The use of experiments in economics emerged in the immediate post-war period, and even before in the 1930s. Furthermore, it experienced a major boom since the 1970s, with a consecration in the 2000s through the award of Nobel Prizes for experimental economics, and more recently the similar recognition of the pioneering work of Esther Duflo on Randomized Controlled Trials (RCTs) in development economics. In this article, we focus on the two most commonly used methodologies: on the one hand those that fall into the category of experimental economics and on the other, RCTs. Although they share the same basic principle of engendering an observation in order to study certain phenomena, they differ in their origin and epistemological status. On the one hand, experimental economics, oriented towards theory, finds its origins in the 1970s in a refutationist approach. On the other hand, RCTs, oriented towards the evaluation of public policies, have confirmationist origins. However, we argue that the joint influence of the two methodologies subsequently put into question the conception of economics described by Hausman [1992] as "an inaccurate and separate science". Indeed, experimental Economics and RCTs have turned economics into a less generalist and more contextualized discipline. They have also forced researchers, both experimentalists and theorists to rethink the very definition of the domain of economics. While some are trying to make it more specific, others are trying to broaden it, based on their different conceptions of interdisciplinarity. Finally, it seems that neither of the two types of experimental methodology really question the theory of rationality as a normative reference.

Apparue dans l'immediat apres-guerre, voire dans les annees 1930, la pratique des experiences en economie a connu un essor important depuis les annees 1970, et une consecration plus recente dans les annees 2000 a travers l'attribution de prix Nobel en economie experimentale, ainsi que la reconnaissance des travaux d'Esther Duflo. Dans cet article nous nous interessons aux deux methodologies les plus couramment utilisees : celles qui entrent dans la categorie de l'economie experimentale et les experiences aleatoires controlees (EAC). Partageant le meme principe de base qui est de « provoquer une observation dans le but d'etudier certains phenomenes », elles n'en different pas moins par leur origine et leur statut epistemologique. D'une part l'economie experimentale, tournee vers la theorie, trouve ses origines dans une tentative refutationniste dans les annees 1970. D'autre part les EAC, tournees vers l'evaluation des politiques publiques, viennent d'une recherche de confirmation. Nous montrons par contre que l'influence conjointe des deux methodologies a ensuite ebranle la conception de l'economie comme science inexacte et separee (Hausman 1992). Economie experimentale et EAC ont oeuvre pour que l'economie devienne une discipline moins generaliste et plus contextualisee. Elles ont aussi oblige les chercheurs, experimentalistes, randomistes et theoriciens a repenser la definition du domaine de l'economie. Quand certains s'efforcent de le preciser, d'autres tentent de l'elargir, selon differentes conceptions de l'interdisciplinarite. Par contre, il semble que les differentes methodologies experimentales ne remettent que peu en question la theorie de la rationalite comme reference normative.

The Vietnamese government is currently attempting to upgrade rice value chains in the Mekong River Delta by encouraging (i) vertical coordination between exporters and farmers through contract farming, and (ii) horizontal coordination among farmers through the "small farmers, large field" program. Previous studies on the determinants of contract farming participation assume that firms offer only a single contract type, whereas in reality, farmers may face a continuum of exclusive contract options. Devising correct and targeted policies for fostering contract inclusiveness hence crucially hinges on deploying correct econometric specification of the decision to participate in contract farming. We model contract farming participation and intensity in four different ways along the vertical coordination continuum: as a discrete, categorical, ordered, and continuous choice. We find that older, smaller and horizontally coordinated farmers with higher levels of trust in buyers tend to secure higher levels of buyer investment through increased vertical coordination. In contrast with the common finding in the literature that contract participation is biased towards larger farms, our findings from Vietnam suggest that the scale bias of contract farming could be successfully relaxed through horizontal coordination and even reversed under increasing levels of vertical coordination as smaller farmers are found to secure higher levels of buyer investment. These findings highlight the role both policies can play in fostering inclusiveness of contract farming in rice value chain upgrading in Vietnam.

