A robust attack was on the basis of much modern investment thinking the Capital Asset Pricing Model (CAPM) was mounted at the Pension Summit 2004, in the Netherlands organised by Fund Partners, at an event where much attention was directed towards the ideal investment for pension funds.
Frank Sortino, director of the San Francisco based Pension Research Institute, hammered at the implicit assumptions of the CAPM approach. “Is there such a thing as a risk free asset?” he asked. “It is never free of investment risk.”
The model says you can immunise your portfolio, by say matching the bond portfolio, for the duration of the liabilities. “But this only holds for small changes in interest rates, such as one percentage point,” he argued.
What about beta, the risk of being in the market? While alpha provides returns above market, it only holds for investors with a certain utility function. Since under the model all return correctly priced, it was pointless to look for a free lunch. It also assumes the distribution of returns are bell shaped.
“But is everyone’s goal ‘beat the market’, as the pensions consultants say?” Sortino wondered. In fact the object of the exercise was rather to fund for pensions with set cost constraints. “Beating the market is just one way of accomplishing that.”
Investors should be risk averse below a certain level, the minimum acceptable return, according to the theories of one US economist Peter Fishburn. “They should be risk neutral above this.”
Behavioural finance theory showed people do not behave in that rational way. “They can be risk averse for small losses, but risk takers for large losses. So people are their own worst enemy. Research findings show that people do not diversify their assets and that they do not know how to handle risk. “So they are scared at bottom of market and foolhardy at top of market.”
Alpha, beta and sigma, had nothing to do with the goal of a funded pension arrangement in his view: “As they leave out the liabilities.”
Some rate of return must be earned on the assets to meet the liability stream in the future. “I call this rate the minimal acceptable return. But what these liability streams are is not known with certainty.” Then in finance there is no certainty.
Based on the work of Fishburn, Sortino believed that uncertainty can be separated into a risk component and an reward component. “There is some return you must have to meet this liability stream in the future. The risk is that you fall below that and the reward is getting more than that.”
Past returns is a bad way of predicting future returns. “But what could have happened in the past, turns out to be a better predictor of what will happen,” he claimed. Looking at recent work of a number of economists, he claimed it was possible to make better estimates of risk and rewards. “Bill Sharpe says look at the style of the manager – as that will tell 90% of the source of the return.”
He said it is possible to generate distribution of returns, which he uses to calculate the inherent characteristics of a manager’s returns. That proves to act as a better predictor of what a manager will do, rather than looking at what he has done in the past.
Sortino indicated how to measure this ‘upside potential’. “If you have to earn 10% minimum, with annual returns generated, give yourself a score of zero for returns below, and calculate the amount the returns are above minimum rate and score appoint for every percentage, so 12% scores 2. Add the scores up and divide by number of observations the result is a figure that combines, frequency with magnitude,” he said.
“The crucial number is the MAR that has to be earned as the minimum to accomplish your goal. Any measure of performance that ignores MAR is not relevant.”
When he looked for evidence to support these concepts, he found that if he ranked managers by their upside potential ratio: “Data over 16 years to 2000, showed that upside potential strategy beats average mutual fund 65% of the time.”
The strategy of maximising upside potential to downside risk ratio could have very positive results, according to Sortino. It can lead to a lower MAR, lower cost of debt and cost of capital, thus yielding higher profits, and higher share price performance, than either just matching cash flows or durations, or a strategy of having low upside potential and low downside risk, he reckoned.
A series of challenging questions were posed to the audience mainly of pension fund delegates, by Keith Ambachtsheer of KPA Advisory Services in Toronto, who also asking what needs to be done to provide post retirement income streams, while dealing with default risk, multiple employer risk, inflation risk and longevity risk. Do suitable securities exist to deliver this? The answer was no.
Without the perfect pension securities, either an ‘intermediary’ must match the risk between the perfect pension security and what the financial markets can provide, or individuals must underwrite this mismatch risk. “But there are not any logical underwriters,” he pointed out
However two steps could be taken to minimise the ‘perfect pension mismatch’.
The first is to “pool longevity/mortality risk among large groups of participants, making self-insurance a practical alternative,” said Ambachtsheer. “Use long tern default risk free, inflation-linked financial securities to lay off the default and inflation risk elements of the perfect pension contract.”
The snag is that the cost of perfect pension deal works out at 25% of pay or more per annum. “That’s hitting the wall, as we cannot afford that!” By taking investment risk and earning risk premium, the extra return could reduce the contribution rate. “But there is risk in doing this, which can bite you as we have found in recent years!”
While there were not as yet “effective pension delivery institutions that can sort out these pension risk measurement and management issues,” he said, adding: “I do not think we are there yet, though a few institutions are working on it. The Netherlands were furthest along in understanding what the issues really are.”
Building better pension plans required understanding the economics of the pension adequacy/cost trade-off and identifying the various risks requiring measurement and management. It involved designing pension delivery organisations capable of assessing stake holder risk tolerances, setting risk budgets, and managing risk exposures dynamically.”
Under the cult of the equity doctrine, the practice in UK and US defined benefit schemes was to invest as much as 75% in equities, said John Ralfe, formerly of Boots in the UK, and now an adviser to company boards on pension issues.
This approach had resulted a bigger pool of risk capital, which gave the UK an extra competitive edge. “The Chancellor Gordon Brown believes this.” But this cult was under challenge.
Ralfe said maturing schemes and accounting standards were hastening a move from DB plans. “Actuaries have started to move away from actuarial smoothing.” Credit rating agencies are looking at pension funds.
But there were more fundamental challenges to the UK equity investing DB model, he said, pointing to the effect on corporate sponsor’s balance sheet. Equity is the same as having debt on the balance sheet. “It is easy to think that pension fund is self-contained.” Inflation linking of benefits was another challenge. Also, the element of discretion in relation to benefits had eroded.
If a company does want to invest in equities directly, then why should it invest in them indirectly? “Current accounting is very much to blame.”
The accounting gets liabilities off balance sheets and assets off balance sheets, but the losses from the mismatch are hidden, while the profit and loss accounts are smoothened with arbitrary amounts.
“This gives the impression that pension fund equity investment has the superior expected return of equities, with a lack of volatility vis-à-vis the underlying liabilities of bonds. Such an instrument is clearly the philosopher’s stone,” Ralfe declared. “If anyone could devise such an instrument, you would have to beat people off, so great would be the demand.”
FRS 17 wants deficits and surpluses brought onto the sponsoring corporates’ balance sheet. “People are keen to take surpluses on, not so keen on deficits.”
To invest in equities is to invest in assets with zero financial value, he said. “The expected cash flow discounted at the risk adjusted discount rate must equal the price.A £1,000 of equities, bonds or cash has the same value. If we think £1,000 worth of equities are worth more what should we do – buy equities and sell bonds.” The equity risk premium is just a return for risk.
“I am not against holding equities, I just question whether they are right for DB plans. With the pension fund investing in equities, it is doing something the individual can do directly. Why should company managements try to second guess the individual shareholder?”
In the UK it is tax efficient for individuals to hold equities thanks to tax credits and tax efficient for pension funds to hold bonds. The credits have a value of at least 10% in favour of individuals. “This is not a good advert for holding equities in a DB fund.”
In previous years, trustees had a great degree of discretion about enhancing benefits. It made sense to hold bonds to match guarantees and equities to obtain the upside potential. But now inflation guarantees were more onerous thanks to trustee action and legislation. “This change makes it much less sensible for trustees to invest in equities. They should be protecting the downside, not looking for the upside.”
The costs of pension contributions do not become any less because you choose to invest in something other than bonds, he said.