Academic journal article Economics, Management and Financial Markets

Do House Prices Impact Consumption and Interest Rate in South Africa? Evidence from a Time-Varying Vector Autoregressive Model

Academic journal article Economics, Management and Financial Markets

Do House Prices Impact Consumption and Interest Rate in South Africa? Evidence from a Time-Varying Vector Autoregressive Model

Article excerpt

ABSTRACT. This paper investigates the existence of spillovers from the housing sector onto consumption and the interest rate for South Africa using a time-varying vector autoregressive (TVP-VAR) model with stochastic volatility. In this regard, we estimate a three-variable TVP-VAR model comprising of real consumption growth rate, the nominal three-months Treasury bill rate and the growth rate of real house prices. The results suggest that, in general, consumption responded positively to a house price shock over the entire sample, with the effect being stronger post financial liberalization. On the other hand, a positive delayed response of nominal interest rate followed a house price shock, with the effect being weaker post financial liberalization until the South African Reserve Bank (SARB) moved to the official inflation-targeting regime. The effect of house prices on both consumption and interest rate was understandably weak during the financial crisis.

JEL Classification: C11, C15, C32, E31, E32, E44, E52

Keywords: Bayesian inference, consumption, house price, Markov Chain Monte Carlo, monetary policy, structural vector autoregression, stochastic volatility, time-varying parameter

(ProQuest: ... denotes formulae omitted.)

1. Introduction

The permanent income hypothesis asserts that house price inflation increases the expected lifetime wealth of homeowners and hence their desired consumption. This is known as the wealth effect. The collateral effect, on the other hand, postulates that fluctuations in house prices relax homeowners' financial constraints, which may in turn affect their actual consumption. Numerous papers indicate a strong positive link between the housing market and consumption in the United States (Green, 1997; Belsky and Prakken, 2004; Carrol, 2004; Iacoviello, 2005, 2011; Case et al, 2005; Learner, 2007; Kishore, 2007; Jarocinski and Smets, 2008; Sousa, 2008; VargasSilva, 2008; Ghent, 2009; Mian and Sufi, 2009; Pavlidis et al, 2009; Ghent and Owyang, 2010; Iacoviello and Neri, 2010, Shirvani et al., 2012), the United Kingdom (Aoki et al, 2002; Campbell and Coco, 2007; Muellbauer and Murphy, 2008; Elbourne, 2008; Attanasio et al, 2011 ) and for other individual countries, such as Australia (Dvornak and Köhler, 2007), China (Chen et al, 2009; Koivu, 2012), Czech Republic (Sec and Zemcik, 2007), Hong Kong (Cheng and Fung, 2008; Gan, 2010), Italy (Paiella, 2004; Guiso et al, 2005; Bassanetti and Zollino, 2010; Bulligan, 2010), Portugal (de Castro, 2007; Farinha, 2008), Singapore (Edelstein and Lum, 2004; Phang, 2004), Spain (Aspachs-Bracons and Rabanal, 2011), Sweden (Chen, 2006) and Turkey (Akin, 2011). A few comprehensive international studies are also available, where a number of countries are studied concurrently (Boone et al, 2001; Bertaut; 2002; Bayoumi and Edison, 2003; Byrne and Davis, 2003; Barrel and Davis, 2004; Catte et al, 2004; Ludwig and Slok, 2004; Case et al, 2005; Aron and Muellbauer, 2006; Fung and Cheng, 2007; Goodhart and Hofinan, 2008; Slacalek, 2009; Aron et al, 2010; Musso et al, 2011; André et al, 2011; Ciarlone, 2011; Peltonen et al., 2012; Sonje et al, 2012).1

As far as South Africa, our country of interest in this paper, is concerned, to the best of our knowledge, there exist four studies2 analyzing the relationship between consumption and real house prices. They are Aron and Muellbauer (2006), Das et al, (201 1), Ncube and Ndou (201 1) and Simo-Kengne et al. (forthcoming a).3 Aron and Muellbauer (2006) indicated that much of the empirical literature assessing the wealth effect of house prices on consumption is marred by poor controls for the common drivers of both house prices and consumption. Given this, the authors suggested an empirical model for the United Kingdom and South Africa grounded in theory, and with more complete controls than generally used. The estimates suggested that in South Africa, unlike the UK, the marginal propensity to spend for housing wealth or collateral is slightly larger than for illiquid financial assets, though the difference is not statistically significant. …

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.