Real estate markets and stock markets: two stories, or just one?

Why does it matter?

Real estate indices generally exhibit a much smoother and less volatile evolution than listed real estate return performance. This is explained by three data measurement issues: the impact of property valuations rather than actual transactions, the aggregation bias of combining results over time, and the effect of financial leverage in listed entities which increases the volatility of equity returns. The net result is that real estate indices do not necessarily reflect returns to levered investors, which cannot be measured through an index^[1].

Investors however want to know whether they can achieve diversification benefits through real estate and stock markets. They also want to know in which direction causality flows, or at least whether stock markets are leading indicators of real estate (as has traditionally been held) or vice versa, as both more recent research and anecdotal evidence from global housing markets have both suggested. They therefore seek to compare returns from stock markets with those of real estate market indices. What does the evidence show?

Previous Research

Deutsche Asset Management’s analysis from 2018^[2] using international MSCI indices suggested that ‘Over a 20-year time period, correlations in developed markets between real estate returns and equities and fixed income markets are relatively low or inversely correlated’, ranging from 0.29 in Japan to 0.47 in the USA. Their research indicated that correlations actually declined from 2014 onwards, probably because of a combination of loose monetary policy following the financial crisis which encouraged stock market performance, with real estate lagging as a reflection of the real economy, including Gross Fixed Capital Investment and job creation. As a result, too, stock markets have exhibited considerably more volatility than real estate markets globally since the Financial Crisis, around half in the US case.

Academic research strongly supports this broad-based research. To take two recent papers on very different markets, for Germany, the correlation between direct property investment index and stock returns has been found to be small (around 0.2) but negative, and likewise decreasing in the more recent five years. The researchers concluded unsurprisingly that the two asset classes were often driven by different factors, particularly in the case of real estate the search for stable yields, such that ‘including real assets into a portfolio improved its overall risk-adjusted performance’^[3] For China, probably the best large-scale example of a developing market over the same long timeframe, researchers again found that co-movement between real estate and stock prices was weak in the short run, except during the financial crisis period^[4]

All this work refers to stock markets overall. By contrast, a recent South African Working Paper^[5] followed many others in examining correlations between REITs and overall equity prices in nineteen markets over 1998-2018. The working assumption was that of a bivariate normal distribution, i.e. that both REIT and equity prices were normally distributed (like heights of students in a class) and therefore combined would also have a normal distribution. The first conclusion was that REITs and equity markets within and across economies have become more integrated over time, with increasingly positive correlations and progressively less diversification benefits for investors, the exception being the US, which may be explained by oscillations in the underlying real estate market. The second conclusion was that ‘contagion’ between equity and REIT prices is now a worldwide phenomenon, and the third conclusion was that REIT prices have become progressively more correlated worldwide.

What does this combination of apparently paradoxical results tell us? That real estate markets cannot be measured by the performance of REITs, but that real estate returns globally are poorly correlated with equity markets which are dominated by many other trends, including but not limited to commodity prices, discretionary consumer spending, B2B investment and global liquidity conditions. Researchers have suggested, however, that if anything, asset markets now drive equity markets rather than the traditional direction of causality.

What we have found

Common wisdom is that UAE real estate markets are evidently much more closely correlated with local stock markets than is typical worldwide, mainly as a function of market capitalisation being dominated by real estate companies. The evidence, however, tells a more nuanced story. The chart below tracks four separate indices: the Dubai and Abu Dhabi Stock Exchanges and Property Monitor data for prices per sq ft in both Dubai and Abu Dhabi.

The Dubai and Abu Dhabi real estate markets themselves are highly closely correlated over the time period, at 0.78. The data show strikingly however different correlation results between the two markets and their respective stock markets.

For Dubai, the overall correlation since January 2016 is a positive 0.68, sufficient to justify the contention that the two markets run in effective tandem, albeit that stock market prices do continue to exhibit greater volatility, partly a function of the inevitable smoothing of the property index but partly due to speculation on liquid markets.

