Oil price and core inflation
The effect of a change in oil price upon core inflation in emerging countries and developed countries.
By Sophie Greuter 10656154
June 2016
University of Amsterdam Faculty Economics & Business
ABSTRACT
This paper examines the effect of a change in oil price upon the core inflation for developed countries and emerging countries. A dataset is obtained for the 2000-2014 period. The regression model includes the variables GDP growth as control variable and the change in oil price. The core inflation is the dependent variable in this paper. The results show that a change in oil price upon core inflation has not a significant effect in developed countries, but has a significant effect in emerging countries. There are several reasons that can explain this conclusion. The most important reason is that the oil-intensity (measured as oil consumption/GDP ratio) is lower in the developed countries than in the emerging countries.
Contents
1. Introduction 4
2. Theoretical framework 6
2.1 Oil developments 6
2.2 The relationship between oil prices and inflation 8 2.3 The difference between emerging countries and developed countries 10
3. Literature review 13
4. Empirical analysis 17
4.1 Data and sample 17
4.2 Variables 17 4.3 Model 19 4.4 Hypothesis 20 4.5 Empirical results 20 4.6 Internal validity 23 5. Discussion of results 24 5.1 Conclusion 24 5.2 Future research 25 6. References 26 7. Appendix 28
1. Introduction
In June 2014 the oil price fell by 40% as a result of a weaker demand and an oversupply in oil (Economist, 2014). The weaker demand was caused by the slow economic growth in Europe and China and the oversupply of oil due to the
overproduction of oil in the OPEC countries (Economist, 2014). A salient feature of the recent oil price shock is that the sharp decline in the oil price had not affected the inflation in June 2014. This event is inconsistent to the period 1970-1980, when the inflation experienced a sharp increase due to the higher oil price.
(Hooker, 2002).
Nowadays, there is an ongoing discussion about the relationship between the oil price and the inflation. Since 1970, the oil price pass-trough into inflation was high, because oil was important for the production and consumption. However, in the last two decades different studies have examined that this effect has decreased
(Gregorio et al., 2007, Killian, 2009, LeBlanc, 2004, Hooker, 2002). Several reasons can explain why the oil price pass-through might have declined around 1980. One reason is that oil is less intensively used due to the fact that oil can be substituted for other energy sources. Another reason is the low inflation environment caused by a more active monetary policy that keeps the inflation around 2% (Kilian, 2009).
Before 1980 the change in oil price upon inflation only had a significant effect in developed countries, but since the 21th century the oil demand increased in emerging countries (Kilian, 2009). The intensity in oil consumption is different per countries and changed over time. Some countries are more dependent on oil, thus more sensitive to a change in the oil price than less intensive countries (Égert, Jimenez-Rodriguez and Morales-Zumaquero, 2010). Gregorio, Landerretche, Neilson, Broda and Rigobon (2007) examined the difference between countries, especially between emerging countries and developed countries. He found that the oil price pass-through into inflation is lower for developed countries than emerging countries since 1980.
The change in oil price does not only have an effect on the inflation but also on the core inflation. Core inflation is the inflation in the long run. The core inflation excludes the change in energy prices and food prices because these are quite volatile. The effect of a change in oil price has a smaller effect upon the core inflation than the inflation (Hooker, 2002). No studies have examined the effect of change in oil price upon the core inflation in the emerging countries and developed countries to give insight in the difference between the two groups of countries since the lower oil price pass-through in 1980.
The goal of this paper is to shed light on the change in oil price upon core inflation and if this effect is different between the developed and emerging countries. The model in this paper takes data from 2000-2014 for 13 developed countries and 13 emerging countries to analyse the effect between the developed countries and emerging countries. This paper doesn’t describe the effect of a change in oil price upon inflation for a specific event but over a period of several years. The main hypothesis is that the oil price pass-through into core inflation has a significant effect on emerging countries and an insignificant effect on developed countries. Since 1980 the oil intensity is decreased in developed countries but increased in the emerging countries due to the change in industry in both groups of countries.
The paper is organized as followed: chapter 2 describes the oil price development resulting from the relationship between the oil price and inflation.
Finally, chapter 2 explains the difference between emerging countries and developed countries with regard to the oil price-inflation relationship. In chapter 3 the literature regarding the oil price and inflation will be discussed. In chapter 4 the empirical model will be discussed. It starts with a short explanation of the data and sample, followed by a clear description of all the variables that are included in the model. Before executing the regression analyze the hypothesis is given. Finally, the
regression results are published and discussed. Chapter 5 describes the conclusion of this paper and discusses some possibilities for future research.
