BOOK REVIEWS
219
the reader as he gets to grip with it: who, apart from the authors' patron, cares at all how great will be Hong Kong's imports of cotton, rice, wheat and other foodstuffs in the years ahead? This intelligence is of no conceivable commercial use, nor does it serve the policy-maker. The Hong Kong Government is unlikely to set long-term plans in motion to provide export capacity designed to generate the foreign currency sufficient to meet these import requirements, nor indeed will it need to do so. The reader is left to ponder the point of the study. Apart from the fact that it answers a research director's need for work to occupy research assistants, the study could form part of a larger study geared to determining the world pattern of commodity trade flows in the future — a useful basis for UNCTAD discussions and U.S. agricultural policy. It is only within a context such as this that the study makes any sense: alone, it looks like a missing piece of a jigsaw puzzle.
Granted that the study has some point, how do the authors set about forecasting in Hong Kong and how well do they do it? The authors calculate imports as the difference between predicted demand and predicted domestic production, where such exists. The main body of the work is taken up by demand predictions which are made on the basis of the standard econometric model which expresses per capita consumption as a function of per capita income and the price of the commodity concerned relative to the general price index. Having estimated these demand relationships on the basis of past data, the authors predict per capita incomes and relative prices for the years 1970, 1975 and 1980 and calculate consumption with the use of the estimated regression coefficients: sound, routine stuff so far. In the course of this exercise, several difficulties emerge which show up the authors' strengths and weaknesses. The Hong Kong economy is wretchedly documented and the authors have worked commendably to build a few bricks without much straw. At all times the discussion and handling of data problems is honest, professional and interesting. This approach is illustrated very well in the reconstruction of a price index which overcomes the problems of unrepresentative weights applied in the construction of the 1947 base.
When these statistics are actually used econometrically the study seems to flounder. The reconstructed price index performs very badly in all the regressions: it is statistically insignificant in all nine
Page 225
Page 226
BOOK REVIEWS
219
the reader as he gets to grip with it: who, apart from the authors' patron, cares at all how great will be Hong Kong's imports of cotton, rice, wheat and other foodstuffs in the years ahead? This intelligence is of no conceivable commercial use, nor does it serve the policy- maker. The Hong Kong Government is unlikely to set long-term plans in motion to provide export capacity designed to generate the foreign currency sufficient to meet these import requirements, nor indeed will it need to do so. The reader is left to ponder the point of the study. Apart from the fact that it answers a research director's need for work to occupy research assistants, the study could form part of a larger study geared to determining the world pattern of commodity trade flows in the future — a useful basis for UNCTAD discussions and U.S. agricultural policy. It is only within a context such as this that the study makes any sense: alone, it looks like a missing piece of a jigsaw puzzle.
Granted that the study has some point, how do the authors set about forecasting in Hong Kong and how well do they do it? The authors calculate imports as the difference between predicted demand and predicted domestic production, where such exists. The main body of the work is taken up by demand predictions which are made on the basis of the standard econometric model which expresses per capita consumption as a function of per capita income and the price of the commodity concerned relative to the general price index. Having estimated these demand relationships on the basis of past data, the authors predict per capita incomes and relative prices for the years 1970, 1975 and 1980 and calculate consumption with the use of the estimated regression coefficients: sound, routine stuff so far. In the course of this exercise, several difficulties emerge which show up the authors' strengths and weaknesses. The Hong Kong economy is wretchedly documented and the authors have worked commendably to build a few bricks without much straw. At all times the discussion and handling of data problems is honest, pro- fessional and interesting. This approach is illustrated very well in the reconstruction of a price index which overcomes the problems of unrepresentative weights applied in the construction of the 1947 base.
When these statistics are actually used econometrically the study seems to flounder. The reconstructed price index performs very badly in all the regressions: it is statistically insignificant in all nine
Page 225Page 226
No comments yet.
Private notes are available after approval.