Empirical Evidence of Housing Wealth Effect on Consumer Spending

Abstract

Understanding the housing wealth effect is important for macroeconomic forecasting and policy setting. The paper uses cointegration theory to study the interplay between consumption, income and housing wealth in 14 countries observed quarterly from 1995 to 2013. Cointegration is found to be present in Canadian, Danish, German, Italian, Japanese and United States data, implying that the consistently estimable propensity to consume out of housing wealth is in the range of two cents per US dollar in Italy and nine cents per US dollar in Denmark. The housing wealth effect tends to be lower in countries where the income elasticity of consumption, wealth volatility and income inequality are higher. The effect tends to be larger in countries where business conditions are superior and access to credit is easier. Examining the short-run relationship between consumption, income and housing wealth highlights that transitory income moves do not Granger-cause consumption growth, and transitory housing wealth changes tend to contain information useful for predicting consumption growth in each sample country. A correlation analysis shows points of policy action aimed at influencing the housing wealth effect, and thus aggregate consumption and the economy.