Delegation to Encourage Communication of Problems

We study a principal's choice to centralize or delegate decisions to an agent when delegation can be used to encourage the agent to communicate potential problems. We find that the principal may choose centralization either to exercise better control over the agent's actions or to provide stronger incentives. Delegation emerges in equilibrium only if the costs of effort to acquire information for both the principal and the agent are sufficiently high. We find that increases in the principal's penalties for an incorrect decision may increase the principal's expected payoff, owing to optimal organizational responses. In addition, catastrophic risk, the risk of incorrectly accepting a defective audit (or product), may be greater under centralization than under delegation. Furthermore, catastrophic risk can be increased by well-intentioned legislative efforts to decrease such risk by, for example, increasing the agent's penalties for failing to take a corrective action, because the organizational structure may change.     

The value of extended delegation in dynamic agency

One of the main advantages of delegation is that specific department level information is used. Its main disadvantage is probably that central management looses direct control over certain actions. In this paper we challenge this widely accepted trade-off. We show that delegation might be favorable even if specific knowledge is completely absent. We consider a firm that lives for two periods. Due to its organizational structure part of the tasks and decision rights is inevitably delegated to a subordinate (agent). The agent performs the tasks assigned to him, tantamount to personal effort, in each of the two periods. Besides this effort the decision to implement a particular project has to be made at the beginning of period two. With regard to the project choice, central management can decide to delegate it to the agent (decentralization). Alternatively it can make it personally (centralization). If the project choice is decentralized it remains unobservable for central management. Along with second period effort it must be motivated via an incentive contract written on period output.We analyze two different contracting regimes: long-term commitment and long-term renegotiation-proof contracts. With full commitment we find that centralization is indeed favorable as compared to delegation if no informational advantage exists. This confirms conventional wisdom. However, the result does not necessarily hold with renegotiation-proof contracts. Renegotiation-proofness may force central management to set too low second-period incentives. Delegation counteracts this effect as it allows central management to implicitly commit to a higher second-period incentive rate. This arises as both, personal effort and the project choice, rather than effort alone need to be motivated. A necessary condition for too low second-period incentives, and thus for delegation to be favorable, is a negative intertemporal correlation of output.     

Performance Incentives within Firms: The Effect of Managerial Responsibility

We show that top management incentives vary by responsibility. For oversight executives, pay‐performance incentives are $1.22 per thousand dollar increase in shareholder wealth higher than for divisional executives. For CEOs, incentives are $5.65 higher than for divisional executives. Incentives for the median top management team are substantial at $32.32. CEOs account for 42 to 58 percent of aggregate team incentives. For divisional executives, the pay–divisional performance sensitivity is positive and increasing in the precision of divisional performance and the pay–firm performance sensitivity is decreasing in the precision of divisional performance. These results support principal–agent models with multiple signals of managerial effort.     

Controlling Investment Decisions: Depreciation- and Capital Charges

This paper examines a multiperiod principal-agent model in which a divisional manager has superior information regarding the profitability of an investment project available to his division. The manager also contributes to the periodic operating cash flows of his division through personally costly effort. We demonstrate that it is optimal for the principal to delegate the investment decision and to base the manager's compensation on the residual income performance measure. Our analysis points to a class of depreciation rules and to a particular capital charge rate which together ensure that a profitable (unprofitable) project makes a positive (negative) contribution to residual income in every period. As a consequence, the compensation parameters for each period can be chosen freely so as to address the moral hazard problems without impacting the manager's investment incentives.     

A note on the interaction effects of non-financial measures of performance

Using principal-agent analyses, the effect of the interactions between two non-financial measures of performance in an agent’s incentive compensation scheme is studied. The agent can allocate effort between “meeting output targets” and “getting output that needs no rework.” The principal trades off (1) a penalty for not meeting output targets, and (2) cost of reworking output that is defective when initially produced. In a compensation mechanism that includes incentives based on measures of output that needs no rework, as well as total output, it is shown that the agent may respond to an increased weight on output that needs no rework by reducing effort allocated towards it. This occurs when the increased weight on the output that needs no rework is accompanied by a sufficiently steep decrease in the weight on total output in the compensation mechanism, leading to a reduction of all effort, and all output. Numerical analyses and implications for the use of multiple measures of performance-based incentives are provided.     

