Organizational Economics Meeting
May 18-19, 2012
Robert S. Gibbons of MIT, Organizer
Canice Prendergast, University of Chicago
The Economics of Wild Goose Chases
Authority, where employees are told what to do by their superiors, is a dominant feature of firms. There is consistent evidence that firms that rely on supervisory authority are less likely to provide formal incentives to their employees than firms in which employees control their own activities. Prendergast addresses the provision of incentives when such authority is present, in cases where authority derives from superiors knowing more about the worker's productivity than the workers do themselves. Workers can carry out two kinds of tasks: those where output is contractible and those where it is not. Prendergast shows that authority can eliminate incentives, and will always do so with enough task assignment options in the baseline model. Furthermore, he shows that when effort is feasible, authority is complementary with incentives when incentive provision is inexpensive, but harms incentive provision when incentives are difficult to provide. He also shows how these authority issues affect intrinsic motivation and bureaucratic allocation in ways that share the theme of authority being beneficial only when contracting on performance is relatively easy.
Eric Van den Steen, Harvard University
A Theory of Explicitly Formulated Strategy
When a CEO tries to formulate 'a strategy', what is she looking for? What exactly is 'a strategy', why does it matter, and what are its properties? Van den Steen defines an explicitly formulated 'strategy' as the 'smallest set of choices and decisions sufficient to guide all other choices and decisions,' which formally captures the idea of strategy as a plan boiled down to its most essential choices. He shows that this definition coincides with the equilibrium outcome of a game where a person can – at a cost – look ahead, investigate, and announce a set of (intended or actual) choices to the rest of the organization. Strategy is also – in some precise sense – the smallest set of decisions that needs to be decided centrally to ensure that all decisions are consistent (by giving a clear direction). Van den Steen analyzes what characteristics make a decision 'strategic' and when and how having a strategy creates value, including when a strategy 'bet' can create value. He shows how understanding the structure of strategy may enable a strategist to develop the optimal strategy without a comprehensive optimization, and derives some broader organizational implications.
Marina Halac, Navin Kartik, and Qingmin Liu, Columbia University
Optimal Contracts for Experimentation
Halac, Kartik, and Liu study dynamic contracts for experimentation in a principal-agent setting with adverse selection (pre-contractual hidden information), dynamic moral hazard, and learning. Under full commitment, an optimal (that is profit-maximizing) menu of contracts generally induces a low-ability agent to terminate experimentation inefficiently early. There is no distortion in the stopping time for a high-ability agent. The structure of optimal contracts is influenced by a variety of considerations, including dynamic agency costs and post-contractual hidden information. The authors derive two explicit menus of contracts that can be used to implement the optimal solution: "bonus contracts" and "clawback contracts". Both feature history-contingent dynamic streams of transfers.
Heski Bar-Isaac and Joyee Deb, New York University
Reputation for a Servant of Two Masters
Classic models of reputation consider an agent taking costly actions to affect a single, homogeneous audience's beliefs about his ability, preferences, or other characteristic. However, in many economic settings, agents must maintain a reputation with multiple parties with diverse interests. Bar-Isaac and Deb study reputation incentives for an agent who faces two audiences with opposed preferences. They ask if the existence of multiple audiences per se changes reputation incentives. Further, should the agent deal with the different audiences commonly or separately? Their analysis yields some new qualitative insights. Specifically, the presence of heterogeneous audiences is more likely to lead the agent towards "pooling" equilibria in which he takes an intermediate compromise action. Instead, dealing with only one audience leads the agent to cater towards that audience's preferences, giving rise to a "separating" outcome or pooling on some extreme action. Reputation acts as an informal contract that enforces desirable behavior through future continuation payoffs rather than explicit contractual terms. This analysis highlights the fact that the presence of multiple heterogeneous audiences can, naturally, lead these rewards to be non-monotonic in an agent's reputation. The authors show different ways that this non-monotonicity arises. In an infinite horizon setting, it can emerge through endogenous interactions between the audiences, through equilibrium expectations of the agent's choice of action. It can also arise, perhaps more trivially, through direct payoff interactions. The analysis has both positive and normative implications and is relevant for numerous applications that we briefly describe.
