The empirical turn in economics

from Lars Syll

The Empirical Revolution in Economics - Business Review at BerkeleyCe qui fait l’unité de la discipline est plutôt l’identification causale, c’est-à-dire un ensemble de méthodes statistiques qui permettent d’estimer les liens de cause à effet entre un facteur quelconque et des résultats économiques. Dans cette perspective, la démarche scientifique vise à reproduire in vivo l’expérience de laboratoire, où l’on peut distinguer aisément la différence de résultat entre un groupe auquel on administre un traitement et un autre groupe semblable qui n’est quant à lui pas affecté.

Les outils statistiques permettraient aux économistes d’appliquer cette méthode en dehors du laboratoire, y compris à l’histoire et à tout autre sujet. Là encore, il faudrait considérablement nuancer ce constat. Mais, il ne me semble pas aberrant de dire que si, pour comprendre les canons de la discipline, tout économiste devait auparavant au moins maîtriser les bases du calcul rationnel, il s’agit surtout aujourd’hui de maîtriser les bases de l’identification économétrique (variables instrumentales et méthode des différences de différences en particulier).

Si les canons de la discipline ont changé, les rapports de l’économie dominante aux autres disciplines n’ont quant à eux pas évolué. Certains économistes se considéraient supérieurs auparavant car ils pensaient que seuls les modèles formels d’individu rationnel pouvaient expliquer les comportements de manière scientifique. Les autres explications tenant de l’évaluation subjective non rigoureuse.

Eric Monnet

Although discounting empirical evidence cannot be the right way to solve economic issues, there are still, as Monnet argues, several weighty reasons why we perhaps shouldn’t be too excited about the so-called ’empirical revolution’ in economics.

Behavioural experiments and laboratory research face the same basic problem as theoretical models — they are built on often rather artificial conditions and have difficulties with the ‘trade-off’ between internal and external validity. The more artificial conditions, the more internal validity, but also less external validity. The more we rig experiments to avoid the ‘confounding factors’, the less the conditions are reminiscent of the real ‘target system.’ The nodal issue is how economists using different isolation strategies in different ‘nomological machines’ attempt to learn about causal relationships. One may have justified doubts on the generalizability of this research strategy since the probability is high that causal mechanisms are different in different contexts and that lack of homogeneity and invariance doesn’t give us warranted export licenses to the ‘real’ societies or economies.

If we see experiments or laboratory research as theory tests or models that ultimately aspire to say something about the real ‘target system,’ then the problem of external validity is central (and was for a long time also a key reason why behavioural economists had trouble getting their research results published).

A standard procedure in behavioural economics — think of e.g. dictator or ultimatum games — is to set up a situation where one induces people to act according to the standard microeconomic — homo oeconomicus — benchmark model. In most cases, the results show that people do not behave as one would have predicted from the benchmark model, in spite of the setup almost invariably being ‘loaded’ for that purpose. [And in those cases where the result is consistent with the benchmark model, one, of course, has to remember that this in no way proves the benchmark model to be right or ‘true,’ since there, as a rule, are multiple outcomes that are consistent with that model.]

For most heterodox economists this is just one more reason for giving up on the standard model. But not so for mainstreamers and many behaviouralists. To them, the empirical results are not reasons for giving up on their preferred hardcore axioms. So they set out to ‘save’ or ‘repair’ their model and try to ‘integrate’ the empirical results into mainstream economics. Instead of accepting that the homo oeconomicus model has zero explanatory real-world value, one puts lipstick on the pig and hopes to go on with business as usual. Why we should keep on using that model as a benchmark when everyone knows it is false is something we are never told. Instead of using behavioural economics and its results as building blocks for a progressive alternative research program, the ‘save and repair’ strategy immunizes a hopelessly false and irrelevant model.

By this, I do not mean to say that empirical methods per se are so problematic that they can never be used. On the contrary, I am basically — though not without reservations — in favour of the increased use of behavioural experiments and laboratory research within economics. Not least as an alternative to completely barren ‘bridge-less’ axiomatic-deductive theory models. My criticism is more about aspiration levels and what we believe we can achieve with our mediational epistemological tools and methods in the social sciences.

The increasing use of natural and quasi-natural experiments in economics during the last couple of decades has led several prominent economists to triumphantly declare it as a major step on a recent path toward empirics, where instead of being a deductive philosophy, economics is now increasingly becoming an inductive science.

Limiting model assumptions in economic science always have to be closely examined since if we are going to be able to show that the mechanisms or causes that we isolate and handle in our models are stable in the sense that they do not change when we ‘export’ them to our ‘target systems,’ we have to be able to show that they do not only hold under ceteris paribus conditions and a fortiori only are of limited value to our understanding, explanations or predictions of real economic systems.

So — although it is good that that much of the behavioural economics research has vastly undermined the lure of axiomatic-deductive mainstream economics, there is still a long way to go before economics has become a truly empirical science. The great challenge for future economics is not to develop methodologies and theories for well-controlled laboratories, but to develop relevant methodologies and theories for the messy world in which we happen to live.