Project design with limited commitment and teams

We study the interaction between a group of agents who exert effort to complete a project and a manager who chooses its objectives. The manager has limited commitment power so that she can commit to the objectives only when the project is sufficiently close to completion. We show that the manager has incentives to extend the project as it progresses. This result has two implications. First, the manager will choose a larger project if she has less commitment power. Second, the manager should delegate the decision rights over the project size to the agents unless she has sufficient commitment power.     

The economics of wild goose chases

Empirical evidence consistently finds that incentive pay is more frequent when authority is delegated to workers than when their superiors hold authority. We provide a model where incentive pay results in the abuse of authority by their superiors, and (under reasonable conditions) implies that (i) incentive pay is higher when an agent holds control rights than when her principal has authority, (ii) effort is less responsive on the margin to incentive pay when the principal holds authority, and (iii) more incentive pay can reduce effort under authority, even on tasks that can be easily measured.     

A theory of delegated search for the best alternative

Searching for the best worker, a reliable supply alternative, or the most profitable investment is frequently delegated to an agent. This article develops a theory of delegated search. We show that the principal’s ability to delegate depends on the agent’s luck, her initial resources, and the contract that governs her search. With moral hazard, the optimal contract is characterized by performance deadlines with bonuses for early completion. If performance cannot be specified, the optimal search is implemented by an option‐to‐buy contract for the principal. If performance is partially specified, the optimal contract is a standard pay‐for‐performance arrangement.     

Relational contracting and endogenous formation of teamwork

We study a relational contracting model with two agents where each agent faces multiple tasks: effort toward the agent's own project and helping effort toward another agent's project. We show that the optimal task structure is either specialization without help or teamwork with a substantial amount of help: teamwork with a small amount of help is never optimal. Specialization with high‐powered incentives can be implemented by relative performance evaluation. However, under teamwork, the evaluation scheme must be substantially different to overcome the multitasking problem. Consequently, a small amount of help is dominated by specialization with high powered incentives.     

Investing in a relationship

A principal can make an investment anticipating a repeated relationship with an agent, but the agent may appropriate the returns through ex post bargaining. I study how this holdup problem and efficiency depend on the contracting environment. When investment returns are observable, informal contracts ex post can be more efficient than formal contracts, as they induce higher investment ex ante: the principal invests not only to generate direct returns, but also to improve relational incentives. Unobservability of returns increases the principal's ability to appropriate the returns but reduces her ability to improve incentives. The optimal information structure depends on bargaining power.     

Contracting for information under imperfect commitment

We study optimal contracting under imperfect commitment in a model with an uninformed principal and an informed agent. The principal can commit to pay the agent for his advice but retains decision‐making authority. Under an optimal contract, the principal should (i) never induce the agent to fully reveal what he knows—even though this is feasible—and (ii) never pay the agent for imprecise information. We compare optimal contracts under imperfect commitment to those under full commitment as well as to delegation schemes. We find that gains from contracting are greatest when the divergence in the preferences of the principal and the agent is moderate.     

Decision Processes, Agency Problems, and Information: An Economic Analysis of Capital Budgeting Procedures

Corporations use a variety of processes to allocate capital.This article studies the benefits and costs of several commonbudget procedures from the perspective of a model with agencyand information problems. Processes that delegate aspects ofthe decision to the agent result in too many projects beingapproved, while processes in which the principal retains theright to reject projects cause the agent to strategically distorthis information about project quality. We show how the choiceof a decision process depends on these two costs, and specificallyon severity of the agency problem, quality of information, andproject risk.     

A Case for Percentage Commission Contracts: The Impact of a “Race” Among Agents

In standard principal-agent problems, the issue at hand is how to align the interests of the agent with those of her principal. A commonly used contract involves the principal paying the agent a percentage of the sale price as commission. With respect to real estate brokerage contracts, it has been argued that percentage commission contracts fail to provide sufficient incentives to the agent. This paper re-evaluates the standard solution to a one seller, one agent agency problem by introducing more than one agent. It is shown that percentage commission contracts can induce first-best effort levels from agents. The result is due to the negative externalities created by the winner-takes-all race among agents. The optimal commission rates in this model are inconsistent, however, with the observed uniformity in commission rates across markets in the USA.     

