Essays on information and markets
- This dissertation is a collection of three empirical studies broadly concerning how information flows and how markets function. The first chapter looks at informed trading in a financial market where asset prices are set in advance and not adjusted for investor demand. I focus on a novel setting, online marketplaces for unsecured consumer debt. Here, a single platform serves as both financial intermediary and market-maker, vetting borrowers and setting prices (interest rates). The platform randomly assigns the loans between two markets, one retail- and one institution-facing, where they are subsequently sold. In both markets, the price is fixed before trading starts and does not fluctuate once investor demand is realized. In both markets, I find strong evidence that there exist sophisticated investors who freely and aggressively purchase underpriced loans given that they face no price impact. This creates a significant adverse selection problem for the comparatively unsophisticated investors in each market. I find that the cost to unsophisticated investors is on the order of several hundred basis points in returns per loan. I then demonstrate that investment decisions---proxied by observable order flow---are an informative signal of agents' private valuations. I show that unsophisticated investors infer private information and mitigate their exposure to adverse selection. Finally, I offer evidence that the platform itself uses observed order flow to identify gross pricing errors or outright borrower fraud and, based on this, cancel loans ahead of origination. The findings suggest that this broad form of information discovery helps sustain investor participation in these markets, despite the potential for adverse selection. The second chapter, coauthored with Ryan Shyu and Ramana Ramakrishnan, considers the informational frictions pervading labor markets. We specifically ask how firms' learn from other firms' human resource allocation decisions. We concentrate on job promotions which, according to theory, serves as an informative signal to outside employers if there is asymmetric information about worker ability. Using variation in the timing of promotion reports on LinkedIn CVs, we implement a differences-in-differences strategy to demonstrate that online promotion reports increase recruiter-initiated worker contacts ("InMails"). The signaling impact of promotions is concentrated among those who have recently attracted previous recruiter interest; this effect is subsumed by firm and job heterogeneity, suggesting that factors idiosyncratic to individual users are less important relative to firm and job quality in providing labor market visibility. The third chapter, coauthored with my advisor, Peter Koudijs, examines the market for margin loans in Amsterdam between 1762 and 1775 to establish a causal relation between leverage and the level of asset prices. Margin loan contracts typically ran for 24 months and had margin calls and margin returns built in. After a margin call, a borrower was forced to delever; after a margin return a borrower could lever up. Margin calls and margin returns were activated at critical price levels that were typically a multiple of 10. We document that price changes crossing a multiple of 10 were larger than price changes crossing any other integer, to the tune of 50--150 bps, suggesting that changes in leverage causally affected asset prices. These price changes persisted for at least 3 months. We also provide evidence suggesting that margin returns allowed the most optimistic agents to lever up, allowing them to purchase more shares and drive up the price.
|Type of resource
|electronic resource; remote; computer; online resource
|1 online resource.
|Degree committee member
|Degree committee member
|Stanford University, Graduate School of Business.
|Statement of responsibility
|Submitted to the Graduate School of Business.
|Thesis Ph.D. Stanford University 2018.
- © 2018 by Bobak Moallemi
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...