How do organizations learn and evolve? Phin Upham looks at the classic work “Opportunity and Constraint.”
In Opportunity and Constraint, Paul Ingram and Joel Baum explore how organizations learn from experience and what sorts of experience they learn from. Their results are counterintuitive and initially confusing – at times finding “own experience” to be helpful and at times hurtful. This lends support for the idea that a deep analysis of learning might be both useful for managers and also provide a causal explanation for many of the other trends related to age/organizational_failure seen in other strategy research. An approach reminiscent in some ways of organizational ecology research is pursued in this piece where organizations are assumed to have a significant amount of momentum in their organizational and behavioral form, learning incrementally or becoming unfit incrementally.
The paper has taken a two stage view of the effects of own experience on survival. The authors hypothesize that own experience helps to promote survival, especially in geographic specialists (the authors operationalize this by looking at hotel chains that operate in a small number of census regions), early on in their development, but experience later causes them to have a higher death rate. The authors speculate that a hardening of organizational routines based on experience causes the hotel group to be less sensitive to environmental changes and thus, over time, less fit and less nimble.
The paper also expands the idea of experience in an interesting way by including learning from experience in an industry or from competitors/allies in an industry. They divide this into operating experience, which can be gained through trade magazines, experienced employees, alliances, exposure to other hotel practices, etc. and competitive experience which is the learning one gains though having faced competition over time. But rather than attempt to measure this on the individual firm level, the authors turn the idea on its head and claim that the level of industry operating and competitive experience will be a relevant factor to a firm either entering or existing in the industry.
This introduction of industry level experience in operational and competitive metrics can be seen in two ways. First, it can be seen to be true because there is much interchange of knowledge between firms of an industry, thus as individual firms become more competent their competencies become general knowledge, thus raising the pool of general knowledge within the firm over time. On the other hand, one can imagine a model where firms with idiosyncratic competencies compete and less competent firms leave the market while more competent firms expand in size. Over time, a more competent population would arise and no less competent firm could enter the market and survive. Thus, in this model, the industry operational and competitive experience is an ever raising bar under which no entrant can afford to be below for long. I think the authors hold the former view more strongly since they speak of intra-firm transfers of knowledge and how operational and competitive knowledge is higher as a function of time in industry, at least initially. On the other hand, industry competitive experience at time of entry turns out to be non-significant in this study.
Some thought on the idea that initial experience is a boon and, over time, this experience becomes a burden is warranted. The authors present various arguments as explanations of this phenomenon, but he most compelling to me is the argument that organizations explore initially, gathering information to gain competencies and then exploit those competencies. During this exploitive period, though, the organization has cemented in a view of the world and developed complex mechanisms and buffers to deal with this perceived environment. Unfortunately for this organization, when the environment continues to change eventually the organization becomes less fit. While it gains high rents initially, the cost of these rents is less adaptiveness (more exploitation). But other explanations, more mundane in structure are possible as well.
Since we are dealing with small three hotel chains, it might be that the original owner, after a few years, retires and taken much of the competencies of the firm with him. This would also explain why a firm would tend to die over time – leadership change.
This paper uses a long period of analysis – from 1896 to 1985 – and analyses a large number of firms – 1135 chains, of which 705 fail. It has gathered founding, size, and death statistics for all of them and some other metrics for a large subset of them, such as geographical distribution of hotels. The authors acknowledge that their data is incomplete at times, such as having only room number statistics for hotels beginning in 1950. Such lacks are acknowledged and some attempts to justify attempts to mitigate them are provided. Chain size is used as a proxy for chain success, a measure of per-unit profits might be a better metric on which to measure firm performance. Lastly, it is unclear how the data treats mergers and acquisitions. Are they treated as failures? Are they dropped from the dataset? This could be a significant aspect of the data set which is overlooked.
Phin Upham has a PhD in Applied Economics from the Wharton School (University of Pennsylvania). Phin is a Term Member of the Council on Foreign Relations. He can be reached at phin@phinupham.com.