The impact of net buying pressure on VIX option prices

• Author: Ming‐Hung Wu,Yi‐Wei Chuang,Wei‐Che Tsai
• Published date: 01 February 2020
• Date: 01 February 2020
• DOI: http://doi.org/10.1002/fut.22060

J Futures Markets. 2020;40:209–227. wileyonlinelibrary.com/journal/fut © 2019 Wiley Periodicals, Inc.

|

209

Received: 1 September 2019

|

Accepted: 1 September 2019

DOI: 10.1002/fut.22060

RESEARCH ARTICLE

The impact of net buying pressure on VIX option prices

Yi‐Wei Chuang

1

|

Wei‐Che Tsai

2

|

Ming‐Hung Wu

3

1

Department of International Business,

Feng Chia University, Taichung,

Taiwan, ROC

2

Department of Finance, National Sun

Yat‐sen University, Kaohsiung,

Taiwan, ROC

3

School of Management, Beijing Normal

University, Zhuhai, China

Correspondence

Wei‐Che Tsai, Department of Finance,

National Sun Yat‐sen University, No. 70

Lienhai Road, Kaohsiung 80424, Taiwan,

ROC.

Email: weiche@mail.nsysu.edu.tw

Funding information

Ministry of Science and Technology,

Taiwan, Grant/Award Number:

106‐2628‐H‐110‐001‐MY3

Abstract

This paper analyzes the impact of intraday trading activity on option prices

in the Volatility Index (VIX) options market. Our results show that there is a

temporal relationship between net buying pressure (NBP) and changes in

implied volatility of VIX options. Moreover, an increase in NBPs lowers the

next‐day delta‐hedged option returns. Using several measures proxying for

limits to arbitrage, the average levels of the implied volatility curve rise when

limits to arbitrage are severe. A trading strategy in the VIX futures market

constructed by using the NBP generates an average annualized return of 10.09%.

KEYWORDS

implied volatility of volatility, limits of arbitrage, net buying pressure, Volatility Index options

JEL CLASSIFICATION

G13

1

|

INTRODUCTION

Unlike in a perfectly efficient market, trading activity usually impacts market prices in real‐world financial markets

around the world. For example, market makers may face inventory control problems during periods of large order

imbalances, and if the trading amount of buy orders is slightly greater than that of sell orders and exceeds the quantities

that they can provideat that time, then the large order imbalance will force them to respond by raising quoted prices (see,

e.g., Chordia, Roll, & Subrahmanyam, 2002; Ho & Stoll, 1983; Spiegel & Subrahmanyam, 1995; Stoll, 1978). Although

higher prices caused by a temporary order imbalance may attract arbitrageurs to step in and help asset prices revert to

fundamental values, mispricing may still persist due to limits to arbitrage (see, e.g., Brunnermeier & Pedersen, 2009;

Gromb & Vayanos, 2002; Shleifer & Vishny, 1997).

Evidence of trading pressure has also been documented in derivatives markets. Bollen and Whaley (2004) find that

option net buying pressure (NBP), constructed as the number of buyer‐motivated contracts traded each day minus the

number of seller‐motivated contracts, contributes to daily changes in the level and slope of implied volatility, which is a

one‐to‐one function of market option prices. Gârleanu, Pedersen, and Poteshman (2009) further provide a theoretical

model of demand pressure effects in the option market, whereas Muravyev (2016) notes that the price impact of daily

order imbalances in option markets attributable to inventory risk is large.

Order imbalance certainly also reflects informed trading in addition to having an inventory risk impact. As

Stoll (2000) and Schlag and Stoll (2005) suggest, informed trading does have a permanent influence on market

prices. Several studies provide support for the existence of informed trading in the options markets, such as

Easley, O’Hara, and Srinivas (1998), Pan and Poteshman (2006), Kang and Park (2008), Cremers and Weinbaum

(2010), Roll, Schwartz, and Subrahmanyam (2010), Xing, Zhang, and Zhao (2010), Johnson and So (2012),

Conrad,Dittmar,andGhysels(2013),An,Ang,Bali,andCakici(2014),Hu(2014),andChesney,Crameri,and

Mancini (2015).

This study aims to provide a comprehensively empirical analysis covering the impact of intraday trading activity of

Volatility Index (VIX) options on their market prices.

