Journal of Infrastructure, Policy and Development, 2018
In this paper, we establish a two-way causality between the phenomenon of infrastructure which is... more In this paper, we establish a two-way causality between the phenomenon of infrastructure which is underused (the so-called “white elephant case”) and the aggregate productivity level (TFP) of the economy. On the one hand, the fact that a transport infrastructure is not used as much as it could be is itself a cause of low TFP because it represents low productivity for an important item of social capital. On the other hand, low aggregate productivity makes firms’ strategies founded on large scale of production and exports riskier, given the possibility that the political decision to build the required transport infrastructure may never be taken.
In the context of school choice, we experimentally study the student-optimal stable mechanism whe... more In the context of school choice, we experimentally study the student-optimal stable mechanism where subjects take the role of students and schools are passive. Specifically, we study if a school can be better off when it unambiguously improves in the students' true preferences and its (theoretic) student-optimal stable match remains the same or gets worse. Using firstorder stochastic dominance to evaluate the schools' distributions over their actual matches, we find that schools' welfare almost always changes in the same direction as the change of the student-optimal stable matching, i.e., incentives to improve school quality are nearly idle.
In an economy where different agents undertake simultaneous and interdependent investments, this ... more In an economy where different agents undertake simultaneous and interdependent investments, this paper models the possibility that the outcome where some players invest and others do not invest is sustained in Nash equilibrium. It is well known that in models where all goods are financed through prices charged by the suppliers ("tolls" in the case of transport infrastructures), there are only two coordination equilibria: the "Big push" equilibrium, where every agent involved invests; and the "Poverty trap", whenever none invests. We consider a two person simultaneous game, where the Government decides whether to build a highway and a firm producing a composite good decides whether to use it. Instead of resorting to tolls, the infrastructure is funded through an income tax that falls on wages. Having the Government supplying the highway and the firm not using it is a Nash equilibrium if the employment generated by the construction of the highway is intermediate and the rate of the wage income tax is high. The proliferation of unused transport infrastructures in Southern Europe seems to be related with low effects of public works upon the demand for labor and with demanddepressing "austerity" macroeconomic policies.
In this note, we start to claim that established marriages can be heavily destabilized when the p... more In this note, we start to claim that established marriages can be heavily destabilized when the population of existing couples is enriched by the arrival of new candidates to marriage. Afterwards, we discuss briefly how stability concepts can be extended to account for entry and exit phenomena affecting the composition of the marriage market.
Duopoly firms engaged in a standard two-stage game of R&D and Cournot competition are caught in a... more Duopoly firms engaged in a standard two-stage game of R&D and Cournot competition are caught in a prisoner's dilemma for their R&D decisions whenever spillover effects are low. This effect works to the advantage of consumers and society. This result provides an interesting perspective on the well-known wedge between private and social incentives for R&D. The prisoner's dilemma is the key effect behind this wedge under low spillovers. The latter take over when sufficiently high, as is widely recognized. This mutually exclusive nature of the prisoner's dilemma and significant spillovers also serves to explain the incentives to form R&D cartels.
Decentralized markets are modeled by means of a sequential game where, starting from any matching... more Decentralized markets are modeled by means of a sequential game where, starting from any matching situation, …rms are randomly given the opportunity to make job o¤ers. In this random context, we prove the existence of ordinal subgame perfect equilibria where …rms act according to a list of preferences. Moreover, every such equilibrium preserves stability for a particular pro…le of preferences. In particular, when …rms act truthfully, every outcome is stable for the true preferences. Conversely, when the initial matching is the empty matching, every stable matching can be reached as the outcome of an ordinal equilibrium play of the game.
