Academic journal article Journal of Real Estate Portfolio Management

A Note on Intracity Geographic Diversification of Real Estate Portfolios: Evidence from Hong Kong

Academic journal article Journal of Real Estate Portfolio Management

A Note on Intracity Geographic Diversification of Real Estate Portfolios: Evidence from Hong Kong

Article excerpt

Executive Summary. There is mixed evidence on the effectiveness of diversifying real estate portfolios geographically. Some suggest disappointing results are traceable to the use of geographic areas that are too crudely defined. We divide Hong Kong into submarkets, and find that this intracity geographic diversification produces marginally improved portfolio performance in some cases, but that several naive diversification strategies are effectively efficient.

Introduction

Modern portfolio theory holds that one can reduce risk for equivalent rates of return (or increase rates of return for equivalent levels of risk) by combining assets affected by economic fundamentals in ways that are less than perfectly correlated (Markowitz, 1952). Diversification benefits may be captured by combining different classes of assets, such as stocks, bonds and real estate. Further improvements in portfolio performance may be possible by mixing on the basis of differentiating characteristics within an asset class. For example, one may vary stock on the basis of capitalization (size) or by industry class. For real estate, refinement often comes by acquiring property in different locations or by acquiring different property types or both.1

The benefits of geographic diversification are not clear. Early work (for example, Miles and McCue, 1982; and Hartzell, Hekman and Miles, 1986) typically divided the United States into four regions, and found little benefit to diversifying among these broadly defined areas. The use of eight regions and then individual metropolitan areas (Hartzell, Shulman and Wurtzebach, 1987; and Giliberto and Hopkins, 1990) produced only marginal improvement. The explanation for these somewhat disappointing results is that "pure" geographic diversification is fundamentally naive; what really matters is the set of underlying economic and institutional differences among the geographic areas selected. These differences are captured only coincidentally by geographic diversification based on political boundaries, often resulting in economic homogeneity between areas or heterogeneity within areas, both of which dilute potential diversification benefits. In response, recent work has focused on diversification based on industry or employment characteristics instead of political boundaries. The encouraging results of Mueller and Ziering (1992) and Mueller (1993), opened the question of whether diversification within even more narrowly defined areas-intracity-would produce comparable (or improved) portfolio performance. The answer is important because the narrowing of investment focus carries with it the potential benefit of reducing both information and management costs (Shilton and Stanley, 1995; and Capozza and Seguin, 1996).

There are reasons to be optimistic about intracity diversification. Grissom, Wang and Webb (1991) found substantial variations in intracity returns in Texas cities, and Rabianski and Cheng ( 1997) found low correlations between office and industrial vacancy rates in intracity submarkets in Atlanta, Boston, Chicago and Denver. Differences in suburban and CBD office markets are frequently observed, and in practice we refer not to the New York market, but rather to the upper, middle or lower Manhattan markets, and not to the Hong Kong market, but instead to the Central or Admiralty or Kowloon market. In the first attempt to measure the impact of intracity geographic diversification on portfolio performance, Wolverton, Cheng and Hardin (1998) produced significantly increased risk-adjusted returns in the Seattle apartment market. They began with fifteen neighborhoods, then reduced that number to five based on similarities in rent (income) levels. This was a hybrid approach that refined pure geography (neighborhoods) to reflect the underlying economic and institutional drivers of rent. Their methodology and results support the expectation that the ability to measure diversification benefits is positively associated with the ability to group investments based on their homogeneity. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.