Geography Reference
In-Depth Information
the same things in a constant environment than to change. They tend to be much better
at self-maintenance and changing in the direction of more of the same than they are at
any other kind of change.
This pattern of behaviour is built up not on the basis of a grand rule of optimising
behaviour founded on utility maximisation but on the basis of satisi cing (Simon, 1955)
regular routines. These include well-specii ed technical routines for producing things
through procedures for hiring and i ring, ordering new inventory, or increasing pro-
duction in items in high demand to policies for investment, research and development,
advertising and business strategies (Nelson and Winter, 1982, p. 14).
Routines are organisational - the competences, capabilities and skills of i rms as a
whole. They are built around the division of labour within i rms and hence the division
of skills between workers in a i rm. The skills basis of a i rm consists in large part of the
experience and tacit knowledge of the workforce. These are hard to codify (Boschma
and Frenken, 2006, p. 5). The importance of these kinds of knowledge, how they are
acquired and transferred, has become a central concern of both evolutionary economics
and geography. Furthermore, the importance of the division of labour within i rms and
industries in the acquisition and transfer of knowledge should also be noted at this stage
and is discussed below.
Once a set of routines has been established, the progress of i rms and industries
employing those routines tends to become path dependent. In these circumstances,
other things being equal, behaviours will continue along well-established trajectories.
In the worst case scenario this can lead to lock-in and continuation along paths that are
increasingly unsuited to market conditions. This is well illustrated by the classic case of
the Swiss mechanical watch industry (Maillat et al., 1996).
The second main concept in evolutionary theory that is required to provide chances
that routines do not become too path dependent and locked-in to out dated behaviours
is search. Searches are the processes by which routines are redirected and changed.
Searches involve organisational activities that are associated with the evaluation of
current routines and may lead to their modii cation, drastic change or even complete
replacement (Nelson and Winter, 1982, p. 400).
Search introduces a behavioural activity that can be one of the main driving forces for
change in i rm behaviours. It involves the development of new knowledge both within
i rms and in relation to their industrial and general environment. Given the presence
of bounded rationality, however, there is no guarantee that all searches will discover
relevant new knowledge for changing i rm routines. They can be undertaken internally
by existing functions within i rms. But, as is argued below, increases in the division of
labour with respect to the knowledge-based economy often lead to search activities being
outsourced to specialised knowledge-intensive business services (KIBS).
Selection is the third key concept in evolutionary theory. Along with search, as dei ned
above, it is part of the simultaneous interacting aspects of the evolutionary process.
These dynamic processes in which i rm behaviour patterns and market outcomes are
jointly determined over time are the core concern of evolutionary theory (Nelson and
Winter, 1982, p. 19). Search and selection are the joint processes that lead to the ways
in which i rms and industries evolve, or indeed fail to evolve, over time. They also mean
that history matters because as a result of past searches and selections, the condition of
an industry in one period provides the basis for its condition in a following period.
