Believers in the benefits of active styles generally acknowledge that both value and growth investment strategies can and do add value in a multi-manager portfolio. Strong believers of a particular style might overweight that style versus the other, but most acknowledge that style diversification offers tangible benefits.

If style diversification can add value for a portfolio of US stocks, then why not for a global equity portfolio? Some argue that since it is proven" that the majority of an international manager's returns are from picking the right country, then a manager should be an active country-picker and a (more-or-less) passive stock-picker. Others argue that "value" stocks have demonstrably outperformed both the country benchmarks and the indices of international "growth" stocks. Still others point to studies that indicate that international "growth" managers have consistently done better than those of the "value" camp.

Over the past four years there have been a number of important studies, by both academics and market practitioners, that address the value versus growth debate. The investigations have been done on at stock, country and manager levels.

One study, International Value and Growth Stock Returns by Carlo Capaul, Ian Rowley and William F Sharpe (published in the January-February 1993 issue of Financial Analysts Journal, analyses returns from portfolios of stocks with high price to book ratios ("growth" stocks) and returns from low price to book portfolios ("value" stocks). These were analysed using data from six countries from 1981-92. While it is clear that managers use many more analytical tools to make portfolio selections, the authors contend that a price to book ratio is a fair dividing line between the value and growth strategies.

The six countries analysed were Japan, France, Germany, Switzerland, the UK and the US. The authors computed the returns for each style portfolio and then subtracted the growth returns from the value returns. In the US, the outperformance of value over growth stocks was 15.9%. In Japan, the cumulative outperformance was 69.5%, in France 73.7%, in Germany 17.7%, in the UK 31.5% and in Switzerland 42.7%. The authors found that the global value-growth spread was almost 40% during the measurement period. They summarised: "Value stocks outperformed over the period covered and was significant on a global basisŠ This suggests that an investor considering a tilt toward value stocks would be well advised to implement the policy globally."

Recognising that the period measured was one of generally rising stock prices, they decided to see if the outperformance phenomenon was solely because the "value" stocks were more sensitive to overall market movements than the "growth" stocks (as measured by stock betas). The results of this analysis indicated that the outperformance was not simply a reflection of differing beta values. In fact they went on to say, "Šin most cases, value stocks had lower beta values than growth stocks. In a period such as the one covered in this study, one would expect the lower beta values to cause them to underperform growth stocks. But just the reverse was the case."

Another study, A Value Based Approach to International Equity Investing, by Paul Jackson of Morgan Stanley, published in June 1994, looks at the growth versus value question on a country level. He analysed the relationship between country valuation levels and subsequent returns for 1975-93. Jackson ranked markets at the beginning of each year according to relative valuation of a number of measures, such as price to earnings, price to cash flow, price to book value and yield. Each ratio was then ranked from lowest to highest valuation and quartiled. Stock performance results were then measured against the local market index.

His findings were consistent with those of Capaul, Rowley and Sharpe. He wrote, "All four measures of value appear capable of generating excess returns, with first and second quartile performance superior to third and fourth. "Price to book value and yield seem to show better results."

So it seems that stock level returns consistently suggest a value orientation might enhance the returns of a portfolio of international stocks. But does this hold true at manager level?

According to consultants InterSec Research Corp, the answer to this question is not as clear. InterSec, recognising the difficulty of categorising international equity managers, has nonetheless constructed peer group universes and published the results of a sample of 24 value managers and 24 growth managers. InterSec uses its knowledge of managers' investment style to construct the performance samples. Each of the managers analysed was managing portfolios of non-US equities benchmarked to the Morgan Stanley Capital International (MSCI) EAFE index.

The results indicate that, on a total portfolio basis, the median value manager did indeed outperform the median growth manager, but by only 70 basis points during the three-year period ended September 1996. Over the 10-year period, the median manager performance is 12.8% for value and 10% for growth. However, the results are decidedly in the growth managers' favour once the portfolio analysis was regionalised during the three-year period.

In many respects, style analysis for international equity managers is still in its infancy. Classification is difficult, style benchmarks are generally quite new and peer group universes tend to be small. Academics and practitioners have studied stock level data and, in general, point toward value stocks for return enhancement and as a portfolio diversifying agent. However, when measuring manager level data the results are in important respects contrary to the stock level data. Lastly, both the stock level and manager level data seems to recommend an active approach over a market-like approach for portfolios of internationally diversified equities.

James Diack is with Geneva-based Brandes International Partners"