Research

Published articles

with Tristan Roger (ICN Business School), Patrick Roger (EM Strasbourg, University of Strasbourg) and Marc Willinger (University of Montpellier) 

Abstract:

Recent empirical research in accounting and finance shows that the magnitude of stock prices influences analysts’ price forecasts (Roger, Roger and Schatt [2018]). In this paper, we report the results of a novel experiment where some subjects are asked to forecast future prices in a continuous double auction market. In this experiment, two successive markets take place: one where the fundamental value is a small price and one where the fundamental value is a large price. Although market prices are higher (compared to fundamental value) in small price markets than in large price markets, our results indicate that analyst subjects’ forecasts are more optimistic in small price markets compared to large price markets. Analyst subjects strongly anchor on past price trends when building their price forecasts and do not mitigate subject traders’ bias. Overall, our experimental findings support the existence of a small price bias deeply rooted in the human brain.

Published in Revue Economique, 72 [CNRS cat. 2/HCERES], September 2021.


with Patrick Sentis  (University of Montpellier) and Marc Willinger  (University of Montpellier)

Abstract:

We examine how the Brexit announcement influenced the long-run market performance of British and European listed firms. Using daily data and a sample composed of 3,015 European listed firms (805 UK and 2,210 non-UK), we find that, over a 12-month horizon, the Brexit announcement affected negatively the long-run market performance of UK firms (regardless of their business activities) and European non-British (non-UK hereafter) firms that conduct most of their business activities within the British area. We also provide evidence that, after the Brexit announcement, analysts’ earnings forecasts and the realized accounting decreased and the return volatility increased for UK firms. These new evidences in the literature show definitively that the Brexit announcement was harmful for the long-run financial and operational performance of firms involved in British businesses.

Published in Bankers, Markets & Investors, 63 [CNRS: 4; FNEGE: 3; HCERES: B], 2020.


with  Patrick Sentis (University of Montpellier) and Marc Willinger (University of Montpellier)

Abstract:

We study experimentally the reaction of asset markets to fundamental value (FV) shocks. The pre-shock and post-shock FV are both constant, but after the shock the FV is either higher or lower than before. We compare treatments with expected shocks (the date and the magnitude are known in advance, but not the direction) to treatments with unexpected shocks (subjects only know that a shock may occur but are unaware of the date and the magnitude). We observe mispricing in markets without shocks and in markets with shocks. Shocks tend to reduce the post-shock price deviation and to increase the difference of opinions (DO), whatever the type of the shock (expected or unexpected) and its direction (upwards or downwards). In contrast to standard predictions, the larger DO after a shock is not accompanied by an increase in transaction volumes, but by sharp depression of share turnover.

Published in Journal of Behavioral and Experimental Finance, May 2019.

Working papers

with Tristan Roger (ICN Business School), Patrick Roger (EM Strasbourg, University of Strasbourg) and Marc Willinger (University of Montpellier)

2017 AFSEE Award (French Experimental Economics Association)

Abstract:

In contrast with financial theory, much empirical evidence shows that stock price magnitude influences portfolio choices or future returns. The usual justifications for such an influence most often refer to stock characteristics (lottery-like features like a high variance and a positive skewness of returns). In this paper, we demonstrate that the stock price level impacts investors behavior not only because it correlates with some stock characteristics but, mainly on its own, as a result of a neuropsychological bias. In the context of experimental markets, we show that subjects process small and large prices differently, even though the probability distribution of returns on the small price asset and the large price asset are identical. Two consecutive treatments are performed, one with a small fundamental value equal to 6 (small price market) and one with a fundamental value equal to 72 (large price market). Small price markets exhibit greater mispricing than large price markets, a result obtained in both between-participants and within-participants analyses. Our findings indicate that price magnitude has a direct impact on how the subjects’ brain perceive the distribution of future returns. Though at odds with standard finance theory, our findings are consistent with: (1) evidence in neuropsychology on the use of different mental scales for small and large numbers, and (2) empirical results in the finance literature.


with Patrick Sentis  (University of Montpellier) and Marc Willinger  (University of Montpellier)

Abstract:

We provide experimental evidence about the impact of unexpected fundamental value (there after FV) shocks on prices, forecasts and volumes. The benchmark treatment “T0” involves a constant FV as in Noussair et al. (2001). In our test treatments we either implement two consecutive upwards shocks “UU” or two consecutive downwards shocks “DD”. The subjects are aware of the possibility of new information being communicated to them. However, they have no clue about the content of this information, i.e. the possible changes in FV and the timing of arrival of new information. In the two test treatments (UU and DD), shocks were implemented in the same periods. Subjects received a message announcing the new FV. We performed 18 experimental markets: 4 baseline markets and 14 test markets (7 for each treatment). Our main findings are as follows. Positive price deviation arises in almost all markets, with and without FV shocks. In the “DD” treatment, shocks lower the trading volume and the price volatility, but increase the price deviation and differences of opinion. In the “UU” treatment, shocks increase price volatility, volume of transactions and differences of opinion, but lower the price deviation. We also find that price volatility is positively related to the relative deviation and differences of opinion, but not related to the differences in risk aversion, average level of risk taking and transaction volume. Finally, we observe a positive relation between the differences of opinion (based on elicited beliefs about future prices) and the volume of transactions only in the baseline treatment “T0”.

with Patrick Sentis  (University of Montpellier) and Marc Willinger  (University of Montpellier)

You can find my  CV here