The change we observe is seldom discontinuous.
One of the surprises in the post-socialist transformation process in Central and Eastern Europe has been the persistence of production structures in the agricultural sectors. Amid the dramatic and sudden changes that ended the continuity of the Central and Eastern European socialist economies, the observation made by Douglass North which serves as the motto of this article remains true for the structure of farms in large parts of the region.
Socialist farms were either collective or state farms, both of which were corporate organizations. There were separations between farm ownership, control over the production process, and implementation of production tasks. These were wage-labor farms, as distinct from the Western-type family farm. In the Western agricultural economics literature preceding the liberal revolutions of 1989-1991 in the region, socialist agriculture had long been identified as cost-inefficient due to incentive problems inherent in the governance structure of wage-labor farms. This view was also the dominant approach in the early transformation years and is expressed explicitly and in most detail in a paper by G. Schmitt (1993), originally written in 1990. The argument was an application to post-communist farming of theoretical work on household production by R. Pollak (1985). For contemporary expositions of it, see Sarris et al. 1999 or Swinnen and Mathijs 1999 (4-8).
In these accounts, the raison d'etre for collective, large-scale production is the existence of economies of scale or scope. However, in most of the socialist economies, farms had a number of workers far exceeding the number of workers on the farms on which generalizations about scale and scope economies are based (Pryor 1992, 147). In addition, in much of agriculture, technical economies of scale and scope were argued to be soon outweighed by diseconomies in the organization of production, and particularly of labor (Ferenczi 1994, 403). The alleged cause is, in the spirit of A. Alchian and H. Demsetz' (1972) approach, a rapid increase in monitoring costs with number of employees. These costs arise because of the information asymmetry problem between managers and workers with regard to the state of the land, crop, and animals, and the amount of labor input required and actually applied.
This asymmetry precludes the use of adequate incentives and necessitates more supervision. However, supervision is costly because food production processes are characterized by sequential and interdependent jobs, seasonal work and labor peaks, and the spatial dispersion of farm work. For these reasons, the farm size (in number of employees) that is, in this view, regarded as optimal with regard to the production process, is not expected to exceed family-farm size. In this mode of farming, organizational diseconomies were argued to be largely excluded. Wage labor is not used and shirking or free riding is supposedly difficult because of the small size, the hierarchical structure, and the externally enforced continuation of relations within the household 'labor force' (Deininger 1995). Based on these considerations and observations, the common perception of socialist-style farm structures in the early transformation years was that "the evident weakness of this organizational form provides the argument for full scale privatization" (IMF et al. 1991, 157-158) and that "privatization in . . . agriculture mainly concerns the breaking up of large units" (World Bank 1995, 2).
The economic goal of agricultural reforms was to raise productivity, for which dc-collectivization was considered a condition. Crucially, and understandably in the above view, de-collectivization was commonly defined as the formation of individually operated (for short, individual) farms from the land and property of formerly collective and state farms (henceforth referred to as traditional farms), This process was expected to follow privatization, that is, reforms in property rights over farmland and other farm assets. The reasoning was that these reforms would create new, or newly effective, individual (rather than collective or state) owners, who could be expected to search for the most efficient and profitable use of their assets. Given the assumed inherent inefficiency of wage-labor farming, owners would use their assets in (or rent them to) individual farms, rather than leave them in traditional farm structures (Machness and Schnytzer 1993, 162; Mathijs 1998, 33). It was this expectation that motivated the farm and land reform programs implemented throughout Central and Eastern Europe in the early l99Os.
However, as early as 1994 it could be noted that "already now it is clear that the process of farm restructuring . . . is taking a course, which appears to be different from the original expectations of many Western European observers. . . . It is remarkable that farm enterprises . . . choose to reorganize as whole entities, without dismantling the collective structure" (Csaki and Lerman 1994, 566, 573). Typically, new farming structures emerged which were based on individual property rights, yet in majority preserved features that had allegedly foreordained collectivism to inefficiency. In many Central and Eastern European countries, a considerable share of agricultural land was, and still is, worked by farming structures other than individual farms. This share is 89 percent in Slovakia, 62 percent in the Czech Republic, 48 percent in Bulgaria, 46 percent in Hungary, 37 percent in Estonia, 33 percent in Romania and Lithuania, and 5 percent in Latvia (Lerman et al. 2002). (1)
The aim of this article is to offer an explanation for this persistence of traditional farm structures. The subject is interesting both in its own right, as an important economic phenomenon in the region, and as a case study of the comparative merits of neoclassical and institutional theorizing and explaining. Two aspects make the issue particularly interesting in the last perspective. First, the initial expectations outlined above were so clearly inspired by a coherent framework characterized by the assumptions of rationality as the leading behavioral motive, by maximizing behavior, and by methodological individualism. The reforms were very much a neoclassical program, and the expected outcomes grounded in neoclassical theory. Now, a decade after the conception of those reform plans, is the time to compare the empirical developments with those early expectations and to reassess the theoretical framework that inspired them.
Second, the objections and corrections to the neoclassical view of economics, as proposed by economists that would label themselves institutionalist, evolutionary, Post Keynesian, or generally heterodox, appear particularly relevant to the post-socialist transformation process. These issues include the complexity of human behavior; the importance of group dynamics; the interaction between economic, political, and other social parameters; and the relevance of history and time. Topics in the post-socialist transformation process, such as (the absence of) de-collectivization in agriculture, seem to be particularly amenable to what can be labeled the institutional approach to economic analysis. (2) If we seek to demonstrate the merits of that approach, this is an excellent opportunity.
Hence, the approach taken here is that farms are productive institutions, and thus de-collectivization is primarily a process of institutional change. Another assumption, and one that differs from the theory sketched above, is that the emergence and viability of farming modes depends not only on internal farm features such as governance structure but also on the wider institutional context of economic and, more generally, social conditions. lam addressing a problem of systemic change rather than the evolution of an isolated institution. Hence, if we want to explain this process, or account for its absence, we need a theory of systemic institutional change specified for post-socialist agriculture. This will be developed in the next section.
This framework will then be applied in a comparative study of agriculture in the Czech and Slovak republics, using information collected by the author from secondary literature, statistical sources, and reports by local and national governments and international institutions. There were several reasons for selecting these countries for a comparative study. Both successor states to Czechoslovakia came from a rigidly centrally planned system with, in agriculture, virtually no private farming other than garden plots before 1989. Both implemented restitution and de-collectivization schemes that were identical in institutional detail and timetable in the crucial and initial phase up to the 1993 split. Both speedily privatized virtually all farmland and assets. However, de-collectivization (in the restricted sense of individualization) of farming was most limited here compared to other Central and Eastern European countries, as was shown by the above figures. Moreover, it has been clearly more limited in the Slovak than in the Czech Republic. The combination of close institutional similarity and different outcomes makes for an interesting comparison.
Systemic Change of Productive Institutions
We need not here develop a systemic, institutional theory of de-collectivization from scratch. Various specialists have written on the subject early in the post-socialist period, some in this journal (Hagedorn 1991, 1993; Schmitt 1993; Pryor 1992). These contributions are based on different strands of theoretical work on institutional change of a more general nature, from various methodological perspectives (North 1990; Hodgson 1988) or on the comparative study of farming modes (Binswanger and Rosenzweig 1986; Hayami and Otsuka 1993). Also, the constant monitoring of developments by institutions and academic...