Les algorithmes sont de formidables outils de gouvernance pour les administrations et semblent aujourd'hui inevitables. Nous voudrions montrer ici, a travers l'exemple de Parcoursup, qu'ils pourraient devenir aussi de puissants outils democratiques. En janvier dernier, le gouvernement decidait de repondre a la forte contestation dont l'algorithme APB avait fait l'objet, par la mise en place hâtive de Parcoursup. Les mecontentements lies a APB avaient fini par se cristalliser autour du manque de transparence. Sans surprise, Parcoursup souffre des memes critiques, demultipliees du fait que chaque etablissement controle a present une partie du processus. Dans une tribune publiee au Monde en decembre 2016, le collectif academique Algocit auquel nous appartenons, prenait position en faveur de la transparence d'APB et notamment de la transparence de son cahier des charges. Notre position reste bien sur inchangee en ce qui concerne Parcoursup et nous voulons ici, a titre personnel cette fois, preciser cette position et les consequences que l'on doit en tirer pour la construction de ces algorithmes. La substitution de Parcoursup a APB ne permettra pas de faire l'economie d'une reflexion sur les procedures de construction des algorithmes utilises par les administrations. Et a l'ere de la republique du numerique, cette reflexion doit prendre la forme d'un debat public. Les questions sont nombreuses. Par qui doivent etre developpes les algorithmes utilises par les administrations ? Comment doivent-ils etre developpes ? Peut-on definir des principes qui encadreraient leur construction ? Leur fonctionnement ? Doit-on en particulier les rendre transparents ? Quelle forme pourrait prendre cette transparence ? L'intention n'est pas ici de repondre a toutes ces questions mais d'eclairer la toile de fond du debat. Il nous semble en effet necessaire de montrer que les reponses qui seront proposees dependront de la conception que l'on se fait de la democratie.

Lorsque l’économie comportementale a voulu s’inspirer des recherches sur l’empathie pour modifier le modèle de l’ homo œconomicus en interaction, elle s’est heurtée à l’impossibilité de penser les émotions dans le cadre d’une théorie des préférences sociales toujours ancrée dans la rationalité du modèle individualiste interactionniste de la théorie des jeux. Cette impossibilité est due au fait que la théorie des jeux tout d’abord, puis l’économie comportementale ensuite, n’ont pas voulu remettre en question l’objectif fondamental de l’individu lorsqu’il agit en interaction avec les autres. Or, se pencher sur la question des émotions et de l’empathie émotionnelle en particulier devait amener à remettre en question les ressorts du comportement. Que nous disent en effet les autres sciences du comportement ? Que si la survie de l’organisme est sans doute un objectif fondamental et commun à tous les êtres vivants, ce dernier tire par ailleurs sa spécificité de mécanismes fondamentaux et pré-câblés qui le spécialisent pour la vie en interaction avec ses semblables. Grâce à ces mécanismes fondamentaux l’espèce humaine a pu élaborer des modes de coopération productifs et sociétaux à grande échelle. Nous proposons en conséquence un changement de perspective en économie comportementale qui permettrait de considérer que l’objectif ultime de l’individu en société n’est pas d’assurer sa survie et ses besoins, mais de créer et d’entretenir le lien aux autres.

In a sequential bargaining game framework with complete information and no discounting, this paper proposes a mechanism in the spirit of a final offers arbitration (FOA). At each step, two parties may stop negotiating in order to implement a lottery between the two current proposals or go on making new proposals. A crucial difference with the FOA is that the list of past proposals is recorded and used by the mechanism. Once a lottery has been implemented, the parties can reject the offer which has been drawn and go back to a lottery between proposals made in the past. At the dominant subgame perfect equilibrium the two parties make gradual concessions leading to a final agreement which coincides with the Raiffa solution. The main incentive to reach an agreement comes from risk-aversion. The necessity to make step by step concessions generates evolving disagreement points and gradualism.

We propose a new non-cooperative approach to implement a cooperative bargaining solution. This approach relies on a mechanism which specifies what happens when, at the end of the bargaining phase, players' propositions are still not compatible. The mechanism uses a random lottery and the history of proposals. At equilibrium, under the threat of this mechanism, provided that the use of it carries a small cost, players reach consensus after a finite number of proposals. No discounting is needed. The equilibrium strategies implement the full Raiffa bargaining process and players agree on a solution arbitrarily close to the Raiffa solution.

