Like us, you probably thought that the purpose of airbags and antilock brakes was to make car-driving safer. It seems that we may have been wrong.
Researchers at Purdue University, West Lafayette have found empirical evidence supporting the ‘Offset Hypothesis’. Under this hypothesis, drivers are happy with the level of risk they are taking. Any technologies which apparently reduce risk are taken instead as an opportunity to drive faster and more aggressively. The probability of having an accident or injury is left unchanged; the speed of driving is increased.
Now we’re as guilty as anyone else of talking about Liability Driven Investment (LDI) primarily in terms of risk. We even define LDI as “efficient risk management”. But maybe this is a bit negative. Maybe we should focus on LDI as a way of increasing expected returns, whilst leaving the risk of serious loss unchanged. Risk and reward are two sides of the same coin after all. So we’re going to use this article to try to redress the balance, and couch LDI primarily in terms of expected returns.
Let’s have one last look at our old risk-based definition of LDI before we tear it up. For us, LDI is not a product or an approach; it is a philosophy. Old-fashioned investment was all about managing a portfolio of assets; LDI is all about managing a set of risks (including those originating from the liabilities). The thinking behind LDI is simple: minimise risks that you don’t expect to get paid for; and spend your risk budget in the areas where you do expect to be rewarded. We have yet to hear a rational argument for spending a risk budget inefficiently (ie not adopting an LDI mindset).
Right then, we’ll have a stab at drafting a new reward-based definition of LDI. How about:
“LDI is about efficiently seeking out opportunities for higher returns. When faced with two alternative investment strategies with equivalent levels of risk to the funding position, LDI implies taking the option which offers the greatest expected return.”
Now for those of you who have read this far and are still paying attention - thanks for that; it’s gratifying to still have an audience six paragraphs into an LDI article – you may have noticed that the new definition is still pretty much equivalent to the old one. If so, you’d be right. We are not looking to change what LDI is, just how we look at it, and the language we use to describe it.
That’s enough of abstract pontificating; let’s take a look at some of the opportunities for increasing returns.
We are talking about active management here.
Alpha-based opportunities are those short-term chances for increasing returns that arise because of market mispricings. Some people believe in the strong form of the Efficient Market Hypothesis (EMH), which would imply that short-term opportunities for adding value do not exist. We find the EMH a bit far fetched, but we are not going to try to argue the case either way here. Far brighter people than us have tried without success to conclusively to prove it one way or the other.
Assuming that it is possible to exploit short-term anomalies to increase returns, then the best way of doing so is to employ active managers. Fund managers often get a bad press, but not all of it is justified. There are one or two extremely smart fund managers about. Fund managers spend all of their time and energy seeking out alpha-based opportunities, so if anyone is going to be able to exploit such opportunities it would be them.
I appreciate that this may have seemed like an obvious thing to say, but as investment consultants, it is amazing the number of pension fund trustees (and investment consultants who should know better!) we meet who apparently have perfect insight into the markets (eg “long-term interest rates cannot possible fall from here”) and take it upon themselves to bet substantial portions of their beneficiaries’ money on these short-term opportunities, rather than delegating such decisions to fund managers.
So how can an LDI mindset influence active fund management?
‘Off-benchmark’ investing, as the name suggests, is where a fund manager switches into assets that are not in her benchmark. For example she might switch into overseas bonds when the benchmark is 100% UK gilts. Whilst this is not exclusive to LDI, it is very consistent with an LDI philosophy - searching for attractive opportunities wherever they may be.
Taking this one stage further is the so-called ‘portable alpha’ approach. Here the active manager exploits short term opportunities in a market that is unrelated to the market you want exposure to. For example you may want exposure to the UK equity market, but feel that (for example) the best opportunities for adding value through active management exist in the Philippine equity market. So get a fund manager to run a Philippine equity fund for you, but hedge out the exposure to the Philippine equity market as a whole using derivatives. You would be left with just the value of the ‘bets’ that the fund manager was placing on your behalf (effectively creating an absolute return fund). Simply add back in exposure to the UK equity market through suitable derivatives, and this ends up as ‘portable alpha’.
One last point to make about alpha-based opportunities is that unlike some beta-based opportunities, such as investing in equities, your actuary may not give you credit in advance for them in the ongoing valuation (by increasing his assumption for investment returns). This may be an issue for some pension schemes.
Alpha-based opportunities exist where the market has mispriced assets. Beta-based opportunities are where the market is ‘correctly’ priced, but the market as a whole is discounting an asset due to a perceived drawback. This is what creates the opportunity as it implies that the returns in the future are expected to be higher. We describe some examples below:
- Equities versus bonds
Equities are expected to outperform bonds over the ‘long-term’. This is because investors demand compensation for the increased price volatility, increased uncertainty and reduced liquidity. The extra return is sometimes called the ‘Equity Risk Premium’ (ERP). However, quite what constitutes ‘long-term’ is debatable. Japanese equities still have not recovered from the crash in 1989. The stock market there is still worth less than half its 1989 peak, whilst Japanese bonds have performed strongly over that period.