There is also strong evidence that these correlations have recently and quite suddenly declined in both markets. The property index smoothing effect reduces the significance of lagged indicators, but there is some evidence to suggest that real estate markets may be leading indicators, at least in Dubai. Using the latest data, the correlation between the PM real estate index and the one-year lagged stock market performance of the DFM is a phenomenal 0.92. The reverse is a vastly lower 0.26, given firm support to the role of the real estate market as a leading indicator. In Abu Dhabi, however, the history has shown almost no effectiveness of real estate as a leading indicator. Moreover, the recent rise in stock market prices, neither preceded by a jump in real estate asset prices nor at least for the time being followed by it, strongly suggests that bank lending generated by the recent revival in oil prices are probably the cause of what is an evident disconnect.

In order to study the degree of dominance of the stock market by real estate, however, the next step is to examine stock market performance of the sub-index of real estate companies by comparison to real estate prices.

Using a selected set of listed real estate companies^[6] produces a similar result: a correlation between listed real estate prices and property market indices of 0.73 for Dubai and an even stronger correlation for Abu Dhabi, at 0.79. The most recent prices, showing a divergence in overall stock market and real estate company performance, provide evidence for another contention – that real estate stock market prices in both markets are largely a function of the performance of the real estate market itself, rather than acting as a proxy for a REIT market and tracking equities more closely or developing a separate market niche of their own.

Conclusions

In the recent past, investors have not been able to achieve diversification in Gulf markets between real estate and stocks. It is very possible that this era is now coming to an end.

Two contentions are now being tested in the Gulf market. Firstly, whether real estate markets now function as leading indicators of stock market performance – here the evidence from Dubai is already strong – the issue now is whether this continues to be the case. Secondly, the degree of independence of stock markets from real estate asset prices, where Abu Dhabi is leading the way and where the next question is the extent to which Dubai follows suit.

If both these contentions prove to be accurate, then widespread understanding of demand and supply constraints in real estate markets, publicly available data and the gradual diversification of the economies of Gulf countries have allowed the delivery of the level playing field with developed country markets that local and international investors have been seeking for so long: the ability to diversify their real estate investments through the use of liquid instruments such as equity derivatives. The benefits to Gulf investors, especially those in real estate, will be very real.

[1] Katzler, S. and Song, (2017) H-S. Public real estate- correlation and volatility dynamics in the U.K mixed-asset portfolio. Available at: https://www.diva-portal.org/smash/get/diva2:1095007/FULLTEXT01.pdf Retrieved 26 April 2019.

[2] Deutsche Asset Management (2018) 2018 Spring Global Real Estate Strategic Outlook.  Available at: https://institutional.dws.com/content/_media/Spring_Global_Real_Estate_Strategic_Outlook_March_2018_.pdf p.17.

[3] Szumilo,N., Wiegelmann, T.,Łaszkiewicz,E., Pietrzak, M.B. and Balcerzak, A.P. (2018) The real alternative? A comparison of German real estate returns with bonds and stocks, Journal of Property Investment & Finance, 36(1).19-31, https://doi.org/10.1108/JPIF-02-2017-0012 The real alternative? A comparison of German real estate returns with bonds and stocks. Available at: https://www.emeraldinsight.com/doi/pdfplus/10.1108/JPIF-02-2017-0012 Retrieved 26 April 2019. p.29.

[4] Su, C-W., Yin, X-C., Chang, H-L and Zhou, H-L. (2018) Are the stock and real estate markets integrated in

China? J Econ Interact Coord (2018). https://doi.org/10.1007/s11403-018-0215-x

[5] Bouri, E. Gupta, R. and Wang, S. (2019) Contagion between Stock and Real Estate Markets: International Evidence from a Local Gaussian Correlation Approach. Department of Economics Working Paper Series. University of Pretoria. Available at: https://www.up.ac.za/media/shared/61/WP/wp_2019_17.zp169435.pdf  Retrieved 26 April 2019..

[6] An unweighted average of: Union Properties, Emaar, Drake & Scull International, Deyaar, DAMAC Arabtec Holdings, Aldar and Al Mazaya Holding Company