1

A closely related paper to this present study is Bollen and Whaley

(2004), but our work is different from theirs in several aspects. First of all, we examine the effect of NBP in the newly

established and rapidly growing VIX options market. Compared to complex volatility trading strategies in the S&P 500

index options market, VIX options offer investors a simple and direct way to trade volatility without dealing with the

other associated risk factors that otherwise affect the overall performance of volatility strategies. Hence, trading VIX

options has become a useful risk management tool for investors wanting to hedge their volatility exposure. If there are

serious limits to arbitrage in the VIX options market, then investors may face expensive hedging costs for insurance,

possibly leading to liquidity spirals, such as seen in Brunnermeier and Pedersen (2009). It is therefore worth exploring

the extent and sources of trading pressure in the VIX options market. Second, we would like to extend Bollen and

Whaley’s (2004) analysis by using intraday data of VIX options so as to have a better understanding of the effect of

trading activity on prices.

Our empirical findings show a temporal relationship between the NBP of VIX options and changes in implied

volatility of volatility (IVoV) from VIX option prices. In this case, increased NBP creates positive pressure on option

prices, supporting the prediction concerning the limits of the arbitrage hypothesis rather than the information

hypothesis. Consistent with the study of Bollen and Whaley (2004), which is associated with limits to arbitrage, the

result affects the pricing of VIX options.

We also examine the relationship between the NBP and delta‐hedge returns in the VIX options market. Our finding

shows that the higher NBP of VIX options is associated with lower delta‐hedge returns of VIX options. These results

suggest that the trading amount of buy orders is slightly greater than that of sell orders, resulting in more expensive

options with higher option prices and lower delta‐hedge returns.

Turning our focus to the analyses of limits to arbitrage on VIX option prices, our paper further examines the

relationship between the level of the implied volatility curve and proxy variables for measures of limits to arbitrage.

These proxy variables include measures for noise‐trader risk (investor attention and sentiment) and funding liquidity

(Libor‐Tbill and Libor‐Repo spreads). Our result documents that the level of the implied volatility curve rises when

there are serious limits to arbitrage in the VIX options market.

We further investigate whether intraday prices of VIX futures are affected when liquidity providers face buying or

selling pressure in the VIX options market. Hu (2014) also shows that liquidity providers in options hedge their risk

exposures in the underlying asset market, leading to changes in the underlying asset price. Thus, we observe the impact

of NBP of VIX options on VIX futures returns by building up a trading strategy. The trading strategy in the VIX futures

market generates an average annualized risk‐adjusted return of 10.09% with a significantly positive t‐statistic of 2.12,

likely implying option trading pressure transmission to VIX futures market via dealers’hedging activity. In other words,

our result suggests that VIX futures prices also adjust to order pressure in the VIX options market.

This article contributes to prior research and focuses on high‐frequency intraday data to gain a better

understanding about the effects of NBP in the VIX options market.

2

In particular, our empirical evidence provides

support for the limit to arbitrage hypothesis in the VIX options market. Our results also show that the trading

amount of buy orders is slightly greater than that of sell orders, resulting in more expensive options with higher

prices and lower delta‐hedge returns of VIX options. Moreover, our paper investigates the relationship between the

proxy variables for measures of the limit to arbitrage and level of the implied volatility curve. Our robust result

documents that the level of the implied volatility curve rises when there are serious limits to arbitrage in the VIX

options market. Consistent with Hu (2014), our results further support the argument that liquidity providers

transfer their exposure from the VIX options market to the VIX futures market through their hedging when large

NBP is initiated in the VIX options market.

The remainder of this article is organized as follows. Section 2 offers the literature review and describes the

hypothesis development. Section 3 presents the data used for analysis and the empirical methodology herein. Section 4

shows empirical results, and Section 5 makes the concluding remarks.

1

In the related literature, Tsai, Chiu, and Wang (2015) and Kao, Tsai, Wang, and Yen (2018) investigate the informational role of intraday trading activity and quote changesin VIX options with regard

to the movements of the VIX index, whereas our study mainly focuses on the trading pressure of VIX options on VIX option price changes.

2

Chung, Tsai, Wang, and Weng (2011) utilize daily market prices of VIX options to provide support for the informational role of VIX options regarding returns, volatility, and density predictions in the

S&P 500 index. Wang (2013) finds that the daily trading volume of VIX call options is informative regarding future realized volatility. Tsai et al. (2015) show that volume imbalances convey no

significant predictive information, whereas quote changes in VIX options can significantly predict changes in the index; this predictive power is especially more pronounced for VIX calls around

periods of monetary policy announcements.

210

|

CHUANG ET AL.