Essays in Energy and Development Economics

Abstract

The three essays in this dissertation investigate the economic and policy challenges of providing access to reliable and affordable electricity in India. Across much of the country, regulated retail electricity prices enable agricultural and residential consumers to buy power at a fraction of the cost of supplying it to them. Forced to sell electricity at a loss to broad segments of their customer base, state-owned electricity distribution companies often resort to rationing electricity through rolling blackouts, a practice referred to as ``load shedding.'' Chapter 1 evaluates the welfare and distributional implications of retail electricity price subsidies and the rationing that accompanies it for customers served by India's largest electricity distribution utility, the Maharashtra State Electricity Distribution Company Limited (MSEDCL), in 2017 and 2018. It begins by documenting how MSEDCL uses peak load shedding to curtail its costs of wholesale power purchase, estimating that in 2017 and 2018 on average the company shed load for an additional approximately 246,000 people in response to a one standard deviation increase in its marginal cost of energy. The paper then develops a dynamic structural model of electricity demand under quantity rationing in order to evaluate the welfare loss to customers from power outages and their willingness to pay for higher levels of reliability. Estimating the model using an original dataset of hourly electricity consumption on 16,633 power lines served by MSEDCL, I find that the welfare losses from power outages in 2017 and 2018 for the categories of customers facing severe rationing of power were equivalent to those of increases in the marginal electricity price ranging between about 20 to 70 percent.

Chapter 2, which is coauthored with Shefali Khanna, estimates the short- and long-run responses of retail electricity consumption and bill payment to electricity prices in Delhi, India from 2015 to 2019. Using billing data from one of Delhi's three private electricity distribution utilities, we reconstruct payment histories for more than 1.5 million retail residential, commercial, and small industrial customers. Our empirical strategies exploit features of the regulated electricity price schedule that generate short- and long-run variation in the average price of electricity for these customers. In addition to providing annual price elasticity estimates, we find evidence that non-payment rates increase when prices rise. The effects are particularly large for the poorest informal customers, for whom arrears more than double in response to a doubling in the average electricity price.

Finally, Chapter 3, which is also coauthored with Shefali Khanna, evaluates residential consumers' electricity consumption and appliance investment responses to power outages from 2015 to 2019 in Delhi, India. Our empirical strategy takes advantage of features of the electricity distribution network in the service territory of one of Delhi's regulated distribution utilities that exposes similar customers to plausibly-exogenous annual variation in electricity reliability. Using original household survey data and four years of billing and power outage records for more than one million customers, we estimate that an additional hour per month of power outages reduced electricity consumption by about 4.85 percent.