Is finance a science? And what about neurofinance?

The new economic season was spiced up by the controversy initiated by the book by Cahuc and Zylberberg, two economists who claim to be orthodox, on what they call "economic negationism". This negationism would be the tendency of certain intellectuals to ignore the results of economic studies or even to reject the economy as an ideology at the service of the dominant class. The economists targeted are notably those who present themselves under the label of heterodoxes. The response of some of them has not been long in coming. They qualify inter alia Cahuc and Zylberberg's book of "pamphlet as violent in tone as it is mediocre in substance". According to them, "never has the attack been of such a low level".

An addition to the intellectual brawl that is unfolding before our eyes, the book raises the interesting question of the "science" character of our economic disciplines in the broadest sense. Indeed, this question applies to both economics and management sciences. As such, finance is at the crossroads of several disciplines (economics, management science, applied mathematics, etc.). We will therefore try to answer this question of scientificity for one branch of the economy: finance.

Classic finance, a science?

First of all, let us recall what we mean by "science". A science consists, through a precise methodology, in understanding and explaining the world. We start from observations that give rise to a falsifiable theory and therefore to hypotheses that we test and that allow us to invalidate or support the theory in question. Until then, finance meets the definition. However, a science also aims to draw accurate predictions from this knowledge. And this is where the stick starts to hurt.

For even in the case of finance, an economic field in the broad sense of the term where data and observations are very abundant, the predictive power of models remains relatively weak. For example, probably the greatest achievement in financial economics is the CAPM, the famous Financial Asset Pricing Model. Without going into detail, once we know the movement of the market as a whole, it allows us to know the expected return on a particular stock. It correctly predicts about 60 % of the variability of the movement of securities. To give the reader an idea, physics models are generally not considered valid below 90 %. Economic systems are indeed complex, depending on the aggregation of decisions of many economic agents, which makes economic prediction difficult.

In addition, finance traditionally uses mathematical demonstrations to create its models. This state of affairs does not fail to annoy some financiers in that they would like to put the human and even the political back at the centre of the debate. Mathematical demonstrations are necessarily based on axioms and hypotheses. One of the main hypotheses is that of the rationality of agents. However, it has been clear since 1953 and the French Nobel Prize winner Maurice Allais, that this hypothesis does not hold at the individual level. For the anecdote, Maurice Allais proposed a questionnaire at an economics conference in Paris, to test some basic axioms of rationality. Luckily, one of the Popes of orthodoxy in economics, Leonard Savage, was there to test some basic axioms of rationality. participa. His choices violated the axiom of substitution. He later acknowledged his error and attributed it to the presentation of the choices that were offered in the questionnaire.

Behavioural finance to the rescue

This has led to the creation of a field of behavioural finance: one can read on this subject the writings of Robert Shillerwho, paradoxically, won the Nobel Prize in Economics in 2013, at the same time as Lars Peter Hansen and Eugène Fama (considered one of the founders of the theory of efficient markets).

Behavioural finance comes from the application of psychology to economics. It is about using psychology to shed light on and better understand economic phenomena. This field is particularly interested in many deviations from rationality in financial markets. These biases go against the famous theory of efficient markets.

The complete list of these deviations is long. For example, let's quote theMonday effect The trend of the markets to be rather up on Friday and rather down on Monday, which could be due to a mood effect of the market operators. And yes, nobody likes Mondays! The weather effects are another example. Instead, the market would tend to be up on good days and down on bad days, again due to a mood effect.

However, most of these biases tend to disappear once they are exposed, a rather pro-market efficiency argument. Moreover, behavioural finance is also often criticised because it lacks an overall theoretical framework and does not really improve the predictive power of conventional models. It is more like a collection of observed phenomena, most of which can be explained by the emotions of economic agents.

The solution through neuro-financing?

This is why we are now witnessing the rise of neuroeconomics and neurofinance. These disciplines use methods borrowed from the neurosciences to better understand the decisions of economic agents. In a sense, they are putting humans, as living beings, back at the centre of the scientific agenda. The stated goal of neuroeconomics is to bring together psychology, economics and neuroscience into a single unified discipline to explain human behaviour. Neuroeconomics and neurofinance are therefore inherently multidisciplinary and encourage scientists from very different backgrounds to communicate with each other, which in the long run can only be beneficial for science and society in general. In a word, neurofinance is intended to be resolutely federative.

Although very young - neuroeconomics dates back to the 1990s, neurofinance dates back to the 2010s - these disciplines have already contributed a great deal. In particular on the role of emotions in rationality, a stumbling block between classical and behavioural finance. Funnily enough, it appears that emotions are essential to rational behaviour. For example, it has been shown that people suffering from a emotional deficit of physical origin could not behave rationally in some cases.

Similarly, two studies of traders have shown that traders experience strong emotions during periods of increased market volatility, regardless of their level of experience. Indicators of stress appeared among the traders studied. Their rhythm conductance of their skin increased while their heart rate increased. cardiac variability was diminishing. Similarly, another study showed that during these periods of increased volatility, their cortisol levels - a hormonal marker of psychological and physical stress - increased significantly. As for their daily profits, they were correlated to their morning testosterone levels!

Similarly, MRI studies, including those published in the Journal of Finance and the Review of Financial Studies have shown the neural origins of some of the biases dear to behavioral finance. The disposition effect, the tendency to sell too quickly a financial asset whose price has risen in relation to its purchase value and, on the other hand, to delay the sale of the losing securities would have its origin in the ventral striatum of the brain. Selling a winning stock would increase its activity, triggering a sense of reward, which would explain this effect.

Similarly, two studies, published in the very serious Journal of Finance and Journal of Financial Economics studied the risk-taking in the financial markets of more than 30,000 Swedish twins. By comparing the risk-taking of genetically identical "identical identical identical twins" with that of "fraternal twins", the researchers found that almost 30 % of the variability in our behaviour towards this type of risk is genetically based.

The list of studies using methods from the neurosciences such as hormone assays, MRIs or genotyping of financial actors is growing. This new branch is young, but it could in the long term bring classical finance and behavioural finance closer together. At the very least, it could answer the question of the origin of the biases observed by behavioural finance.

Will it, however, succeed in improving the predictive power of financial models? Perhaps. Neurofinance methods are so revolutionary compared to those of the past that breakthrough innovations are possible.

Science? Yes, but not exact...

Finance and economics are sciences, in that they seek to understand the world. They made significant advances during the 20th century.^e century, in particular thanks to the theory of expected utility.

However, even in finance, where data is abundant, we have just seen that models often have rather weak predictive powers. Many areas of economics do not have as much data. Moreover, economic systems are by nature changeable, making it even more difficult to model and therefore to predict.

It is in this sense that neuroeconomics and neurofinance can contribute a lot. By bringing together experts from several fields, we may be able to better understand what is inherent in human beings. Ultimately, we may obtain models with a stronger predictive capacity.

In conclusion, while we must remain humble about our models - economics and management sciences belong to the field of human and social sciences, where prediction is difficult - we must not dismiss the results already obtained. This is certainly the message to be retained from this controversy surrounding the book by Cahuc and Zylberberg.

Luc MeunierProfessor of Finance and PhD student, Grenoble École de Management (GEM); François Desmoulins-Lebeaultfinance professor, Grenoble École de Management (GEM) and Jean-François GajewskiProfessor of Finance, Savoie Mont Blanc University

The original text of this article was published on The Conversation.