Butterfly Economics: A New General Theory of Social and Economic Behavior

Butterfly Economics: A New General Theory of Social and Economic Behavior

by Paul Ormerod

View All Available Formats & Editions

Why did VHS, an inferior video recording technology, succeed in the marketplace, driving the superior Betamax out of business? Why do big-budget, acclaimed movies sometimes flop at the box office, while low-budget, idiosyncratic films become huge hits? The answers to these questions, says Paul Omerod, remind us that economics is a science based on the workings of


Why did VHS, an inferior video recording technology, succeed in the marketplace, driving the superior Betamax out of business? Why do big-budget, acclaimed movies sometimes flop at the box office, while low-budget, idiosyncratic films become huge hits? The answers to these questions, says Paul Omerod, remind us that economics is a science based on the workings of human society, as unpredictable an entity as there is. "Conventional economics is mistaken," claimes Omerod, "when it views the economy as a machine, whose behavior, no matter how complicated, is ultimately predictable and controllable."

In this cogently and elegantly argued analysis of why human beings persist in engaging in behavior that defies time-honored economic theory, Omerod also explains why governments and industries throughout the world must completely reconfigure their traditional methods of economic forecasting if they are to succeed and prosper in an increasingly global marketplace.

Editorial Reviews

New York Times
It is accessible and even entertaining...Mr. Ormerod not only writes about capital, but also points to a way it may at last be understood.
Publishers Weekly - Publisher's Weekly
Combining sophisticated economic analysis with a gift for lucid explanation, Ormerod deepens and expands upon the points he made in 1994's The Death of Economics. He starts with an elegant critique of conventional economics, arguing that the prevailing thinking mistakenly ignores insights from other fields (notably biology, psychology and literature) and that practitioners of the dismal science pay too little attention to empirical verification and see the world through narrow theoretical blinders. Ormerod, head of the economic assessment unit at the Economist, then presents his alternative approach, Butterfly Economics, an interdisciplinary view that takes its cues from sources as diverse as ant behavior and the mathematics of chaos theory. The mathematics, relegated to three appendices, are simplified to a high-school algebra level. But Ormerod's argument is easy enough to follow without the numbers as he applies Butterfly Economics to explain why VHS beat out Betamax to become the VCR standard and why low-budget movies often outperform the most expensive Hollywood features. At the core of Ormerod's thinking is the observation that human behavior is not nearly as neatly predictable as prevailing economic models assume, that economic life is more like a living organism than like a machine. His book is amusingly written, and every page offers surprising facts or strikingly new ways of looking at well-known facts. Anyone who likes to think about people and how they act will find much of interest, and probably something to love or hate, in this book. (Jan.) Copyright 1999 Cahners Business Information.
Christopher Lehmann-Haupt
"Butterfly Economics" is penetratingly lucid. It argues with particular strength that governments ought to stop trying to control and predict what they can't and start concentrating instead on enhancing the process by which markets make important decisions, or, in the terms of Mr. Ormerod's informing experiment, how ants choose to go to one food source or the other.
Mr. Ormerod even illuminates to what extent Marx was wrong in predicting the end of capitalism's growth, and why Marx's mother, in an anecdote the author insists is true, wrote to a relative, "If only little Karl had made some capital rather than just writing about it."
Mr. Ormerod not only writes about capital, but also points to a way it may at last be understood.
The New York Times

Product Details

Knopf Doubleday Publishing Group
Publication date:
Sold by:
Random House
File size:
4 MB

Read an Excerpt

Chapter 3: To Catch A Thief

Margaret Thatcher famously declared, 'There is no such thing as society.' Her intent was ideological rather than intellectual, to assert the primacy of the interests of the individual over that of the state. Our emphasis on the concept of interacting agents, exemplified in our ants, does not in any way lead to a presumption that the interests of society as a whole should be favoured at the expense of the individual. Neither does it imply that the collective outcome is necessarily a desirable one. These remain a matter of personal and political preference.

But it does mean that, in an important sense, there i's such a thing as society. Analytically, once the principle that the behaviour of individuals can be affected directly by the behaviour of others is accepted, the properties and features of the collective whole can no longer be deduced simply from the conduct of a typical individual. The outcome for the colony in its entirety will not be the same as it would be if it consisted of a single individual operating in complete isolation on a desert island.

