This dissertation deals with the design of optimal incentive schemes in the presence of collaborative assignments. The later are considered as either intentional, in the form of teamwork, or unintended, in the form of knowledge spillovers.In the first part, the present study investigates optimal incentive contracts for R&D divisions in the presence of spillovers on R&D-outputs and downstream market competition. It examines the effect of spillovers on cost-reducing R&D incentives and addresses the question of whether the standard result that profits are higher under full information applies in this setting. Spillovers necessitate the use of relative performance evaluation schemes even in the absence of correlation between the random factors in the R&D process. Such compensation schemes can effectively filter out the knowledge externality from employees reward by placing a negative weight (short position) on rival-firm performance. These contracts introduce competition between agents and promote efficiency in designing incentives for risk-averse employees.The analysis is performed in a framework where the feedback mechanism in the R&D process is positive and self-reinforced as happens mainly in science-based industries and technological parks. Empirical evidence reveals that technical advance in knowledge-driven industries go hand in hand with feedback mechanisms which display increasing returns to spillovers. Given such technological interactions, product market rivals exert higher efforts as spillovers increase. In particular, each rival seeks to realize a slightly lower marginal costs from its competitor so as to extent its businesses vis-à-vis the market. Given that all have the same incentives and research has a greater impact on innovators’ profits, firms enter into a rat race. In turn, intensified spillovers speed up the R&D activity. The agency problem gets also increased implying that there are far more distortions in incentives.This study argues that, in highly competitive industries, firms, that are involved in this prisoners dilemma-type of race and are driven by business-stealing incentives, may enjoy higher profits under asymmetric information. It is so because, in a full-information world, rivals exert a fairly high level of R&D-efforts burning up their profits. Higher risk and uncertainty about individual production induce the employers to moderate such incentives and their appetite for innovation. Cost-savings by exerting lower efforts due to risk-sharing may yield higher profits for rivals. In a sense, principals are better-o¤ as the incentive-insurance trade-o¤ is shifted towards the later. This is a simple reason why one might derive a positive relationship between insurance and net profits which counters the prediction of the principal-agent theory. This result indicates firms’ incentives to engage in collusive- like behavior. Delegating the R&D decisions under asymmetric information and thus, committing themselves ex-ante to lower effort can work as a colluding-like device that affects firms responses in the proceeding stages. As a result, firms earn higher profits than under the full information benchmark.For a more complete discussion, this study considers centralized cases where spillovers are internalized. It examines the incentives of individual firms to form a strategic alliance in R&D by agreeing upon collusive wage agreements and the social R&D incentives of the government. This work anticipates the government’s intention to intervene in the R&D market in order to affect competition in the product market. Social R&D incentives induce individual firms to innovate more implying that, taking the decisions unilaterally, there are unexploited productive opportunities due to misallocation of resources. More notably, if the R&D process is not too costly, the cost of misallocation that arises in the presence of the moral hazard problem exceeds the agency cost R&D competitors incur. The distortion in private R&D incentives seems to be less severe than the social losses due to allocative inefficiency. Policy implications are discussed.In the second part (Chapter 5), this study extends the career concerns literature by investigating the incentives to built up reputation when an employee’s task outcome depends on her own ability and effort as well as on the quality of the team she belongs to - i.e. the ability and effort of her fellow team members. It addresses the question of how the learning process about a worker’s ability and the intensity of career concerns depend on the degree of teamwork interactions. It also examines how effectively explicit contracts can be used in the presence of teamwork and incentives to help. This study argues that implicit incentives to work and help arise. In this setting, by exerting effort and providing support, an employee is better able to manipulate the market’s assessment about her own ability. Her actions can influence her own and colleague’s outputs and thus, there are more tools available she can use in order to bias the learning process in her favor. Moreover, intensified dependence among team members increases the returns of supplying labor and thus, the intensity of career concerns. If explicit contracts are provided, this study argues that, reputation and sabotage- like incentives arise. The employer’s benefits by using disaggregate performance measures are also demonstrated.