This paper analyzes the relationship between democracy and redistribution, focusing on how tax revenues are allocated to local public goods in different tax structures. In a probabilistic voting model with three jurisdictions where residents have different income levels, it is shown that it is always optimal to distribute tax revenues as public goods to only the residents of the richest and median income jurisdictions. The study compares the overall welfare of citizens under a one-bracket tax structure, where the poor contribute to taxes but do not receive public goods, with a progressive two-bracket or three-bracket tax structure, where the poor face no taxes but also do not receive public goods. In situations where the government extracts tax revenues as political rents and maximizes expected payoff rather than the probability of re-election, there is a possibility of complete extraction, implying taxing away all private income with no allocation of public goods if electoral uncertainty is high. The paper explores how income, rather than numbers, plays a more central role in resource allocation. It shows that in a one-bracket tax structure, the jurisdiction with the poorest individual receives no local public good, while the jurisdiction with the median income receives the largest allocation. In a two-bracket tax structure, the jurisdiction with the poorest individual is completely left out of the political system, and in a three-bracket tax structure, the jurisdiction with the poorest individual receives no local public good, while the jurisdiction with the richest individual receives the most efficient level of local public good. The study also analyzes the overall welfare under different tax structures and finds that a three-bracket tax structure generally provides higher welfare than one- or two-bracket structures, especially when the difference in income between the richest and poorest individuals is substantial. However, in situations where the difference in income is small, a one-bracket tax structure may be more efficient. The paper concludes that in a situation where the government extracts a surplus, it may go for complete expropriation of all private income if electoral uncertainty is high and the loss from non-provision of local public good is low. In such cases, a two-bracket or three-bracket tax structure is preferable to minimize such extortions and ensure minimum provisions of local public goods to all jurisdictions.This paper analyzes the relationship between democracy and redistribution, focusing on how tax revenues are allocated to local public goods in different tax structures. In a probabilistic voting model with three jurisdictions where residents have different income levels, it is shown that it is always optimal to distribute tax revenues as public goods to only the residents of the richest and median income jurisdictions. The study compares the overall welfare of citizens under a one-bracket tax structure, where the poor contribute to taxes but do not receive public goods, with a progressive two-bracket or three-bracket tax structure, where the poor face no taxes but also do not receive public goods. In situations where the government extracts tax revenues as political rents and maximizes expected payoff rather than the probability of re-election, there is a possibility of complete extraction, implying taxing away all private income with no allocation of public goods if electoral uncertainty is high. The paper explores how income, rather than numbers, plays a more central role in resource allocation. It shows that in a one-bracket tax structure, the jurisdiction with the poorest individual receives no local public good, while the jurisdiction with the median income receives the largest allocation. In a two-bracket tax structure, the jurisdiction with the poorest individual is completely left out of the political system, and in a three-bracket tax structure, the jurisdiction with the poorest individual receives no local public good, while the jurisdiction with the richest individual receives the most efficient level of local public good. The study also analyzes the overall welfare under different tax structures and finds that a three-bracket tax structure generally provides higher welfare than one- or two-bracket structures, especially when the difference in income between the richest and poorest individuals is substantial. However, in situations where the difference in income is small, a one-bracket tax structure may be more efficient. The paper concludes that in a situation where the government extracts a surplus, it may go for complete expropriation of all private income if electoral uncertainty is high and the loss from non-provision of local public good is low. In such cases, a two-bracket or three-bracket tax structure is preferable to minimize such extortions and ensure minimum provisions of local public goods to all jurisdictions.