There are potential benefits when including public real estate assets in a mixed asset portfolio. Public real estate securities can provide investors with access to another asset class, lower trading costs compared to direct real estate investment, liquidity and divisibility.

However, it is has been argued in the academic literature that public real estate securities provide little diversification benefit in a broad equity portfolio. The evidence on this point is mixed.

It must be remembered that real estate in general is a defensive asset and as such posits significant diversification benefits particularly during periods of strong market movement. Previous academic work has shown that during the 1997 stock market decline, listed real estate firms had less price decline, faster recovery and lower bid-ask spread widening than typical stocks. Additionally, other research has established that real estate provides cross-country diversification even though the primary aspects of the diversification are from location diversity.

We believe that whether or not European real estate adds value to a mixed asset portfolio is an important question. Thus, in this study, we examine a set of European data supplied by the European Public Real Estate Association (EPRA) to determine the diversification benefits and associated characteristics with real estate during the time period 1990 through June 2006.

Our overall results suggest that real estate can contribute strongly to a mixed asset portfolio in terms of return enhancement and risk reduction. While there were variations in returns by country and by market conditions, European real estate securities have performed strongly over the last decade.

European real estate provided strong performance benefits for investors in the period from 1990-2005. Calculations show high Sharpe ratios for most countries (the Sharpe ratio should be zero for a fair priced stock or asset). All countries, except Belgium and Sweden, have positive Sharpe ratios on a nominal basis. France and Spain showed very strong positive Sharpe ratios implying that an investor would earn excess return based upon risk. Only in France, Italy and Spain do the Sharpe ratios for real estate exceed that of equities over the full-sample period.

When the 10-year risk adjusted performance figures are calculated, the risk adjusted performance of real estate is found to exceed that of equities in all European countries (see figure 1). France has the highest risk adjusted performance followed by Spain, the Netherlands and Sweden.

If real estate is added to a portfolio of stocks, bonds and cash, the benefit from holding real estate in a mixed asset portfolio can be clearly seen. In figure 2, the impact of adding an increasing amount of real estate to a portfolio containing equities, bonds and cash is shown. For all countries, as more real estate is added to the portfolio the average return of the portfolio increases (moving from portfolio 1 to portfolio 3). In portfolio 1, the asset allocation does not include real estate, 50% of the portfolio is in equities, 40% in bonds and 10% in cash. In portfolio 2, the proportion in equities and bonds drops to 45% and 35% respectively, and 10% of the portfolio total is invested in the stocks that make up the EPRA index in each country. In the portfolio 3 the proportion invested in real estate is increased to 20% of the total, while the amount in equities and bonds falls to 40% and 30%. With the exception of Italy, Spain and the UK, the total risk (standard deviation) of the portfolio falls.

In order to gain a more detailed understanding of the relationship between the different asset classes, a complete correlation matrix between real estate, equities, bonds and cash returns is presented for each country in the full report. A summary of the average correlation values is presented below in Exhibit 2.

Figure 3 shows a summary of the average correlation values between real estate, equities, bonds and cash. The most notable trend is the high correlation between equities and real estate securities. This reflects the fact that real estate and equities are key assets in an economy and likely to be driven by similar economic factors. However, as the analysis in the section above demonstrates, real estate securities do provide diversification benefits to investors despite this high correlation. The apparently low correlations reported between the stock market and measures of the performance of direct (or private) real estate markets are likely to be subject to substantial measurement error (this problem is known as valuation smoothing).

Real estate has a small positive correlation to the bond market and this may reflect the income characteristics of public real estate securities and the relationship between property yields and bond yields. Real estate is also more sensitive to increasing interest rates than is the overall stock market (correlation of -0.09 compared to -0.06 for stocks), although this difference may not be statistically significant.

In the last 10 years, public real estate markets in Europe have performed very strongly. On a risk adjusted basis real estate markets have outperformed equities in all of the major European markets. However, we must caution that this may be related to the specific period of analysis. Longer-term analysis seems to indicate that real estate has similar performance to equities on a risk-adjusted basis. However it has the key attribute of performing well when the market does badly and this has been observed in the period of analysis of this study. The research data has also shown that despite a large correlation with equities, investment in public real estate markets provided European investors with strong growth and the ability to reduce risk by combining it with other assets in a well-diversified portfolio over the sample period

John Glascock is Grosvenor Professor of real estate finance at the University of Cambridge department of land economy and Shaun Bond is senior lecturer in real estate finance