In short, the assembled beast behaves differently from its myriad individual component cells, a concept articulated several millennia ago in the book of the prophet Ezekiel: 'son of man, can these bones live?' Or, as the popular song puts it more graphically, 'dem bones, dem bones, dem dry bones'.

The principle of interacting agents, of our ants, ought to be able to illuminate not just economic issues, but questions in the social sciences. The discipline of sociology is, after all, meant to involve the study of society.

A light-hearted, but neverthelessserious, example might be the best way to proceed. The so-called discipline of psephology, the study and prediction of voters' behaviour, is one of the few which is able to make economic forecasting look respectable. Enormous errors in the prediction of election results can be made by experts only days before the actual event, and even on the day itself.

From time to time, a conventional wisdom emerges about what determines the outcome of elections, only for it to be rapidly contradicted by events. Bill Clinton declared that 'it's the economy, stupid', and many psephologists took these words to heart in thinking about the outcome of the British election in 1997. The economy was growing rapidly, unemployment was falling sharply, and inflation was lower than it had been for over thirty years. Yet the ruling Conservatives suffered one of the biggest defeats in their entire history, which stretches back over two hundred years.

The behaviour of Western electorates seems to have two striking features. In terms of the overall division of votes between the leading political parties, there can be long periods of reasonable stability, punctuated by large and rapid changes. The changes in some European countries during the course of this century have been truly dramatic, but the point can be illustrated even in the generally more stable climate of the United Kingdom.

The Liberal Party won an enormous victory in the election of 1906, but were almost defeated in 1910 and by the. early 1920s had been reduced to the status of a small minority party, where they have remained ever since. The massive Labour victory Of 1945, overthrowing Winston Churchill, was overturned by 1951 and the Conservatives were brought back to power. A large Conservative victory in 1959 had been changed into an equally large Labour one by 1966. Yet there have also been periods of stability with, for example, the Conservatives remaining in power between 1951 and 1964, and again from 1979 to 1997.

The second point relates to evidence at the level of the individual voter. Even when the overall share of votes in opinion polls is reasonably stable, studies which follow the views of the same panel of voters over time indicate a fairly high level of change by individuals. By definition, if the aggregate share of votes is stable, the various changes of mind by individuals must cancel each other out. But these changes of opinion by individuals most definitely exist, even at times of stability in the overall outcome. Psephologists even have their own word to describe such behaviour, the faintly unsettling 'churning'.

In the scientific spirit, we are searching for a model of individual voter behaviour which is compatible with these two pieces of empirical evidence. A steady incidence of switching opinion by individuals and an overall pattern of voting for the major parties in which periods of stability are followed by swift and substantial changes.

As a first approximation, we need look no further than our friends the ants. To simplify the exposition, imagine that there are only two parties, A and B, and that voting is compulsory, on pain of a penalty so huge and Draconian that everyone does actually vote (it could be, for example, a life sentence condemning non-voters to listen to every single debate which takes place in the national parliament). The model can readily be extended to incorporate more parties, including a nominal one to represent those who choose not to vote at all. But it is easier to start with just two.

An individual who supports party A in any given period of time can, in the next period, do one of three things. He or she can simply continue to support A. The voter may be persuaded, of his or her own volition, or by some piece of news, to switch to party B. We do not need to specify the nature of the news, which we assume arrives at random. Both anecdotal and more structured evidence suggests that voters can and do make up their minds on the most bizarre grounds. For example, I was told in the early 19gos in all seriousness that someone had decided to abandon the Conservatives on the grounds that the then finance minister 'looked like an owl'. Finally, someone may decide to switch as a result of being influenced by the views and decisions of others, such as friends or colleagues at work.

This is exactly the model of ant behaviour. An individual can continue to be loyal to the same party, can decide to switch with a given probability on his or her own account, or can change opinion, with a different probability, as a result of being influenced by others. So the model involves, inherently, a constant stream of changes of opinion by individuals. And, as we saw in the previous chapter, this generates in aggregate the type of behaviour which is actually observed. Periods of relative stability in the division of support between the two parties are interspersed with sharp and rapid changes.

A typical solution of such a model of voting behaviour in a two-party...

Meet the Author

Paul Omerod has been head of the Economic Assessment Unit at The Economist and a visiting professor at the Universities of London and Manchester. He lives in Londom.

Customer Reviews

Average Review:

Write a Review

and post it to your social network


Most Helpful Customer Reviews

See all customer reviews >