Three essays on agricultural futures market liquidity: exchange rules, Covid-19, and market stress

Peng, Kun

Permalink

https://hdl.handle.net/2142/116071 Copy

Description

Title

Three essays on agricultural futures market liquidity: exchange rules, Covid-19, and market stress

Author(s)

Peng, Kun

Issue Date

2022-07-15

Director of Research (if dissertation) or Advisor (if thesis)

Robe, Michel A

Doctoral Committee Chair(s)

Robe, Michel A

Committee Member(s)

• Irwin, Scott H
• Janzen, Joseph P
• Adjemian, Michael K

Department of Study

Agr & Consumer Economics

Discipline

Agricultural & Applied Econ

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Agricultural commodities
• Maximum order size
• Liquidity
• Volatility
• Calendar spread
• Implied trade

Abstract

This dissertation studies how commodity market liquidity responds to external shocks, both market-specific (e.g., exchange rule changes) and economy-wide (e.g., the COVID-19 pandemic). The first chapter exploit an exchange-mandated increase of the maximum order size, in the U.S. corn and soybean calendar spread market, in order to investigate the connection between exogenous constraints on order placement and execution, volatility, and liquidity. We show that the old maximum of 2,500 contracts was binding, and that demand exists for placing and executing much larger orders. The limit-order book depth (at the best bid and ask) increases dramatically after the exchange quadruples the maximum order size to 10,000 contracts. Intraday realized volatility is not statistically significantly different before and after the rule change. Amid increased market depth and stable volatility, we document that quoted spreads narrow significantly and that the price impact of large trades is smaller. In sum, market quality is higher after the maximum order size change. The second chapter documents the impact of the early stages of the COVID-19 pandemic on liquidity in U.S. agricultural markets. Notably, we show that soybean futures-market depth collapses weeks before the U.S. financial markets’ crash of March 2020. Soybean futures liquidity is affected the earliest, the most, and the longest. Soybean depth drops by half for outright futures and by over 90 percent for calendar spreads, and soybean bid-ask spreads increase significantly. This liquidity pullback starts on the night of February 12 to 13, 2020—a full two weeks before (i) liquidity evaporates in U.S. bond and equity markets and (ii) soybean prices start to fall sharply. The start of the soybean liquidity pullback coincides with overnight news of bleak COVID-19 developments in China (a dominant source of world demand for oilseeds). Following a series of emergency interventions by the U.S. Federal Reserve in March and April 2020, liquidity recovers in the soybean outright futures market—but depth remains abnormally low for calendar spreads. These patterns cannot be explained by other factors, such as seasonalities or changes in soybean futures trading volume and price volatility: the COVID-19 shock was novel, and it destroyed soybean-market liquidity in a way that foretold financial-market developments two weeks later. In contrast to soybeans, we find little evidence of a drop in corn or wheat futures liquidity until U.S. financial and crude oil markets sink in early March. Soybeans were truly the canary in the coal mine. The third chapter investigates, using 2019 MDP data for electronically-traded CBOT soybean futures, the flow of liquidity between the outright and calendar spread books. It provides empirical evidence on “implied trades,” in which outright soybean futures orders are matched with one leg of a calendar spread limit order, and documents the contribution of those “implied trades” to the total trading volume responds to intraday price volatility. Since the slope of the soybean futures term structure is directly related to the commodity’s cost of carry, it is much less variable than the price level itself. We therefore hypothesize that, even if the share of liquidity provision going from the spread book to the outright book typically averages only about 5-10 percent of the trading volume, this contribution should increase when prices become more volatile. Consistent with this intuition, we first show that the contribution of implied trades to the total trading volume in the outright market increases when the outright price level is more volatile, i.e., when outright futures trading becomes riskier. This finding supports the above intuition. However, we also find that the opposite is true in periods of unusually high volatility. Specifically, when the intraday volatility of the futures price is two standard deviations greater than its sample average, we show that the contribution of implied trades to the total trading volume in the outright market falls significantly. We find a similar pattern (i.e., a significant decrease in the implied trades share flowing from the outright book to the spread market) in the calendar spread market when the slope of the term structure becomes highly volatile. Our results point out that one must account for volatility patterns in both outright and calendar spread markets to understand the total liquidity provision in each compartment.

Graduation Semester

2022-08

Type of Resource

Thesis

Copyright and License Information

Copyright 2022 Kun Peng

Owning Collections