UCLA

This dissertation consists of three independent chapters. The first chapter is about worker productivity and the final two chapters are on fiscal spending.

The first chapter studies worker productivity. A growing literature explores the impact of home-based versus office-based work. Differences in productivity may arise due to a treatment effect of the office or from workers with different abilities sorting into office or home work. If more productive workers find working in the office less costly (a selection effect) or more complementary to their skills (a selection on treatment effect), we expect positive selection into office work. But if more productive workers have stronger preferences for home work or face more severe constraints on working outside the home, the selection effect could be negative. We conduct an RCT in the data entry sector in India that exogenously allocates workers to the home or office. We find that the productivity of workers randomly assigned to working from home is 18% lower than those in the office. Two-thirds of the effect manifests itself from the first day of work with the remainder due to quicker learning by office workers over time. We find negative selection effects for office-based work: workers who prefer homebased work are 12% faster and more accurate at baseline. We also find a negative selection on treatment: workers who prefer home work are much less productive at home than at the office (27% less compared to 13% less for workers who prefer the office). These negative selection effects are partially explained by subgroups that likely face bigger constraints on selecting into office work, such as those with children or other home care responsibilities as well as poorer households.

In the second chapter, we explore how the opinions of economists regarding the impact of a fiscal stimulus have evolved over the last century. In describing the evolution of economists’ opinions, we contrast deductive methods based on words, pictures, and math including calibrated models versus inductive methods using data images, tables of statistics, and estimated models.

In the third and final chapter, we highlight the problem of not accounting for spillover effects while estimating local fiscal multiplier models. We survey various methodologies from empirical macroeconomics, network econometrics, and trade literature to account for spillovers. We provide tentative evidence for the spillover effects of tax cuts given at the state level by re-estimating Zidar (2019)