When the mathematician John von Neumann and the economist Oskar Morgenstern wrote their groundbreaking classic on game theory at Princeton in the 1940s, they based their work on axioms—basic assumptions—which postulated that players are completely rational. The two theoreticians assumed that so-called homo oeconomicus (economic man) possessed all the information about his environment, was able to solve even the most complicated calculations in split seconds, and was not influenced in any way by personal preferences or prejudices. He would always make the mathematically correct decision.

A few years later the French economist Maurice Allais, Nobel laureate for economics in 1988, recognized that when answering questionnaires about situations that involved very low probabilities and very high sums of money, test persons tended to make “incorrect decisions.” Their real-life decisions contradicted conventional expected utility theory. A few decades later Amos Tversy, from Stanford University, and Daniel Kahneman, from Princeton University, found that market participants—be they businesspeople, doctors, or just regular consumers—make decisions that contradict the axioms established by theoreticians in both regular and momentous situations. (Kahneman was awarded a Nobel Prize for Economics in 2002.)

The theoreticians were not easily put off by this contradiction between theory and reality. Homines oeconomici who did not behave according to the postulated axioms were simply labeled as irrational. The theory is correct, claimed the scientists. It must be that a large part of society simply reacts incorrectly. What these economists

did not realize was that by stubbornly sticking to this belief they distanced themselves more and more from reality.

Herbert Simon, winner of the Nobel Prize for Economics in 1978, attempted to explain why financial markets conformed only poorly with people’s behavior as predicted by game theory. He introduced the theory of “bounded rationality.” By noting that agents must bear costs when acquiring information, must face uncertainties, and are not able to calculate like machines, he came closer to the truth. But this development was no cure-all either, and the anomalies observed in the financial market became more and more obvious. Above-average gains and losses were more common than classical theory would have led one to expect, volatility was larger than predicted, and exaggerated expectations led to inflated prices. Time and again market players who did not give a toss about the Von Neumann–Morgenstern axioms achieved higher profits than their more rational colleagues. Scientists had to look further afield for other explanations.

During the past century, economists often turned to other academic disciplines for tools to help them answer their questions about decision science and financial theory. A new fad among a young generation of financial theorists is evolutionary biology. Professors at renowned universities have made the theory of evolutionary finance an important topic for their research. Rumor has it that fund managers also make use of recent research results in this new field.

In early summer 2002 the Swiss Exchange invited scientists and practitioners from all over the world to a seminar in Zurich. Presenting their newest results, those present did not hesitate to criticize classical game theory for being removed from reality. While classical financial theory assumes that investors attempt to maximize their discounted income streams over the long run through clever investment strategies, evolutionary theorists suggest that investors follow just a few simple rules that they adapt from time to time to the varying circumstances.

In analogy to biological processes, economists set up

models of socioeconomic developments—complete with selection, mutation, and heredity—as a stream of learning processes and surges in innovation. In games that are rapidly played one after the other, the investment strategies assume the role of the animal species, and capital is allocated to the different strategies by the rules governing natural selection. Investment funds employing profitable strategies attract much capital and thrive, while funds with inferior investment strategies eventually disappear. Furthermore, surviving strategies must continuously adapt themselves to the changing market environment in accordance with the laws of natural selection.

The central question is, which investment strategies survive in an environment that is shaped by uncertainty and is prone to disasters? If several strategies operate side by side initially, which ones will survive in the long run? How do traders react to unexpected disturbances from outside? A paper by Alan Grafen of Oxford University may serve as an example for the approaches presented at the conference. In his model market players are equated to living organisms. To maximize their own level of fitness, they are subject to the process of natural selection and they adapt their behavior to the environment and to the strategies employed by competitors.

What Grafen observed was that agents will not get involved in any complicated calculations as they would in the classical theory. All they do is follow simple rules. If these rules prove successful and yield satisfactory results, their pervasiveness in the market increases. Under certain circumstances their dominance could, however, prove detrimental. Once weak strategies have disappeared, even successful strategies will no longer yield high returns, since nobody will be left to be exploited. Then the successful strategies will also disappear, just as predators become extinct once there is no prey for them to hunt anymore.