The theory of storage in the crude oil futures market, the role of financial conditions

• Published date: 01 July 2020
• Author: Niaz Bashiri Behmiri,Matteo Manera,Maryam Ahmadi
• Date: 01 July 2020
• DOI: http://doi.org/10.1002/fut.22113

J Futures Markets. 2020;40:1160–1175.wileyonlinelibrary.com/journal/fut1160

|

© 2020 Wiley Periodicals, Inc.

Received: 4 September 2019

|

Accepted: 23 February 2020

DOI: 10.1002/fut.22113

RESEARCH ARTICLE

The theory of storage in the crude oil futures market, the

role of financial conditions

Maryam Ahmadi

1

|Niaz Bashiri Behmiri

2

|Matteo Manera

3

1

Department of Economics, Management

and Statistics (DEMS), University of

Milan‐Bicocca, Milan, Italy

2

Department of Economics and Finance,

University of Stavanger, UiS Business

School, Stavanger, Norway

3

Department of Economics, Management

and Statistics (DEMS), University of

Milan‐Bicocca and Fondazione Eni Enrico

Mattei (FEEM), Milan, Italy

Correspondence

Maryam Ahmadi, Department of

Economics, Management and Statistics

(DEMS), University of Milan‐Bicocca,

Piazza dell'Ateneo Nuovo, 1, 20126

Milano, Italy.

Email: mar.ahmadi@gmail.com

Abstract

This study examines the impacts of inventory and financial instability on the

basis of the crude oil market. The results show that, first, the basis rises with

inventory, and this effect is higher during low inventory regimes. This vali-

dates the theory of storage in the crude oil market. Second, the basis rises with

financial instability, and this effect is higher during turbulent regimes. These

results warn the oil market participants that, to make decisions based on the

basis variation, traditionally known as a signal of scarcity or abundance, the

underlying cause of the variation has to be considered.

KEYWORDS

crude oil futures, financial stress, inventory, theory of storage, threshold SVAR

JEL CLASSIFICATION

G13; G15; G31; Q41

1|INTRODUCTION

The markets for energy products are characterized by high levels of fluctuations in prices and inventories. These

fluctuations are in part unpredictable, and this leads market participants to use futures contracts and inventories as

vehicles for reducing risk. Thus, the spread between futures and spot prices, the basis, is an important signal of the

marginal value of storage for commodity participants (Knittel & Pindyck, 2016).

The relationship between price variation and the holding inventory of commodities is described by the theory of

storage. While this theory is widely accepted by energy market participants, the literature on the analysis of the theory

of storage in the market for crude oil is inconclusive.

1

This is an important issue, since the variation of the basis is an

important factor in setting the efficient hedging strategy, improving profitability, and deciding the time when to sell or

buy in oil futures markets. It is also a signal for refiners to decide the timing of crude oil purchases and the production

and delivery of their products (Cho & McDougall, 1990; Serletis & Hulleman, 1994). This paper examines the theory of

storage and its implications in the crude oil market, by considering, for the first time, the role of financial market

conditions.

The theory of storage makes two main predictions related to the quantity of a commodity held in inventory. First, in

times of scarcity, the spot price rises more than the futures price, mainly because market participants believe that a

higher price in the long term will stimulate the supply of the commodity, a situation which is called “backwardation.”

Conversely, in times of abundance, the spot and futures prices tend to decrease in equal measure, a situation which is

1

See, for example, Serletis and Hulleman (1994); Cho and McDougall (1990); Geman and Ohana (2009); Symeonidis, Prokopczuk, Brooks, and Lazar

(2012); and Nikitopoulos, Squires, Thorp, and Yeung (2017). We return to this, in more detail, in the next section.

known as “contango.”Second, in times of scarcity, not only the spot price, but also the variation of the spot price rises,

since in an uncertain situation, all news about supply, demand or inventory affects the spot market. However, this effect

is less evident in the variation of longer‐term futures contracts, as market participants are aware that, in the long term,

a higher price stimulates supply, and inventory will be rebuilt (Geman & Smith, 2013). In times of abundance, there

will be no significant difference between the variation of spot and futures prices (see, e.g., Brennan, 1958; Gao &

Wang, 2005; Geman & Smith, 2013; Telser, 1958).

