In the Republic, pension fund trustees have tended to focus on analysing past performance (typically over three to five years) when monitoring managers and making decisions in relation to the hiring and firing of managers.

However, there are several reasons why one manager may outperform another manager over time. Luck and skill are both factors and there are cycles in investment management, which means that it is very difficult, if not impossible, to predict a manager's future performance from analysing past performance data alone.

One of the factors that makes the assessment of investment managers a complicated pursuit, when based on the analysis of past performance compared to the average, is the issue of style bias.

So what is style?

Investment "style" is a term used in investment management to describe the way in which a particular manager's philosophy (beliefs about what factors drive stock prices) and investment process are combined to try to add value to portfolios in order to outperform the benchmark.

The two most common style groups to which managers profess membership are growth and value, although there are many others. Brief explanations of the growth and value are given below:

Growth investor: This is a label given to investors who favour growth stocks, i.e. those companies that are expected to show above-average rates of earnings growth and/or above-average profits growth (historical and/or prospective).

Typical characteristics of a growth portfolio include an emphasis on growth industries (e.g. technology) and an underweight position in cyclical stocks and defensive stocks (e.g. utilities). This is an area that performed strongly in the late 1990s but which has come under pressure in the past 12 to 18 months.

Value investor: This is a label given to investors who are primarily concerned with the price of a security. As such, investments are made in securities considered underpriced and which are viewed to have substantial potential for appreciation.

Typical measures used to assess value are price/earnings, price-to-book, price-to-sale or market yields. The large bluechip companies in mature industries are probably an obvious example of these types of stock that may be favoured.

For example, financial stocks such as banks, had a rough time in the late 1990s but have now begun to come back into favour.

It is important to realise that a manager's "style" does not dictate exclusively the type of stocks that they will hold but just gives an indication of the emphasis maintained by the manager on certain stocks and industries in constructing its portfolio.

Performance in investment markets has been very volatile over the past 18 months. Technology stocks were the big performers in 1998, whilst in the second half of 1999 the bottom seemed to fall out of this market.

We have recently seen "old economy" stocks return to favour with a corresponding effect on performance for some investment managers.

Take for example Bank of Ireland Asset Management (BIAM), a large Irish investment manager that operates with a "value" bias in its portfolios. Having performed well for much of the early 1990s, BIAM's performance took a turn for the worse in the second quarter of 1997 relative to the average manager in the Irish market.

They continued to underperform, with results for its rolling annual performance placing them in the lower quartile for the next 18 months. This trend was reversed briefly in 1999 before another period of underperformance and then finally good performance for the past nine to 12 months, with figures in the upper quartile of the managers surveyed in the Irish Pensions Managed Fund business.

In contrast, KBC Asset Management (KBCAM), a manager with a "growth" bias, had strong performance over the period when BIAM underperformed. However, KBCAM's fortunes reversed in 2000 and it has since been struggling to produce returns above the average.

The following scenario is not difficult to envisage:

An individual or group of trustees appoint BIAM in 1996 on the back of good five-year performance figures. After three years of disappointing performance, BIAM is replaced by KBCAM in 1999 on the back of their performance figures.

In addition to incurring the cost of the transfer, the investor has moved from an underperforming manager to an underperforming manager.

Many Irish pension funds have had the experience of "buying high and selling low" in this way. There is no way to avoid this risk completely; however, it should be possible to reduce the risk of making such decisions by placing more emphasis on monitoring other factors that may affect the future investment performance of an asset manager.

These may include changes in the team of investment professionals, changes in the ownership or business philosophy of a manager, or changes to its investment process.

In addition, when analysing past performance, this should be done within the context of what was expected from the manager and, in particular, in the context of the manager's professed style.

Style biases are not specific to the Irish market and are widely acknowledged in more developed investment markets such as the US.

The accompanying chart gives an analysis of how growth stocks globally performed relative to value stocks for rolling five-year periods over the past 10 years.

As can be seen from the chart, "value" outperformed "growth" for much of the past 10 years except between the periods ending August 1998 up until December 2000. However, as can be seen from the table, performance is very dependent on the timing of the investment.

There is, in fact, little evidence to suggest that any one style will consistently do better than any other. The quality of the investment manager's investment process, the depth and efficiency of its resources and its ability to implement decisions effectively are all just as important as the underlying philosophy. However, it should be recognised that an investment manager's style will often have an effect on its relative performance in particular phases of the market cycle.

It can be seen that the cycle identified in the chart coincides with the relative out/ underperformance of the "growth" and "value" of managers in our hypothetical example. If the managers' performance had been monitored in the context of their professed style rather than just against the average, then a switch of managers at an inappropriate time could have been avoided.

Performance should be taken into context and manager performance should be monitored relative to what would be expected from the manager given the current market conditions. The past performance should be consistent with its investment style.

In addition, an investment manager should be analysed with "soft" factors in mind, i.e. changes to business structure, people and process in combination with the hard performance data.

In summary, investors need to be aware of the investment philosophy they have bought into when selecting a manager. Managers should then be monitored relative to how they are expected to have performed given the market conditions in that period.

Trustees need to take a long-term view to investment performance and should not take any rash decisions, especially if based on short-term performance. In theory, investment performance does follow cycles where, at some stages, value managers will outperform and at other times of the economic cycle growth Managers will be in favour.

What trustees should be aware of is the timing issue in hiring and firing managers following periods of strong or poor performance, as this can prove fatal. Hiring a manager simply because recent performance has been good can be a recipe for disaster, as can sacking a manager after a relatively short period of poor performance.

The trustees need to understand the reasons for this occurrence and where exactly the manager added or lost value before any decisions on future investment is made.

Evelyn Ryder and Joseph O'Dea are investment consultants with Watson Wyatt