The paper by Gustavo Manso analyzes the optimal incentive schemes for motivating innovation in a principal-agent framework. It shows that innovation requires a different approach from standard performance-based incentives, which often penalize early failures. Instead, the optimal scheme should tolerate or even reward early failures and reward long-term success. Key elements include long-term compensation, job security, and timely performance feedback. The paper also discusses how managerial compensation can be structured using stock options with long vesting periods, option repricing, golden parachutes, and managerial entrenchment to encourage innovation. The study uses a bandit problem model to represent the tension between exploration (trying new ideas) and exploitation (repeating known strategies). The model considers two periods and two possible outcomes (success or failure) in each period. The agent can choose between shirking, exploiting a known approach, or exploring a novel approach. The paper shows that the optimal contract for innovation differs from standard contracts used to motivate effort or prevent resource tunneling. The paper also examines the role of termination in influencing incentives. Termination can help prevent shirking and encourage exploitation, but its effect on exploration is ambiguous. The paper argues that termination may be optimal for exploitation but not necessarily for exploration. It also highlights the importance of feedback and long-term commitment in motivating innovation. The paper discusses the implications of these findings for corporate governance, executive compensation, bankruptcy laws, and intrapreneurship. It suggests that practices like golden parachutes and managerial entrenchment, often criticized for protecting managers after poor performance, may be part of an optimal incentive scheme that motivates innovation. The paper also notes that regulations restricting these practices may have adverse effects on innovation. The study concludes that innovation requires a different approach from standard performance-based incentives, emphasizing the need for tolerance of early failures, long-term success rewards, and long-term commitment. The paper provides a framework for understanding the incentives for exploration and exploitation in innovation processes and offers insights into how these incentives can be structured in practice.The paper by Gustavo Manso analyzes the optimal incentive schemes for motivating innovation in a principal-agent framework. It shows that innovation requires a different approach from standard performance-based incentives, which often penalize early failures. Instead, the optimal scheme should tolerate or even reward early failures and reward long-term success. Key elements include long-term compensation, job security, and timely performance feedback. The paper also discusses how managerial compensation can be structured using stock options with long vesting periods, option repricing, golden parachutes, and managerial entrenchment to encourage innovation. The study uses a bandit problem model to represent the tension between exploration (trying new ideas) and exploitation (repeating known strategies). The model considers two periods and two possible outcomes (success or failure) in each period. The agent can choose between shirking, exploiting a known approach, or exploring a novel approach. The paper shows that the optimal contract for innovation differs from standard contracts used to motivate effort or prevent resource tunneling. The paper also examines the role of termination in influencing incentives. Termination can help prevent shirking and encourage exploitation, but its effect on exploration is ambiguous. The paper argues that termination may be optimal for exploitation but not necessarily for exploration. It also highlights the importance of feedback and long-term commitment in motivating innovation. The paper discusses the implications of these findings for corporate governance, executive compensation, bankruptcy laws, and intrapreneurship. It suggests that practices like golden parachutes and managerial entrenchment, often criticized for protecting managers after poor performance, may be part of an optimal incentive scheme that motivates innovation. The paper also notes that regulations restricting these practices may have adverse effects on innovation. The study concludes that innovation requires a different approach from standard performance-based incentives, emphasizing the need for tolerance of early failures, long-term success rewards, and long-term commitment. The paper provides a framework for understanding the incentives for exploration and exploitation in innovation processes and offers insights into how these incentives can be structured in practice.