Risk diversification gains from metropolitan housing assets

• Published date: 01 October 2019
• DOI: http://doi.org/10.1002/rfe.1062
• Author: MeiChi Huang
• Date: 01 October 2019

Rev Financ Econ. 2019;37:453–481. wileyonlinelibrary.com/journal/rfe

|

453

© 2019 University of New Orleans

1

|

INTRODUCTION

This study explores diversification opportunities across a variety of assets, including houses of metropolitan statistical areas

(MSAs), REITs, stocks, bonds, and riskless assets. The empirical fact that different types of assets experience different price

fluctuations is worthy of close observations, particularly for the 2002–2005 period, which the Michigan Survey of Consumers

specifies as the housing boom phase: an early boom from 2002 to 2003 and a second stage from 2004 to 2005. During the

housing boom, some MSAs displayed a sharp rise in housing prices, while others had mild housing market fluctuations

and stable housing prices. For instance, the real monthly housing price of Los Angeles1 approximately doubled from 2002

to 2005. In contrast, New Orleans, whose housing boom–bust cycle is milder than other cities, experienced a 16% decline

in its housing price dynamics during the same period (Figure 1). Additionally, from 2002 to 2005, the real monthly equity

REIT price index soared, increasing 95%, but the real monthly S&P 500 stock price index rose slightly, increasing only 4%

(Figure 2). Subsequently, from 2006 to June 2009,2 the housing price in Los Angeles plunged by 40%, whereas that of New

Orleans decreased by only 20%. During the same period, the real REIT and stock price indexes dropped by about 46% and

33%, respectively. The remarkable price booms were followed by dramatic price busts, particularly for the housing assets of

Los Angeles and REITs.

Motivated by these empirical facts, this paper analyzes the 20 most populous cities in the United States (the selected MSAs

along with their population are exhibited in panel (b) of Table 1) for single- housing portfolios, which include a single- housing

asset as well as REITs, stocks, bonds, and risk- free assets. Houses play dual roles because they provide both the utility of liv-

ing space and potential capital gains, and their dual roles encourage households to establish optimal portfolios by choosing

multiple- housing assets as investment diversification vehicles. The study investigates the disaggregate- level housing markets

Received: 26 October 2018

|

Revised: 9 January 2019

|

Accepted: 10 January 2019

DOI: 10.1002/rfe.1062

ORIGINAL ARTICLE

Risk diversification gains from metropolitan housing assets

MeiChiHuang

Department of Business

Administration,National Taipei University,

New Taipei City, Taiwan, R.O.C

Correspondence

MeiChi Huang, Department of Business

Administration, National Taipei University,

151, University Rd., San Shia District, New

Taipei City 23741, Taiwan, R.O.C.

Email: meichihuang@mail.ntpu.edu.tw

Funding information

Ministry of Science and Technology of

Taiwan, Grant/Award Number: MOST

103-2410-H-305-044

Abstract

The study analyzes the roles of metropolitan housing assets in risk diversification by

assessing intertemporal hedging demands for multi- asset portfolios, which include

metropolitan houses, REITs, stocks, bonds, and riskless assets. Investors substitute

housing assets in high- population MSAs with those in low- population cities, and

they switch their holdings of housing assets to less risky bonds in the 2007–2008

housing bust. The findings from the multi- period portfolio choice problem provide

evidence for momentum reversal since forward- looking investors substitute bottom

metropolitan housing assets for top ones in the housing boom, and the GTTB index

and the lagged REIT price return have negative impacts on various asset returns.

KEYWORDS

intertemporal hedging demand, metropolitan housing asset, momentum reversal, multi-period portfolio

choice problem, risk diversification

JEL CLASSIFICATION

C32; G11

454

|

HUANG

(namely, at MSA levels),3 which enhance our analysis of multiple- housing portfolios. People often buy “vacation houses”

(Goodman and Thibodeau (2008), Miao, Ramchander, and Simpson(2011), and Zhu, Füss, and Rottke (2013), etc.) located in

different areas from their homes. Thus, the study analyzes multiple- housing portfolios, which include two metropolitan housing

assets as well as REITs, stocks, bonds, and risk- free assets, in an attempt to explore diversification opportunities for households

who are planning to buy or have potential demands for additional houses.

The study incorporates both direct properties and securitized (REIT) assets into the portfolio choice problem with an infinite

horizon, which is designed for the analysis of dynamic and forward- looking portfolio allocations. Campbell, Chan, and Viceira

(2003) and Rapach and Wohar (2009) address multi- period portfolio choice problems for investors with an infinite horizon and

Epstein- Zin- Weil utility and characterize asset price returns using a vector autoregressive (VAR(1)) model. In their framework,

the total demand for assets is decomposed into a myopic demand and a hedging demand: the former lasts for only a single

period, and the latter arises to hedge against expected adverse shocks to the asset markets. The study is methodologically moti-

vated by the multi- period portfolio choice problem.

The following concerns of households and investors are addressed in the paper: are the hedging demands higher for hous-

ing assets located in high- income (population) MSAs? Do they significantly hedge against adverse shocks to financial markets

by investing housing assets experiencing high price appreciation? Which variables can serve as good predictors for asset price

returns? To answer these questions, the study considers three grouping criteria for multiple- housing portfolios. Pairs of MSA-

level housing assets are grouped together and included in the portfolios based on their income levels (economic features),

population levels (demographic features), and median housing price growths (housing market conditions). The findings from

multiple- housing portfolios elucidate the diversification opportunities for real estate investors, and they are informative for

FIGURE 1 Real housing price indexes: Los Angeles vs. New Orleans

20

40

60

80

100

120

140

1975 1980 1985 1990 1995 2000 2005 2010

LOSANG ELES NEWORLEANS

FIGURE 2 Real Stock & REIT Price Indexes

Notes: Figure 1 shows the real monthly housing price indexes of Los Angeles and New Orleans (Freddie Mac House Price Index); Figure 2 shows

the real monthly stock price index (S&P 500) (right axis) and the monthly REIT index (equity REIT total index of the FTSE NAREIT US Real

Estate Index Series) (left axis). The data series span from January 1975 to December 2012.

0

1,000

2,000

3,000

4,000

5,000

6,000

0

200

400

600

800

1,000

1975 1980 1985 1990 1995 2000 2005 2010

REITS SP500