http://repub.eur.nl/res/aut/545/ List of Publications en http://repub.eur.nl/static-eur/img/logo.png http://repub.eur.nl http://repub.eur.nl/res/pub/38972/ 2012-03-16T00:00:00Z Wat me opvalt aan de publieke discussies over wetenschappelijke integriteit en aan de vragen die mij daarover gesteld worden zijn een drietal zaken. Ten eerste de nadruk op data. Ten tweede de aandacht voor promovendi als groep die fraude kan identificeren. En ten derde de nadruk op controle. Volgens mij zijn er twee redenen waarom er zoveel aandacht is voor data: Stapel en het idee dat wetenschappelijk onderzoek gelijk is aan empirisch onderzoek. De aandacht komt dus voort uit een begrijpelijke, maar eenzijdige reactie op een recent schandaal en een misvatting over wat wetenschappelijk onderzoek is. http://repub.eur.nl/res/pub/26791/ 2011-10-01T00:00:00Z This paper analyses the optimal combination of costly and costless messages that a Sender uses in a signaling game if he is able to choose among all equilibrium communication strategies. We provide a complete characterization of the equilibrium that maximizes the Sender's ex ante expected utility in case of uniformly distributed types and quadratic loss functions. First, the Sender often wants to avoid money burning by using the most informative cheap talk communication strategy. Second, if he does burn money, he avoids separation and only re-arranges the existing intervals of the most informative cheap talk equilibrium, possibly adding one extra interval. Money burning takes place in the second interval only. http://repub.eur.nl/res/pub/18544/ 2010-03-01T00:00:00Z We study two aspects of globalization. It allows a decision-maker to go beyond his own local experience and to learn from other decision-makers in addressing common problems. This improves the identification and diffusion of best practices. It also provides extra information to `markets' that evaluate decision-makers: comparisons become possible. We identify conditions under which the globalization of markets helps or hurts (i) the communication among decision-makers about their own experience and (ii) the quality of the decision that is taken next. An important mediating factor is whether decision-making is centralized or decentralized. http://repub.eur.nl/res/pub/15253/ 2008-09-01T00:00:00Z It is often assumed that bad corporate performance means a bad CEO. The task of a board of directors is then simple: dismiss the executive. If it fails to do so, the board is said to be indolent. We take a kinder approach to observed board behavior and point to the problems even well-intended boards would encounter. They face the twin task of disciplining and screening executives. We analyze the nature of the retention contract a board uses to discipline and screen executives. Consistent with empirical observation, we find that executives may become overly active to show their credentials, and that the link between bad performance and dismissal is weak. http://repub.eur.nl/res/pub/28632/ 2008-04-01T00:00:00Z Some committees are made up of experts, persons who care both about the matter at hand and about coming across as able decision makers. We show that such committees would like to conceal disagreement from the public. That is, once the decision has been reached, they show a united front to the outside world. Also, if such committees are required to become transparent, for example, by publishing verbatim transcripts of their meetings, members will organize pre-meetings away from the public eye. A large part of the article is dedicated to a case study of the U.S. Federal Open Market Committee in the United States. It provides suggestive evidence supporting our theory. http://repub.eur.nl/res/pub/11731/ 2008-03-01T00:00:00Z We analyse the implications of endogenizing information collection and reputational concerns for the performance of a sequential decision structure. In this model, two agents decide in a sequence whether to implement a public project. The cost of gathering information is private. We derive two results. First, endogenizing information replaces the herding problem with a free-rider problem. Second, endogenizing information aggravates the distortionary effect of reputational concerns. http://repub.eur.nl/res/pub/11728/ 2007-10-01T00:00:00Z An important role of managers is to motivate subordinates. Monetary rewards have been the focus of economic analysis. Managers also use other means to influence subordinates. If a manager can better assess the subordinate's ability than the subordinate himself, and if ability and effort are complements, the manager finds it hard not to overstate a junior's ability. Talk is cheap. We analyze under what conditions a manager can use organizational practices such as delegation and the selective provision of attention to credibly communicate his assessment. We compare their desirability. Delegation is preferable in case the manager's assessment is fairly accurate; attention is inescapable if it is inaccurate. http://repub.eur.nl/res/pub/10440/ 2007-07-17T00:00:00Z Transparent decision-making processes are widely regarded as a prerequisite for the working of a representative democracy. It facilitates accountability, and citizens may suspect that decisions, if taken behind closed doors, do not promote their interests. Why else the secrecy? We provide a model of committee