Comments by Bhanupong Nidhiprabha, on International Co-movement of Housing Price Cycles in East Asia and Greater China

Abstract

Bhanupong Nidhiprabha: The authors have extended the understanding of housing price cycles in East Asia by introducing two measures of dynamic co-movement based on dynamic correlation (cohesion) and time-varying dynamic conditional correlation. The problem with correlation analysis is that correlation does not necessarily imply causation. Even when the Granger causality test is used (Table 13), it does not mean causality in the usual cause–effect reasoning, because Granger causality is based on prediction criterion. There may be other factors that influence housing prices to move in tandem in East Asia, such as expectations, uncertainties, panic, or animal spirits. These factors are difficult to measure but it is not impossible to find some proxies for such expectations.The authors need to provide a convincing explanation of the co-movement of international housing prices rather than interpreting the statistical result from ten marching tables on variance decomposition in various countries. In addition, readers should be guided by the authors on how the booms started and peaked before their bust. Detailed explanations are required for different strength of the driving forces. For example, the authors should provide reasons why Germany, Taiwan, and Hong Kong have the lowest world factor determining housing price growth (Table 1). What are those country factors? Are they different from one country to another? The authors are quite right in pointing out in the last paragraph of the paper that demographic structure, supply side regulations, and socioeconomic determinants play a crucial role in determining housing prices. According to Table 13, housing prices in Taiwan is Granger-caused by China, Hong Kong, and Singapore. Therefore the China Effect is strong in “Greater China.” There is also a question on the definition of Greater China. I am wondering if the rent and housing prices are determined by the number of Chinese expatriates who work in Taiwan. A discussion of the demand and supply side constraints can make a coherent explanation supporting the Granger causality test.It is obvious from Figure 1 that the housing price cycles in Hong Kong and Singapore are synchronized and the movements reflect the world business cycle. The trade openness variable in tables showing variance decomposition actually reflect exposure to shocks transmitted from the rest of the world rather than indicating trade integration. It is well known that monetary policy, rather than fiscal policy has a direct impact on the housing sector through credit availability and cost of borrowing for long-term loans. As such, it is difficult to completely separate the monetary policy factor (broad money supply) and credit supply factors. Deposits and credits are simply on different sides of bank balance sheets. The paper provides and interesting statistical finding, but it would be more readable if the authors support the finding with more convincing arguments.