Organizational Economics Working Group

November 20-21, 2009
Robert S. Gibbons, Organizer

Claudine Desrieux, University of Paris 2 Pantheon; Eshien Chong, University of Paris XI; and Stephane Saussier, IAE - University of Paris I Sorbonne
Putting All One's Eggs in One Basket: Relational Contracts and the Provision of Local Public Services

The contracting out of local public services has kept increasing over the past decades. Desrieux, Chong, and Saussier observe that local governments regularly choose the same private operator for their different services, that is. they "bundle" and delegate them to a single firm. The researchers develop a model of relational contracts that shows how this strategy may lead to better performance at lower cost for the public authority. They test and corroborate the proposition using an original database concerning the contractual choices made by 5000 French local public authorities in 1998, 2001 and 2004.


David McAdams, Duke University
Performance and Turnover in a Stochastic Partnership

McAdams characterizes the welfare-maximizing equilibrium performance and duration of stochastic partnerships, in an economy in which each period partners will choose costly observable efforts, voluntary wages, and whether to leave the relationship to be re-matched. In this economy individuals' lives tend to transition between a few qualitatively distinct phases: "dating" at birth or between relationships; "honeymoon";"hard times"; "good times"; and "golden years", from which partners are parted only by death. Given an exogenous stochastic process, higher states are associated with higher stage-game and continuation payoffs, as well as longer-lasting relationships.


Ricard Gil, UC, Santa Cruz; and Justin Marion, University of California Santa Cruz
The Role of Repeated Interactions, Self-Enforcing Agreements and Relational [Sub]Contracting: Evidence for California Highway Procurement Auctions

Gil and Marion empirically examine the impact of relationships between contractors and subcontractors on firm pricing and entry decisions in the California highway procurement market. They use data from auctions conducted by the California Department of Transportation. In this market, relationships are valuable if they mitigate potential hold-up problems and incentives for ex post renegotiation attributable to contractual incompleteness. An important characteristic of informal contracts are that they must be self-enforcing, so that the value of relationships between firms and suppliers depends on the extent of possibilities for future interaction. The atuhors construct measures of the stock of contractors' prior interactions with relevant subcontractors. They find that a larger stock of relationships leads to lower bids and a greater likelihood of entry. Importantly, this relationship does not hold in periods of time and areas with little future contract volume, suggesting that the self-enforcement mechanism is crucial in providing value for informal contracts.


Hideshi Itoh, Hitotsubashi University; and Hodaka Morita, University of New South Wales
Formal Contracts, Relational Contracts, and the Holdup Problem

Can formal contracts help resolve the holdup problem? Itoh, and Morita address this question by studying repeated transactions between a seller and a buyer in which the seller can make relation-specific investments in each period. They demonstrate that a simple fixed-price contract based on product delivery can help to mitigate the holdup problem even when relation-specific investment is purely cooperative. In particular, there is a range of parameter values in which a higher investment can be implemented only if a formal fixed-price contract is written and combined with an informal agreement on additional payments contingent upon investments. Furthermore, they show that under an additional natural assumption, focusing their attention on fixed-price contracts as a form of formal contracts is without loss of generality. The key driving force of their result is a possibility that relation-specific investment decreases the value of no-trade surplus.


David Cooper, Florida State University; and Kai-Uwe Kuhn, University of Michigan
Communication, Renegotiation, and the Scope for Collusion

Cooper and Kuhn use experiments to analyze what type of communication is most effective in achieving cooperation in a simple collusion game. Consistent with the existing literature on communication and collusion, even minimal communication leads to a short-run increase in collusion. However, in a limited message-space treatment where subjects cannot communicate contingent strategies, this initial burst of collusion rapidly collapses. When unlimited pre-game communication is allowed via a chat window, an initial decline in collusion is reversed over time. Content analysis is used to identify multiple channels by which communication improves collusion in this setting. Explicit threats to punish cheating prove to be by far the most important factor to successfully establish collusion, consistent with the existing theory of collusion. However, collusion is even more likely when the authors allow for renegotiation, contrary to standard theories of renegotiation. What appears critical for the success of collusion with renegotiation is that cheaters are often admonished in strong terms. Allowing renegotiation therefore appears to increase collusion by allowing for an inexpensive and highly effective form of punishment.


Giacomo Calzolari, Universita di Bologna; and Giancarlo Spagnolo, University of Rome, Tor Vergata
Relational Contracts and Competitive Screening

Calzolari and Spagnolo study the tension between competitive screening and contract enforcement where a principal trades repeatedly with one among several agents, moral hazard and adverse selection coexist, and non-contractible dimensions are governed by relational contracting. They simultaneously characterize optimal relational contracts and competitive screening policies which are interdependent. When non-contractible dimensions are important, the principal optimally restricts competitive screening to a subset of "loyal" agents, giving up performance bonuses and, when such dimensions are crucial, negotiates an indefinitely renewable contract with one agent. To enhance enforcement, explicit contract duration is also reduced. However, these policies facilitate collusion among agents, which induces an additional trade-off between reputational forces and collusion. When non-contractible dimensions are very important, this last trade-off may disappear, as collusion allows more efficient enforcement of better performance.

