Is Oil Price Still Driving Inflation?

• Author: Renou-Maissant, Patricia

1. INTRODUCTION

   The oil shocks and stagflation that characterized the 1970s led to a great deal of research on the effects of oil prices on the economy. Many empirical studies have shown that oil price shocks affect output and inflation. However, there is no consensus on the magnitude of the oil price effect in explaining recession episodes, and many studies have indicated that the relationship between oil price and the macroeconomy has changed over time. First, several studies (Mork, 1989; Mork et al. 1994; Hamilton and Herrera, 2004) support the position that the relationship between oil prices and macroeconomic aggregates broke down in the mid-1980s, the oil price collapse that occurred in 1986 having not produced an economic boom. Other authors, among them Kilian (2009), Blanchard and Gali (2007a), Hamilton (2009), Segal (2011), and Blanchard and Riggi (2011) point to a reduced impact of oil price shocks on macroeconomic aggregates over time. Indeed, since the late 1990s, the global economy has experienced two oil shocks comparable to those of the 1970s, but in contrast with the earlier shocks GDP growth and inflation remained relatively stable in much of the industrialized world until the financial crisis. Finally, authors including Bernanke et al. (1997), Barsky and Kilian (2004), Kilian and Lewis (2011), and Blinder and Rudd (2008) argue that oil price shocks have never been a major factor in macroeconomic cycles, even in the 1970s.

   Some studies have focused exclusively on the issue of pass-through of oil prices into inflation, including for example Hooker (2002), LeBlanc and Chinn (2004), van den Noord and Andr6 (2007), De Gregorio al. (2007), Blanchard and Gali (2007a), Chen (2009a, 2009b), Clark and Terry (2010), and Fukac (2011). This issue is particularly important for the implementation of monetary policy, and remains relevant in a context of low oil prices. Indeed, it is now widely accepted that one of the main goals of monetary policy is the pursuit of price stability. To ensure this objective, those responsible for monetary policy must assess the impact of oil price changes on inflation and utilize the appropriate tools to control inflation. Understanding inflation dynamics is especially important today due to the very low inflation levels prevailing in many countries. When inflation is far below target, central banks' tolerance for further negative inflation impulses can be low. Furthermore, there may be a high risk that inflation expectations will fall as a result of the drop in oil prices. In a context of falling oil prices with a weak global growth environment and with nominal interest rates constrained by the zero lower bound in the advanced economies, monetary policy should react forcefully to stimulate economic activity, maintain the anchoring of inflation expectations, and prevent deflation risks.

   In this paper we investigate the effects of oil price changes on inflation using an augmented Phillips curve framework. As widely suggested in the literature, the pass-through of oil prices into inflation has evolved over time and is still very much in a state of change. That is why, in line with Chen (2009a), we assume that the instability of the oil price pass-through may be gradual and we use a time-varying coefficients model which is particularly well adapted to take an on-going process into account. In addition, we implement an unobserved components model (Harvey, 2011; Stella and Stock, 2015) to take into account the persistence of inflation. We consider eight industrialized countries, namely the United States, Canada, Japan, Australia, Germany, France, Italy, and the United Kingdom over the period 1991-2016. This period is characterized by the adoption of inflation targeting by most central banks in the early 1990s, and by a low level of inflation. The aim of our work is to better understand the dynamics of inflation in a context of low inflation.

   Our analysis differs from the existing literature on several points. First, the analysis is based on a new methodology using an unobserved components model in a state space framework. This allows a time-varying pass-through and avoids some difficulties in the specification of the Phillips curve, which involves a regression on unobserved variables, namely the output gap and inflation expectations. Second, our data set extends to 2016 and includes both the financial crisis and the falling oil prices initiated in June 2014, events that are not covered by previous studies. Third, because we employ a common methodology across countries and hold the sample period fixed, our results are directly comparable across countries, so we can assess both how the oil price pass-through evolves over time and what are the common features between the countries under study.

