Publications

This paper examines the effect of weather shocks on violent crime using disaggregated data from Brazilian municipalities over the period 1991–2015. Employing a distributed lag model that takes into account temporal correlations of weather shocks and spatial correlation of crime rates, I document that adverse weather shocks in the form of droughts lead to a significant increase in violent crime in rural regions. This effect appears to persist beyond the growing season and over the medium run in contrast to the conventional view perceiving weather effects as transitory. To explain this persistence, I show that weather fluctuations are positively associated not only with agriculture yields, but also with the overall economic activity. Moreover, evidence shows the dominance of opportunity cost mechanism reflected in the fluctuations of the earnings especially for the agriculture and unskilled workers, giving credence that it is indeed the income that matters and not the general socio-economic conditions. Other factors such as local government budget capacity, (un)-employment, poverty, inequality, and psychological factors do not seem to explain violent crime rates.

This study aims to evaluate people’s willingness to provide their geospatial global positioning system (GPS) data from their smartphones during the COVID-19 pandemic. Based on the self-determination theory, the addition of monetary incentives to encourage data provision may have an adverse effect on spontaneous donation. Therefore, we tested if a crowding-out effect exists between financial and altruistic motivations. Participants were randomized to different frames of motivational messages regarding the provision of their GPS data based on (1) self-interest, (2) pro-social benefit, and (3) monetary compensation. We also sought to examine the use of a negative versus positive valence in the framing of the different armed messages. 1055 participants were recruited from 41 countries with a mean age of 34 years on Amazon Mechanical Turk (MTurk), an online crowdsourcing platform. Participants living in India or in Brazil were more willing to provide their GPS data compared to those living in the United States. No significant differences were seen between positive and negative valence framing messages. Monetary incentives of $5 significantly increased participants’ willingness to provide GPS data. Half of the participants in the self-interest and pro-social arms agreed to provide their GPS data and almost two-thirds of participants were willing to provide their data in exchange for $5. If participants refused the first framing proposal, they were followed up with a “Vickrey auction” (a sealed-bid second-priced auction, SPSBA). An average of $17 bid was accepted in the self-interest condition to provide their GPS data, and the average “bid” of $21 was for the pro-social benefit experimental condition. These results revealed that a crowding-out effect between intrinsic and extrinsic motivations did not take place in our sample of internet users. Framing and incentivization can be used in combination to influence the acquisition of private GPS smartphone data. Financial incentives can increase data provision to a greater degree with no losses on these intrinsic motivations, to fight the COVID-19 pandemic.

In this paper we introduce a definition of approximate Pareto efficient solution as well as a necessary condition for such solutions in the multiobjective setting on Riemannian manifolds. We also propose an inexact proximal point method for nonsmooth multiobjective optimization in the Riemannian context by using the notion of approximate solution. The main convergence result ensures that each cluster point (if any) of any sequence generated by the method is a Pareto critical point. Furthermore, when the problem is convex on a Hadamard manifold, full convergence of the method for a weak Pareto efficient solution is obtained. As an application, we show how a Pareto critical point can be reached as a limit of traps in the context of the variational rationality approach of stay and change human dynamics.

Redistributive justice is based on the premise that it is unfair for people to be better or worse off relative to others simply because of their fortune or misfortune. It assumes equal opportunities arising from four factors: social circumstances, effort, option luck and brute luck. This paper seeks to investigate how differences in perceived brute luck influence individual preferences for redistribution in favour of two public policies: “health intervention” and “environmental actions”. These policies are viewed somewhat differently: the environment is considered a pure “public good” and health, more as a “private good” with a strong public good element. Consequently, potential self-serving biases inherent in the preferences for redistributive policies are expected to differ, more likely favouring health than the environment. The perceived degree of brute luck may capture such a difference—those perceiving themselves as luckiest should be less amenable to redistribution in favour of health than the unluckiest. Data from the three waves (2000, 2006 and 2008) of a French population survey are used to examine this self-serving bias. A Generalised Ordered Logit (GOL) model is found to be statistically more relevant compared to other logistic regression models (multinomial and ordered). We find that a perceived low degree of brute luck is significantly associated with a decreased preference of redistributive environmental policies but the reverse is true for redistributive health policies, i.e., association with an increased preference. Assuming that all inequalities due to differing luck are unjust, this empirical validation gives redistributive justice grounds for equalisation policies regarding health.