Essays on Hong Kong's container handling industry

A structural vector error correction model (VECM) of the container throughput volume of Hong Kong's container terminal is estimated to obtain both the long-run and contemporaneous relationships between it and the container throughput volume of midstream and Singapore's container terminals. The resulting forecasts obtained from the structural VECM are then compared with those from a Bayesian vector autoregression (BVAR) model. A sensitivity analysis is conducted to determine how much the growth path of Kwai Chung's container throughput is affected by the growth rate of foreign trade and Shenzhen port's container throughput volume. Among others, I have found that in most cases both the structural VECM and the BVAR model produce higher growth paths than the forecast of the Hong Kong Port Development Board.;The government has two instruments to affect the output decision of the terminal operators. One of which is the so-called "trigger point mechanism" (TPM) and the other is the number of terminal operators allowed to enter the market. The TPM triggers new terminals on the basis of the gap between forecast demand and existing terminals' container handling capacity. If forecast demand is formed with adaptive expectation, then the government may adjust the user cost of terminal's capacity and control the sensitivity of the "trigger" by changing the fraction of last period capacity shortfall that will be filled up by next period new capacity. By choosing the two parameters appropriately, the government can achieve the socially optimal solution under the TPM. The simulation exercise suggests that, if terminal operators engage in competition non-cooperatively, and under reasonable sensitivity of the "trigger" and user cost of capacity, the number of terminal operators that achieves the socially optimal solution is two.;Statistical analysis has revealed that there has been a structural change in the profitability of shipping lines in 1990 and that the level of "terminal handling charges" (THC), i.e., fees set by shipping lines for moving container cargoes from the shore to ships, and the shipping lines' profitability were positively related. In addition, the empirical tests suggest that the demand for Hong Kong's container handling services experienced a negative structural change since the introduction of THCs, and that it was negatively related to the level of THC. The above results can be explained by the increase in market power of shipping lines. Furthermore, by using the locational rent approach, I have found that the existing level of THC should be reduced by 53% on average in order to maximize Hong Kong's GDP.