In the literature, the validity of the theory of storage in the commodity markets is examined by direct and indirect

tests. The direct tests model the convenience yield as a nonlinear function of the inventory. Previous empirical studies

mostly suggest that the theory of storage is valid in different commodity markets (see, e.g., Brunetti & Gilbert, 1995;

Geman & Ohana, 2009; Geman & Smith, 2013; Ng & Pirrong, 1994). The alternative indirect tests proposed by Fama

and French (1988) investigate the validity of the predictions of the theory of storage by applying the sign of the basis

that is adjusted for a risk‐free interest rate, to proxy the high and low inventory periods. In this method, a negative

(positive) sign for the basis corresponds to a low (high) inventory period.

While the theory of storage was initially developed for agricultural commodities that are subject to supply sea-

sonality due to harvest seasons (see, e.g., Kaldor, 1939; Working, 1948), its implications have been investigated in other

commodity markets, such as metals and energy (see, e.g., Brunetti & Gilbert, 1995; Cho & McDougall, 1990; Fama &

French, 1987; Gorton, Hayashi & Rouwenhorst, 2013; Ng & Pirrong, 1994). The theory is widely accepted by energy

market participants, whereby, as Cho and McDougall (1990) argues, the sign of the basis is known by energy market

participants as a signal to decide whether to draw energy products out of storage or to store commodities.

In the analysis of the basis behavior, the relationship between commodity and other financial markets is often

ignored. Instability in the financial conditions of an economy can be a driving factor behind the commodities’basis

variation, as it often affects price dynamics. With the growing financial liberalization of commodities, financial market

conditions have become an important factor in the behavior of commodity markets (see, e.g., Adams & Glück, 2015;

Basak & Pavlova, 2016). The association between the commodity and other financial markets became more important

in particular after the 2000 crash in equity markets. As commodities had a negative correlation with the stock market,

after the 2000 collapse in the equity markets, commodities became a popular asset class for portfolio diversification (see,

e.g., Gorton & Rouwenhorst, 2006). Therefore, the financialization of commodities increased rapidly, and institutional

investors built their positions in commodity futures (Basak & Pavlova, 2016). The U.S. Commodity Futures Trading

Commission (CFTC) reports a dramatic boost in the institutional holdings of commodities, which increased from

15 billion dollars in 2003 to more than 200 billion dollars in 2008. In this regard, since 2004, commodities have been

recognized as a new financial asset class. With this boost in the financialization of the commodity markets, the

association between commodity futures and equity markets surged. This has led to an increase in information trans-

mission between the financial and commodity markets. Consequently, in addition to the inventories, financial market

conditions also became one of the key factors which influence the commodity futures prices. Moreover, as Nazlioglu,

Soytas, and Gupta (2015) argue, the oil price and financial stress are related through their impact on economic activity

and on investors’behavior. The effect of financial shocks is found to be asymmetric, depending on whether the

economy is in a high or a normal financial stress period (Evgenidis & Tsagkanos, 2017; Nazlioglu et al., 2015).

Sari, Soytas, and Hacihasanoglu (2011) show that a higher uncertainty in stock markets adversely affects the energy

futures prices, and Cheng, Kirilenko, and Xiong (2015) used the volatility index (VIX) to predict changes in trading

patterns in the commodity futures markets.

To our knowledge, there is no previous study in the literature that investigates the effect of financial stress on the

crude oil basis. To address this issue, this study examines the theory of storage and its implications in the crude oil

market, by considering, for the first time, the role of financial market conditions. We proxy financial instability by the

financial stress index (FSI) of the Federal Reserve Bank of St. Louis. This index represents the financial stress in the

whole market. Moreover, we control for the stock market uncertainty by the VIX. The applied data cover the period

from January 1994 to August 2017 in monthly frequency. First, we use the Fama and French (1988) indirect tests of the

theory of storage. Then, we perform a direct estimation to investigate the validity of the theory and the relative

importance of the financial market conditions. To achieve this, we apply the threshold structural vector autoregression

(TSVAR) approach proposed by Balke (2000) to examine the time‐varying responses of the interest‐adjusted basis in the

crude oil market to the U.S. petroleum inventory shocks, financial stress shocks, and shocks to stock market un-

certainty. The TSVAR approach has two main advantages. First, it enables us to estimate the model during different

regimes, namely: high and low inventory, and high and low financial stress regimes. Second, it decomposes the

responses of the basis to positive and negative shocks. Consequently, the nonlinearity and asymmetry of the basis

AHMADIA ET AL.

|

1161