By Dominique Jacquet

Last week, I had the pleasure of participating to the defense one of my PhD students.

This moment is always very special (especially for the doctoral student…), because it sets a crucial stage in an academic career, but also because it allows a dialogue around a theoretical and, sometimes, practical advance in an important subject.

The thesis was devoted to the use of “mezzanine” financing in the development of renewable energy projects. One of the central questions was: is the use of such sophisticated financing a determining factor in the success of the financial arrangement? By observing a representative portfolio of projects, the candidate concluded that there was a robust correlation between the mobilization of the tool and the success of the operation.

This first observation was particularly pleasant to hear. Indeed, the financial world is often accused of developing financial innovations with the aim of appropriating significant profits, sometimes to the detriment of the value created among stakeholders, customers, States or citizens… In this case, the structured finance brings significant support to the management of climate change, it is a nice conclusion.

On a theoretical level, the debate (friendly and constructive!) with the jury focused on the transition from a statistical correlation to the establishment of a causal relationship.

This question is critical for the finance world. We observe financial actors, in this case investors, and their investment decisions, and conclude that these are consistent with behavioral assumptions. But, one must always be modest in the conclusions. If A and B are correlated, it can mean that A implies B, or B implies A, or that there is C which implies A and B, a third factor sometimes difficult to identify.

However, so-called “modern” finance is built on this type of inference. The portfolio theory observes that one can significantly reduce the risk, that is to say the variability of the return, by constituting a diversified portfolio of securities, and deduces from this that the wise investor will not put all his eggs in the same basket (disappearance of the specific risk) to optimize the risk-return couple, and will only require a risk premium on its residual and non-diversifiable part, the correlation with the market, the systematic risk.

This considerable advance, due to Markowitz, Sharpe and others, is called the CAPM and constitutes the basis of calculation widely used by the financial departments to estimate the discount rate of investment projects, to evaluate acquisitions and to measure the performance of operational activities. Why do we need a model describing an investor’s return requirement? Because we cannot interview them all, sending out a questionnaire would give incomplete and biased results, etc. In fact, we try to “spy” on them by analyzing their behavior to deduce their “utility function”. To be honest, the result isn’t brilliant, as anyone who has calculated ‘betas’ in their career has noted that the explained variance is often a measly 20% or 30% of the total variance (remember the R² of your statistics course?), which means that the motivation of the investor is, in fact, largely unknown to us…

Financial theory offers models, sometimes very useful, but which are only limited attempts to explain observed phenomena. We must, therefore, be very modest about the scope of our “knowledge” despite the very categorical remarks made by some dogmatic colleagues who confuse econometric results and revealed Truth…

The defense ended with the jury’s congratulations, the audience’s applause and a very friendly and convivial lunch. Bravo, Sylvère, and welcome to the community of professors-researchers, there is still a lot of work ahead of us!