International Transmission Mechanisms and Contagion in Housing Markets

• DOI: http://doi.org/10.1111/twec.12288
• Published date: 01 July 2016
• Author: Jia‐Huey Yeh,Anupam Nanda
• Date: 01 July 2016

International Transmission Mechanisms

and Contagion in Housing Markets

Anupam Nanda

1

and Jia-Huey Yeh

2

1

Henley Business School, University of Reading, Reading, UK and

2

University of Reading, Reading, UK

1. INTRODUCTION

THE standard theories in financial economics (capital asset pricing model (CAPM) and the

modern portfolio theory (MPT)) suggest that a diversified portfolio can reduce risk as

shocks from one market may not transmit to other markets if these markets are uncorrelat ed.

However, economic linkages between countries have become much stronger over the last two

decades due to a multitude of factors, that is financial deregulation, advancement of informa-

tion technology and the growth of global financial transaction volume. At the sam e time, glo-

bal financial activities have become increasingly clustered within a small number o f financial

centres. Such coalescing of financial activities may strengthen linkages among the centres.

Moreover, locational attributes that cannot be substituted easily across regions may generate

significant diversification benefits due to spatial fixity. Such locational heterogeneities may

become more dominant in the case of real estate markets due to a number of special features

such as inherent spatial fixity feature, lumpy investment volumes, relatively illiquid market

and the importance of local market knowledge and skills. Therefore, the diffusion of real

estate price changes (e.g. house price changes) may follow spatial and temporal patterns.

The explanations for the house price diffusion effects may be sought through several plau-

sible transmission channels. First, an increase in the correlations among the economies may

cause diffusion of the house price movements. Second, international, inter-regional and intra-

regional migrations may contribute to the diffusion effects due to their significant contribution

to housing demand and resulting house price variations. Third, information asymmetries,

structural differences and housing wealth as well as collateral effects may cause house prices

to change first in one area, followed by changes in other areas with varying length of spatio-

temporal lags.

With the growth of financial market activities across Asia, major financial centres in the

region come under heavy competition for business location and consequent investment flows

along with increasing demand for housing. The flurry of investment activities, recent global

economic melt-down and ever-changing macroeconomic dynamics in these markets provide

an interesting case study to test domestic as well as global macroeconomic influences in the

housing market. The aim of this paper was to explore dynamics of key macroeconomic vari-

ables on house prices and test house price diffusion effects across six major Asian cities,

namely Hong Kong, Tokyo, Seoul, Singapore, Taipei and Bangkok. We select these cities due

to data availability over a long time period. Several other markets could not be included due

to short time series for the house price information. For example, the housing price index is

only available from 2000 for Kuala Lumpur. For Beijing, the house price inform ation is

restricted and available from 1999 due to the housing reform in 1998 and the free market

becoming the main channel for providing residential housing (Zhang et al., 2011). Neverthe-

less, the six cities chosen represent a major part of the Asian market. The main contribution

of this paper lies in offering plausible explanations of house price diffusion effects within a

global vector autoregressive model (GVAR) that is informed by the theoretical framework of

©2015 John Wiley & Sons Ltd 1005

The World Economy (2016)

doi: 10.1111/twec.12288

The World Economy

macroeconomic transmission channels (Goodhart and Hofmann, 2008) and the Balassa–Sam-

uelson hypothesis of international trade. Although the house price diffusion effects have been

examined variously across a number of developed countries, the evidence from the housing

markets across these Asian cities is thin.

The findings of this paper are compelling and those not only indicate important roles of

macroeconomic and international factors on house price fluctuations but also provide strong

evidences of the diffusion of these fluctuations from one capital city to the others. The GVAR

method used in this study combines domestic variables and country-specific foreign variables

in individual country VAR systems. This technique measures generalised impulse response

functions to estimate house price diffusion over space and time. The implications of the find-

ings in this paper are twofold and imperative: (i) it provides better understanding of the mar-

ket dynamics which would facilitate formulation of investment strategies and (ii) it identifies

causal relationships across these housing markets which should help the policymaking agen-

cies devise focused and effective tools to prevent any contagion effects during the times of

economic upheavals and channelise public resources more effectively.

The structure of the rest of the paper is as follows: in Section 2, we discuss theoretical

underpinnings. Section 3 presents our empirical framework including the data description and

the GVAR model. We then discuss results in Section 4. Finally, Section 5 provides conclud-

ing remarks.

2. THEORETICAL UNDERPINNINGS

The co-movement can be viewed as a process through which diffusion of house price vari-

ation occurs over spatial units. Mishkin (2007) has discussed various channels through which

monetary transmission mechanisms may work their ways into the housing markets. He identi-

fies following channels: interest rates directly influence the user cost of housing capital,

expectations of future house price movements and housing supply; interest rates indirectly

influence the real economy through standard wealth effects from house prices and credit-chan-

nel effects on consumer spending and housing demand. In this paper, we offer plausible theo-

retical explanations through two complementary routes: the housing wealth and collateral

effect, and the Balassa–Samuelson effect.

a. Housing Wealth and Collateral Effects

There may be two comparable channels through which changes in the house prices can

transmit through the economy (Goodhart and Hofmann, 2008). The monetarist view of trans-

mission mechanism suggests that increases in money supply affect the stock and marginal

utility of liquid assets relative to the stock and other assets. The resulting portfolio rebalanc-

ing process leads to increases in asset prices through adjustment in the demand for monetary

assets (Meltzer, 1995). The other comparable link is between credit level and house prices,

which could be established through interaction between the demand for credit and the level of

credit supply.

According to the permanent income hypothesis and the life cycle model, consumption of

households depends on their lifetime wealth constraints, which limits homeowners’ ability to

smooth consumption over the life cycle. With increased house prices, the home equity value

rises and the homeowners feel wealthier to spend more through realised wealth effect by refi-

nancing or selling off the house, or unrealised wealth effect through higher expected wealth.

©2015 John Wiley & Sons Ltd

1006 A. NANDA AND J.-H. YEH