decision-making that explains the public’s demand for transparency, and committee members’ aversion to it. In line with case study evidence, we show how pressures to become transparent induce committee members to organize pre-meetings away from the public eye. Outcomes of pre-meetings are less determined, more anarchic, than those of formal meetings, but within bounds. We characterize feasible deals that are credible and will be endorsed in the formal meeting. http://repub.eur.nl/res/pub/10159/ 2007-04-11T00:00:00Z Organization differ in the degree to which they differentiate employees by ability. We analyse how the effect of differentiation on employee morale may explain this variation. By comparing employees using ordinary talk, a manager boosts the self-image of some, but hurts that of others. Whether the net effect is positive for the organization depends on the degree of synergy between employees and on the shape of their objective function. An implication for relative performance pay is that it yields a double dividend or constitutes a double-edged sword. http://repub.eur.nl/res/pub/11215/ 2007-01-01T00:00:00Z A committee makes a decision on a project on behalf of "the public." Members of the committee agree on the a priori value of the project, and hold additional private information about its consequences. They are experts who care about the value of the project and about being considered well informed. Before voting on the project, members can exchange their private information simultaneously. We show that reputational concerns make the a priori unconventional decision more attractive and lead committees to show a united front. These results hold irrespective of whether information can be manipulated or not. Also, reputational concerns induce members to manipulate information and vote strategically if their preferences differ considerably from those of the member casting the decisive vote. Our last result is that the optimal voting rule balances the quality of information exchange and the alignment of interests of the decisive voter with those of the public. http://repub.eur.nl/res/pub/11751/ 2007-01-01T00:00:00Z It is often assumed that bad corporate performance means a bad CEO. The task of a board of directors is then simple: dismiss the executive. If it fails to do so, the board is said to be indolent. We take a kinder approach to observed board behaviour and point to the problems even well-intended boards would encounter. They face the twin task of disciplining and screening executives. To perform these tasks directors do not have detailed information about executives’ behaviour, and only infrequently have information about the success or failure of initiated strategies, reorganizations, mergers etc. We analyse the nature of (implicit) retention contracts boards use to discipline and screen executives. Consistent with empirical observation, we …nd that executives may become overly active to show their credentials; that the link between bad performance and dismissal is weak; and that boards occasionally dismiss competent executives.It is often assumed that bad corporate performance means a bad CEO. The task of a board of directors is then simple: dismiss the executive. If it fails to do so, the board is said to be indolent. We take a kinder approach to observed board behaviour and point to the problems even well-intended boards would encounter. They face the twin task of disciplining and screening executives. To perform these tasks directors do not have detailed information about executives’ behaviour, and only infrequently have information about the success or failure of initiated strategies, reorganizations, mergers etc. We analyse the nature of (implicit) retention contracts boards use to discipline and screen executives. Consistent with empirical observation, we …nd that executives may become overly active to show their credentials; that the link between bad performance and dismissal is weak; and that boards occasionally dismiss competent executives. http://repub.eur.nl/res/pub/11755/ 2007-01-01T00:00:00Z Some committees are made up of experts, persons who care both about the matter at hand and about coming across as able decision- makers. We derive two propositions about the way members of such committees interact with the outside world. First, they would like to conceal disagreement from the public. That is, once the decision has been reached, they show a united front to the outside world. Second, if such committees are required to become transparent, e.g., by publishing verbatim transcripts of their meetings, members will organize pre-meetings away from the public eye. Large part of the paper is dedicated to a case study of the U.S. Federal Open Market Committee in the United States. It provides suggestive evidence supporting the two propositions. http://repub.eur.nl/res/pub/7834/ 2006-06-19T00:00:00Z Boards of directors face the twin task of disciplining and screening executives. To perform these tasks directors do not have detailed information about executives' behaviour, and only infrequently have information about the success or failure of initiated strategies, reorganizations, mergers etc. We analyse the nature of (implicit) retention contracts boards use to discipline and screen executives. Consistent with empirical observation, we find that executives may become overly active