Reconciling Alchian and Penrose: Competitive Strategy, Organizational Capabilities and Firm Performance as Evolutionary Learning

Strategy research has often – though not always – focused on the connections between the elements of strategy, structure, conduct and performance (e.g. Chandler 1962, Rumelt 1974). However, issues in competitive strategy research are how to link individual-level perceptions and actions to firm-level strategic actions and these with the resulting effects on firm performance. These are usually empirically analyzed using statistical methods relating firm-level, business unit and (top-)management characteristics as proxies for effects and inferred causal relations on performance. However, the link remains weak in that the myriad of actions taken by actors in companies is usually not linked to performance effects either on the company or business unit level even today, even decades after criticizing this approach (e.g. Miller 1979). We require a genetic framework for a causal theory of the interactions between actors' mental representations and their actions that is measurable - as demanded by Low and Macmillan (1988) and reiterated about a decade later by Aldrich and Martinez (2001). The interactions of social actors and events in and outside of organizations play a crucial role in this. For instance, Mintzberg (1994) has argued that a companies´ realized strategy is the outcome of the interaction between the intended and an emergent strategy resulting from the tension between deliberate (planned) strategy and the impact of environmental forces. This process results in evolutionary paths of strategy-making behaviour and resultant outcomes, following patterns developed in the past, getting stuck in dead ends or exploring new domains or forms of organizational change (Meyer, Gaba and Colwell 2005). While it may be desirable to disentangle myriads of individual actions and link them to their performance effects, e.g. via approaches such as strategy-as-practice (Whittington 1996, Hendry 2000) in the long run, on a shorter time horizon, it may be helpful to focus on the observation of meso-level interactions between individual strategic actions and performance outcomes using constructs such as ‘routines’ and ‘dynamic capabilities’. Thus for instance, Rumelt (1991) has called for an evolutionary approach to the analysis of the development of industries, Teece (1994), and Teece, Pisano and Shuen (1997) have argued for the existence of dynamic capabilities affecting firm performance and linking these to evolutionary ideas (Teece 2007). Likewise, Winter and Nelson's (1982) routines have been used to explain the stability in firm behavior affecting performance. And indeed, on the macro-level of observed outcomes, an evolutionary approach to strategy can build on the knowledge developed in ecology and evolutionary theory (e.g. Hannan and Freeman 1989, Nelson and Winter 1982), as well as linking routines to to dynamic capabilities (Zollo and Winter 2002) on an explanatory meso-level. Thus, while it is usually argued that evolutionary approaches to social phenomena on the aggregate level can be useful to get to grips with change and development processes, here it is argued that an evolutionary approach to social phenomena is particularly useful on a micro-level. This approach allows to counter criticisms of the application of Neo-Darwinian evolutionary concepts that neglect intentionality and complexity of social interactions.