The value of simple (hopefully analytical) description should not be underestimated, especially in view of the fact that sufficiently scrutinised empirical data are not always readily available at least in many regions of the world. The deficiency of the data is all the more serious if one looks for an objective (value-neutral) source of describing development performance from a comparative perspective.
More importantly, third, one would naturally be interested in proceeding to conduct hypothesis testing, for which long-term economic statistics will provide rich source materials. In fact we face quite a few unanswered questions in the world that surrounds us. If we are lucky, some of these riddles may be resolved by resorting to historical statistics. In the following, the present writer proposes to illustrate this point by referring to some interesting questions which have been posed by historians in the past.
Illustrative Questions in Comparative Economic Development
Growth Performance
It is simply fascinating to uncover mere facts regarding changes in GDP and its industrial components over time. One may ask in this respect questions such as: When did MEG begin, why, and how? What are the characteristic features of the MEG in the region concerned? Were "long swings" (Abramovitz 1961) or "trend acceleration" (Ohkawa and Rosovsky 1973, pp.39-42) observable outside of the USA or Japan, respectively? What were the initial conditions of MEG? How does one evaluate the role of the initial conditions in (or their impacts on) the country's industrialisation? How about the hypothesis of concurrent growth, as has been found applicable to Japan?
In order to answer questions like these, one needs to accumulate, critically evaluate, and estimate basic statistical information. This process often involves long, tedious work which still requires the sound judgement of a well-trained researcher. A specimen example of such an effort is currently being made in order to come up with a new estimate of manufacturing production and its historical path by taking a new look at the long deserted statistical documents (Guan 1997; see Figure 1).
