|dc.description.abstract||There is a long history of examining the relationship between consumption and wealth. The recent subprime mortgage crisis in the US and the European sovereign debt crisis associated with the remarkable fluctuations in both stock and housing markets have brought new concerns over the response of consumer spending to asset price shocks. This thesis re-investigates the relationship between consumption, income, financial and housing wealth: specifically, the wealth effect on consumption based on both time-series data in an examination of the US and panel data in an examination of OECD countries. It is argued that nonlinear estimation might provide a better explanation of fluctuations in the relationship between consumption and wealth, given the nature of the studied variables and the complexity of economic systems. The econometric methods employed include the Markov regime-switching approach, the quantile autoregressive distributed-lag framework, panel unit root and cointegration tests, and a panel vector autoregressive procedure. In terms of the US market, only weak evidence of a linear cointegrating relationship is found between consumption, income and wealth. It suggests that the consumption-wealth relation is better characterised in regime-specific terms. Furthermore, the transition probabilities between regimes are time-varying and driven by monetary indicators such as interest rates. In addition, different speeds of adjustment across the range of quantiles are identified in the long-run relationship between consumption, income and wealth. Wealth effects are found to be larger in lower quantiles. The findings imply that asymmetric monetary policies should be responsible for the movement in asset prices in analysing future inflation and aggregate demand due to the sensitivity of financial and housing wealth effects in different economic states.
In OECD studies, cointegration evidence is only observed in market-based countries, not in bank-based economies. Moreover, the wealth effects are found to be larger in market-based than in bank-based countries. In addition, since a positive wealth effect caused by an increase in capital value of housing might be partly offset by a negative price effect caused by an increase in the cost of housing services, the ‘net housing wealth effect’ actually is found to be slightly smaller than the stock market wealth effect in OECD countries. However, due to the significant boom in real estate markets since the 1990s, the housing wealth effect has clearly exceeded the share market wealth effect over the past decade. The results show that asset wealth has asymmetric effects on consumption, with stronger and more persistent effects from positive asset wealth shocks.||