Animal Spirits: How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism.

Hoag, John H.

Animal Spirits: How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism, George A. Akerlof and Robert J. Shiller, Princeton University Press, Princeton and Oxford, 2009.

In this book, Akerlof and Shiller make the case that a successful macro model cannot simply be based on the rational economic agent paradigm that is rooted solely in economic factors. They contend that both non-economic factors and non-rational (in the economic sense) decisions are part of the story. Animal spirits refers to these not economically rational, non-economic based factors and takes its name from a quotation of John Maynard Keynes. Before an examination of strengths and weaknesses, let us proceed with a word about the details of the book.

The authors identify five aspects of animal spirits, confidence, fairness, corruption and bad faith, money illusion, and stories. These aspects are part of the non-rational decision making process involving non-economic factors. Each of these has one chapter devoted to it. There are then eight chapters that follow each of which develops how animal spirits can be used to better understand a specific economic question. The chapters are: Why do economies fall into depression, why do central bankers have power over the economy, why are there people who cannot find a job, why is there a trade-off between inflation and unemployment in the long run, why is saving for the future so arbitrary, why are financial prices and corporate investments so volatile, why do real estate markets go through cycles, and why is there special poverty among minorities.

For the most part, the chapters devoted to the aspects of animal spirits are well done. They amplify on the idea and provide some rationale for why this might be a useful concept. The applications in the eight chapters that follow are a bit more uneven, and it is not always easy to understand exactly how the animal spirit concept is being applied. The author's fundamental point is that for policy purposes, we cannot rely on macro models that only include the standard staples of macroeconornic theory. Models that rely on rational expectations and optimizing behavior are found particularly wanting. Reliance on models that arise from the view that markets are uniformly the best way to organize economic behavior is seen as a particular sin. In the view of Akerlof and Shiller, models need to take a broader and deeper view of the world, a view that includes animal spirits as they understand them. Make no mistake, they are well aware of the power of the market and applaud the successes that markets bring. But they also see the problems that can arise from unfettered markets run amok. With Keynes, they see the best outcome as one where the excesses of the market are controlled by careful, thoughtful government interference.

What's not to like about this approach? First, I wonder how much this view explains what happens. One analogy for animal spirits is a flock of birds flying first one way, then another. The exact moment of turning somehow decided without, as it seems to the observer, an active discussion on the part of the flock. To say that the flock changes direction because of a change in confidence (of because of a perceived unfairness or because the change is dictated by our story) seems like a non-explanation. What causes the very first change in confidence and how is it transmitted through the flock? What is the trigger that causes the turning of the core group? How does this change come to be? It seems to me that this is an important, if not key, question that needs to be addressed if we are to make policy based on this theory. In short, in what way does this theory go beyond description of what happens and in some deeper sense explains what happens and why it happens. If we are, as Keynes would claim, to make policy based on our understanding for the betterment of the world, we need this deeper understanding. It may be in this book, but I was not able to ferret it out.

Second, Akerlof and Shiller are careful to point out that economic models as we know them have limits. Yet one would expect that the model that they propose also have limits. What are the limits? Under what conditions would we expect this model to not explain well? In the absence of these limits, we have no way of knowing when to expect the model to do well and when it would be expected to fail.

Third, in a sense, and building on the second point, there is a deeper methodological issue at stake. As one tests a theory against the data of the world, it is likely that the testing process will require some measurable attributes representing the components of the theory. If they are not measurable, it will be difficult, but probably not impossible, to falsify the theory. Theories of this kind abound in humanities where the methodology is unlike the sciences. In this sense, it seems to me that Akerlof and Shiller are advocating a methodological movement away from the more scientific models that come from optimization and rational behavior which generate falsifiable outcomes. The exact nature of the methodology they propose is not clear. Upon what basis would we agree that their theory is right? On what basis would we want to argue that they had missed the boat? While there are methodologies in other areas where this kind of argument is well practiced, most students of economics would need some training to come to a comfortable level of interchange. I am aware that there is considerable question about whether economists really do practice the falsifiability requirement they preach. But I also suspect that most economics students are taught only one paradigm.

Fourth, I am unclear about who they are writing for. For a while, I thought they might have an educated layman as a target, but they seem to have bigger fish to fry. They seem to want professional economists to change their approach. The book is well documented with excellent references and frequently both sides of the argument are well represented. But, based on this book, I cannot imagine that a follower of Friedman will be persuaded to drop what they are doing and develop models including animal spirits. Those who already see that markets need some controls will appreciate the added dimensions brought by their analysis. But will economics change as a result?

So what's the final verdict? While there are some concerns about how one would actually use the theory Akerlof and Shiller propose, they make some important points. Even if one does not buy the particular additions they want to make, there is plenty here to provoke discussion and debate. While this would not completely satisfy the authors, one suspects that they would be happy with that as a first step.

John H. Hoag

Professor of Economics

Bowling Green State University

Bowling Green, Ohio