orgtheory.net

competitive advantage – part two

Teppo

In an earlier post I argued that we are missing an important (or perhaps the most important) component of competitive advantage by not giving consideration to individual-level factors such as appropriation (incidentally: there are many other individual-level factors as well). I cited Russ Coff's paper as a classic based on the fact that it draws attention to these very factors.

Brayden's response (he is doubtful whether Russ' piece is a classic) misses the point. He highlights that Russ' article only pieces together existing theories (agency, resource-based etc.), without meaningfully creating new theory. Fine, though I think he is right only in part. However, the original and more important point still remains unaddressed – what is competitive advantage mostly about? I believe Russ' paper is getting at these very issues.

One way to potentially think about this is in terms of variance decomposition (e.g., see Anita McGahan's research). Clearly all levels are implicated in competitive advantage – e.g., individual, organizational, networks, alliances, industry, geography, country (we have research in many of these areas, presently a strong focus on alliances/networks) – but where does most of the variance reside? Theories implicitly (or not so) take sides and make quite strong claims about the source of variance (both Powell et al. and Dyer & Singh are explicit about the role they assign to interorganizational networks or alliances as the source of competitive advantage). I don't think this by any means is just an empirical question, but a critical theoretical question as well! Theories at different levels compete and are not complementary, though we may simply wave our hand and say: well, its a multi-level phenomenon. Of course its multi-level, but is it too controversial to privilege one level over another? We all do it.  With the perfect dataset, I would argue that we would find increasingly more variance at lower levels, thus giving primacy to the individual-level (of course recognizing variance will also exist elsewhere).

Now, there is of course a much deeper and critically important sociology of knowledge-type question embedded in this whole discussion (Brayden also alludes to it), but I will save that for a later post.

Note: Russ Coff weighs in on the discussion in his comments to Brayden's post.

Written by teppo

April 28, 2006 at 9:01 pm

Posted in strategy, teppo

%d bloggers like this: