seemingly represents an insignificant item (be it a commodity, an industry, or an occupation). The value of this category may even look questionable since it is a hodgepodge of items all of which were regarded ad of minor value either in quantity or in value at the time of the compilation of the statistics.
From the point of view of scholars interested in economic changes, however, the category may represent a source of extraordinary value. For instance, the "miscellaneous goods" were a very important source of foreign earnings in the early phases of Japanese industrialisation, as they included contained valuable exportables such as matches, toys and some other small items of seemingly little significance. Moreover, the category may contain some items which may later on prove to be highly important (in some sense) in the life of the economic community. Certain items in the chemical industry, for instance, were often labelled as "miscellaneous" in the nineteenth century sources partly because the chemical industry had not been firmly established as yet. But the industry later turned to be such a growing sector of the economy that it needed a separate classification of its own.
By the same token, one notes that the historical records of the mining industry often contain valuable information on the early activities of manufacturing, as it was often essential for the mines, which were generally located in far-away, often mountainous locations, to have machine shops at hand in order to engage in maintenance and repair services. Mining must have contributed much more than usually believed to the early development of manufacturing capabilities of the so-called advanced, industrialised countries, especially in the late nineteenth and the early twentieth centuries.
Speaking of changes in industrial structure, one cannot possibly neglect the importance of the rise of the tertiary sector. This point, in fact, is a repetition of what was recognised centuries ago by Sir William Petty. Nonetheless, this phenomenon causes a serious problem for contemporary statisticians, since the statistical documents of the service industry leave much to be desired. In addition, there are certain conceptual difficulties in measuring some important economic concepts such as labour productivity.
