We take four arguments from the actor-oriented theory of institutions (North 1990) and model growth as an increase in the number of strategies and informations about these strategies. In the conventional games which are the base of discontinuities in institutional change, individuals take their immediate experiences on individual or organizational level as natural focal points. However, growth and swelling information make easy focal points suboptimal, and optimality proceeds towards more abstract levels. In the sequence which is defined by the existence of transaction cost, focal points for defining institutions move from the individual level in the pre-industrial phase to the organizational level in the industrial and to the institutional level in the (still emerging, more than 40 years after Bell 1977) post-industrial phase.
Using additional assumptions about the nature of relevant institutional fields, we can derive a grid which allows to understand a lot of institutional innovation in the past, present, and immediate future, from Romantic Love to not-yet designed schemes of computerized voting.