Critique and Review of Butterfly Economics

by Kevin Tapp

My initial reaction to Brian Arthur’s article, “Increasing Returns and the Two Worlds of Business” (assigned reading for class 1), was that the article was not really earth shattering news. To me Arthur was saying that today’s world is dominated by knowledge-based, tech-heavy industries, where it is more possible than ever to make extraordinary profits and sustain them for longer periods than our beloved robber barons of days gone by, and that it should be the policy of governments to encourage this. Needless to say my first thought was to pull the latest proxy on Microsoft and look for Arthur’s name in the beneficial owners section. Then I read Paul Ormerod’s book Butterfly Economics.

In Butterfly Economics, Ormerod adds his two cents (more like two dollars by the end of the book) to the current state of economic thought. His basic premise, in a nutshell, is that traditional economic theory is at best misleading and at worst blatantly wrong on almost all scores for the simple reason that orthodox economic models cannot factor in the concept that individual agents are influenced at the lowest levels by other agents—that we simple folk are actually influenced by the ideas of our friends and neighbors when making decisions about which beer to buy, which candidate to vote for or even whether or not we should take up a life of crime. Ormerod calls this the principle of interacting agents and suggests that this principle follows a pattern he calls non-linear signal processing. Amusingly, he utilizes a study of ant behavior to illustrate this point. Furthermore, he asserts that since the whole of traditional economic thought is based on linear models, where the aggregate is simply a sum of many individual micro-level models, that it is unequipped to deal with this kind of concept.

Ormerod goes on to demonstrate how the simple study of ant behavior has vast implications on everything from understanding crime rates to the destruction of the nuclear family in the Western world. He introduces concepts and models beyond the scope of this reviewer, namely chaos theory and complex systems, but one of the conclusions he reaches did reverberate with me and brought me back to the aforementioned article by Brian Arthur. Ironically, Brian Arthur’s earlier work on rules for replacement is quoted liberally in this book, which helped me gain a new appreciation for the ideas he put forth in his article, chiefly that ‘increasing returns are the tendency for that which is ahead to get further ahead, for that which loses advantage to lose further advantage.’ Paul Ormerod expounds on this idea in much greater detail and with remarkable clarity in his book. This book has far greater insights and achieves much loftier goals than merely this, but for me, reading this book and seeing it develop into an argument in defense of Arthur’s article afforded me my greatest insight.

Ormerod comes to the conclusion—one of many—that ‘diminishing returns are not a realistic description of the world’, and that if we are to understand the workings of world economies we must embrace the idea that the concept of increasing returns is the true result of dealings at the firm or micro-level, as well as the aggregate level across countries. Furthermore, Ormerod argues that this isn’t just another way of understanding the complex machinations of political economies, but rather the distinct result of empirical evidence collected from the Western economies since the beginning of the industrial revolution. And the real kicker is that this concept has no place in any classical econometrician’s model, for the orthodox models can only conclude a world of diminishing returns, where eventually everybody catches up. Now that, to me, resounded a little louder than finding out we live in a “casino of technology”, which is all I seemed to get from Arthur’s short article.

The implications of increasing returns as they apply to firms competing for dominance of the Internet are obvious, and also illuminates why so many investors have been so willing to lay down so many bets. If we can picture the Internet landscape as a big grid, like Ormerod’s model, we could envision similar results in the e-commerce space as those exhibited by the model. This means that we should see a few clusters emerge as clear winners and those that emerge will not only become bigger but will create an ever-widening gap between themselves and what could only be called the losers. Lastly, this transformation should also have implications on the mergers and acquisitions world, as this activity will—as it was the case in Ormerod’s model—help enable the large firms to become larger, in effect aid them in increasing returns. While this concept, increasing returns, is but one of many conclusions drawn by Ormerod, it is among the most salient issues raised in Butterfly Economics, and one I hope to elaborate on in class.