Reshaped returns help to unscramble risk puzzle
The Best Finnish Pension Fund in this year's IPE Awards is VR Pension Fund (Division B), one of a range of pension plans for employees of Finnish Railways. VRPF has won the accolade for its achievements in efficiently reshaping the return distribution of its portfolio, as well as minimising asset/liability risks, following a change in the law.
Earlier this year, changes in Finnish pension law came into force, aimed at encouraging more equity investment by pension funds. Some new solvency requirements were also brought in. The changes raised the maximum percentage of equities allowed in a pension fund portfolio from 25% to 35%. However, the government did not want to specify where these equity investments should be made on a geographical basis.
Instead, the law stated that a proportion of a pension fund's liabilities should grow in line with the average fund's equity portfolio. This meant that it would grow over time, from a starting-point of 2% of a fund in 2007, until it reached 10% of the fund's total holdings.
The result of this change is that the pension fund system, as a whole, can also increase its equity weighting by 2% a year, without any liability risk. So the maximum for equity investing will rise by 2% a year over a five-year period, starting this current year.
However, for individual funds, there is a growing tracking risk. VRPF has assets of around €450m and returned 9.2% on its portfolio in 2006. The fund decided that the identified tracking risk could only be mitigated by replicating the aggregated portfolio of Finnish pension funds across the board. To do this, VRPF contacted a Finnish asset management firm, which dug up the required information about Finnish pension fund portfolios, and launched two mutual funds specifically to meet VRPF's needs. One invests in Finnish equities, one in international equities.
This solved the fund's ALM problem. Furthermore, this solution was made available to other Finnish pension funds, and has improved the specific asset/liability risk for many corporates.
But this was only the start of VRPF's strategy for reshaping its return distribution and minimising asset/liability risks. In May, as valuation levels globally had become quite stretched, within both the equity markets and credit markets, there seemed to be more downside risk than usual, and there were still very few good opportunities for getting a reasonable return. The fund says that buying puts would have been a sensible idea, but as the offered cross-asset correlations were very high, this was not feasible.
In June, VRPF initiated a project to set up algorithmic portfolio management, a multi-asset class portfolio with a dynamic asset allocation, aiming for at least a 0% return. The fund says that as the portfolio strategy does not involve any derivatives, the solvency treatment is no worse than for a traditional portfolio.
By tuning the algorithm to give up any annual return above 20%, the downside protection became very cheap, leaving the portfolio with an expected return of around 7%, with no notional downside.
Reshaping the return distribution to non-normal means the chances of producing spectacular returns in individual years are gone, but so are the risks of seeing years when there is a disastrous performance.
In VRPF's hedge fund portfolio, meanwhile, the emphasis has been on distinguishing alpha from different sources of beta. Even though at first glance the difference may seem to be merely a matter of semantics, it is important to have a deeper understanding of the substance, while at the same time estimating risks and judging appropriate fee levels.
CHANGE ONE, CHANGE ALL
Analysing hedge funds by strategy and risk factors is like solving one side of a Rubik cube - in other words, changing one parameter changes all the others.
The largest part of what is called alpha within the hedge fund industry is, in essence, alternative beta - that is, exposures to quite liquid risk premiums that are available for every investor, without the expensive hedge fund wrapping.
Even though these risk premiums tend to act as portfolio diversification tools, under normal circumstances their correlation to mainstream markets tends to grow in times of turmoil. In general, they are a good addition to the portfolio, although their ability to bring diversification should not be overrated.
One other source of market risk which tends to be marketed as alpha is what VRPF refers to as virgin beta. Virgin beta means investments in assets which are less liquid, have a higher risk premium and by their very nature are relatively undeveloped, so that they are associated with quite high return expectations. However, a certain proportion of virgin beta investments will always give negative returns, especially in down markets.
Virgin beta is a very valuable addition to the portfolio, but it shares the conditional correlations with the alternative beta in times of stress.
Alternative beta and virgin beta are both easily and cheaply achievable, and should therefore be cheap to invest in, which the fund says puts pressure on parts of its hedge fund portfolio.
However, by identifying these betas and placing them in their appropriate baskets, the fund has lowered its total expense ratio and solved its Rubik cube of risk.
HIGHLIGHTS AND ACHIEVEMENTS
The change in Finnish pension fund law led to a need for better downside risk management, as well as a deeper focus on alpha and beta and their related costs. VRPF has tackled these challenges in a highly innovative way. To combat the growing tracking risk brought about by linking part of the growth in liabilities with equities, VRPF has replicated the aggregate holdings of Finnish pension funds within two specially constructed mutual funds. It has also set up a system of algorithmic portfolio management, which should give a return of about 7%, with cheap downside protection. In its hedge fund portfolio it has identified cheap forms of beta in a way which has allowed it to lower its total expense ratio and solve the Rubik cube of risk.