Divergent performance provides diversification opportunities in UK real estate

We highlighted earlier in the year that UK real estate investments have delivered competitive returns over the long-term when compared with equities and government bonds. Real estate can also provide diversification benefits to multi-asset investors if it performs well when other asset classes perform poorly. However, the last two years were challenging for equities, bonds, and real estate alike owing to the upward shift in interest rates that occurred in the UK and other major economies.

When investing into UK real estate, the retail, office and industrial sectors historically followed similar paths in terms of their investment performance. However, the total returns from these sectors have diverged in recent years, varying not only in magnitude, but also in direction as economic and social changes have affected each sector differently. This is illustrated in Figure 1. While total returns rose in the office and industrial sectors after 2017, retail returns fell. More recently, 2023 saw some recovery in retail and industrial returns that was not shared with the office sector.

Figure 1: Total return indices by sector

Divergence in total return can represent a risk for real estate investors, especially if portfolios are weighted to sectors that perform poorly. Yet it can also present an opportunity for more effective risk reduction. This is because variation in portfolio returns depends on the volatility of the investments chosen and on how closely their returns are related – and hence the likelihood that poor returns from one investment will be balanced by better returns from others. Therefore, the risk reduction benefits from exposure to different sectors should increase if they no longer perform in a similar manner to each other.

A correlation coefficient measures the extent to which two variables move in the same direction at the same time, and it can range from +1 to -1. A coefficient close to +1 between the returns for two different real estate sectors indicates a strong positive correspondence in their performance, while 0 indicates no correspondence, and -1 a negative correspondence between them. Correlation coefficients are a useful signal for which sectors might offer diversification benefits based on historical performance. However, they do not formally prove that a relationship exists or reveal what the drivers for that relationship might be.

With that in mind, how strongly correlated have the returns been for different segments of the UK commercial real estate market? We compare the 10 years to May 2014 with the following 10 years to May 2024. For the earlier 10-year period, there was a high level of correlation in the returns of different segments in our Monthly Index. The average correlation coefficient was 0.88, with a range from 0.79 to 0.95 depending on the two segments that are compared.

Figure 2: Correlation matrix based on monthly total returns for the 10 years to May 2014

The returns delivered by Shopping Centres and Other in the decade to May 2014 were mildly less correlated with those of other market segments. However, the overall impression is one of a strong correspondence in how monthly total returns were changing across different parts of the UK real estate market. This does not mean that sector choice was unimportant in this period, since some parts of the market still delivered higher total returns than others, but it does suggest that risk reduction benefits from investing in different segments were limited.

In contrast, results to May 2024 suggest that potential risk reduction benefits have increased in recent years, with segment returns exhibiting more divergence in when and how they change. While the average correlation coefficient is moderately high, at 0.72, it is below the lowest correlation found for the previous ten-year period. Moreover, the range in correlation coefficients is much wider. The lowest, at 0.35, is between Shopping Centre and Industrial total returns, while the highest, at 0.94, is between Outer London / M25 Offices and Rest of UK Offices.

Shopping Centre returns continue to have the lowest correlations with those from other segments, but the average monthly total return for the more recent ten-year period was negative. This reflects that Shopping Centres were especially impacted by structural changes affecting the retail sector. While this tempers the diversification benefit suggested by Figure 3, it could be positive for future investment if these structural changes have produced a lasting shift in how retail returns behave relative to other sectors.

Figure 3: Correlation matrix based on monthly total returns for the ten years to May 2024

The Retail Warehouse segment also emerges as one whose returns have had lower correlations with those of other segments. As Retail Warehouses have seen stronger performance than other retail property types, this suggests that this segment should have been attractive from a portfolio perspective over the last decade.

For investors in UK real estate, the step of constructing or rebalancing a portfolio involves determining those sectors and regions in which to deploy capital. This is not simply a case of identifying the segments with the highest expected returns, as these choices could also involve a high degree of risk. Instead, investors should be able to achieve better risk-adjusted performance by combining investments whose returns are expected to have a low correlation. This is an important part of asset allocation, whether investing in real estate directly or when investing in the asset class through specialist funds.

Our key finding is that returns from different segments of the UK commercial real estate market have diverged considerably over the last ten years and have become less correlated as a result. While there is no guarantee that these correlations will persist in the future, it raises the interesting prospect that risk reduction benefits from a multi-sector strategy have increased. Combined with research into the underlying performance drivers, investors can use the knowledge of how segment performance is related to ensure their investment goals are achieved under various market conditions.