Volatility patterns of short-term interest rate futures

Abstract

A general question in finance is whether the volatility of the price of futures contracts follows any particular trend over the contract’s life. In this study, we contribute to the debate by empirically analyzing the trend of the term structure of the volatility of short-term interest rates (STIR) futures prices. Using data on the Eurodollar, Euribor, and Short-Sterling futures contracts for the period between 2000 and 2018, we model the volatility of each individual contract considering time to expiration and trading activities. Furthermore, we investigate whether these trends change according to overall economic conditions. We find that STIR futures behave differently than futures on other underlying assets and that, most of the time, STIR futures price volatility declines as the contract approaches expiration. Moreover, the relation between volatility and time to maturity depends on market conditions and trading activities, and it is non-linearly related to the observation period.

KEYWORDS:

• Futures contracts
• short-term interest rates futures
• maturity effect
• Samuelson hypothesis
• interbank rates
• volatility

JEL Classifications:

• G12
• G13
• G15

Acknowledgement

We are grateful for the valuable comments of two anonymous referees, Steve Figlewski, Janko Hernandez, Oldrich Vasicek and Aurelio Vasquez. We thank participants at ITAM brown bag seminar, FMA 2018 Annual Conference, FMA 2018 Asia Conference, and the European Financial Management 2018 Annual Conference. Renata Herrerias gratefully acknowledges the support of Asociacion Mexicana de Cultura, A.C. Pedro Gurrola has worked on this research while he was at the Bank of England. The views expressed here are entirely ours and do not necessarily reflect those of the World Federation of Exchanges or any other institution. All errors are our own.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 WFE IOMA Derivatives Report 2018, World Federation of Exchanges 2019. Available at: https://www.world-exchanges.org/storage/app/media/statistics/WFE%202018%20IOMA%20Derivatives%20Report%20FINAL%2010.04.19.pdf.

2 The Short-Sterling and the Euribor contracts were originally traded in the London International Futures and Options Exchange (LIFFE). After several mergers and acquisitions transactions, in 2014 LIFFE became part of the Intercontinental Exchange Inc. (ICE) and is currently part of the ICE NYSE group under the name ICE Futures Europe. A small share of Euribor contracts are also traded at Eurex.

3 In early 2014, NYSE Euronext took over the administration of Libor from the British Bankers Association. The new administrator was NYSE Euronext Rates Administration Limited. On November 2013, the Intercontinental Exchange (ICE) Group acquired NYSE Euronext and the NYSE Euronext Rate Administration Limited was renamed ICE Benchmark Administration Limited.

4 Traders will roll over futures contracts that are about to expire to a longer-dated contract in order to maintain the same position following expiration.

5 Detailed results of each regression model, including all values of regression coefficients and adjusted R²s are available from the authors upon request.

6 Seasonality dummies deliver non-significant coefficients in at least 80% of contracts for all models and all underlying interest rates. For example, in the case of Short-Sterling futures contracts, only 1 contract reports a positive and significant coefficient for the seasonality dummy. When the seasonality dummy is significant, it tends to be positive, indicating that sometimes futures volatility increases on expiration months. For the sake of brevity, we do not report these results in Table 2, but detailed results are available from the authors upon request.

7 The traditional assumption was that determining the short-end of the yield curve was the appropriate tool to eventually influence the longer terms. This, of course, has changed after the Great Financial Crisis and the implementation of quantitative easing approaches.

8 For a summary of the empirical literature documenting how central bank operations affect money market volatility see, for example, Osborne (2016).

9 Several studies have analyzed the question of whether the volatility of the overnight rate is correlated with that of longer-term rates. Both in the euro-area and in the U.S., there is evidence that volatility in overnight rates is transmitted to longer-term rates, at least up to 6-months in maturity and using data up to 2007 (Moschitz 2009; and Colarossi and Zaghini 2009). In the case of the UK, however, Osborne (2016) found no evidence that the volatility of overnight rates affected the volatility of the 3-month Libor rate in the period 2006–2014.

10 It is also worth mentioning that, from a theoretical standpoint, some models of the forward rate predict decreasing volatility as maturity approaches. In the variant of the Heath-Jarrow-Morton (HJM) model proposed by Zhou (2002), for example, volatility tends to decrease as maturity approaches.

11 See https://www.federalreserve.gov/monetarypolicy/openmarket.htm.

12 See https://www.bankofengland.co.uk/monetary-policy/the-interest-rate-bank-rate.

13 The increase in uncertainty was also reflected in the dislocation observed since 2007 between the theoretical implied forward rates and the forward rate implied by the Eurodollar and Euribor contracts. Since then, interbank lending has dried up, as banks try to avoid counterparty credit risk and prefer secured funding through repo. But while Euribor and Libor quotes have become largely arbitrary, STIR futures continue to actively trade and have become the proxy for the three-month forward curve (Aikin 2012).