Essays on behavioral economics and experimental design
OA Version
Citation
Abstract
This dissertation consists of three essays that study experimental design and behavioral economics. I discuss novel tools for running adaptive survey experiments and use these tools to measure economic preferences across a series of experiments that address questions in labor and public economics.
In Chapter 1 (joint with Fernando Payró, Neil Thakral, and Linh T. Tô), we propose the use of a dynamic experimental method, which we call Bayesian Adaptive Choice Experiment (BACE), to elicit preferences efficiently in online choice experiments. BACE generates an adaptive sequence of menus from which subjects make choices. Each menu is optimally chosen, according to the mutual information criterion, using the subjects' previous choices. Using simulations, we demonstrate that BACE significantly improves convergence relative to existing discrete choice methods with randomly generated menus. Beyond efficiency gains, BACE addresses a bias in estimating population-level average preference parameters that stems from combining data across individuals who differ in their tendency to make inconsistent choices. Given that BACE requires the calculation of a Bayesian posterior as well as the solution to a non-trivial optimization problem, several computational challenges arise. We address such challenges by using Bayesian Optimization and Monte Carlo techniques and provide a package for researchers to conduct adaptive experiments in an online setting. The separation between a front-end survey interface and a back-end computational server allows the BACE package to be portable for research designs in a wide range of settings.
In Chapter 2 (joint with Neil Thakral, and Linh T. Tô) we study the interplay between how much workers value workplace flexibility, whether they have such amenities, and how the presence of amenities affects their wages. We implement BACE, the method described in Chapter 1, to collect data on the joint distribution of wages, work arrangements, and willingness-to-pay (WTP) for different forms of flexibility. We document a series of facts related to the distribution of WTPs for different types of workplace flexibility. We also examine how WTP and access to workplace flexibility differ across the wage distribution. We then introduce a model in which workers may face different prices for job amenities depending on their productivity, extending the Rosen (1986) model of compensating differentials. The model predicts key patterns in the data, including (i) the relationship between wages and having amenities, (ii) inequality in workplace amenities across the earnings distribution even when workers value these amenities similarly, and (iii) the tradeoffs workers face when choosing between different forms of flexibility.
In Chapter 3, I study how past support for a charity affects how people acquire and respond to information about its quality. I conjecture that past giving motivates donors to process information about a charity differently than non-donors. I test this hypothesis in a series of online experiments in which subjects are randomly assigned to complete a real-effort task that can generate a donation to a non-profit. Prior to allocating money between themselves and charity, subjects receive different information about the relative impact of two charities. Without new information, prior support does not influence giving across charities. However, when subjects learn which charity is more effective, those who already supported the more effective charity respond more positively to this information. This study demonstrates that prior support can influence future giving by changing how people respond to the information present in donation requests. I also examine how past support affects beliefs about relative quality, the reasons why people give, and demand for information and discuss potential improvements to the experimental design.