We present an experimental study where we analyze three wellknown matching mechanisms-the Boston,... more We present an experimental study where we analyze three wellknown matching mechanisms-the Boston, the Gale-Shapley, and the Top Trading Cycles mechanisms-in three di¤erent informational settings. Our experimental results are consistent with the theory, suggesting that the TTC mechanism outperforms both the Boston and the Gale-Shapley mechanisms in terms of e¢ ciency and it is as successful as the Gale-Shapley mechanism regarding the proportion of truthful preference revelation, whereas manipulation is stronger under the Boston mechanism. In addition, even though agents are much more likely to revert to truthtelling in lack of information about the others' payo¤s-ignorance may be bene…cial in this context-, the TTC mechanism results less sensitive to the amount of information that participants hold. These results therefore suggest that the use of the TTC mechanism in practice is more desirable than of the others.
We experimentally investigate in the laboratory two prominent mechanisms that are employed in sch... more We experimentally investigate in the laboratory two prominent mechanisms that are employed in school choice programs to assign students to public schools. We study how individual behavior is influenced by preference intensities and risk aversion. Our main results show that (a) the Gale-Shapley mechanism is more robust to changes in cardinal preferences than the Boston mechanism independently of whether individuals can submit a complete or only a restricted ranking of the schools and (b) subjects with a higher degree of risk aversion are more likely to play "safer" strategies under the Gale-Shapley but not under the Boston mechanism. Both results have important implications for the efficiency and the stability of the mechanisms.
In the college admissions problem, we consider the incentives confronting agents who face the pro... more In the college admissions problem, we consider the incentives confronting agents who face the prospect of being matched by a random stable mechanism. We provide a fairly complete characterization of ordinal equilbria. Namely, every ordinal equilibrium yields a degenerate probability distribution. Furthermore, individual rationality is a necessary and su¢ cient condition for an equilibrium outcome, while stability is guaranteed in ordinal equilibrium where …rms act straightforwardly. Finally, we relate equilibrium behavior in random and in deterministic mechanisms.
Our aim is to characterize perfect equilibria in matching markets. Ordinal preferences require th... more Our aim is to characterize perfect equilibria in matching markets. Ordinal preferences require the use of an ordinal perfect equilibrium concept. We show that, in the game induced by a random stable mechanism, an ordinal perfect equilibrium strategy lists all the acceptable partners. Moreover, when either the firm" optimal or the worker" optimal mechanisms are considered, truth telling is a very prudent form of behavior and is the unique ordinal perfect equilibrium that may emerge. Finally, in the game induced by these ...
We report data from controlled laboratory experiments on two-sided matching markets in which part... more We report data from controlled laboratory experiments on two-sided matching markets in which participants interact in a decentralized way, without having to refer to a central clearinghouse. Our treatments have been designed to evaluate the effect of information, search costs, and binding agreements on the final outcome and also on the individual strategies that lead to it. We find that these features affect the level and pace of market activity as well as the identity of those who receive proposals. While the lack of information ...
Instituto Superior de Economia e Gestão-DE Working papers, 2006
We consider centralized matching markets in which, starting from an arbitrary match¬ ing, firms a... more We consider centralized matching markets in which, starting from an arbitrary match¬ ing, firms are successively chosen in a random fashion and offer their positions to the workers they prefer the most. We propose an algorithm that generalizes some well-known algorithms and explore some of its properties. In particular, different executions of the algorithm may lead to diferent output matchings. We then study incentives in the rev¬ elation game induced by the algorithm. We prove that ordinal equilibria always exist. Furthermore, every ...
We analyze two well-known matching mechanisms-the Gale-Shapley, and the Top Trading Cycles (TTC) ... more We analyze two well-known matching mechanisms-the Gale-Shapley, and the Top Trading Cycles (TTC) mechanisms-in the experimental lab in three different informational settings, and study the role of information in individual decision making. Our results suggest that-in line with the theory-in the college admissions model the Gale-Shapley mechanism outperforms the TTC mechanisms in terms of efficiency and stability, and it is as successful as the TTC mechanism regarding the proportion of truthful preference revelation. In addition, we find that information has an important effect on truthful behavior and stability. Nevertheless, regarding efficiency, the Gale-Shapley mechanism is less sensitive to the amount of information participants hold.
