This paper analyzes the role of separation of powers and checks and balances in enhancing political accountability in democracies. It argues that while elections are a key mechanism for disciplining public officials, they are not sufficient on their own. Separation of powers between the executive and legislative branches helps prevent abuse of power, but only when there are appropriate checks and balances. These checks create a conflict of interests between the executive and legislature, requiring both bodies to agree on public policy, thus disciplining each other in the voters' interest. Separation of powers also helps voters elicit information from officials. The paper focuses on two main sources of rents from office: power and information asymmetries. Power refers to the temporary delegation of decision-making authority to officials, which creates opportunities for abuse between elections. Information asymmetries arise when policymakers have better information about policy outcomes than the public, enabling potential abuse. The paper models the interaction between the executive and legislature as an extensive-form game, showing that separation of powers can improve accountability by creating checks and balances. A key result is that separation of powers enhances accountability only when policies require joint agreement between the executive and legislature. This prevents unilateral implementation and reduces the ability of officials to capture rents. The paper also shows that separation of powers allows voters to extract private information from officials, improving their ability to hold them accountable. This is similar to results in mechanism design literature. The paper further highlights the importance of agenda-setting power in separating decision-making processes and aligning the interests of the weaker party with those of voters. The paper compares presidential and parliamentary systems, noting that direct control by voters in presidential systems keeps the executive more accountable. In parliamentary systems, the executive is indirectly controlled through the legislature, which can lead to collusion. The paper argues that separation of powers helps voters control abuse by requiring joint agreement on policies and by creating a "common pool" problem that limits the ability of officials to capture rents. The paper concludes that separation of powers improves accountability and voter utility, but only under appropriate checks and balances. It also highlights the importance of agenda-setting power in designing effective separation of powers mechanisms. The results are robust to different assumptions about voter information and are consistent with the incomplete contract framework.This paper analyzes the role of separation of powers and checks and balances in enhancing political accountability in democracies. It argues that while elections are a key mechanism for disciplining public officials, they are not sufficient on their own. Separation of powers between the executive and legislative branches helps prevent abuse of power, but only when there are appropriate checks and balances. These checks create a conflict of interests between the executive and legislature, requiring both bodies to agree on public policy, thus disciplining each other in the voters' interest. Separation of powers also helps voters elicit information from officials. The paper focuses on two main sources of rents from office: power and information asymmetries. Power refers to the temporary delegation of decision-making authority to officials, which creates opportunities for abuse between elections. Information asymmetries arise when policymakers have better information about policy outcomes than the public, enabling potential abuse. The paper models the interaction between the executive and legislature as an extensive-form game, showing that separation of powers can improve accountability by creating checks and balances. A key result is that separation of powers enhances accountability only when policies require joint agreement between the executive and legislature. This prevents unilateral implementation and reduces the ability of officials to capture rents. The paper also shows that separation of powers allows voters to extract private information from officials, improving their ability to hold them accountable. This is similar to results in mechanism design literature. The paper further highlights the importance of agenda-setting power in separating decision-making processes and aligning the interests of the weaker party with those of voters. The paper compares presidential and parliamentary systems, noting that direct control by voters in presidential systems keeps the executive more accountable. In parliamentary systems, the executive is indirectly controlled through the legislature, which can lead to collusion. The paper argues that separation of powers helps voters control abuse by requiring joint agreement on policies and by creating a "common pool" problem that limits the ability of officials to capture rents. The paper concludes that separation of powers improves accountability and voter utility, but only under appropriate checks and balances. It also highlights the importance of agenda-setting power in designing effective separation of powers mechanisms. The results are robust to different assumptions about voter information and are consistent with the incomplete contract framework.