Pedro Dal Bo, Brown University and NBER; and Guillaume Frechette, New York University
The Evolution of Cooperation in Infinitely Repeated Games: Experimental Evidence

An usual criticism of the theory of infinitely repeated games is that it does not provide sharp predictions because there may be a multiplicity of equilibria. To address this issue, Dal Bo and Frechette present experimental evidence on the evolution of cooperation in infinitely repeated prisoners' dilemma games as subjects gain experience. They find that cooperation decreases with experience when it cannot be supported as an equilibrium outcome. More interestingly, the converse is not necessarily true: cooperation does not always increase with experience when it can be supported as an equilibrium outcome. Nor is a more stringent condition, risk dominance, sufficient for cooperation to arise. However, subjects do learn to cooperate when the payoff to cooperation and the importance of the future is high enough. These results have important implications for the theory of infinitely repeated games. While the authors show that cooperation may prevail in infinitely repeated games, the conditions under which this occurs are more stringent than the sub-game perfect conditions usually considered.


Simon Board, UC, Los Angeles
Relational Contracts and the Value of Loyalty

Board examines the tradeoff between maintaining an ongoing relationship and exploiting the gains from trade. He supposes that a firm (principal) repeatedly chooses among a set of potential trading partners (agents) under the threat of holdup. The possibility of ex{post opportunism allows agents to collect rents, which act like a fixed cost of initiating a new relationship. The principal responds by dividing agents into "insiders", with whom she has previously traded, and "outsiders", with whom she has never traded. The principal then uses insiders efficiently, while being biased against outsiders. This contract is self-enforcing if the principal is suffciently patient. It can also be implemented by an "employment contract" that is robust to asymmetric information, thereby providing a theory of the firm, where both firm size and growth are endogenous.


Bjorn Bartling, University of Zurich; Ernst Fehr, University of Zurich; and Klaus Schmidt, University of Munich
Screening, Competition and Job Design

In recent decades many firms have offered more discretion to their employees, which can increase the productivity of effort but also provides more opportunities for shirking. Bartling, Fehr, and Schmidt show that complementarities between high effort discretion, rent-sharing, screening opportunities, and competition may be important driving forces behind these new forms of work organization. In particular, the y show the endogenous emergence of two fundamentally distinct types of jobs. Employers either implement a control strategy, consisting of limited effort discretion, low wages, low effort requests, and little or no rent-sharing, or they implement a trust strategy, which stipulates high effort discretion, high wages, high effort requests, and substantial rent-sharing. If employers cannot screen employees, then the control strategy prevails. The possibility of screening causes a substantial increase in the prevalence of the trust strategy. The introduction of competition further fosters the trust strategy and induces a substantial increase in welfare such that both employers and employees are better off.


Yuk-fai Fong, Northwestern University
Relational Contracts, Limited Liability, and Employment Dynamics

Fong and Li study a model of relational contracts of imperfect public monitoring when the agent has limited liability. The optimal relational contract provides joint and definitive implications on worker's job security, average earning, and the sensitivity of pay to performance over time as the employment relationship progresses. In addition, the authors analyze how the employment dynamics change with respect to the surplus in the relationship and firm's ability to commit to long-term long contract. Their results shed light on empirical findings that relate employment dynamics to firm characteristics.


David Miller, UC, San Diego; and Joel Watson, UC, San Diego
A Theory of Disagreement in Repeated Games with Renegotiation

Miller and Watson analyze repeated games with transferable utility, in which the players may negotiate cooperatively over their continuation strategies at the beginning of each period. In contrast to the renegotiation proofness literature, their model gives an explicit account of whether the players have reached an agreement in a given period, so there are feasible paths of play along which the players disagree. On a disagreement path, play may be jointly suboptimal. In a contractual equilibrium, the players cooperatively negotiate to play the continuation game optimally, splitting the surplus (according to fixed bargaining weights) relative to what they would have played under disagreement. Contractual equilibrium outcomes also arise in a class of models with noncooperative bargaining, under several assumptions on the endogenous meaning of cheap talk message(refining the set of subgame perfect equilibria). Contractual equilibria exist for all discount factors, and all such equilibria attain the same aggregate utility. This paper provides necessary and sufficient conditions for patient players to attain efficiency, as well as simple sufficient conditions. The allocation of bargaining power can dramatically affect aggregate utility. The theory here extends naturally to games with imperfect public monitoring.