Still Going: Recent Debates on the Goldschmidt Hypothesis 267The authors describe changes in post-war farming patterns typical of each region,
then examine the effects of these changes on indicators of county well-being gener-
ally for the 1970–1980 period (Buttel et al, 1988; Flora and Flora, 1988; MacCan-
nell, 1988; Skees and Swanson, 1988; van Es et al, 1988).
Also by the mid-1980s, two US Department of Agriculture regional research
projects (S-198, S-246) centred on the changing structure of agriculture and its
effects on communities. As part of these projects, Wimberley (1986, 1987) devel-
oped indicators of agricultural structure from a factor analysis of census agriculture
data. His purpose was to describe empirically the changes that had occurred in
post-war agriculture but that could not be captured through the use of single indi-
cators of farm structure. To describe national patterns of farming, he created
indexes for all counties with census-reported farming, the criteria generally being
ten or more farms. While the vast majority (more than three-quarters) of these
counties are non-metropolitan, Wimberley (1986, 1987) purposefully did not
exclude any metropolitan counties for which farming was reported. Such counties
typically contain highly capitalized operations that reflect important present and
future structural trends. Wimberley found three dimensions of farm structure that
correspond to post-war dualistic tendencies noted empirically and theoretically in
the literature – small, part-time farming; highly capitalized, family-labour farm-
ing; and commercial-scale, hired-labour-dependent farming. Lobao (1990)
employed Wimberley’s indicators in a national study of socioeconomic inequality
across counties. She viewed differences in local well-being as a consequence of
economic structure (farm and non-farm), the power of workers and their house-
holds vis-à-vis employers, and spatial or locational characteristics. She also exam-
ined the extent to which generalized, national patterns hold for various regions. As
most of the sample counties are non-metropolitan, and as a subsequent study
(Lobao and Schulman, 1991) showed that the relationships stay the same when
the few metropolitan counties are excluded, the study has particular policy impli-
cations for rural areas.
With regard to the relationship between farm structure and well-being, the
OTA (Swanson, 1988a) and Lobao (1990) studies stated that their aim was to
address the limitations of earlier studies. As a consequence, they focused on devel-
oping empirical indicators in accordance with their conceptualization of farm
structure, Lobao (1990) through the use of Wimberley’s (1986, 1987) indicators
and the OTA studies through attention to specific regional patterns. Both sets of
studies examined non-linear relationships under the assumption that the effects of
moderate-size, family-operated farms may vary from those of larger and smaller
farms. Non-farm economic structure was controlled in the OTA studies by indica-
tors such as the percentage of the population employed in services and manufac-
turing. Lobao (1990) used segmented economy theory to classify employment
into high-wage, low-wage and state sectors and also included an enterprise-size
variable. Other non-farm variables known to have important effects on well-being,
such as educational levels and ethnicity, were controlled and the net effects of
farming versus the non-farm economy were examined.