This paper starts from the premise that diversification of assets, activities, and incomes is important to African rural households, in that diversification into nonfarm income constitutes on average about 45 percent of incomes, and the push and pull factors driving that diversification are bound to persist. From that premise, we noted that the empirical study of diversification has been beset by practical problems and issues relating to (1) definitions and concepts, (2) data collection, and to (3) measurement of the nature and extent of diversification. The paper addressed each of those problems. Two points are of special interest to the overall conceptualization of diversification research. The first is that empirical studies have exhibited a wide variety - bordering on confusion - of systems of classification of assets, activities, and incomes as pertains to diversification behavior. We argued that the classification should conform to that used in standard practice of national accounts and macro input-output table construction, classifying activities into economic sectors that have standard definitions, and the classification of which does not depend on the location or functional type (wage- or self-employment) of the activity. We further argued that given a sectoral classification, it is useful to make a functional and locational categorization of the activity, and keep each of these three dimensions of the activity - sectoral, functional, and locational - separate and distinct so as to avoid confusion. The second is that it is useful to have an image of a production function in mind when analyzing the components of diversification behavior: (1) assets are the factors of production, representing the capacity of the household to diversify; (2) activities are the ex ante production flows of asset services; (3) incomes are the ex post flows of incomes, and it is crucial to note that the goods and services produced by activities need to be valued by prices, formed by markets at meso and macro levels, in order to be the measured outcomes called incomes. "Livelihoods" is a term used frequently in recent diversification research, and while its meaning differs somewhat over studies, it generally means household and community behavior, with respect to holdings and use of assets and the productive activities to which the assets are applied. The link between livelihoods and incomes needs to be made by valuing the output of livelihood activities at market (and/or virtual) prices. That valuation permits an analytical link between household/community behavior (thus a micro view of diversification) and the aggregate functioning of markets (thus a link with the meso and macro levels and the policies pertaining thereto).

Issue Date:
Publication Type:
Working or Discussion Paper
DOI and Other Identifiers:
Record Identifier:
PURL Identifier:
Total Pages:
JEL Codes:
O; Q12
Series Statement:
Working Paper 2000-19

 Record created 2017-04-01, last modified 2020-10-28

Download fulltext

Rate this document:

Rate this document:
(Not yet reviewed)