Comment lutter contre la fraude fiscale ? Les enseignements de l’économie comportementaleBookNicolas Jacquemet, Stéphane Luchini and Antoine Malézieux, CEPREMAP, 2020, Number 53, 104 pages, Rue d'Ulm, 2020

La fraude fiscale est un sujet qui se dérobe aux outils de l’analyse économique traditionnelle. D’une part, comme toute activité illégale, la fraude fiscale échappe à l’observation du chercheur en même temps qu’elle se dissimule aux autorités : l’analyse empirique de son ampleur, de ses déterminants et de la manière dont différents dispositifs l’affectent est nécessairement très limitée. D’autre part, sur le plan théorique, l’application simple du calcul coût-bénéfice auquel est supposé se livrer le contribuable « rationnel » conduit à un paradoxe : contrairement à une idée largement répandue, les bénéfices de la fraude fiscale sont tellement élevés, et le risque de sanction est tellement faible, que l’on peut s’étonner qu’elle soit aussi peu pratiquée dans l’ensemble des économies développées. Plutôt que la fraude fiscale, c’est donc la «soumission fiscale» qui en constitue le pendant, la disposition à payer l’impôt, qu’il convient d’expliquer pour en comprendre les déterminants.

Le double défi que posent les décisions de fraude fiscale à l’analyse économique n’a pu être relevé que très récemment, grâce à l’émergence, au cours des vingt dernières années, d’une nouvelle approche, l’économie comportementale, qui s’appuie sur la psychologie pour mieux comprendre les comportements économiques ; et, conjointement, d’une nouvelle méthode, l’économie expérimentale, qui permet d’étudier empiriquement les comportements économiques sur lesquels il est difficile de collecter des données convaincantes.

Cet opuscule rend compte des résultats de ces travaux et présente un panorama des outils de politique fiscale qui s’en dégagent.

Who’ll stop lying under oath? Empirical evidence from tax evasion gamesJournal articleNicolas Jacquemet, Stéphane Luchini, Antoine Malézieux and Jason F. Shogren, European Economic Review, Volume 124, pp. 103369, 2020

Using two earned income/tax declaration experimental designs we show that only partial liars are affected by a truth-telling oath, a non-price commitment device. Under oath, we see no change in the number of chronic liars and fewer partial liars. Rather than smoothly increasing their compliance, we also observe that partial liars who respond to the oath, respond by becoming fully honest under oath. Based on both response times data and the consistency of subjects when several compliance decisions are made in a row, we show that partial lying arises as the result of weak preferences towards profitable honesty. The oath only transforms people with weak preferences for lying into being committed to the truth.

Truth-telling under OathJournal articleNicolas Jacquemet, Stéphane Luchini, Julie Rosaz and Jason F. Shogren, Management Science, Volume 65, Issue 1, pp. 426-438, 2019

Oath taking for senior executives has been promoted as a means to enhance honesty within and toward organizations. Herein we explore whether people who voluntarily sign a solemn truth-telling oath are more committed to sincere behavior when offered the chance to lie. We design an experiment to test how the oath affects truth telling in two contexts: a neutral context replicating the typical experiment in the literature, and a “loaded” context in which we remind subjects that “a lie is a lie.” We consider four payoff configurations, with differential monetary incentives to lie, implemented as within-subjects treatment variables. The results are reinforced by robustness investigations in which each subject made only one lying decision. Our results show that the oath reduces lying, especially in the loaded environment—falsehoods are reduced by 50%. The oath, however, has a weaker effect on lying in the neutral environment. The oath did affect decision times in all instances: the average person takes significantly more time deciding whether to lie under oath.

Clinical and economic burden of head and neck cancer: a nationwide retrospective cohort study from FranceJournal articleAntoine Schernberg, Luis Sagaon-Teyssier, Michael Schwarzinger, Stéphane Luchini, ClinicoEconomics and Outcomes Research, Volume 11, pp. 441-451, 2019

To evaluate the clinical and economic burden of head and neck squamous cell carcinoma (HNSCC) in France.

All 53,255 incident adult patients discharged with a first diagnosis of HNSCC in 2010–2012 were identified from the 2008–2013 French National Hospital Discharge (PMSI) database. We conducted a retrospective longitudinal analysis of prognosis and direct costs attributable to HNSCC.