The theory of social preferences expanded the definition of the utility function in order to reproduce the pro-social behavior observed in experiments. Does this then mean that this is the route towards a positive theory of morality in economics? We do not think so. Our claim is that there is an epistemic contradiction between methodological individualism which assumes that the economic agent’s rationality is autonomous from society, and the nature of social consciousness. Therefore, we argue that a positive theory of morality should rely on social mechanisms that could not fit in this framework. We give two examples of lines of research that go in this direction. The first one is drawn from 18th century moral philosophy, this is the approach adopted by Adam Smith in The Theory of Moral Sentiments. The other one is drawn from a very recent domain of research, that of social neuroscience. We show how, in both cases, the objective is not only to understand how moral judgements shape behavior but also to get an understanding of how people form these moral judgements. Smith’s thought and recent developments in social neuroscience seem to be mutually illuminating on this second aspect. Smith’s model is an essential model of a moral agent embedded in society. The sympathy operator and the impartial spectator find an echo in the way in which social neuroscience tries to understand how emotional and cognitive empathy intermesh. Furthermore, social neuroscience attempts to go further in the understanding of the complex empathy mechanism, by considering it as a learning process.

We run a series of experiments to compare the well known arbitration scheme FOA (Final Offer Arbitration) with a new arbitration scheme, non compulsory, we proposed in a companion paper (Tanimura and Thoron (2008)): ROC (Recursive Offer Conciliation). The two mechanisms are also compared with a negotiation without arbitration. We observe that the ROC mechanism seems to cumulate the advantages of the two other procedures, it avoids the high frequency of impasses observed under the FOA procedure and it is as efficient as the Free procedure in this respect. Furthermore, in an asymmetric treatment, it helps the subjects to find an agreement around the equal split of the surplus, like the arbitrator of the FOA procedure does, but without imposing anything on them.

The objective of this paper is to give the prospective outcome or "prospect" of a negotiation when the players' value can only be realized if a certain threshold is reached. A threshold is usually defined as a certain total level of contribution. Here, I define a threshold in a cooperative game by the players who are necessary to reach it. It can take different forms such as a certain number of players, or certain coalitions of players. An important example is the minimum participation constraint in international environmental agreements. A value is proposed, which is defined for a given game and a given threshold. The threshold players may bear a commitment cost which decreases the benefit they can obtain from the negotiation.

We investigate experimentally whether “binding agreements” can provide a solution to the social dilemma that arises in the presence of pure public goods. Signing a binding agreement can prevent players from free riding on the contributions to the public good. However, a well known theoretical result is that the outcome of the endogenous formation of agreements is not necessarily efficient. Our setting is a bargaining game in which agreements form sequentially. The individual level of contribution to the public good increases with the size of the coalition reaching an agreement and the global agreement is always the socially optimal structure. There are two equilibrium outcomes, the global agreement and an asymmetric structure, which consists of two coalitions of different sizes, the small one free riding on the contributions of the larger one. We run an experiment which lends force to the theoretical result that outcomes may be inefficient. However, subjects do not play Nash and the experimental outcome is, on average, even more inefficient than the theory predicts. Our analysis leads to the conclusion that different types of behaviour co-exist.

We propose a mechanism for resolving bargaining problems. The mechanism allows two players to make a sequence of simultaneous propositions. At any step, as long as the players have not reached an agreement, they can choose to implement a lottery between the different propositions. In this aspect, the mechanism is similar to the so called final others arbitration. However, contrary to the existing scheme, our mechanism is not compulsory. The history of the negotiation process is recorded and players can refuse an offer and go back in the process to a previous step. This generates an evolving sequence of status-quo points and results in a sequence of equilibrium others of the two play-ers that gradually converge towards each other. Our model assumes no discounting and complete information. Rather than time preferences, the main incentive to reach an agreement under our mechanism comes from risk aversion. Players have an incentive to avoid the uncertainty related to the lotteries that occur when offers do not result in an agreement. Rather than incomplete information, the process gradualism is driven by the necessity to make step by step concessions in order to generate evolving threat points. We show that under this mechanism, the unique subgame perfect equilibrium that does not use weakly dominated strategies coincides with a well-known static solution concept, the Raiffa solution.