   We establish that even in a low and stable inflation period, oil prices play a significant role in the dynamics of inflation. Furthermore, the results show evidence of significant time-varying oil pass-through, as well differences between countries. In all the countries, except Germany the oil pass-through into inflation increased from the early 2000s up until the global financial crisis, and in the last fifteen years it has nearly doubled in the United States.

   The paper is organized as follows. Section 2 offers a literature review. Section 3 provides a brief description of the evolution of oil prices and inflation. Section 4 briefly reviews the theory and the empirical methodology. Section 5 reports and discusses the empirical results. The final section concludes.

2. LITERATURE REVIEW

   The pass-through from oil prices to inflation is usually examined using a vector autoregression (VAR) or an augmented Phillips curve (APC) with oil prices. Whatever the model used, there is some evidence that the pass-through has sharply declined since the early 1980s. This evolution suggests that a linear, constant coefficient specification may not accurately capture the effects of oil price fluctuations on inflation. Consequently, time variation has to be allowed for in order to adequately model the pass-through from oil prices to inflation, and to explore how this relationship has evolved over time. Some authors have argued that this breakdown of the relationship reveals that the relationship between the variables is non-linear, and thus have proposed different specifications of it, particularly asymmetric ones. These specifications can be implemented either by introducing two separate variables for oil price increases and decreases, as in the models suggested by Mork (1989) and Mork et al. (1994), or by estimating Markov regime-switching models characterized by high and low inflation periods (Catik and Onder, 2011), or even by using a quantile regression framework to estimate the marginal effect on inflation in the distribution (Chortareas et al, 2012). Other authors have suggested different ways to take time variation into account, either by splitting the sample into two sub-periods assuming a structural break in the early 1980s, by estimating regressions over rolling time windows, or by using time-varying parameters models. Table A1 in Appendix presents the key features of main studies carried out since the early 2000s.

   As mentioned above, a broad consensus emerges regarding the decline of the pass-through from the mid-1980s. The structural break appears robust to a variety of specifications and for many industrialized countries. Hooker (2002) showed strong evidence of a structural break, the oil shocks having contributed substantially to core inflation (inflation excluding energy and food prices) until 1981, but the pass-through having become negligible since. LeBlanc and Chinn (2004) concluded that the sharp oil price increase experienced in the 1990s had a modest effect on inflation, although differences in the size of the effects exist across countries. Van den Noord and Andre (2007) and Clark and Terry (2010) showed that the effects of energy shocks on core-inflation were sharply diminished in comparison with the 1970s and remain muted. De Gregorio et al. (2007) identified a drop in the average estimated pass-through for industrial economies and, to a lesser degree, for emerging economies. Blanchard and Gali (2007a) have reported much larger effects of oil price shocks on inflation in the first part of the sample, i.e. before 1984. Chen (2009a) found evidence of declining pass-through for almost all the countries considered. However, he noted that the percentage change in pass-through is negligible for some countries, especially in the United States and Australia. Fukac's results (2011) are consistent with previous studies: they pointed to the decline in the pass-through of oil prices to inflation after 1983.

   Many arguments have been put forward to explain the decline in pass-through since the mid-1980s (see Chen, 2009a). However, more recent studies, based upon up-to-date data, suggest that oil prices have been playing a larger role in the inflation process since the late 2000s. Fukac (2011) showed that the pass-through exhibits a structural break before the great recession of 2007, and has increased over the period 2000-2010 in the United States. Although the pass-through is still low compared with its level in the 1970s and 1980s, its increase since the early 2000s is statistically significant, and it has almost doubled over the last ten years. Fukac offers various explanations in explaining these patterns. First, the share of consumer spending on oil and petroleum products, which fell in the 1990s, has increased to levels last observed in the 1970s in the United States. Second, the "financialization" of commodity markets may have contributed to the increase in oil pass-through. Since the early 2000s, non-energy commodity prices have become increasingly correlated with oil prices. Fuka[pounds sterling] noted that this situation is similar to the 1970s and early 1980s when non-energy...