Rajkamal Iyer, MIT, and Antoinette Schoar, MIT and NBER
Ex Post (In)efficient Negotiation and the Breakdown of Trade
One of the central assumptions of incomplete contract theories is that the contract parties will engage in ex post efficient renegotiation in the case of unforeseen shocks. However, if parties to the trade are concerned about reputation or norms against price gauging, then efficient renegotiation might break down. To test these ideas, Iyer and Schoar conduct a field experiment among tailoring stores in Chennai, a city in Southern India. They send trained auditors acting as customers to the stores and have them place an order to be picked up in several days. The "customer" then returns the next day to ask for urgent completion of the order within one day, giving the tailors an opportunity to renegotiate the contract. The authors find that tailors overall do not use the increase in their bargaining power to ask for a higher price. In 44 percent of the cases, they fill the urgent order but do not ask for any increased payment. In the remaining 56 percent of the cases, trade breaks down -- that is, tailors refuse to fill the order. In these cases, they instead offer customers a chance to take back the material and find another tailor, but they never ask for additional compensation. When offered a higher price, the majority of tailors were happy to fill the urgent order however. This suggests that without the customer offering the mark up, tailors forego an efficient renegotiation option and trade breaks down. When comparing customers from out of state versus local ones, there is still a strong resistance to ask for higher prices in the case of an urgent order. The authors can also rule out that the equilibrium behavior of a buyer in this market does not require voluntarily offering a premium for urgent services. The average customers do not offer a premium, the authors find, and in fact there is breakdown of trade. They conjecture that either norms or reputational concerns prevent tailors from proactively suggesting a mark up for urgent delivery. These results put into questions the notion that ex post renegotiation can be achievedeasily.
Raghuram Rajan, University of Chicago and NBER
The Corporation in Finance
Rajan points out that to produce significant net present value, an entrepreneur has to differentiate her enterprise from the ordinary. To take collaborators with her, she needs to have substantial ownership, and thus financing. But it is hard to raise finance against differentiated assets. So an entrepreneur has to commit to undertake a second transformation -- standardization -- that will make the human capital in the firm, including her own, replaceable. Then outside financiers can obtain control rights that will allow them to be repaid. A vibrant stock market helps the entrepreneur commit to these two transformations. Rajan concludes that the nature of firms and financing are intimately linked.
Guido Friebel, University of Toulouse, and Michael Raith, University of Rochester
Silo or Market: Internal Labor Markets in Multi-divisional Firms
Large corporations are increasingly concerned with assigning their top performers to the most productive positions. In particular, they try to facilitate mobility not only within divisional job ladders ("silos") but across divisions as well. Friebel and Raith argue that middle managers play a key role in how internal labor markets operate: they are responsible for mentoring people, and want to hold on to their best performers and get rid of the weakest ones. The authors develop a model of a firm with two divisions in which managers invest in training their workers, who then may be eligible for promotion to manager of a different division. They show that even in the absence of information problems, a market with cross-divisional transfers leads to a more efficient allocation of employees, but comes at the cost of weaker incentives for managers to train their workers. This negative incentive effect is worse when managers also have private information about their workers' ability, and wage contracts must be designed to induce managers to communicate truthfully about their workers. Depending on the parameters, either a "silo" solution or a "market" can be optimal for the firm. The authors embed their firm model within a model of oligopolistic product market competition, and show that greater product market competition helps explain why the trend for firms to establish cross-divisional internal labor markets is a relatively recent phenomenon.
Birger Wernerfelt, MIT
The Equilibrium Organization of Labor
Wernerfelt looks for the equilibrium mix of trading institutions when manufacturers need sequences of labor services. The environment has two critical features: 1) Multilateral matching allows gains from specialization, but players incur specific set-up costs each time they are matched with a new trading partner. 2) Bilateral relationships economize on set-up costs, but are burdened by bargaining costs. Under weak conditions, four mechanisms weakly dominate all others: markets; employment with negotiated wage;, employment with market wages; and bilateral sequential contracting. In labor market equilibrium, markets are used for services that take longer to perform, are in less demand, require fewer partner-specific investments, and have larger cost differences between sellers. The most efficient sellers work as specialist employees; those of medium efficiency sell their services as professionals in markets; and the least efficient become employees. Large firms hire specialists; medium sized firms use employees or the market; and small firms use the market exclusively. Lowered trade barriers cause a shift towards market governance.
