Econometric Analysis of the Real Estate Market and Investment

Peijie Wang

• Publisher: Taylor & Francis
• Year: 2003
• eBook ISBN: 9781134548767
• Edition: 1
• Topic: Business
• Subtopic: Business General
• Index: Business

This book provides an economic and econometric analysis of real estate investment and real estate market behaviour. Peijie Wang examines fluctuations in the real estate business to reveal the mechanisms governing the interactions between the industry and other sectors of the economy.

Part I
Preliminaries

This book is on the economic and econometric analysis of commercial real estate investment and commercial real estate market behaviour in the UK. Throughout the book, the term ’real estate’ refers to commercial real estate as in the US and commercial property in the UK for abbreviation.

Part I is an introduction to research in the book. It contains preliminary analysis of UK real estate markets and discusses the research theme and framework of the book. Chapter 1 outlines the background of research in this book and the current state of research in the area. An initial analysis of UK real estate investment and markets and real estate performance over the last two decades is carried out to present a general picture of UK real estate markets and the players. This, together with a literature review on the contemporary research in the area, forms the background of research in the book, and advocates and supports why research in this book is required.

Broad descriptive statistics are estimated and reported in Chapter 2, to gain more intuitive understanding of the real estate market and its relationship with the other sectors in the economy. In particular, this chapter examines the links between the real estate market and a large number of the real sectors of the economy and financial markets, in both the short term and the long run. Chapter 3 set out the research framework and methodologies for the book. Three closely-related theories of dynamics, drawn from studies of business cycle, the rational expectations hypothesis, and the efficient market hypothesis, are discussed with specific reference to real estate.

1
Background

This book studies real estate market behaviour in the UK in an economic and econometric analytical framework. The book will examine patterns of real estate market fluctuations, and reveal its relationship with the economy. In particular, it will investigate the ways in which real estate influences, and is influenced by, the other sectors of the economy, and identify the sources and mechanism of shock transmission in the real estate market. As part of this approach, economic and econometric models will be developed to study real estate fluctuations and dynamics in both the short term and the long run and in a multivariate environment. Topics including common cycles and common trends, sources and persistence of shocks, and cointegration, error correction and market efficiency will be covered.

Research in this book is motivated by the need for a reconsideration of real estate market behaviour which, until recently, has generally been viewed as different and distinctive from other financial markets. This separation has failed to develop satisfactory explanations, and arises from a lack of economic studies on the UK real estate market utilising modern finance and economic theory. The research will investigate real estate market behaviour in three related areas: the efficient market hypothesis; rational expectations; and business cycle theory. These three theories, with specific reference to real estate market, eventually lead to Real Estate Dynamics – the theme of this book. Real estate dynamics provides economic and econometric modelling for, and explanations of real estate’s cyclical movement. It deals with the underlying mechanism which, in this book, is to be theoretically derived and empirically tested.

Current research in real estate

Most real estate research has been carried out differently and distinctively, compared with other research are as in finance or economic science in general. Research in real estate has been largely professional. However, the last decade has witnessed the reorientation of the study on real estate from professional towards academic and from surveying and valuation towards finance and economics. Accompanying these trends are the development and use of real estate performance indices in con-formity with stock market research; the analysis of real estate market efficiency in a modern investment analysis framework; the study of real estate market behaviour, expectations and rationality; and the investigation of the cause and nature of the real estate cycle with business cycle theories. The studies are sometimes sporadic but encouraging.

Most studies have been documented in America. The operation and institutional structures of the American real estate market are by no means the same as in Britain. However, the implications and methodology of such studies should be useful elsewhere.

Reviewing literature in real estate research, Fisher and Webb (1992) have summarised six current issues in the analysis of commercial real estate, three of them closely related with economic and financial theories. They are:

• The macroeconomic and other factors affecting returns, which emphasises APT (arbitrage pricing theory);
• Risk and return characteristics of commercial real estate. Most recent studies challenge the traditional view that real estate outperforms other financial assets. Fisher and Webb claim this could have resulted from either the use of performance measures that are not really comparable with other financial markets, or the use of incorrect measures of risk; and
• Commercial market rental index. Again, one of the major difficulties with estimating the returns on commercial real estate is the lack of transaction data. A tremendous step towards improving the ability to monitor real estate market would be achieved by the development of an index of effective market rental rates.

In fact, the interwoven research topics can be broadened and categorised as following: real estate in institutional portfolios; risk, return and performance comparisons; the derivation of transaction-driven data or indices; tests on real estate market efficiency; behaviour of real estate market and prices; and analysis of the real estate cycle. Most of the studies rely on a reliable measurement of real estate prices and performance.

Webb, Miles and Guilkey (1992) generate a transaction-driven commercial real estate return series to determine whether the reliance on appraised values in the estimation of real estate returns is the source of the reported underpricing of real estate relative to shares, bonds and government securities when analysed in the traditional mean-variance point of view, an equilibrium pricing model such as the CAPM (capital asset pricing model), or a Markowitz efficient portfolio. While they find that the transaction-driven real estate returns have greater variance than appraisal-based returns for individual properties, most of the individual real estate risk is idiosyncratic and diversified away at the portfolio levels, which makes them claim that real estate continues to be a dominant asset class even when represented with transaction driven indices. In a sense, this suggests that smoothing in real estate return indices is, in fact, not a big issue at the portfolio level – a finding similar to Wang (1995a, 1995b) which allege that the weight of real estate in a three asset portfolio is not sensitive to the degree of unsmoothing applied to the appraisal based index.

