Financial Stress and Basis in Energy Markets.

• Author: Behmiri, Niaz Bashiri

1. INTRODUCTION

   Increased financialization of commodity markets developed the association between financial and commodity markets. Therefore, besides the physical inventories, the changes in financial market conditions became an important infuential factor on commodity futures prices. In this study, we suspect that the conventional inventory-based models, only measuring the relationship between the commodities basis and inventory under the theory of storage, are not sufficient to explain and predict the changes in the spread between spot and futures prices, the so-called basis. The goal of this study is to understand the relationship between physical inventory, the US financial stress and the basis in the crude oil, heating oil and natural gas markets. Therefore, this study contributes into two groups of literature. The first group contains studies investigating the theory of storage, measuring the relationship between commodities bases and their physical inventories; the second group measures the association between financial condition and commodities futures prices.

   The first group that investigates the relationship between commodities futures prices and their physical inventories has a long tradition in the theory of storage literature. Confirming the main implications of the theory of storage, a large number of studies have found a positive effect from a higher physical inventory on the commodity markets bases (See e.g., Cho and McDongall, 1990; Ng and Pirrong, 1994; Geman and Ohana, 2009; Gorton et al., 2012; Geman and Smith, 2013; Buyn, 2017). In crude oil market, Alquist and Kilian (2010), based on the analysis in Pindyck (1994, 2001), propose a theoretical model of the oil spot and futures markets that incorporates the convenience yield to explain the causes of the changes in basis, or equivalently, the difference in the variability of oil futures and spot prices. Their results show that, first, the difference between the spot and futures prices is equal to the convenience yield; and second, consistent with the findings of Kilian (2009), it is the uncertainty about shortfall of future supply relative to future demand, rather than supply or demand shocks themselves, that provides economic explanation for the fuctuations in the basis.

   However, besides the physical inventory, the energy commodities bases can be affected by the financial condition. When financial stress increases, economic activity slows down, and this decreases the energy demand and their prices. Financial stress can affect economic activity through the bank lending channel, as higher financial stress decreases the available amounts of credits via changes in creditworthiness of borrowing businesses (Nazlioglu et al., 2015). On the other hand, increased financial stress will cause investors to change their portfolios and this will have an impact on energy prices. Consequently, changes in the energy prices can lead to a change in their basis, as we expect a higher change in the nearer futures prices than the more distant ones.

   The relationship between financial condition and energy commodities bases increased during the recent years as participation of investors in commodity markets increased. After the 2008 equity market collapse, commodities, as an asset class, became very popular for portfolio diversification, due to their negative correlation with stock market (see, e.g., Gorton and Rouwenhorst, 2006). As a result, since 2003 the financialization of commodity markets grew and institutional investors rapidly built their positions in commodity futures markets (Basak and Pavlova, 2016; Byun, 2017). The US Commodity Futures Trading Commission (CFTC) shows that institutional holdings increased from $15 billion in 2003 to over $200 billion in 2008. With increased financialization of commodity markets, the correlations amongst the commodity futures markets, as well as between the equity and commodity futures markets increased. This significantly enhanced the information transmission between financial and commodity markets. Therefore, besides the physical inventories, the changes in financial market conditions became an important infuential factor on commodity futures prices, especially after the 2008 global financial crisis.

   In this study, we examine whether the relationship between the energy commodities bases, inventory and financial stress changed during the time span of this study. To better clarify our contribution, we discuss the group of literature that examines the association between financialization and commodities prices or financial condition and commodities prices. In this regard, for instance, Singleton (2014) focuses on the relationship between financialization and crude oil price, investigates the effect of growth of investors in commodity index funds and spread trades by hedge funds on excess returns of futures market contracts of crude oil. He concludes that after controlling for several factors, there is a significant effect from investors order flow on crude oil futures prices. The author argues that information frictions and the associated speculative activity cause the crude oil prices to drift away from their fundamental values. Ready (2018) focuses on the relationship between crude oil price and risk shocks in stock markets and finds that most of the crude oil price variation is explained by structural supply and demand shocks; however, there is also a role for risk shocks during and after the crisis period. Some more studies find a negative effect of higher stress in financial markets on commodity futures prices (see e.g., Sari et al., 2011; Chen et al., 2014). Moreover, a set of studies find a higher correlations between the stock and commodities futures markets during or after the 2008 crisis (see e.g., Tang and Xiong, 2011; Daskalaki and Skiadopoulos, 2011; Delatte and Lopez, 2013; Silvennoinen and Thorp, 2013; Nazlioglu et al., 2015; Cheng et al., 2015; Buyuksahin and Rob, 2014; Junttila et al., 2018). Buyuksahin and Rob (2014) argue that the increased correlation between commodity and equity markets is in part due to higher hedge funds' participation in futures markets. Among the studies concentrating the bases-financial markets rather than futures prices-financial markets relationship, Bailey and Chan (1993) examine the association between the commodity markets bases and financial markets and find the presence of common influences across them. They relate the strength of this correlation to the exposure of spot commodity prices to macroeconomic shocks. However, in a recent article, Byun (2017) finds that the crude oil price-inventory relationship is stable over time and the contribution of financial investors' participation to the crude oil market is weak.Therefore, majority of previous literature supports the presence of a relationship between financial markets and commodity futures prices. This causes the conventional basis-inventory models to be biased and misleading for the energy market participants about their decision makings in this market. As Cho and McDougal (1990) and Serletis and Hullerna (1994) argue, 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. Therefore, the variation of the basis is an important factor in setting the efficient hedging strategy and improving profitability, and an incorrect perception of the basis behaviour can lead to unfavorable hedging results.

   In this study, we focus on the effects of physical inventory and the US financial stress on the bases in energy commodities. We chose crude oil, natural gas and heating oil as the leading energy commodities. Based on the CME (1) group data, these three commodities are the most actively traded energy commodities and crude oil is the most actively traded commodity in the world. Therefore, the correct prediction of their basis behavior is crucial for the traders in these markets. Moreover, it is important for companies of oil and gas exploration and production, oil refiners and marketing players, the airlines, and companies involved in petrochemicals.

   Our study consists of different estimations. First, we estimate the effect of financial stress on the basis in energy commodities markets while controlling for inventory, before and after the 2008 global financial crisis. The results provide a primary insight on how the effects of inventory and financial stress on the energy commodities bases change after the 2008. However, there is a consensus in the literature that the responses of many economic factors to the financial stress shocks are nonlinear and depend on the level of financial stress in the economy (see e.g., Nazlioglu et al., 2015; Evgenidis and Tsagkanos, 2017; Aboura and Roye, 2017). To account for this phenomenon, we estimate the models including the quadratic terms of financial stress and inventory, as well as the interaction term between financial stress and inventory. The results enable us to understand the nonlinear dynamics of the energy commodities bases with respect to inventory and financial stress during the turbulent periods of the economy. Finally, we enrich our analysis by applying a very recent time varying parameter model developed by Bitto and Fruhwirth-Schnatter (2019), which provids several contributions to our study. The results visualize the gradual changes in the effects of inventory as well as financial stress on the energy commodities bases. Moreover, we observe the effects of financial stress on the bases before, during and after the turbulent periods of financial markets. We apply the end of period monthly data for the time span from January 1994 to December 2018.

   Our results imply that, first, only after the 2008 financial crisis, there are evidences for a positive effect from the increasing level of financial stress on the energy market commodities bases. The effect of inventory...