to show their credentials; that the link between bad performance and dismissal is weak; and that boards occasionally dismiss competent executives. http://repub.eur.nl/res/pub/7814/ 2006-05-29T00:00:00Z Some committees are made up of experts, persons interested in both the (subject) matter at hand and in coming across as able decision-makers. Such committees would like to conceal disagreement from the public. We present a theory that describes the reaction of experts to the requirement to publish verbatim transcripts of their meetings: the emergence of an informal ‘premeeting’; the move of the real debate from the formal meeting to the premeeting; and the drop in disagreement in the formal meeting. We analyse what the effect is on accountability and quality of decision-making. Finally, we present evidence suggesting that our model describes the way members of the Federal Open Market Committee in the United States responded to the publication of verbatim transcripts of their meetings. http://repub.eur.nl/res/pub/6592/ 2005-03-10T00:00:00Z We consider a committee that makes a decision on a project on behalf of 'the public'. Members of the committee agree on the a priori value of the project, and hold additional private information about its consequences. They are experts who care both about the value of the project and about being considered well informed. Before voting on the project, members can exchange their private information simultaneously (so no herding). We show that reputational concerns make the a priori unconventional decision more attractive and lead committees to show a united front. These results hold irrespective of whether information can be manipulated or not. Next, we show that reputational concerns induce members to manipulate information and vote strategically if their preferences differ considerably from those of the member casting the decisive vote. Our last result is that the optimal voting rule balances the quality of information exchange and the alignment of interests of the decisive voter with those of the public. http://repub.eur.nl/res/pub/11734/ 2004-12-01T00:00:00Z Can reputational concerns do any good? Economists have shown how they lead agents to ignore valuable information, to herd, and to become overly risk averse. We explore how they may be a social blessing. An agent may exert effort to become informed about the uncertain benefits of a project. A smart agent's efforts make him better informed; a dumb agent's efforts are to no avail. If an agent does not know his type, reputational concerns are socially beneficial. If an agent knows his type, reputational concerns may be socially beneficial. A dumb agent takes inefficient, unconventional decisions to mimic a smart agent. The latter exerts more effort in order not to be mistaken for a dumb. This holds whether or not project rejection is a safe haven for the dumb. http://repub.eur.nl/res/pub/6650/ 2004-04-08T00:00:00Z This article analyzes under which conditions a manager can motivate a junior worker by verbal communication, and explains why communication is often tied up with organizational choices as job enlargement and collaboration. Our model has two important features. First, the manager has more information about a junior's ability than the junior himself. Second, the junior's effort and ability are complements. We show that the manager has an incentive to exaggerate the junior's ability. We discuss two ways in which the manager can make credible statements about the junior's ability. First, the senior can delegate a task to the junior for which it is important that the junior has a correct perception of his ability. Information is shared through a costless signal. Second, the senior can spend more time on a junior she perceives as able than on a junior she perceives as less able. Information is then shared through a costly signal. http://repub.eur.nl/res/pub/6656/ 2004-02-24T00:00:00Z Organizational theory has construed complexity as an objective characteristic of either the structure or the behaviour of an organization. We argue that to further our understanding it should be understood in terms of human cognition of a structure or behavior. This cognitive twist is illustrated using two theoretical approaches, whose relationship is discussed. http://repub.eur.nl/res/pub/6701/ 2003-08-22T00:00:00Z Democracies delegate substantial decision power to politicians. Using a model in which an incumbent can design, examine and implement public policies, we show that examination takes place in spite of, rather than thanks to, elections. Elections are needed as a carrot and a stick to motivate politicians, yet politicians who are overly interested in re-election shy away from policy examination. Our analysis sheds light on the distance created in mature democracies between the political process and the production of policy relevant information; on the role played by probing into candidates' past; and on the possibility of crowding out desirable political behaviour by increasing the value of holding office. http://repub.eur.nl/res/pub/6723/ 2003-03-14T00:00:00Z In models of sequential decision making herd behaviour occurs if the signals smart (dumb) agents receive are (un)correlated and if agents have reputational concerns. We show that introducing costly effort to become informed about project payoffs (i) eliminates herd