe is still treated separately to UK.”
Peters admits that the US was one area where asset allocation went wrong, as happened with other UK funds. “UK managers underestimated the longevity of the US economic upturn, which has been going on for nine years and accompanied by low inflation rates. The development of the ‘new economy’ companies has been something UK investors missed out on.” But the funds have consistently been overweight in the UK and European markets in the last few years, so they made up there what was missed in US. “We manage portfolios on a matrix of country, geographical and sector allocations, with more emphasis on sector allocation than country, compared with five years ago, but we have not removed countries altogether in Europe as yet,” he says.
The key factor in stock selection is sustainable above average returns on capital, and a focus on shareholder value, in BG’s reckoning. “We look at these in context of what changes are likely, monetary policy, inflation and so on. Since valuation techniques seem to be in state of flux at the moment, as no one measure can apply to all companies, we examine quite a number of different measures. But then we would go overweight in a stock we find attractive.”
With such a high fixed income portfolio, it is not surprising that Peters reckons its one area where they have a lot of experience. “We used to have international bonds but now with a customised benchmark based on the gilt index, if we have overseas bonds we must hedge the currency. Anyway, up to now we have preferred domestic bonds to overseas bonds.”
Commenting on the UK gilt scene, he says: “There is a bubble in both UK long gilts and index linked, with real yields down to 1.7% recently. You basically have the situation where the supply of long dated government paper has dried up. However, with mature schemes and MFR issues the demand for gilts is increasing. We are having to buy index linked gilts with such low real yields and long dated gilts at very low nominal yields, the question is whether this is in the best interests of the pension scheme in 10 years’ time.”
Up to 30% of the bond portfolio can be in corporates, and currently about 15 to 20% is in this area.
“Our portfolio turnover is generally pretty low, though it increased in the last few months with a series of basket trades in the UK and elsewhere generated by portfolio restructuring.” This of course involved the use of derivatives, which is done in-house. “We use futures extensively when implementing asset allocation decisions. Frequently, we build up a short equity future position and then use a basket trade to unwind that, handing over the short position to the broker.”
The fund also has an active tactical asset allocation, where the implementation requires futures. “We have done well on the TAA side and the greater part of our outperformance over a considerable period has come from asset allocation.”
While hedge funds are not on the horizon for Peters, he says “we want to become more aware of some of the techniques these funds use as market neutral strategies may have a part to play in an active stock selection programme”.
As to other alternative areas, the exposure to private equity is a minimal 1%, which he would like to take higher and the allocation has been increased by using a fund of fund approach. “In the past, we did find monitoring funds very time consuming.” And if property is considered alternative, it makes up 6% of assets, with two external managers running the portfolio.
All in all, these amount to around 9% of total assets and they would require a considerable team to manage in-house, says Peters. As it is the internal team consists of 31 people, including 17 investment professionals. Though very committed to the in-house approach, he considers the internal/external debate essentially as one of a balance of advantages and disadvantages.
The major advantage of do-it-yourself is the focus on just one client, in his view. “We are not being diverted into marketing for new business and can focus only on investment. The other advantage is economics. Cost of management for equities is about four basis points and so considerably below the cost of external management.”
He adds: “Generally, in-house funds are staffed by people who just want to manage investments without ambitions to develop a global fund management business. Some of them might be uncomfortable in an external fund management environment. Also the packages which in-house funds offer these days are much more competitive and they are no longer seen as a small department trying to fit into an industrial organisation.”
He acknowledges external managers pay a premium not so much in terms of salary but in being able to offer higher bonuses, option schemes and so on. “Our staff here do get a bonus, dependant on the funds’ performance. We are looking, as other internally managed funds are, at having some form of longer term incentive scheme to motivate and retain staff.” And staff turnover has been low. “If we do loose staff, it is to external managers so we have to be aware of trends in remuneration.” One factor that being internally managed does require extra attention is that of research and analysis. While the scheme does not have a research team, the fund managers specialise by sectors so build up detailed knowledge but rely on stockbrokers for fundamental research. “In term of macro research we do buy in a number of third party services. And as leading institutional investors we see a lot of managements. In first six months of last year, for example, our managers met with 500 companies.”
He finds it hard to state a minimum size that funds need to be to justify the in-house route. “People used to say £1bn. But to handle 80 to 90% internally, you would probably need around £5bn.”
Though not a debate actively under discussion in BG at present, Peters says it is never an absent consideration. “We never take the BG trustees for granted as they don’t have to have fund management delivered in this way.”
He points to the fact that in the past year, the ownership of the pension management company has been transferred from the corporation to the trustees of the pension scheme. “We are now similar to the BA scheme, where the fund managers are owned by the scheme rather than the plc. In conjunction with the ownership change, the trustees asked consultants Watson Wyatt to do a due diligence exercise on us, since their responsibilities had changed from being our client to also being our owners.”
One of the reasons for this change was the view from the corporate’s perspective that as a result of the Pensions Act, more of the responsibilities for managing the fund were now with the trustees rather than the sponsoring company. Also, the shift meant the costs of running the scheme were borne by the trustees and as a result of the fund management operation sat more comfortably with trustees.
Another bit of restructuring is that the two major BG schemes are merging as from this April, meaning a disappearance of the old white and blue collar distinction. “The two are identical in terms of benefits and asset allocation policy. The aim is to merge and obtain some administrative savings as a result.” One practical implication was a review of the custody arrangements with the result that Citibank will now handle all the international custody, while Bank of New York look after the UK custody.