Property has a vital role to play in adding to the diversification of a pension fund portfolio, as Mark Long and Tim Francis report

We believe that the benefits of holding a proportion of your investment portfolio in commercial property are unequivocal. Strong and stable performance with a relatively low correlation to other asset classes makes property the actuarial holy grail of investment assets; it both enhances performance and reduces risk. We have analysed direct property performance in the UK and continental Europe and conclude that:

Increasing allocations to direct property can significantly improve the risk-adjusted performance of multi-asset portfolios in most European countries;(1) The most valuable benefit which property investment brings to multi asset investors is diversification and risk reduction; Weak correlations between international property markets strongly support diversification through cross-border investment, which enables investors to take advantage of cyclical opportunities to boost portfolio performance; The high correlations between international equity and bond markets suggests that cross border investment adds little to portfolio diversification; Over the medium term the listed property market also offers significant diversification benefits to investors but over the short term these investments suffer a higher level of volatility and correlation with wider equity markets.

Commercial property is recognised as a hybrid asset, sharing some of the performance characteristics of both equities and bonds. It has exhibited a sustained weak correlation with these asset classes over long periods of time but the relationships are not fixed in the short term.

During periods of static rental growth, property can begin to take on fixed-income performance characteristics, similar to bonds. Conversely when rental growth accelerates, usually the result of strong occupier demand, property performance can be more closely correlated to equities, as we have observed in the UK.

In this short article we present our analysis of direct property performance data, sourced to IPD (Investment Property Databank), in 10 European countries, and also draw together conclusions from research into indirect (listed) property performance.The UK data series dates back to 1971 and allows us to make a long-term comparison of property performance against other assets. While IPD data series for other European countries are more limited, we nonetheless believe that sensible conclusions can be drawn about their medium-term performance.

UK commercial property has an enviable track record, having out-performed UK equities and UK bonds over the previous one, three, five, 10 and 20-year horizons. Such a strong relative performance is unlikely to be maintained since much of the outperformance has resulted from movements in investment yields brought about by the re-rating of property relative to other assets. Looking solely at the performance, however, understates the true value of property - its diversification characteristics. Over an extended period of time UK commercial property has exhibited a lower level of performance volatility than equities and bonds and has a low correlation with both.

The performance benefits of using commercial property in the UK can be quantified using the historical performance data of some widely recognised asset benchmarks: equities as defined by the FT All Share Index; bonds in the form of the FTA five-15 year Gilt series; and property according to the IPD Annual Index. In the table we have constructed four hypothetical portfolios, each with an equal weighting of equities and bonds into which we have introduced an increasing share of property (see figure 1).

Over the period 1971-2006 the addition of direct property to a portfolio of equities and bonds would have increased returns and reduced risk, ultimately resulting in a 50% improvement in risk-adjusted returns, had all three assets been held in equal proportion.
This improvement in risk-adjusted returns derives mainly from the diversification benefits that property brings, not the strength of its performance. So, the most powerful argument for multi asset investors to invest in property in the UK is the benefits of diversification it brings with it.

There is less published research exploring the diversification characteristics of direct property in continental Europe. Here, we present the findings of our own analysis, based on total returns for country level IPD indices, stock market benchmarks and seven- to 10-year government bond indices.(2) At the outset, we have the following observations to make concerning the IPD data:

Data series for the selected continental European markets vary in length from six to 12 years, which is considerably shorter than for the UK and Ireland (each are in excess of 20 years). The analysis for countries with the shortest time series (Denmark, Norway, Portugal and Spain) is therefore not likely to incorporate the full length of a property cycle, which is typically 10 years from trough to trough or peak to peak. Some countries (notably Germany, Norway and Sweden) have had a high, albeit reducing, office sector component (circa 50-70%) within their overall IPD portfolio structure. We believe this helps to explain the abnormally high correlations (70-80%) we have observed between property and equity returns in these countries. Capital growth in the German IPD index has been very weak throughout the life of the index, falling by up to 3% in all but two years since 1996. We believe this is due to the local valuation methodology, which tends to limit the growth of estimated rental values and therefore reduces both the level and volatility of returns.

Notwithstanding these caveats, our findings make for interesting reading as they confirm the diversification benefits of holding direct property within European multi-asset portfolios. Figure 2 shows that higher allocations to property can increase risk-adjusted portfolio returns (measured over the longest possible period) in each country.

