are embedded. The same point was made for the evolution of the automobile
industry in the major manufacturing countries. There has been a further
segmentation of market segments; and which kinds of manufacturers did well
in which kinds of segments is explained by the societal setting which they
already inhabited. Of particular importance, again, are the above-mentioned
dimensions of organization, human resources, labor market and industrial
structures (Streeck, 1992: ch. 6).
Broadening institutionalism across levels
in the socio-economic orderThe examples show how we can extend an analysis initially focused on the
more ‘micro’ setting of relations within plants and within enterprises to take
account of more ‘macro’ settings. To recapitulate how this works, let us take a
simple example from the intra-European differences mentioned above, in Table 5.1.
French job and managerial hierarchies are more finely graded than German
ones; this goes with a more detailed subdivision of labor and therefore
narrower job content; this is the corollary of lesser job enrichment by apprentice-
ship training; the greater gap between vocational education at school or college
and the world of work gives rise to the necessity to develop practical compe-
tence gradually, and this is done precisely by slotting people into more specific
jobs when they come from school or college; hierarchies are accordingly flatter
and heavier at the bottom in Germany, whereas they are taller and heavier
around the middle range in France. This set-up in France also means that
upward socio-occupational mobility is higher throughout France. This is
clearly inevitable if we think through the differences already identified: where
the point of gravity is higher in the hierarchy, there is greater space for
climbers! We can see that a detailed comparison along a multitude of dimen-
sions then makes specific results ‘fall into place’. They do this because differ-
ences across different dimensions ‘hang together’, i.e. they cohere. On this
basis, we can more or less predict what happens if, say, Germany were to reduce
the practical orientation of occupational training and its embeddedness in or
close to work settings; it would move closer to a French institutional profile –
if nothing changes in France.
In research, such differences have been attributed to government law, col-
lective agreements, custom and practice, financial incentives and restrictions,
ideologies, symbols, preferences and social values. In the case of the chemical
production units discussed above (Maurice et al., 1980), it emerged that they
themselves had already investigated the extent to which their structures dif-
fered and could be assimilated. The managers of the sites had been baffled by
the number of differences that were impossible to overcome. They had also
concluded that there was no real need to assimilate structures and practices,
since they were able to achieve comparable performances in strikingly different
128 International Human Resource Management