Journal of Infrastructure, Policy and Development, 2018
In this paper, we establish a two-way causality between the phenomenon of infrastructure which is... more In this paper, we establish a two-way causality between the phenomenon of infrastructure which is underused (the so-called “white elephant case”) and the aggregate productivity level (TFP) of the economy. On the one hand, the fact that a transport infrastructure is not used as much as it could be is itself a cause of low TFP because it represents low productivity for an important item of social capital. On the other hand, low aggregate productivity makes firms’ strategies founded on large scale of production and exports riskier, given the possibility that the political decision to build the required transport infrastructure may never be taken.
In the context of school choice, we experimentally study the student-optimal stable mechanism whe... more In the context of school choice, we experimentally study the student-optimal stable mechanism where subjects take the role of students and schools are passive. Specifically, we study if a school can be better off when it unambiguously improves in the students' true preferences and its (theoretic) student-optimal stable match remains the same or gets worse. Using firstorder stochastic dominance to evaluate the schools' distributions over their actual matches, we find that schools' welfare almost always changes in the same direction as the change of the student-optimal stable matching, i.e., incentives to improve school quality are nearly idle.
In an economy where different agents undertake simultaneous and interdependent investments, this ... more In an economy where different agents undertake simultaneous and interdependent investments, this paper models the possibility that the outcome where some players invest and others do not invest is sustained in Nash equilibrium. It is well known that in models where all goods are financed through prices charged by the suppliers ("tolls" in the case of transport infrastructures), there are only two coordination equilibria: the "Big push" equilibrium, where every agent involved invests; and the "Poverty trap", whenever none invests. We consider a two person simultaneous game, where the Government decides whether to build a highway and a firm producing a composite good decides whether to use it. Instead of resorting to tolls, the infrastructure is funded through an income tax that falls on wages. Having the Government supplying the highway and the firm not using it is a Nash equilibrium if the employment generated by the construction of the highway is intermediate and the rate of the wage income tax is high. The proliferation of unused transport infrastructures in Southern Europe seems to be related with low effects of public works upon the demand for labor and with demanddepressing "austerity" macroeconomic policies.
In this note, we start to claim that established marriages can be heavily destabilized when the p... more In this note, we start to claim that established marriages can be heavily destabilized when the population of existing couples is enriched by the arrival of new candidates to marriage. Afterwards, we discuss briefly how stability concepts can be extended to account for entry and exit phenomena affecting the composition of the marriage market.
Duopoly firms engaged in a standard two-stage game of R&D and Cournot competition are caught in a... more Duopoly firms engaged in a standard two-stage game of R&D and Cournot competition are caught in a prisoner's dilemma for their R&D decisions whenever spillover effects are low. This effect works to the advantage of consumers and society. This result provides an interesting perspective on the well-known wedge between private and social incentives for R&D. The prisoner's dilemma is the key effect behind this wedge under low spillovers. The latter take over when sufficiently high, as is widely recognized. This mutually exclusive nature of the prisoner's dilemma and significant spillovers also serves to explain the incentives to form R&D cartels.
Decentralized markets are modeled by means of a sequential game where, starting from any matching... more Decentralized markets are modeled by means of a sequential game where, starting from any matching situation, …rms are randomly given the opportunity to make job o¤ers. In this random context, we prove the existence of ordinal subgame perfect equilibria where …rms act according to a list of preferences. Moreover, every such equilibrium preserves stability for a particular pro…le of preferences. In particular, when …rms act truthfully, every outcome is stable for the true preferences. Conversely, when the initial matching is the empty matching, every stable matching can be reached as the outcome of an ordinal equilibrium play of the game.
