Routines or Practices?
| Nicolai Foss |
I am growing increasingly skeptical of the extremely popular and influential notion of routines, a central construct in large parts of management, notably organization and strategic management, and in evolutionary economics. My problems with the construct are these (among others):
1) There are still no clean definitions around of “routine.” Proponents of the routine notion sometimes delight in pointing out that it took transaction cost economics almost 4 decades to arrive at its unit of analysis, dimensionalize it, etc. However, with respect to TCE it was only when the unit was finally decided on, defined and dimensionalized that real progress beyond Coase (1937) began to take place. Those who work with routines have not been so patient, and have not hesitated to introduce all sorts of derived concepts. Thus, capabilities are often defined in terms of routines, so that something undefined is defined in terms of something badly defined.
2) Although no clean definitions seem to exist, different views of routines are proffered in the literature. Indeed, there has a notable drift in the dominant conception of routines, from the standard operating procedures of Cyert and March to the emergent/undesigned, collectively held, largely tacit routines of Nelson and Winter.
3) Routines are (partly because of 1)) too often used as a catch-all category that aims at capturing everything (at any level of analysis) about an organization that has some degree of stability/permanence. For example, the much cited “organizational learning” paper by Levitt and March (it has a whopping 1,655 hits on Google scholar) includes everything from individual-level heuristics (“rules of thumb”) to corporate strategy under the routine heading.
4) There are, at least in management, no good stories of how routines emerge and change as a result of individual action and interaction, as Teppo Felin and I have argued in a number of papers, e.g., this one. (Here is an evol econ paper that tries to tell a story of emergence).
5) Routines seem to bring us back to the economist’s charicature of sociological man, that is, somebody who mindlessly follows rigid rules (see also Michael Cohen’s recent semi-skeptical essay), making it difficult to account for creativity and indeed the change of routines (except in terms of meta-routines).
Yet there is, of course, something to the notion of coordinated, recurrent action patterns in organizations, and it would be silly to argue that all such patterns had been rationally designed a priori (although advocates of the routine construct seems to me to badly under-estimate the extent to which routines (or more properly, standard operating procedures) are actually designed in organizations).
It seems to me that if advocates of the routine concept really want to theorize those organizational elements that are semi-rigid, a property of the social system defined by the firm, largely tacit, etc., they wouldn’t do too badly by linking up with the practice stream in strategic management, exemplified by such writers as Paul Duguid, John Seely Brown, JC Spender, Richard Whittington, and others (e.g., this paper).
I disagree with much in this literature, but it has the advantage relative to the routines literature of taking micro-foundations serious and of thinking seriously about micro-macro links. It has also has substantial backing in sociology, notably in the work of Pierre Bourdieu which — somewhat to my surprise — I have recently found to be a much more interesting source of inspiration when it comes to the understanding of organizational practices than the contemporary management stuff on routines and capabilities (please, no jokes about pomo periscopes) (here is the wiki on B.). For example, the extreme behavioral rigidity implied by the routine notion gives way in Bourdieu’s work to a much more subtle understanding of the interaction of action, dispositions, and social structure. Indeed, in Outline of a Theory of Practice Bourdieu explicitly challenges the view that much behavior is rule-governed, and offers instead his own notion of “habitus,” that is, “the durably installed generative principle of regulated improvisations.” This notion can accommodate both constraints in the form of institutions, culture, habits of thought, etc., and rational action — which the notion of routine seems incapable of.