Peter Mitchell looks at the latest research exploring the links between listed and direct property in Asia

There is a view in the real industry that over a period of time the returns of REITs tend to approximate the returns of the underlying real estate, making REITs a good substitute for investing directly and with the added advantage of liquidity and much lower transaction costs. 

Much research has been undertaken in the US and Europe on this topic, the most recent being a report commissioned by the European Public Real Estate Association (EPRA) and released in February 2014. That report aimed to investigate the similarity of public and private real estate returns and risks over the relatively long horizon using data for the US and the UK. The findings indicated that public and private real estate investments can be considered to work as good substitutes in an investment portfolio with an investment horizon of several years. More particularly, the results showed evidence of a one-to-one relationship between publicly traded REIT performance and privately traded direct real estate investment performance in three out of four US real estate sectors and one out of two UK sectors included in the analysis. The return volatilities generally did not differ significantly between the REIT and direct real estate markets, regardless of sector and investment horizon.

This longer-term analysis has not been possible in Asia due to a number of factors including lack of availability of adequate data and shorter possible time series to conduct an analysis (in the case of REITs due to their relatively recent advent in Asia). Such work that has been undertaken with respect to Asian real estate has not had the benefit of the same longer-term time horizons (generally limited to five or six-year time series) and has not identified any similar linkage between returns.

However, with the benefit of access to more data and being able to undertake a more long-term analysis, APREA has recently published a landmark new research report aimed at understanding if there are any trends indicating the same sort of linkage as exists in the US and Europe. The report, written by IPD, indicates that the direct/listed linkages in Asia are consistent with research in the US and Europe.

One of the major challenges of the research was to establish robust data series on which to conduct the analysis. Most Asian institutional property markets are not as old or developed as those in North America or Europe, so data availability is limited. 

Data availability varies by market and by country, but this study focused on the nine-year period from Q1 2004 to Q4 2012 where sufficient data was available across all markets covered in the research. For the direct markets, IPD data was used, supported by CBRE market data. TR/GPR/APREA Composite Index data was used for listed real estate. A bespoke measure was created from the index to remove developers because their performance is less intrinsically linked to the performance of the underlying property assets compared to asset owners and managers. Excluding developers in this manner made the listed indexes more comparable to the direct indexes which only comprise standing investments.

The research is unique, and particularly important, in these respects:

The length and uniqueness of the time series across these Asian markets, enabling robust comparisons of listed and direct markets. The comparison of a ‘matched’ sample of the performance of the direct assets of listed companies and the listed company performance represents a unique piece of analysis, and demonstrates (through the higher correlations for the matched sample) the value of this approach.

Why is a matched sample comparison important? One problem common to all studies of listed and direct market relationships is the potential for an underlying sample bias. This problem could arise where the listed and direct indexes are constructed from different sub-sets of commercial property. If there is a substantial difference between the properties underlying the two indexes, any divergence may simply be due to sampling issues. To test this, two bespoke ‘matched’ return indexes were created where the property underlying both of them is the same. This exercise was conducted at a pan-Asia level to give some indication of the overall regional linkage between listed and direct property markets.

The particular methodology employed. The paper uses simple statistical analysis supplemented by qualitative analysis of the relationship between listed and direct markets. It focuses on correlations between the performance series as well as comparing overall performance metrics (risks and returns). In doing so, adjustments are made for appraisal smoothing leverage, and lag. Not making these adjustments would underestimate the level of linkage between listed and direct property markets. Why are these adjustments important?

Appraisal smoothing: Because direct property indexes are valuation based, they incorporate smoothing that does not exist for listed indexes. It is this smoothing that leads direct indexes to be less volatile than listed indexes in most markets;

Leverage: Listed returns are subject to leverage, but direct property returns are unlevered. Leverage is accretive during periods of growth, but it is dilutive when returns are negative. In other words, adding leverage will increase the magnitude of both positive and negative returns, though the impact is not equal due to the cost of debt. When comparing the absolute returns and volatilities between markets it is important to take this into account. If ignored, unlevered returns will be lower and less volatile than levered returns in most markets;

Lag: Direct property markets are somewhat slower to react to information than listed markets. With valuations often being annual and somewhat backwards looking, it can take longer for direct indexes to respond to external events than listed indexes. This means direct property will likely lag listed property in most markets.

Within the global context, the results are consistent with those from previous studies in the North American and European markets, and suggest that Asian is developing in a manner similar to those markets. The results represent a significant step forward in understanding the performance of Asian real estate markets. 

Peter Mitchell is a consultant and former CEO of APREA