Doctoral Dissertation Research in Economics: ETFs and Arbitrage under Liquidity Mismatch

Harvard University, Cambridge MA

Investigators

Abstract

This award funds doctoral dissertation research that will be conducted by coPI Kevin Pan. He wants to understand how the market for corporate bonds is affected by the existence of corporate bond exchange traded funds (ETFs). These funds maintain liquidity by establishing authorized participants (APs) who are the only market participants allowed to transact with the ETF issuer to create and redeem shares of the ETF. The coPI will examine how AP trading behavior affects underlying bond returns, liquidity, and price volatility. The goal is to understand whether ETF arbitrage might fail under certain circumstances. ETF shares are owned by individual investors in the U.S.; understanding conditions under which the market could fail is important, since a market failure could have significant effects on many Americans' retirement savings. The results will be useful for regulators, who have been concerned about ETF liquidity during financial stress. They will also be useful to the many individual investors who hold ETF shares; the lack of scientific research on ETFs means that individual investors have few resources for disinterested information on the benefits and risks of ETF investments. The project generally seeks to understand how financial institutions, specifically authorized participants, can affect financial markets and asset prices. The coPI will construct a detailed panel set of (1) ETF prices, holdings, and arbitrage ability, (2) AP lists by ETF, and (3) corporate bond trades and characteristics. The plan to is to examine several research questions. First, how are excess volatility and co-movement of underlying corporate bond returns affected by the intermediation friction of broker-dealers in this market? Second, how have changing market trends and conditions affected the market and trade of corporate bonds and corporate bond ETFs? Third, how do APs use the creation/redemption process to relax their inventory or financial constraints? What are the resulting implications for their trading behavior in corporate bonds and ETFs, and how is this behavior different than what is observed for financial intermediaries who are not APs? Fourth, how and under what circumstances might ETF arbitrage fail? What are the implications of failure and what role do APs play in this failure? The coPI's theory suggests a new channel of transmission via APs' inventory constraints and ETFs' liquidity mismatch. The coPI will also use the data to test various theories about arbitrage in this market.