We investigate whether "binding agreements" can provide a solution to the social dilemma that arises in the presence of pure public goods. By signing a binding agreement, players can prevent free riding by the contributors to a public good. However, a well known theoretical result is that the outcome of the endogenous formation of agreements is not necessarily efficient. In our setting, the individual level of contribution to the public good increases with the size of the coalition reaching an agreement and the global coalition is always the socially optimal structure. Agreements form sequentially and the equilibrium outcome is an asymmetric structure, which consists of two coalitions. Our experiment therefore lends force to the theoretical result that outcomes may be inefficient. In fact, we observe an outcome which is even less efficient than that predicted by the equilibrium agreement structure. However, it seems that when subjects reach agreements they do so with the intention of cooperating rather than free riding. Furthermore, it seems that they “learn to cooperate” over time and reach the global agreement more often towards the end of sessions.

An “open source” software is a software whose source-code, that is the sequence of instructions that forms the program, is openly available to anyone and cannot be privately appropriated and sold or rented. The development of open source software, also called “free software”, is based on the contributions of voluntary and benevolent developers. This within a cooperative mode of organization that largely draws its efficiency from the organization and communication commodities which have arisen from the Internet structure.

Conditions under which endogenous processes of jurisdiction formation entail wealth-stratification are examined in a model where unequally wealthy households with identical preferences form jurisdictions in order to produce a public good financed by proportional taxation. We define a stable jurisdiction structure to be a partition of the households into jurisdictions that is immune to individual deviations. We define a jurisdiction structure to be wealth-stratified when each jurisdiction is composed of households who form an interval with respect to the ordering of their wealth. We show that a necessary and sufficient condition for the stratification of any stable jurisdiction structure is for the individual preferences for the public and the private good to exhibit a relation of gross substitutability/complementarity between the public good and the private good that is independent from prices and wealth.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The procedure for implementing any international treaty necessarily involves two steps. The negotiation phase which culminates in the signature of the treaty is followed by a ratification phase. This last phase is governed by a rule which determines how far the ratification process has to advance before the treaty can come into effect. The purpose of this paper is to analyse, using a game theoretical approach, the possible consequences of this minimum participation rule for the ratification phase and for the negotiation phase. I consider the case of International Environmental Agreements in which, during the negotiation phase, the different parties have to decide on the level of a global target and on how to share the efforts necessary to reach it. I use a cooperative approach to define what is called the threshold value (T-value). For a given coalition of parties, the T-value gives the expected outcome of the negotiation over sharing a global target, when the parties take into account the minimum participation rule. Given this T-value, I use a non-cooperative approach to determine which coalition will sign the agreement and what will be its global target. The minimum participation constraint has in fact no impact on the ratification phase because it is always better to refuse to sign rather than to sign and then refuse to ratify. However, I show that the minimum participation constraint can modify the outcome of the negotiation phase. Indeed, it plays a role in a mechanism which can be used by a coalition to signal its leadership commitment. I analyse the conditions under which, at the equilibrium, the leading coalition can provoke an expansion of the signing coalition.

The theory of coalition formation which has been developed in a symmetric framework, explains that the problem of participation in the negotiations on climate change can be attributed to the existence of positive externalities and the incentive to free ride. The three papers on negotiations in this volume present the other e¤ects which arise when asymmetries between countries are taken into account. Here, the instability of the grand coalition and the emergence of climate-blocks are discussed. The question is whether the latter are observed in reality and is this harmful? JEL Classi…cations: C68, C70, C71, C72, H23, H41, Q25. Key words: climate change, emissions trading, ‡exibility mechanism, country-speci…c targets, climate-blocs, agreement, coalition formation, negotiation, membership, stability. GREQAM and University of Toulon, Institute for Advanced Study, Princeton.