Getting the mixture right
Alternative investments are the ‘new kids on the block’ in the investment arena. Although much has been written about them, there has been only limited actual investment activity by pension funds in the area. This article goes back to basics to consider the argument for including alternative assets at all within a pension fund portfolio, a theoretical framework for assessing their relative proportions and finally the practical issues of implementation.
Why include alternatives?
It is perhaps best to consider this question in the context of the rationale for holding any ‘risk investments’ within a pension fund portfolio. Trustees can regard their portfolio as being made up of two components:
q ‘matching’ assets such as bonds, specifically those bonds which behave in the same way as the liabilities, and
q ‘risk assets’ which are there to produce a return in excess of that of the matching assets to reduce the long-term cost of funding.
The total investment risk depends on the proportion of scheme assets invested in the ‘risk assets’, and the riskiness of those investment assets.
There are different ways of combining these two groups of assets. We can arrive at the same level of risk by adopting quite different strategies. We could construct a portfolio whereby the majority of the assets were invested in ‘matching assets’ with a small component invested in equities (which are quite high risk compared with the matching assets). Alternatively, we could invest all the assets in corporate bonds (which have a low degree of risk relative to the matching assets) and produce a similar level of aggregate risk.
In many ways, this is analogous to balancing containers on a see-saw (see figure). The length of the lever represents the amount of the exposure to risky assets while the size of the container represents the riskiness of those assets. For the see-saw to remain in balance, the size of the container will need to decrease as the exposure to risk assets increases (or vice versa).
The lowest risk approach would be to invest all of the fund in ‘matching assets’. This might turn out to be relatively expensive on the basis that the matching assets are likely to have a lower expected long term return compared with riskier assets. So the question comes back to one of achieving an appropriate balance between risk and long term return.
Getting the right mix
In finance jargon, the buzzwords are ‘managing the risk budget’. But the essential challenge for trustees is to seek a strategy that maximises the potential long-term return given the level of risk they are comfortable with. This is where alternative investments may help.
To make any objective assessment of the merits of any asset class (alternative or traditional) we can quantify in a theoretical basis the overall impact of each asset class in the overall risk/return profile of the pension scheme. To do so, we need the following inputs to our model:
q expected long-term return;
q volatility/risk statistics, and
q correlation with other asset classes (ie, the extent to which the asset rises or falls in line with other assets).
However, we then need to overlay any such theoretical framework with practical considerations, as we discuss below.
In most cases, we find that the inclusion of alternative asset classes (even private equity) can help contribute to a reduction in overall risk within a pension fund portfolio. However, it does depend on the level of exposure made in the alternative asset classes. A high exposure to private equity would result in an increase in the risk profile of the scheme; however, our theoretical analysis suggests that an exposure of around 5% can help reduce total portfolio risk (although this depends on the types of other assets held).
If trustees can stomach the drawbacks of private equity (which we discuss below) then not only can adding private equity help reduce total risk, it can potentially increase the long-term return to the pension fund. This appears to offer the best of both worlds – potential higher long-term returns with lower risk. There are of course no guarantees and there are a number of strings attached – private equity is a long-term investment, liquidity is poor and the risk associated with individual investments is high. Nevertheless, it does appear on theoretical grounds to be an attractive class to consider.
The ‘best’ alternative asset classes are thosethat are genuinely ‘alternative’ to traditional assets. One could regard private equity as merely a sub-set of the total equity universe and therefore not a genuinely different asset class. However, private equity might offer investors access to parts of the economy that are not readily accessible via the quoted market.
In the same vein, the corporate bond market could be regarded as the natural extension of the government bond market – both markets generally react in the same ways to world events even though there have been periods when corporates have moved in a different direction to gilts. But, in general, they are closer to the behaviour of a scheme’s ‘matching assets’ than many other investments. The corporate bond market is now relatively mature; in fact, it is as big as the gilt market, in stark comparison to the position only three or four years ago. Many pension funds are increasing their exposure to corporate bonds in their search for additional returns without a markedly increased level of risk.
Some funds, having had their fill of corporate bonds, are looking for even greater excitement in the bond market. There is now some attention being applied to ‘high-yield’ bonds – a more acceptable name for ‘junk bonds’. High-yield bonds are essentially bonds that are not within the normal universe of investment-grade stocks. They can offer very high yields relative to the equivalent government gilt but with high risk. One of the main practical issues is in identifying managers who have a skill in managing these esoteric instruments.
Another popular alternative asset class is hedge funds (and we include all manners of overlay asset allocation or currency strategies in this class). Hedge funds invest in the traditional range of stocks and shares that investors have been buying and selling for decades. What is new is that hedge funds are intended to enable investors to benefit from the skill of a number of individuals who are able to exploit relative inefficiencies within the market. It is more difficult to quantify any relationship between risk and return for a hedge fund manager in comparison with the more traditional asset classes, where there is generally some logical underpin that links risk and return. As such, we find it harder to model hedge funds using the traditional financial techniques discussed above.
Any theoretical framework for the determination of the ‘optimal’ mix of traditional assets and alternative asset classes must take account of the practical issues related to each asset class. There are a number of common issues that apply to alternative asset classes, including:
q Liquidity – private equity in particular is illiquid in comparison with the traditional asset classes and investors should seek a premium for this inconvenience.
q Achieving a broadly diversified exposure is critical. This often means that a pooled approach is the most efficient way of investing in such asset classes. Even then, there is generally a minimum level of investment that can act as a barrier to entry for all but the larger funds.
q Market intelligence – information about alternative asset classes is usually less forthcoming than it is for traditional asset classes and trustees do not have the same access to verified performance figures as they do for the other asset classes. There is therefore a cost of time premium that trustees need to factor into the equation when assessing the appropriate balance of such investments.
It takes an exposure of at least 5% to have any meaningful impact on the total fund. We might see exposures of this order to private equity and hedge funds within pension funds in due course. However, corporate bonds have already stolen the march in these asset classes and typically can represent a third or more of a pension scheme’s bond component. However, such trends are likely to be overshadowed by defined benefit pension schemes moving their general asset allocation more towards bonds (the ‘matching assets’ as described above) as they mature.
Clare Gardner is a senior investment consultant and co-heads manager research at Hymans Robertson