A key question in development economics is why productivity differs so much across regions and countries. Part of the explanation is \textit{misallocation of resources}, where capital and labor are not used in the most productive way. Financial frictions, such as high and unequal borrowing costs, can prevent resources from flowing to the most productive uses. For instance, households paying higher interest rates may invest less or choose lower-return projects, while households with lower borrowing costs may access larger loans, even if their projects are less productive. This can slow aggregate productivity growth and structural transformation.

In many developing countries, formal credit has expanded, yet productivity growth remains limited. One reason may be that borrowing costs are still highly uneven across households, especially by caste, income, and geographic location. In India, lower-caste households often face higher interest rates, reducing their ability to invest in productive activities. This creates persistent inequalities in income and limits the economy's ability to shift resources toward more productive sectors. Understanding how these disparities arise and persist is essential for designing policies that improve financial inclusion and productivity.

This project asks: how large is the variation in borrowing costs across households in rural India? How much of this variation can be explained by access to formal banking, and how much persists due to social or structural barriers? Finally, what would happen if all households had equal access to credit? Addressing these questions requires combining precise micro-level data with a structural framework that captures household decision-making under borrowing constraints.

Politicians face competing pressures when allocating public infrastructure: maximizing economic returns versus securing political support. I estimate the efficiency cost of political considerations in infrastructure allocation using quasi-experimental variation in canal placement across 224 villages in Gujarat, India. Exploiting close elections and politician-voter caste alignment as instrumental variables, I find that while politicians predominantly allocate canals to high-productivity villages, political targeting of opposition caste groups distorts this allocation at the margin. Approximately 14% of canals are misallocated for political reasons—either given to lower-productivity villages or withheld from high-productivity villages. Using a counterfactual simulation based on household production function estimates, I calculate that a purely productivity-maximizing allocation would increase aggregate agricultural output by an additional 12.6%, equivalent to 26.5 million kilograms annually. These findings quantify the trade-off between political incentives and economic efficiency in public infrastructure investment decisions, with implications for infrastructure policy in electoral democracies.

Evidence from the vast literature suggests that the decision-making power of females in the household plays an important role to ameliorate household welfare. We analyze the impact of having a bank account and the channels it triggers, which improves the female’s decision-making power in the household. In this study, we utilized the data from the SEPRI (2016) and JPM (2016) for 12 districts of the Uttar Pradesh state and use the 3SLS approach in our analysis. We concluded in this study that a female owning a bank account increases her share of household decision-making by 33.8%. Therefore, it helps in reducing the vulnerability of the household to fall below the poverty line.