Moving from a 0% to a 33% allocation to direct property improves risk-adjusted returns by a significant 48% (on a weighted average basis). This is in line with our observation regarding the 50% improvement in long-term risk-adjusted returns in the UK. The improvement is less notable in Germany (+22%), Sweden (+29%) and Norway (+33%) than it is in Spain (+61%), France (+66%) and especially Portugal (+114%).

We also note an additional, albeit connected, diversification benefit of holding direct property: low correlations between direct property and other assets in all European countries. Figure 3 illustrates total return correlations in the nine years to end-2006 across Europe. We have highlighted all property related correlation coefficients, noting that 67 out of 87 observations are lower than 0.5, indicating a weak correlation, and that 38 observations have negative correlations. In particular, we note the strong diversification characteristics of the UK, German and French property markets, which have low correlations to almost every other asset class in our analysis.

At the aggregate level we also note that the average (unweighted) correlation between property markets (0.29) is significantly lower than that between equity markets (0.79) and bonds (0.68). This implies that holding property can deliver diversification benefits in two different ways: firstly, via the property's low correlation with domestic equities and bonds, and secondly, through its low correlation with international property markets. It therefore strongly supports a multi-asset investment programme incorporating both domestic and cross-border property markets, enabling investors to benefit from favourable of cyclical opportunities.

This simplified analysis is not without its problems. The highly liquid equity and bond market performance numbers are based on actual transactions data, whereas the less liquid property market must rely on valuations extrapolated from previous transactions. This introduces an element of error into property market information and tends to smooth performance statistics, understating the actual level of risk in the market. That said, a large number of studies in the US and UK have found that even de-smoothed property performance retains its diversification benefits against other asset classes.

A number of studies analysing the diversification benefits of indirect listed property have been undertaken in recent years, and here we draw together the headline conclusions. Due to the easy availability of performance data, most literature on the subject tends to focus on the performance of real estate investment trusts (REITs) and other widely traded listed property companies.

Listed property securities are generally much more liquid than their underlying property investments, and as a result the securities tend to trade at a premium or discount to the underlying value of the properties in the company. The benefit of this enhanced liquidity is offset slightly by increased share price volatility and higher correlation with the wider equity market over the short term. However, studies in Europe and the US suggest that over the medium term there is strong evidence that the performance of listed property shares is more closely correlated with direct property, and that listed property investment can therefore deliver equivalent diversification benefits.

In 2007, the European Public Real Estate Association examined the correlations of European property equity indices with other asset classes.(3) Looking at rolling five-year correlations from 1990 to 2006, EPRA found that "property stocks show correlations significantly below 1.0, and in some cases negative readings occur".
Of particular note were the negative correlations to bond markets in Germany and the UK, and increasingly strong correlations to emerging European equity markets and global hedge funds.

An earlier study (4) by EPRA found that in every European country analysed (5), risk-adjusted returns from listed property securities over the full period were higher than in the respective equity market. In other words, it is over the longer term that the diversification benefits of direct property are more fully reflected in listed property security performance.

A similar conclusion was drawn in the US in 2006 by NAREIT (the National Association of Real Estate Investment Trusts) whose study (6) showed that the addition of REITs to a portfolio of cash, bonds and equity not only increased returns but also reduced risk between 1972 and 2005.

While the EPRA research showed that listed property held the strongest diversification characteristics against bonds, our analysis indicates that direct property offers even stronger diversification benefits over the medium term.

(1) Denmark, France, Germany, Ireland, Netherlands, Norway, Portugal, Spain, Sweden, UK
(2) IPD Annual Index 2007
(3) Source: EPRA, ‘Correlations of Property Stocks with other Asset Classes - Broadening the Investor Base Study', Sebastian and Sturm, May 2007
(4) Source: EPRA, ‘The Performance and Diversification Benefits of European Public Real Estate Securities', Bond and Glascock, March 2006.
(5) Countries studied: Belgium, France, Italy, Spain, Netherlands, Sweden, Switzerland, UK
(6) Source: NAREIT, ‘The REIT Experience in the US: Lessons and Opportunities', Wechsler, 2006

Mark Long is director of investment strategy and research at Invista REIM and Tim Francis is director of continental European research at Invista REIM.