In any work involving PIC statistics the general question should always be asked: are the "international standard" classifications and concepts relevant for the needs of PIC countries? There are always obvious advantages for using existing classifications eg, international comparability, the ready availability of well-defined concepts, easy access to expert advice. But to what extent do these advantages outweigh the need for measures aimed specifically at the needs of each country in its own right?
In an attempt to open some debate on this subject I have set out below some early thoughts, with particular emphasis on the classification of agricultural and fishing activities.
Without wishing to downplay the general importance of international standards, I would suggest that the data needs of PICs own analysts and planners are not always met fully by simply adopting concepts and definitions developed for some hypothetical "standard" country. By all means let us use international standards as a default, and certainly as a starting point in any review - but let us also include in any survey or census design a review of the underlying classifications to ensure their relevance for local purposes. Even though there would always be some "cost" in departing from accepted standards, if there were significant associated benefits to local users then the use of modified classifications should be seriously considered.
Indeed, the need to sometimes adapt international classifications for national purposes is often clearly recognised by the classifications themselves. For example, the notes to the 3rd revision of the International Standard Industrial Classification of All Economic Activities (ISIC) include Chapter III: Application of the classification, with separate sections dealing with this issue: B. Use of ISIC in establishing related national classifications, and C. Expansion or contraction of ISIC.
The latest version of ISIC was released by the UN in 1990, and has 292 component "classes" of economic activity. This latest revision was developed by the UN Statistics Division after extensive consultation with expert statisticians throughout the world. However, it is inevitable that global decision making will require compromises over conflicting priorities - an obvious concern for PIC statistics is that our low ranking according to population and economic size means that our voice is not always heard in full.
One reason for raising this issue now is that it may be even more important in the future. The international trend towards more sophisticated service industries, particularly in fields such as electronic commerce, is likely to lead to increased numbers of identified "industries" in these areas, with a corresponding pressure to collapse the number of industries with relatively small contributions to value added. Any such move would be of considerable concern to economies centred on more traditional activities such as agriculture and fishing.
Relative importance of agriculture and fishing
There are a number of areas within the current version of ISIC where it clearly reflects economic structures which are not those of a typical PIC, particularly in the classification of agriculture, forestry and fishing activities. One telling example: in ISIC all fishing activities are classified to just one of the 292 classes (0500, Fishing), and yet in countries such as Palau fisheries activities account for over 25% of their total GDP (An assessment of the role of women within fishing communities in the Republic of Palau, SPC, 1999). The previous version of ISIC did have 2 classes for Fishing and related activities - but these were collapsed into a single class with the introduction of Revision 3 in 1989.
Agriculture and fishing are a far more dominant feature of most PIC economies than they are to larger, more "developed" economies. For example, Australia has a huge land mass and vast shoreline but agriculture, forestry and fishing activities still accounted for only 3% of its total GDP in 1999 (ABS 5206.0, Dec Qtr 1999). The most recent comparable PIC data is for Samoa, where agriculture, forestry and fishing activities accounted for over 16% of GDP in the September quarter 1999 (Quarterly Economic Review, Feb 2000). So, even after several years of very strong growth in its services sector, agriculture and fishing activities are still more than 5 times as important to the Samoan economy as they are to Australia. And that is measuring "importance" only in monetary terms, it does not make any allowance for the very significant cultural and social importance of traditional food production and consumption.
Classification of Agriculture activities
Most agricultural censuses and surveys are essentially commodity-based, with little or no attempt to classify producing units to a predominant activity. But there are often very large differences in the nature of the underlying activities - if one were to try to allocate industry classifications according to ISIC the types of agricultural activity throughout the Pacific would quickly run into the problem that quite diverse activities were classified to the same class.