Direct medical costs attributable to HNSCC care amounted to 665 million euros in 2012 in France. The majority (62%) of incident patients were 64 years old or less at HNSCC diagnosis and incurred 1.3-fold higher mean direct costs as compared to elderly patients (41,909 vs 32,221 euros over 3 years, respectively; p<0.001). HNSCC stage at initial treatment was the major driver of mean (SD) direct costs over 3 years (p<0.001): 19,819 (23,150) euros in 31% patients diagnosed at early stage; 46,791 (34,841) euros in 60% patients diagnosed at locally advanced stage; and 43,377 (33,953) euros in 9% patients diagnosed with distant metastasis. About half patients died over 3 years at a median (IQR) age of 63 (56–75) years resulting in 10.9 years-of-life lost on average per incident patient.

The present study suggests that the clinical and economic burden of HNSCC is substantial in France.

Estimating health state utility from activities of daily living in the French National Hospital Discharge Database: a feasibility study with head and neck cancerJournal articleMichael Schwarzinger, Stéphane Luchini, Sylvain Baillot, Mélina Bec, Lynda Benmahammed, Caroline Even, Lionnel Geoffrois, Florence Huguet, Béatrice Le Vu, Laurie Lévy-Bachelot, et al., Health and Quality of Life Outcomes, Volume 17, Issue 1, pp. 129, 2019

Background :
Health state utility (HSU) is a core component of QALYs and cost-effectiveness analysis, although HSU is rarely estimated among a representative sample of patients. We explored the feasibility of assessing HSU in head and neck cancer from the French National Hospital Discharge database.

An exhaustive sample of 53,258 incident adult patients with a first diagnosis of head and neck cancer was identified in 2010–2012. We used a cross-sectional approach to define five health states over two periods: three "cancer stages at initial treatment" (early, locally advanced or metastatic stage); a "relapse state" and otherwise a "relapse-free state" in the follow-up of patients initially treated at early or locally advanced stage. In patients admitted in post-acute care, a two-parameter graded response model (Item Response Theory) was estimated from all 144,012 records of six Activities of Daily Living (ADLs) and the latent health state scale underlying ADLs was calibrated with the French EQ-5D-3 L social value set. Following linear interpolation between all assessments of the patient, daily estimates of utility in post-acute care were averaged by health state, patient and month of follow-up. Finally, HSU was estimated by health state and month of follow-up for the whole patient population after controlling for survivorship and selection in post-acute care.

Head and neck cancer was generally associated with poor HSU estimates in a real-life setting. As compared to “distant metastasis at initial treatment”, mean HSU was higher in other health states, although numerical differences were small (0.45 versus around 0.54). It was primarily explained by the negative effects on HSU of an older age (38.4% aged ≥70 years in “early stage at initial treatment”) and comorbidities (> 50% in other health states). HSU estimates significantly improved over time in the “relapse-free state” (from 8 to 12 months of follow-up).

HSU estimates in head and neck cancer were primarily driven by age at diagnosis, comorbidities, and time to assessment of cancer survivors. This feasibility study highlights the potential of estimating HSU within and across severe conditions in a systematic way at the national level.

A Psychometric Investigation of the Personality Traits Underlying Individual Tax MoraleJournal articleNicolas Jacquemet, Stéphane Luchini, Antoine Malézieux and Jason F. Shogren, The B.E. Journal of Economic Analysis & Policy, Volume 19, Issue 3, 2019

Why do people pay taxes? Rational choice theory has fallen short in answering this question. Another explanation, called “tax morale”, has been promoted. Tax morale captures the behavioral idea that non-monetary preferences (like norm-submission, moral emotions and moral judgments) might be better determinants of tax compliance than monetary trade-offs. Herein we report on two lab experiments designed to assess whether norm-submission, moral emotions (e.g. affective empathy, cognitive empathy, propensity to feel guilt and shame) or moral judgments (e.g. ethics principles, integrity, and moralization of everyday life) can help explain compliance behavior. Although we find statistically significant correlations of tax compliance behavior with empathy and shame, the economic significance of these correlations are low–—more than 80% of the variability in compliance remains unexplained. These results suggest that tax authorities should focus on the institutional context, rather than individual preference characteristics, to handle tax evasion.

Fairness in cost-benefit analysis: A methodology for health technology assessmentJournal articleAnne-Laure Samson, Erik Schokkaert, Clémence Thébaut, Brigitte Dormont, Marc Fleurbaey, Stéphane Luchini and Carine Van de Voorde, Health Economics, Volume 27, Issue 1, pp. 102-114, 2018

We evaluate the introduction of various forms of antihypertensive treatments in France with a distribution-sensitive cost-benefit analysis. Compared to traditional cost-benefit analysis, we implement distributional weighting based on equivalent incomes, a new concept of individual well-being that does respect individual preferences but is not subjectively welfarist. Individual preferences are estimated on the basis of a contingent valuation question, introduced into a representative survey of the French population. Compared to traditional cost-effectiveness analysis in health technology assessment, we show that it is feasible to go beyond a narrow evaluation of health outcomes while still fully exploiting the sophistication of medical information. Sensitivity analysis illustrates the relevancy of this richer welfare framework, the importance of the distinction between an ex ante and an ex post approach, and the need to consider distributional effects in a broader institutional setting.

