The Williamson hypothesis. The content, criticism and afterlife of a concept

Abstract

In a 1965 article, J. G. Williamson published a remarkable hypothesis about an ‘inverted U patterned’ connection between the economic development and regional disparities of various countries. Thanks to its simplicity and elegance, the concept has gained significant popularity among experts dealing with spatial issues and has become a major theoretical concept of regional studies.
The first aim of this paper is to give a thorough evaluation ofWilliamson’s concept from a critical point of view. We try to reveal the methodological and theoretical limits of the hypothesis. Attention is drawn to two disputable methodological assumptions: a reliable comparability of different countries’ income-type indicators and an acceptable analogy between longitudinal and cross-sectional data sets. In the discussion we show that these issues were indeed realised and considered, although not necessarily resolved by Williamson, while they went unnoticed by most of his successors. Up to date statistics are also used to test the validity ofWilliamson’s main arguments. Our findings suggest different connections between economic development and regional disparities within various countries and groups of countries, questioning the universal relevance of a single curve.
Secondly, we evaluate several papers from the last three decades that have added useful new elements to the original concept. This illustrates how new findings have modified the shape of the ‘inverted U curve’ and made it asymmetrical, revealed the relation of economic development and spatial inequalities ‘in a higher stage of development’, opened the way for an extension of the concept beyond a determinate time-frame, and pointed at the unique paths followed by certain countries and groups of countries. Finally, we describe the main characteristics of disciplinary politics and their effects, and the way how different scientific interests have influenced the reception of the original concept over time. In conclusion, it could be demonstrated that Williamson’s concept, unlike conceptually similar positivistic models, has never become a ‘black box’, an ‘undisputable truth’ for researchers. Yet, it also has not become a subject of serious criticism from non-positivistic approaches in the social sciences. This phenomenon can be traced back to some specific circumstances in disciplinary politics, which are also analysed in this paper.