behaviour and (ii) shifts attention from the incentives for agent 2 to herd to agent 1 to exert effort. While the first agent anticipates the second agent's behaviour, his influence is only partial. The unique equilibrium either implies delegation to the first agent; to the second agent; or has both agents participating. http://repub.eur.nl/res/pub/6792/ 2003-03-13T00:00:00Z Can vanity do any good? It may seem obvious to answer this question in the negative, as economists have shown how reputational concerns lead agents e.g. to ignore valuable information, to herd, and to become overly risk averse. We explore how proud agents may be a social blessing. An agent may exert effort to become informed about the uncertain benefits of aproject. A smart agent's efforts make him better informed; a dumb agent's efforts are to no avail. If an agent does not know his type, pride is socially beneficial. If an agent knows his type, a dumb agent takes inefficient, unconventional decisions to mimick a smart agent. The latter exerts more effort in order not to be mistaken for a dumb. This holds whether or not project rejection is a save haven for the dumb. http://repub.eur.nl/res/pub/6813/ 2002-05-28T00:00:00Z This paper analyses the relationship between organizational complexity ( the degree of detail of information necessary to correctly assign agents to positions), robustness (the relative loss of performance due to mis-allocated agents), and performance. More complex structures are not necessarily more profitable, but are less robust. One of the least complex structures always performs worst. Superior organizational performance may vanish completely due to mis-allocated agents. Organizational performance can be enhanced through training agents; re-assigning them when adequate knowledge about their characteristics is obtained through monitoring; simplifying the organizational structure; and influencing the environment. The trade-offs involved are analysed. http://repub.eur.nl/res/pub/11758/ 2002-01-01T00:00:00Z A new measure of the complexity of optimal economic decisions is introduced. It is based on the level of detail of information (no information; ordinal; and cardinal information) that is required to establish optimality. A detailed example involving sequential group decision making is provided. It is shown that the type of links between successive agents determines the degree of complexity. The measure is also illustrated in the realm of matching problems. http://repub.eur.nl/res/pub/6830/ 2001-12-10T00:00:00Z Collective decision procedures should balance the incentives they provide to acquire information and their capacity to aggregate private information. In a decision problem in which a project can be accepted or rejected once information about its quality has been acquired or not, we compare the performance of a delegation structure with that of two voting procedures. Delegation makes one's acceptance decision pivotal by definition. The decisiveness of one's vote in a voting procedure depends on the other agent's vote. This in turn determines the decision to acquire information. In the debate about a rational choice foundation of Condorcet's Jury Theorem, the distribution of information was left exogenous. Mixed (acceptance) strategies were required to validate the Theorem. Endogenizing information acquisition as we do reveals mixed (acceptance) strategies to be detrimental for welfare as they lead to indifference between buying and not buying information. http://repub.eur.nl/res/pub/11741/ 2000-06-01T00:00:00Z The effect of organizational communication structure on performance is studied using a project selection framework in which heterogeneous, rational agents can reject or accept projects. A hierarchy performs better than a polyarchy in tough environments, whereas the reverse holds in friendly environments. Hierarchies can be too strict from a social welfare point of view. Indeed, if the value of profitable but rejected projects is taken into account, a single agent can perform better than a hierarchy. http://repub.eur.nl/res/pub/11743/ 1999-01-01T00:00:00Z Abstract. Multi-product firms are modelled as locally interacting entities that gather information on the profitability of product combinations in an environment defined in terms of their currently supplied markets. They learn from their own past play. Local information gathering may slow down convergence and may prohibit profit rates from becoming equal. Cycles show parts of the economy that are in rest, while others remain in a state of flux. The first two results stem from the endogeneity of the interaction structure, while the third follows from the interplay of learning and information gathering. This paper was written while I was a Ph.D student at the European University Institute, Florence. I would like to thank Andrzej Baniak, Etienne Billette de Villemeur, Aedin Doris, Giovanni Dosi, Iouri Kaniovski, Spyros Vassilakis, Bruno Versaevel and two anonymous referees of this journal for their comments on previous drafts of this paper. I have greatly benefitted from the pleasant working atmosphere at IIASA, Laxenburg, where I stayed during the YSSP96. Financial support from the Fundatie van de Vrijvrouwe van Renswoude te's-Gravenhage, the Stichting Dr Hendrik Muller's Vaderlandsch Fonds and the Stichting "Noorthey" is gratefully acknowledged.