Contribution of alcohol use disorders to the burden of dementia in France 2008-13: a nationwide retrospective cohort studyJournal articleMichael Schwarzinger, Bruce G. Pollock, Omer S. M. Hasan, Carole Dufouil, Jürgen Rehm, S Baillot, Quentin Guibert, F Planchet and Stéphane Luchini, The Lancet Public Health, Volume 3, Issue 3, pp. e124-e132, 2018

Dementia is a prevalent condition, affecting 5-7% of people aged 60 years and older, and a leading cause of disability in people aged 60 years and older globally. We aimed to examine the association between alcohol use disorders and dementia risk, with an emphasis on early-onset dementia (<65 years).

We analysed a nationwide retrospective cohort of all adult (≥20 years) patients admitted to hospital in metropolitan France between 2008 and 2013. The primary exposure was alcohol use disorders and the main outcome was dementia, both defined by International Classification of Diseases, tenth revision discharge diagnosis codes. Characteristics of early-onset dementia were studied among prevalent cases in 2008-13. Associations of alcohol use disorders and other risk factors with dementia onset were analysed in multivariate Cox models among patients admitted to hospital in 2011-13 with no record of dementia in 2008-10.

Of 31 624 156 adults discharged from French hospitals between 2008 and 2013, 1 109 343 were diagnosed with dementia and were included in the analyses. Of the 57 353 (5·2%) cases of early-onset dementia, most were either alcohol-related by definition (22 338 [38·9%]) or had an additional diagnosis of alcohol use disorders (10 115 [17·6%]). Alcohol use disorders were the strongest modifiable risk factor for dementia onset, with an adjusted hazard ratio of 3·34 (95% CI 3·28-3·41) for women and 3·36 (3·31-3·41) for men. Alcohol use disorders remained associated with dementia onset for both sexes (adjusted hazard ratios >1·7) in sensitivity analyses on dementia case definition (including Alzheimer's disease) or older study populations. Also, alcohol use disorders were significantly associated with all other risk factors for dementia onset (all p<0·0001).

Alcohol use disorders were a major risk factor for onset of all types of dementia, and especially early-onset dementia. Thus, screening for heavy drinking should be part of regular medical care, with intervention or treatment being offered when necessary. Additionally, other alcohol policies should be considered to reduce heavy drinking in the general population.


Coordination with communication under oathJournal articleNicolas Jacquemet, Stéphane Luchini, Jason F. Shogren and Adam Zylbersztejn, Experimental Economics, Volume 21, Issue 3, pp. 627-649, 2018

We focus on the design of an institutional device aimed to foster coordination through communication. We explore whether the social psychology theory of commitment, implemented via a truth-telling oath, can reduce coordination failure. Using a classic coordination game, we ask all players to sign voluntarily a truth-telling oath before playing the game with cheap talk communication. Three results emerge with commitment under oath: (1) coordination increased by nearly 50%; (2) senders’ messages were significantly more truthful and actions more efficient, and (3) receivers’ trust of messages increased.

Individual Uncertainty About LongevityJournal articleBrigitte Dormont, Anne-Laure Samson, Marc Fleurbaey, Stéphane Luchini and Erik Schokkaert, Demography, Volume 55, Issue 5, pp. 1829-1854, 2018

This article presents an assessment of individual uncertainty about longevity. A survey performed on 3,331 French people enables us to record several survival probabilities per individual. On this basis, we compute subjective life expectancies (SLE) and subjective uncertainty regarding longevity (SUL), the standard deviation of each individual’s subjective distribution of her or his own longevity. It is large and equal to more than 10 years for men and women. Its magnitude is comparable to the variability of longevity observed in life tables for individuals under 60, but it is smaller for those older than 60, which suggests use of private information by older respondents. Our econometric analysis confirms that individuals use private information—mainly their parents’ survival and longevity—to adjust their level of uncertainty. Finally, we find that SUL has a sizable impact, in addition to SLE, on risky behaviors: more uncertainty on longevity significantly decreases the probability of unhealthy lifestyles. Given that individual uncertainty about longevity affects prevention behavior, retirement decisions, and demand for long-term care insurance, these results have important implications for public policy concerning health care and retirement.