Kohmei Makihara*, Florian Roeser**, Guillaume Bataille***
Kenza Elass: kenza.elass[at]univ-amu.fr
Camille Hainnaux: camille.hainnaux[at]univ-amu.fr
Daniela Horta Saenz: daniela.horta-saenz[at]univ-amu.fr
Jade Ponsard: jade.ponsard[at]univ-amu.fr
*This paper analyzes a local public goods game on network in which agents can be of two types: standard maximizer, or Kantian. A Kantian behaves following the Kantian morality, i.e. they assume that everyone contributes the same amount as themselves. Under this setting, we discuss how the aggregate contribution of public good changes when replacing one standard maximizer by a Kantian. Our study shows that, contrary to previous results on public good games with Kantian agents, introducing Kantian agents in a network game does not necessarily have a positive effect on provision of public goods, i.e. there are some cases in which the aggregate contribution decreases when replacing a standard maximizer by a Kantian. However, we also provide sufficient condition such that replacing a standard maximiser by a Kantian player will have an increase of the aggregate contribution, regardless of the number of each type of player in the game.
**We analyze the role of educational spending on growth and inequality in a dynamic R&D growth model with endogenous education and automation. We assume low-skilled workers to be perfect substitutes to machines, while high-skilled workers are complements. Automation increases income inequality and the share of college graduates over time. Without government spending on education, an increase in taxes to reduce inequality always dampens growth. Educational spending is able to break this inequality-growth trade-off. Public educational spending increases the productivity of both low- and high-skilled workers allowing for redistribution towards the low-skilled which reduces inequality. We calibrate the model using US data and provide the conditions on basic and higher education spending, and on transfers, such that higher growth is compatible with lower inequality. We show that such a result can be achieved either using taxes of labor, a robot tax, or a combination of the two.
***This paper explores harvesting behaviors in a predator-prey system where the economic value of a biomass unit is lower when moving-up to a higher trophic level. Precisely, the prey species is a more valuable product for harvesters than the same prey transformed in predator after biomass conversion. We compare the optimal exploitation with an unregulated limited access fishery. We show that the optimal harvesting of the predator species must happen with negative profit. The limited access fishery is characterized by lower prey stock, higher predator stock and under-effort on both species compared to the social optimum. We propose a global tax on the harvest of each species to obtain the first best allocation. For the predator industry, the tax is always a subsidy with decreasing amount as the number of vessels owner grow. For the prey industry, the tax is either positive, equal to zero or negative depending on the number of vessels owner. The optimal top-down regulation is to subsidize or at least unlimited access to harvest in the predator sector to favor predator harvesting and to set exactly the right number of license in the prey industry. These results show that in our context current public policies with the aim of controlling over-exploitation (e.g Individual Transferable Quotas) profoundly degrade the economic efficiency of fisheries, which may explain the failure of fisheries management tools.