UC Berkeley

The relationship between democracy and inequality and the role of democratic participation in fostering redistribution has been a critical question of interest to economists over time. My dissertation focuses on a key aspect of democratic participation - the ability of citizens to form political parties and contest elections - and empirically identifies the relationship between the distribution of political power, and the distribution of public resources. To this effect, I study the economic impacts emanating from the political mobilization of historically underprivileged and socio-economically disadvantaged citizen groups in India - the low caste and indigenous communities - which has occurred over the past three decades. While comprising over half the nation's population, citizens from these groups have faced institutionalized discrimination through the hierarchical caste system, severely restricting their access to education, public institutions and basic services. This process of political mobilization by low caste citizens resulted in the formation of powerful caste-based political parties which contested elections in an attempt to capture political power through the democratic process, and subsequently, use their nascent state power to redirect public resources to their communities. The core chapters of this dissertation empirically document the economic and social impacts emanating from this change in the distribution of political power through democratic processes, resulting in the political empowerment of marginalized citizen groups.

The first chapter of my dissertation identifies the economic impacts of this process of political mobilization. Using the outcome of close elections between parties exclusively representing low caste interests and other mainstream parties as the source of exogenous variation, the paper shows that the marginal legislator from low caste parties increases the share of state expenditures allocated towards targeted and untargeted welfare schemes benefitting low caste citizens by 2 and 1 percent. In a departure from existing studies in the literature which have exploited the system of mandated representation of low caste politicians in India, my paper shows that the mandated representation of low caste politicians has no impact on welfare transfers for low caste citizens when unaccompanied by a strong ex-ante party commitment in favour of low caste welfare. This paper also highlights the trade-offs emanating from the redistributive spending choices of fiscally constrained legislators by identifying the negative impact of the marginal legislator from low caste parties on public investments in road construction and electricity generation. Using tests of heterogeneity, the paper argues that the redistributive spending preferences of legislators from low caste parties is one of the mechanisms explaining the negative causal impact between the electoral success of these parties and regional inflows of private investment.

Low caste citizens in India have often been victims of targeted violence by social elites in an attempt to preserve the existing social hierarchy and consolidate economic rents. This forms the motivation for the second chapter of my dissertation, identifying the causal impact of low caste parties on the incidence of targeted crimes against low caste citizens. The empirical results from this paper show that the marginal legislator from low caste parties reduces the incidence of violent crimes against low caste citizens by 4-14 percent and contribute significantly to the protection of civil rights. The paper also identifies the mechanisms explaining the reduced form impact - first, through higher rates of arrest, prosecution and diligence with which law enforcement agencies pursue such cases of targeted violence; second, through changes in the attitudes of low-caste citizens towards state institutions which affect the enforcement of criminal law.

Finally, the third chapter of my dissertation focuses on a major question of interest in the field of development economics and economic growth - the role of agricultural growth in the process of structural transformation. Using data from a panel of 273 Indian districts over a twenty-five year period and rainfall shocks as an instrument for agricultural growth, I present causal evidence that rainfall-induced agricultural growth has a significant impact on the structural transformation process by increasing the share of rural workers employed in the manufacturing sector. My results show that the non-farm employment gains from higher agricultural growth is concentrated amongst unskilled workers with relatively low levels of education, signalling the poverty alleviating aspect of agricultural growth. By studying the differential impact of agricultural productivity across regions with initially low levels of agricultural productivity and urbanization, I also document the potential of agricultural growth to facilitate convergence across regions in terms of rural employment diversification.