Similar research is conducted by Miles et al. (1990) who inquire why commercial real estate only makes up a relatively small percentage of most institutional portfolios, even though the existing literature has consistently reported attractive risk-return characteristics that would suggest much larger allocations. This discrepancy has been explained by a perceived lack of comparability between return series calculated for real estate and those calculated for other asset classes. They make adjustments, using sales data from real estate that help comprise the NCREIF/FRC (National Council of Real Estate Investment Fiduciaries/Frank Russell Company) index, to generate a ‘transaction-driven’ commercial real estate return series. They find that: risk adjusted real estate returns of the transactionbased series are more consistent with other asset classes than the appraisal-based returns; real estate investment still presents an attractive diversification opportunity for shares, bonds and government security portfolios; and real estate is not a homogeneous asset, advising there is a need for further research on the performance of subcategories within this broad asset class.

Liu et al. (1990) investigate the consequences of several imperfections associated with real estate markets on pricing and optimal investor portfolios in a CAPM context. CAPM assumptions are relaxed to recognise illiquidity and segmented market structure, and that illiquidity reduces the extent to which investors hold real estate in their portfolios.

Chan et al. (1990) employ a multi-factor arbitrage pricing model using prespecified macroeconomic factors to analyse monthly returns on equity REITs (real estate investment trusts) which are not appraisal based. Their findings are, when a single factor CAPM model is used, excess returns still seem to exist as in the case of appraisal based returns; when a five factor APT model is utilised, the evidence of excess returns disappears; and real estate is not seen to be a hedge against inflation.

The relationship between stock market and real estate market returns is examined by Gyourko and Keim (1992). They analyse the risks and returns of real estate related firms traded on NYSE (NewYork Stock Exchange) andAMEX (American Stock Exchange), and the relation between real estate share portfolio returns and returns on a standard appraisal-based index. It has been found that lagged values of traded real estate portfolio returns can predict returns on the appraisal-based index after controlling for persistence in the appraised series. The stock market reflects the information on the real estate market that is later imbedded in infrequent real estate appraisal.

Empirical tests on real estate market efficiency can be found in Guntermann and Norrbin (1991), who use a dynamic multiple indicator, multiple cause (DYMIMIC) model to test market efficiency; in Case and Shiller (1989), McIntosh and Henderson (1989), Rayburn et al. (1987), and Gau (1984), who adopt a forecasting approach; and in Guntermann and Smith (1987) and Linneman (1986), who employ a traditional financial analysis approach. These studies generally find some evidence of market inefficiency, but the results are more or less mixed.

A recent study on real estate market efficiency and rational expectations is by Tegene and Kuchler (1993). The methodology employed is cointegration tests on present value models, as proposed by Campbell and Shiller (1987). They conduct tests for the contribution of speculative bubbles to farmland prices. These tests are carried out under the hypothesis that farmland investors rationally form expectations. They infer whether farmland prices are determined by market fundamentals – discounted returns from the highest economic land use, or whether rumours about farmland price movements are self-fulfilling. They find little to reject the hypothesis that market fundamentals determine farmland prices.

Tegene and Kuchler (1991) also use Campbell and Shiller’s present value models to test the way in which the expectations are formed, and the implications of the present value model. They find, combined with rational expectations, the present value hypothesis is strongly rejected, while combined with adaptive expectations, the hypothesis is accepted.

In the UK,Wang and Matysiak (1994) investigate the regional patterns in commercial real estate rent fluctuations and movement. Their study confirms that there is regional segment, and London and the Southeast lead other regions in rent adjustments. Following Wang and Matysiak (1994) and Matysiak and Wang (1995), Campeau (1994) examines the geographical patterns and the relationships between unsecuritised real estate investment and securitised real estate investment, by investigating the valuation process and the data generating process underlying these investments. Barkham and Geltner (1995), in a Granger causality framework, find a long-run price discovery mechanism to link direct investment in real estate and investment in real estate company shares. Similar findings have also been documented in Lizieri and Satchell (1997) and Wang et al. (1997). In real estate index research, Blundell and Ward (1987) were the first to address the issue of smoothing and pioneered an approach to correcting for smoothing in the appraisal based real estate indices. Valuation accuracy is discussed in Lizieri and Venmore-Rowland (1991, 1993), and Matysiak and Wang (1995), among others. The diversification potentials of real estate in multi-asset portfolios are studied by MacGregor and Nanthakumaran (1992), confirming US studies that real estate is a dominant asset class for diversification.

It can be seen from the above review that tremendous efforts have been made in real estate research, and significant results and findings have been achieved in recent years. However, frontiers remain and the task is as large and hard as before. Price and return dynamics has yet to be examined thoroughly in a rigorous economic and econometric analytical framework. Specifically, factors which influence real estate performance and the ways in which they influence and are influenced by real estate performance have yet to be inspected; the mechanism of adjustments between real estate and other sectors in the economy has yet to be probed; and the patterns of real estate in response to shocks of a different nature have yet to be observed. This book is, therefore, a timely project in the course of search for knowledge, understanding, and solutions.

Real estate investment and markets

Institutions

The institutions are the main investors in the real estate market with their real estate holdings amounting to £53 billion at the end of 1993, according to Financial Statistics. It would then be helpful to look into asset structures of the institutions, especially their real estate holdings over time as they control a majority of total real estate investment. Callender and Key (1996) estimate that about 45 per cent of commercial real estate in the UK is held by ‘investors’, and roughly half of this is held by institutions. Of the remaining 55 per cent, not all is owne...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Figures
• Tables
• Abbreviations and variables
• Preface
• Part I: Preliminaries
• Part II: The dynamic behaviour of economic and financial time series
• Part III: The dynamic behaviour of real estate
• Bibliography