We present an experimental study where we analyze three wellknown matching mechanisms-the Boston,... more We present an experimental study where we analyze three wellknown matching mechanisms-the Boston, the Gale-Shapley, and the Top Trading Cycles mechanisms-in three di¤erent informational settings. Our experimental results are consistent with the theory, suggesting that the TTC mechanism outperforms both the Boston and the Gale-Shapley mechanisms in terms of e¢ ciency and it is as successful as the Gale-Shapley mechanism regarding the proportion of truthful preference revelation, whereas manipulation is stronger under the Boston mechanism. In addition, even though agents are much more likely to revert to truthtelling in lack of information about the others' payo¤s-ignorance may be bene…cial in this context-, the TTC mechanism results less sensitive to the amount of information that participants hold. These results therefore suggest that the use of the TTC mechanism in practice is more desirable than of the others.
We experimentally investigate in the laboratory two prominent mechanisms that are employed in sch... more We experimentally investigate in the laboratory two prominent mechanisms that are employed in school choice programs to assign students to public schools. We study how individual behavior is influenced by preference intensities and risk aversion. Our main results show that (a) the Gale-Shapley mechanism is more robust to changes in cardinal preferences than the Boston mechanism independently of whether individuals can submit a complete or only a restricted ranking of the schools and (b) subjects with a higher degree of risk aversion are more likely to play "safer" strategies under the Gale-Shapley but not under the Boston mechanism. Both results have important implications for the efficiency and the stability of the mechanisms.
In the college admissions problem, we consider the incentives confronting agents who face the pro... more In the college admissions problem, we consider the incentives confronting agents who face the prospect of being matched by a random stable mechanism. We provide a fairly complete characterization of ordinal equilbria. Namely, every ordinal equilibrium yields a degenerate probability distribution. Furthermore, individual rationality is a necessary and su¢ cient condition for an equilibrium outcome, while stability is guaranteed in ordinal equilibrium where …rms act straightforwardly. Finally, we relate equilibrium behavior in random and in deterministic mechanisms.
Our aim is to characterize perfect equilibria in matching markets. Ordinal preferences require th... more Our aim is to characterize perfect equilibria in matching markets. Ordinal preferences require the use of an ordinal perfect equilibrium concept. We show that, in the game induced by a random stable mechanism, an ordinal perfect equilibrium strategy lists all the acceptable partners. Moreover, when either the firm" optimal or the worker" optimal mechanisms are considered, truth telling is a very prudent form of behavior and is the unique ordinal perfect equilibrium that may emerge. Finally, in the game induced by these ...
We report data from controlled laboratory experiments on two-sided matching markets in which part... more We report data from controlled laboratory experiments on two-sided matching markets in which participants interact in a decentralized way, without having to refer to a central clearinghouse. Our treatments have been designed to evaluate the effect of information, search costs, and binding agreements on the final outcome and also on the individual strategies that lead to it. We find that these features affect the level and pace of market activity as well as the identity of those who receive proposals. While the lack of information ...
Instituto Superior de Economia e Gestão-DE Working papers, 2006
We consider centralized matching markets in which, starting from an arbitrary match¬ ing, firms a... more We consider centralized matching markets in which, starting from an arbitrary match¬ ing, firms are successively chosen in a random fashion and offer their positions to the workers they prefer the most. We propose an algorithm that generalizes some well-known algorithms and explore some of its properties. In particular, different executions of the algorithm may lead to diferent output matchings. We then study incentives in the rev¬ elation game induced by the algorithm. We prove that ordinal equilibria always exist. Furthermore, every ...
We analyze two well-known matching mechanisms-the Gale-Shapley, and the Top Trading Cycles (TTC) ... more We analyze two well-known matching mechanisms-the Gale-Shapley, and the Top Trading Cycles (TTC) mechanisms-in the experimental lab in three different informational settings, and study the role of information in individual decision making. Our results suggest that-in line with the theory-in the college admissions model the Gale-Shapley mechanism outperforms the TTC mechanisms in terms of efficiency and stability, and it is as successful as the TTC mechanism regarding the proportion of truthful preference revelation. In addition, we find that information has an important effect on truthful behavior and stability. Nevertheless, regarding efficiency, the Gale-Shapley mechanism is less sensitive to the amount of information participants hold.
Uploads
Papers by Joana Pais