One, a hundred, thousands of industrial districts. Organizational variety in local networks of small and medium-sized enterprises

Organization Studies, Fall, 1998 by Ivana Paniccia

Despite more than a decade of studies on industrial districts, carried out by different scholars in various disciplines and regions of the world, a systematic and comparative study is still missing, as the pioneer in this field also had to admit (Becattini in: Becattini and Rullani 1993).

Behind the so lightly used term 'industrial district', many different forms of organization of labour and many different socio-cultural fabrics are hidden. In what may be regarded as a common definition, industrial districts are taken as forms of organization governed by trust and co-operation. This also explains why the term is easily confused with the concept of network - a similarity that adds no clarification.

What is striking in the extensive case-study literature on industrial districts is the variety of approaches and the different phenomena that are taken into account. The resulting impression is that, on the one side, specific and non-replicable characters of particular industrial districts are compared with a general - although not always specified - model; on the other side, general characteristics of the model are automatically applied - without any empirical assessment - to areas where there are just high concentrations of (small) firms. The concepts of both industrial district and network suffer from 'semantic ambiguity'.

Most of the literature has followed a case-study approach, which is very useful in shedding light on the internal functioning of industrial districts, and in particular on the rules governing the relationships between firms and individuals. What is wrong in case studies, which, by definition, focus on particular features, is their use in illustrating the characteristics or evolution of industrial districts in general. In addition, so-called 'districts' are chosen according to personal preferences or convenience. For example, Harrison's (1994) choice of 'district' confirmed his idea of devolution; other researchers have focused on the most successful 'districts'.

Given such shortcomings in the literature, a necessary starting point for studies aimed at building a sound theory of industrial districts is first to use an agreed and operationalized definition and then to conduct a comparative and longitudinal study.

As a first definition, industrial districts may constitute a variant of local productive systems of small- and medium-sized enterprises (henceforth, SMEs). Garofoli (1983, 1991) and Sforzi (1990) provided the first rigorous attempts to find criteria to identify industrial districts. Their methodology, although not definitively, is an important hallmark for our research, as will be explained later. These two authors were not really concerned with operationalization, whereas it is my conviction that operational concepts provide objective criteria for describing and comparing the organizational structure and performance of different local productive systems (industrial districts, networks, and clusters or agglomerations of small firms) and to discriminate between them. The preference here is also for a comparative approach based on gathering a large amount of data for different features and periods of time. Comparative studies can make explicit the structural characters of industrial districts, while a dynamic analysis is an aid to understanding their formation, change and adaptability. A rigorous attempt to classify and compare different local systems of production leads to a more ambitious aim - to propose a theory on industrial districts. Becattini himself failed to carry this out, and I would not presume to do so here.

Aims and Approach

This paper is aimed at correcting some of the flaws in the existing literature on industrial districts. It is part of wider research that aims to verify whether industrial districts - starting from the analytical translation of the concept - are effective organizational modes and to discuss more accurately some of the main rules governing inter-firm relationships that they are said to have, such as 'co-operation' and 'extended division of labour'. In contrast to previous studies, the organizational approach favoured here is one that is able to illuminate the organizational texture of industrial districts and thus the different typologies hidden behind the label 'industrial district'.

The paper is organized as follows: the next section illustrates the concept of the 'canonical' industrial district, the second section describes the methodological steps of our empirical research, and the third section presents some of the general results. In particular, the second part of the third section focuses on the different models of labour division among our local productive systems, followed by a discussion on vertical line co-operation and on co-operation between peers. Some concluding remarks are given at the end of the paper.

The Concept of the Industrial District

The conceptualization of industrial districts referred to as 'canonical' is that proposed by Becattini (1987, 1989, 1990) and complemented by his scholars (Bellandi 1989, 1992; Dei Ottati 1994; Sforzi 1989, 1990). The sociologist Trigilia (1986, 1990) and Brusco (1982, 1990) can also be included in the 'canonical' school, although the latter built up the concept of industrial district on a different basis.