Sylvain Chassang, Princeton University, and Christian Zehnder, University of Lausanne
A Theory of Informal Justice
Chassang and Zehnder develop a positive theory of informal justice based on social preferences. They consider a principal and an agent whose relationship is mediated by a third-party arbitrator that decides on ex post transfers between the two parties. Importantly, formal contracts are not available and the arbitrator chooses transfers according to her own social preferences and sense of fairness. The authors characterize the implicit incentive schemes induced by the arbitrator's preferences and interpret them as systems of informal justice. They show how qualitative properties of informal justice, such as the role of punitive damages, or the importance of intentions, vary with the relative weights that the arbitrator puts on ex post (or allocative) fairness, versus ex ante (or procedu ral) fairness. Finally, they show that these implicit incentive schemes satisfy surprisingly robust efficiency properties, which suggests an interpretation of social preferences as heuristics for informal contracting.
Supreet Kaur, Harvard University, and Michael Kremer and Sendhil Mullainathan, Harvard University and NBER
Self-Control at Work: Evidence from a Field Experiment
Self-control problems change the logic of agency theory: workers not only fail to work as hard as employers would like, they also fail to work as hard as they themselves would like. In response, firms can use incentive pay to affect the self-control problem, not just moral hazard. Kaur, Kremer, and Mullainathan describe the results of a year-long field experiment on data entry workers that was designed to test the empirical importance of these ideas. First, they find that workers will choose dominated contracts - which pay less for every output level but have a steeper slope - in order to motivate themselves. Second, their effort increases significantly as the (randomly assigned) payday gets closer. Third, these two effects are linked: the demand for dominated contracts (and their benefits) is concentrated amongst those with the highest payday effects. Finally, as workers gain experience, they appear to learn about their self control problems: the correlation between the payday effect and the demand for the dominated contract grows with experience. Both payday and contract effects are quantitatively large when benchmarked against the impact of a change in the slope of incentives or of a year of education. These results together suggest that self-control, in this context at least, meaningfully alters the firm's contracting problem.
Valerie Smeets and Frederic Warzynski, Aarhus University, and Michael Waldman, Cornell University
Performance, Career Dynamics, and Span of Control
An extensive theoretical literature is based on what is called the scale-of-operations effect -- that is, the idea that the return to managerial ability is higher the more resources the manager influences with his or her decisions. This idea leads to various testable predictions including that higher ability managers should supervise more subordinates or have a larger span of control. Although some of this theory's predictions have been empirically investigated, there has been little research on the predictions concerning span of control. Smeets, Warzynski, and Waldman extend the theoretical literature on the scale-of-operations effect to allow firms' beliefs concerning a manager's ability to evolve over the manager's career, focusing largely on the determinants of span of control. They then empirically investigate testable predictions using a unique single-firm dataset that contains detailed information about the reporting relationships at the firm. Their investigation strongly supports the model's predictions concerning wages, wage changes, and the probability of promotion, as well as predictions concerning span of control, including those derived from the learning component of the model. Overall, their investigation supports the notion that both the scale-of-operations effect and additionally learning are important determinants of the internal organization of firms.
Camelia M. Kuhnen, Northwestern University, and Paul Oyer, Stanford University and NBER
Exploration for Human Capital: Theory and Evidence from the MBA Labor Market
Drawing on insights from corporate finance and personnel economics, Kuhnen and Oyer show that firms consider potential employees using a real options approach, much as they do when making other types of capital investment decisions. Firms' hiring decisions are influenced by the uncertainty in workers' productivity, competition in the labor market, adjustment costs, and re-deployability concerns. Firms value probationary employment arrangements, which provide the option to learn about the productivity of potential hires before permanent investment occurs. Higher uncertainty and adjustment costs hinder permanent investment and increase the value of the option to learn. Greater competition for workers speeds up firm investment and increases the value of probationary employment. Higher worker re-deployability leads to more investment, if firms face low competition. After setting out these results in theory, the authors test and confirm the predictions empirically using a novel dataset with detailed recruiting information from the labor market for MBA graduates.