Incentives for experimenting agents

We examine a repeated interaction between an agent who undertakes experiments and a principal who provides the requisite funding. A dynamic agency cost arises—the more lucrative the agent's stream of rents following a failure, the more costly are current incentives, giving the principal a motivation to reduce the project's continuation value. We characterize the set of recursive Markov equilibria. Efficient equilibria front‐load the agent's effort, inducing maximum experimentation over an initial period, until switching to the worst possible continuation equilibrium. The initial phase concentrates effort near the beginning, when most valuable, whereas the switch attenuates the dynamic agency cost.     

Managerial discretion in the choice among financial reporting alternatives

This paper considers the role of an agent choosing among reporting alternatives when that choice is unobserved by the principal, and the agent's compensation contract is optimal. The agent is allowed to take post-outcome costly actions which lead to more precise reports of actual profits than would be yielded by less precise, but costless conventional translations of outcomes (e.g., GAAP). The extent to which the principal allows the agent this discretion depends upon the improvement in profits as an indicator of the agent's pre-outcome effort when post-outcome actions take place versus the attendant cost of these actions.     

Rewarding improvements: optimal dynamic contracts with subjective evaluation

We study a T‐period contracting problem where performance evaluations are subjective and private. We find that the principal should punish the agent for performing poorly in the future even when the evaluations were good in the past; at the same time, the agent should be given opportunities to make up for poor evaluations in the past with better performance in the future. Optimal incentives are thus asymmetric. Conditional on the same number of good evaluations, an agent whose performance improves over time should be better rewarded than one whose performance deteriorates.     

Contracts offered by bureaucrats

We examine the power of incentives in bureaucracies by studying contracts offered by a bureaucrat to her agent. The bureaucrat operates under a fixed budget, optimally chosen by a funding authority, and she can engage in policy drift, which we define as inversely related to her intrinsic motivation. Interaction between a fixed budget and policy drift results in low‐powered incentives. We discuss how the bureaucrat may benefit from stricter accountability as it leads to larger budgets. Low‐powered incentives remain even in an alternative centralized setting, where the funding authority contracts directly with the agent using the bureaucrat to monitor output.     

Options in Compensation: Promises and Pitfalls

We derive the optimal compensation contract in a principal–agent setting in which outcome is used to provide incentives for both effort and risky investments. To motivate investment, optimal compensation entails rewards for high as well as low outcomes, and it is increasing at the mean outcome to motivate effort. If rewarding low outcomes is infeasible, compensation consisting of stocks and options is a near‐efficient means of overcoming the manager's induced aversion to undertaking risky investments, whereas stock compensation is not. However, stock plus option compensation may induce excessively risky investments, and capping pay can be important in curbing such behavior.     

The form of incentive contracts: agency with moral hazard,risk neutrality,and limited liability

The analysis obtains a complete characterization of the optimal agency contract with moral hazard, risk neutrality, and limited liability. We introduce a “critical ratio” that indicates the returns to providing the agent with incentives for effort in each random state. The form of the contract is debt (a capped bonus) when the critical ratio is increasing (decreasing) in the state. An increasing critical ratio in the state‐space setting corresponds to the hazard rate order for the reduced‐form distribution of output, which we term the “decreasing hazard rate in effort property” (DHREP). The critical ratio also yields insights into agency with adverse selection.     

Specific knowledge and performance measurement

I examine optimal incentives and performance measurement in a model where an agent has specific knowledge (in the sense of Jensen and Meckling) about the consequences of his actions for the principal. Contracts can be based both on “input” measures related to the agent's actions and an “output” measure related to the principal's payoff. Whereas input‐based pay minimizes income risk, only output‐based pay encourages the agent to use his knowledge efficiently. In general, it is optimal to use both kinds of performance measures. The results help to explain some empirical puzzles and lead to several new predictions.