Graduation Year

2026

Date of Submission

4-2026

Document Type

Campus Only Senior Thesis

Degree Name

Bachelor of Arts

Department

Economics

Second Department

<--Please Select Department-->

Reader 1

Eric Hughson

Terms of Use & License Information

Terms of Use for work posted in Scholarship@Claremont.

Abstract

The call-put implied volatility spread (CPIV) has been shown to predict cross-sectional stock returns. This signal could reflect informed trading in the options market but also the mechanical effect of stock borrow costs on put-call parity. This distinction rem­ains contested. This paper tests whether CPIV retains return predictability once borrow costs are controlled for. Using ORATS near-end-of-day option quotes on U.S. common equities from January 2007 through early 2025, I compute implied repo rates directly from put-call parity as an options-derived measure of stock borrow costs. This requires no proprietary lending data. I then test CPIV predictability within a low borrow cost group and a high borrow cost group using monthly double portfolio sorts and Fama-MacBeth regressions for controls. CPIV produces no statistically reliable return spread in the low-fee environment where borrow cost contamination is minimized. It also has no incremental explanatory power in cross-sectional regressions when implied repo is included as a control. The evidence is consistent with CPIV being a noisy proxy for borrow cost information rather than an independent informed trading signal. This follows directly with Muravyev, Pearson, and Pollet (2025). Implied repo itself predicts large negative returns in the extreme upper tail of the borrow cost distribution with a magnitude of roughly −1.34% per month at the one-month horizon, though this result is not implementable as a trading strategy given that the stocks driving the return are also the most expensive to short.

Download

This thesis is restricted to the Claremont Colleges current faculty, students, and staff.

Share

COinS