According to the Barclays Capital Equity-Gilt Study 2000, the average outperformance of UK equities over gilts was 4.7% per year over the whole of the last century. Few commentators would expect the ERP to be as high over the whole of this century, so that 4.7% pa could be taken as an upper bound for the ERP.
Actuaries generally allow for the additional expected future return on equities over gilts when they calculate ongoing valuations. Typically their estimate of the ERP will lie in the range 1-3%pa. This is viewed as a reasonably conservative estimate, so could be taken as a lower bound.
- Overseas equities versus UK equities
There is no fundamental reason why UK equities should outperform equities from other industrialised economies in the long-term (or vice versa). The reason for investing in overseas equities is one of diversification rather than increased expected return. However, like a car driver with a newly fitted airbag, a trustee board could use the risk reduction from diversification of their equity portfolio as a great opportunity to drive faster (ie take additional risk elsewhere).
- Other equities
Small cap, emerging market and private equities are all expected to outperform (to varying degrees) the FTSE All-share index over the long-run to compensate for the increased risk and reduced liquidity.
Many pension fund trustees overestimate their need for liquidity. Most schemes could comfortably tie up at least 50% of their assets for the next ten years without impairing their ability to meet their liabilities. In our view this makes the illiquidity premium received from investing in property a very attractive opportunity for pension schemes to enhance their returns over the long-term.
- Corporate bonds versus gilts
The credit risk premium is another attractive opportunity for pension schemes. Even after allowing for the expected losses from default I would expect a well diversified portfolio of corporate bonds to outperform their gilt equivalents by 0.5% to 1.0% pa. Whilst this is small compared to the equity risk premium, the amount of risk taken to achieve this makes it an attractive proposition.
- Hedge funds
Given the absolute return objective of most hedge funds, one way of thinking about them is as an extremely actively-managed cash fund. In our view, hedge funds offer no beta-based opportunity. One could hardly expect to be paid excess returns for the ‘risk’ taken from investing in cash. Hedge-funds can be a valuable LDI tool, but it is alpha-based opportunities they are exploiting.
We know, our new return-based definition of LDI has almost fallen at the last hurdle. ‘Unrewarded opportunities’ is a bit of an oxymoron. Even a marketing consultant with a PhD in spin would struggle to sell these. But maybe that’s the point. ‘Unrewarded opportunities’ aren’t opportunities at all, they are threats and are best avoided.
- Betting on rising long-term interest rates
This is an ‘opportunity’ that most pension funds are currently set up to exploit, although not normally as an intentional ‘bet’. Small increases in long-term interest rates (gilt yields) would cause a dramatic decrease in the value of a pension scheme’s liabilities, and hence an improving funding position. That’s great, but obviously the converse is also true. Unless this risk is hedged out, the pension fund is exposed to this ‘opportunity’.
Many would argue that this is a rewarded opportunity as long-term interest rates are bound to rise from here.
Well, this is certainly not a beta-based opportunity. Long-term interest rates do not systematically rise (or fall for that matter) over long periods of time, unlike say equity market levels where there is a clear upwards bias and trend over the long-term. If there was a systematic bias towards rising, then one would expect current long-term interest rates to be substantially above where they were in 1900, which in turn would be materially higher than the 1800 level.
Is this an alpha-based opportunity? Maybe. But we do get concerned about the scale of the ‘bets’ that many trustees place on this ‘opportunity’. If anyone can consistently call the market direction correctly it would be gilt fund managers. These are the guys who get up in the morning solely to study and predict the market. It’s interesting that most credible gilt managers shy away from taking material duration (ie market direction) bets, and yet often pension fund trustees and investment consultants, talented amateurs at best, are comfortable in taking ‘bets’ a factor of 10 or 20 times larger on less rigorous analysis.
- Betting on falling inflation expectations
We’d concede that there’s a better case that this opportunity might be rewarded. Demand for inflation-proofing has outstripped supply, and arguably index-linked gilts are expensive compared to their conventional counterparts. However, the scale of some of the unintentional positions taken by pension schemes is still hard to justify. More importantly, index-linked gilts could get a lot more expensive before they get cheaper if other pension funds decide to hedge this risk. The old adage that the market can stay irrational longer than you can stay solvent is worth remembering.
Some people believe that LDI is all about minimising risk; cashflow matching and locking into the current funding level. This is a misconception. Cashflow matching is just one (very limited) version of LDI. Hedging out the interest rate and inflation risks has been grabbing all of the headlines, but this is only part of LDI. Most pension schemes cannot afford to lock into their current funding level so they have to seek out opportunities for higher expected returns. This part of LDI is just as important as the hedging part and should not be overlooked.