Graduate Thesis Or Dissertation


Three Chapters in Finance: New Advancements in Financial Narratives, Investor Incentives, and Consumer Biases Public Deposited

Downloadable Content

Download PDF
  • In the first chapter, with lab experiments, Shah, Shafir, and Mullainathan (2015) conjecture that resource scarcity could lead consumers to deviate from heuristics and become less susceptible to decision biases. Reversely, I test if observed behavioral biases can help identify financially constrained borrowers. Taking advantage of a rich data set from the world's largest peer-to-peer FinTech consumer lending platform, I empirically test this hypothesis in a real world field setting by examining the relationship between the deviation from the heuristics of choosing a round-number loan and financial constraints. Controlling for all borrower and loan characteristics, I robustly find that borrowers who deviate are 2 percentage points more likely to default (more financially constrained) than conformists who choose round-number loans. Separating lenders by type, I find that institutional lenders do largely mitigate this extra default risk through their screening process, whereas the cost of this mispricing is transferred mostly to retail investors. In the second chapter, we study a standard machine learning algorithm (Taddy, 2013) to measure sentiment in financial documents. Our empirical approach relies on stock price reactions to color words, providing as output dictionaries of positive and negative words. In head-to-head comparisons, our dictionaries outperform the standard bag-of-words approach (Loughran and McDonald, 2011) when predicting stock price movements out-of-sample. By comparing their composition, word-by-word, our method refines and expands the sentiment dictionaries in the literature. The breadth of our dictionaries and their ability to disambiguate words using bigrams both help to color finance discourse better. In the third chapter, we develop a testable model to evaluate the concern that illiquidity in corporate bonds may generate a first-mover advantage among bond fund investors. A key insight of the model is that first-movers, if present in bond funds, must impose non-trivial costs on non-redeeming investors. We use estimates from a structural VAR applied to daily data spanning five periods of turbulent bond markets to calculate the "cost of remaining invested'' (CRI), in terms of reduced returns, on days when the market drops. We find little evidence of an economically meaningful incentive for first-movers to redeem. In aggregate, CRI is typically very small, less than 4 basis points on an annualized basis, even for funds that hold less liquid bonds. Our hypothesis test for the presence of first-movers in individual funds signals that, for a subset of high yield bond funds, the elasticity of fund flows to market returns is higher than would be expected by chance alone. However, the assets in these funds are too small to impose meaningful price pressure on the high yield bond market.

Date Issued
  • 2022-04-13
Academic Affiliation
Committee Member
Degree Grantor
Commencement Year
Last Modified
  • 2022-12-13
Resource Type
Rights Statement