Taking the risk out of pension funds

Traditionally UK pension funds have focussed on six main asset classes: UK equities, overseas equities, UK bonds, overseas bonds, cash and sometimes property. The basic idea is that these asset classes have historically shown that their performance, both in terms of return and risk (ie volatility) differ over time and in different economic environments. Equities are perceived to be a rather good hedge against inflation, whereas bonds fulfil an income requirement as well as offer a form of protection on the down side as the default risk for government bonds is minimal.

The argument for international diversification has been that equity markets and to a certain extent bond markets in different countries do not perform in line, that is when for example the UK market is flat or even falling hopefully the overseas markets are rising. This approach is based on the assumption that different equity markets tend to react to different, often domestic, factors and hence have a low correlation. Historically this has also been the case as illustrated by the correlation coefficient.

A correlation coefficient between two asset classes close to +1 indicates the risk diversification achievable between the two asset classes is very low as they tend to move in line. A coefficient close to zero indicates there is no correlation between the performance of the two asset classes whatsoever, in other words, a risk diversification is achievable. Finally, a correlation coefficient close to -1 indicates the two asset classes offer a perfect diversification, so that if one market goes up the other goes down.

Using monthly observations for the equity market indices in the UK and US we have calculated the correlation between the UK equity markets and the US equity markets using the FTSE 100 and S&P 500 indices. Between 1978 and 1985 the coefficient was as low as 0.48, indicating that a UK pension fund which invested in both markets was able to achieve a significant risk diversification. If we look at the latest 10-year period, 1986 – 1995, we will find the correlation coefficient has increased to 0.77.

The implication is obvious; the risk diversification achievable by investing in the blue-chips stocks in the UK and the US stockmarkets has been reduced rather dramatically.

The explanation to this change in the ability of pension fund trustees to reduce the overall risk in their pension funds by investing in the larger and well known companies on the different national equity markets is rather straight forward and can be expressed in one word: globalisation.

The globalisation of companies has its origin in a fast growing international trade and the abolition of different kinds of trade barriers. This is especially true for sectors such as pharmaceuticals, oil and oil-related, car and consumer staples. In all these sectors we experience companies generating the largest part of their revenues from abroad and having subsidiaries all around the globe. The main players in these sectors are the well known multi- (or even trans-) national companies. It is therefore not surprising the larger companies on the stockmarkets tend to be concentrated in global sectors, whereas smaller companies tend to be concentrated in sectors which have most of their earnings generated from the domestic economy.

The flip-side is obviously that as the larger companies become more and more international the smaller companies with their domestic exposure will offer a quite interesting risk diversification. Between 1976 and 1995 the correlation coefficient between the UK equity markets and its small component, measured by the FT-All Share index and the Hoare Govett Smaller Companies Index, was 0.76. This correlation has fallen to 0.72 during the last 10 years. The implication is obvious. UK pension funds can achieve a high level of risk diversification by investing in smaller companies in the domestic market without taking on board the currency risk any overseas investment implies.

It is, in this context, worth noticing that the risk diversification achieved for an investor investing in smaller UK companies was better than investing in the main US equity market. (If one chose to invest in smaller US equities the risk diversification would have been better, the correlation coefficient would have fallen to 0.7.)

But what about the return? Have smaller companies produced the same return as the larger ones or would pension fund trustees sacrifice long-term return in order to reduce the overall volatility in the fund by investing in smaller companies? Available statistics suggest otherwise. As is illustrated in the graph, smaller companies, as defined by the Hoare Govett Smaller Companies Index, have been the best performing asset class on the UK equity market over the last 42 years. Over this period, 1955 to 1996, the Hoare Govett Smaller Companies Index has outperformed the FT-All Share Index in 30 years of the 42. On average, smaller companies have produced a yearly outperformance of approximately 4%. Anyone involved in the pension fund business would appreciate the implications of such a long-term outperformance on especially the funding costs for the plan sponsor.

As one of the main arguments for including smaller companies in a pension fund is risk diversification, the question of how much to invest in this asset class is logical. By studying the return/risk profile for larger and smaller companies on the UK equity market over 10 years we have calculated the efficient frontier. This indicates all the efficient combinations of larger and smaller companies in a portfolio. This research suggests the optimal combination is 60% in larger companies and 40% in smaller companies. This should be compared with the fact that smaller companies as an asset class only constitutes 7% of the market capitalisation of the UK equity market.

Based on the date for correlation as well as return it seems to be almost imprudent for UK pension funds not to have a significant exposure to smaller companies and indeed to treat them as a special asset class in line with, for example, overseas equities. Whether the dominating balanced fund managers in the UK are able to offer the pension funds that exposure or if a specialist manager is required is obviously another discussion. But to refrain from an exposure to smaller companies just because it is convenient to use just one balanced manager seems to be indefensible.

Peter Dencik is managing director of Singer & Friedlander International Asset Management.