theory underlying evolutionary economics is based on routine behaviour rather than rational choice
(Nelson and Winter 1982; Nelson 1995b). Instead of describing the behaviour of individuals or firms as if
they optimise an objective function given budgetary and other constraints, evolutionary economists start
from the premise that the larger part of human behaviour including organisational behaviour is routinised.
In this, they follow Simon’s (1955) concept of ‘bounded rationality’ stressing that cognitive constraints of
human organisations render them incapable of optimisation in most real-world relevant situations.
Consequently, individuals and firms will differ in their knowledge and beliefs, and accordingly, will show
heterogeneity in their choices and behaviour (which can no longer be approximated by a ‘representative
agent’). It is this variety that fuels the evolutionary selection process as an open-ended and out-of-
equilibrium process of economic development.
The concept of routine in evolutionary economics is still quite broadly and loosely defined, and the
concept needs to be worked out more thoroughly both conceptually and empirically (on this, see Hodgson
2002). Routines can be understood as organisational skills (Nelson and Winter 1982), which cannot be
reduced to the sum of individual skills. Routines are manifest at the firm level due to division-of-labour
and hereby, division-of-skills between the people working for a firm. Organizational routines, as for
individual skills, consist for a large part of experience knowledge (learning-by-doing) and tacit knowledge
(which is hard to codify). Both aspects of organizational routines render them difficult to imitate by other
firms. As a result, they are competencies to the firm that largely determine the competitiveness of a firm
(Teece et al. 1997).
Adaptive behaviour of firms can take on two forms, which have been stressed by scholars at an early
stage (Alchian 1950). First, firms learn from their own mistakes through trial-and-error. When routines do
not work well, this failure induces active search for other routines, for example, by investing in Research
and Development. The successful replacement of routines by fitter routines can be considered an
innovation. Evolutionary theory predicts most firms to innovate incrementally (that is, to change routines
of minor importance) to continue to exploit the knowledge built up in the past. Empirical research tends to
show that where incremental innovations typically increase the life chances of firms, major organizational
transformations tend to decrease the survival probabilities of firms (Anderson and Tushman 1990).
Second, organisations are able to observe successful behaviour of others and try to imitate their successful
routines. The difficulty here for firms is to find out which routines are crucial in explaining the success of
fellow firms and, therefore, should be imitated, and which routines are detrimental for fellow firms, and
therefore should be ignored. Moreover, attempts to imitate successful behaviour are failure-prone, because
routines partly consist of tacit components that are hard to copy by imitators.
Apart from the intelligent, adaptive behaviour of firms as evidenced by their ability to get rid of
unsuccessful routines and to exploit the opportunities of innovation and imitation, ‘intelligence’ also exists
at the level of an industry as a whole (similar to the population level in biology). In this, evolutionary
economics has similarities to the population ecology approach in organizational sociology (Carroll and
Hannan 2000). As long as firms show routinised behaviour, which only sporadically changes, market
competition acts as a selection device, which causes the ‘smart’ fit routines to diffuse and ‘stupid’ unfit
routines to disappear. The assumption of routinised behaviour is crucial because selection for fitter
routines can only take place if the speed of change of organizations is substantially lower than the time-
span of selection. A number of processes render fitter routines to become more dominant in an industry,
including differential growth of firms (Alchain 1950; Nelson and Winter 1982), imitation (Alchain 1950;
Nelson and Winter 1982), labour mobility, and through spin-offs (Klepper 2002a,b).