2. Theoretical Framework
The oil shocks and inflation are mostly positive correlated. For a clear understanding of the relation of the change in oil price and inflation first some background
information is given below. This chapter gives an overview of the oil price
development from 1973 until 2014, to give insight in the relation between the oil price and inflation. The last paragraph describes the difference between emerging
countries and developed countries, concerning the relationship between the change in the oil price and inflation.
2.1 Oil price development
Oil became an important source for energy over last decades. To analyse the oil price development, the years after 1945 are relevant. Before 1945 the oil
consumption and production were only a small part of the energy consumption and production (Yan, 2012). Also the oil price had a smaller impact on the world economy than it does now (Yan, 2012). Since 1945, oil became more important in the world economy, the technology was at a higher level and the oil consumption and
production increased due to the increase in usage of automobiles as the main source of transportation (Yan, 2102).
By 1967 oil became the major source of energy in the world and took up more than forty percent of the total spending on energy (Yan, 2012). But on October 17th 1973, the first oil crisis began and the oil price increased dramatically. The cause of this crisis was the ongoing Yom Kippur war. The OPEC members, Egypt and Syria announced a 5% reduction of the oil production (Kesicki, 2009). From October 1873 to January 1974 the oil price increased from 2.59 dollar to 11.65 dollar per barrel (Kesicki, 2009). Refer to figure one, in 1979, the second oil crisis followed, due to the Iranian revolution. The global oil supply decreased by 4%, driving the price far higher (Kesicki, 2009). In 1980 the oil price increased further, due to the Iran-Iraq war, the oil production was stopped in Iran. The two oil crises had a great impact on the world economy because of the important role that oil had played in the industry and
contemporary life since 1965 (Yan, 2012). The three oil events preceded by a high economic growth followed up by an economic recession (Leblanc and Chinn, 2004).
After 1980, the oil price decreased and in 1986 the oil price decreased even more from $67.01 in November 1985 to $22.31 in March 1986, see figure 1. The oil production decreased in Saudi Arabia: one of the large export countries of oil,
relatively constant. However, in 1999 the oil price started rising and peaked in the summer of 2008 (Hem and Kamel, 2015). Hem and Kamel (2015) give different reasons for the increase in the oil price. Firstly, in this period the emerging countries used far more modern transportation, which led to an increase in the demand of oil. Secondly, the pressure on the demand of oil increases, because the production in both developed and emerging countries increase. And third, in developed countries the agriculture sector started using more mechanized tools that increased the demand for oil. But after the summer of 2008, the oil price fell to $46.86 in January 2009. The weak demand in Europe and the stronger exchange rate of the US dollar are explanations for this phenomenon.
Recently, in 2014 the oil price fell by 40% (Economist, 2014). An explanation for the sharp decline in the oil price might be that the oil supply exceeds the demand. The demand for oil is low because of the slow economic growth in the Eurozone and China, that leads to a smaller consumption of oil. But on the supply side the OPEC, the Organization of Petroleum Exporting Countries, decided to maintain the
production at the same level, 40% of the world market (Economist, 2014) In figure 1 the development of the oil price is shown. The horizontal axis presents the years from 1970-2015 and the vertical axis is the price of WTI oil per barrel in dollars.
2.2 The relationship between inflation and oil price
The direct relationship between the change in oil price and inflation became important in the 1970s. There was an increase in interest for the impact of the
change in oil price upon inflation. The international oil price became important for the price behavior of a company, because oil is an essential source of energy for the production and consumption.
The change in oil price and inflation are often seen as a causal relationship. The oil price can affect the general price level, the inflation, because oil is widely used as input for production and distribution. An increase in the price of oil raises the costs of firms. The firm can charge the cost to the general price level leading to an increase in the inflation rate. If the oil price increases there will be other items that are close substitutes. The demand for this substitutes increase thus leading in a higher price. The overall price level will increase to an increase in the oil price (Cavallo, 2015). This has influence on the expenditure pattern of households.
Figure 2 shows the price movement of oil and the inflation rate in the U.S. The right axis shows the percentage of the change in oil price and the left axis shows the percentage of change in the inflation rate. The oil price and the inflation rate in the U.S. have an almost similar pattern of increasement and decreasement.
The correlation coefficient of the U.S. between the change in oil price and inflation is calculated to show the relationship between the inflation rate and the change in oil price. For the period of1970-2014, the correlation coefficient for the U.S. is 0.66. This coefficient proves a strong relation between a change in oil price and the inflation. However, the oil price fluctuations are more volatile than the inflation rate in the U.S (LeBlanc and Chinn, 2004). The reason for this difference is that more factors cause the inflation rate, for instance the exchange rate and the unemployment rate. In 1973 the oil price rose from $3 per barrel to more than $40 per barrel in 1979 and the price index doubled from 41.2 to 86.3 during this period. This high inflation rate was mainly caused by the sharp increase in the oil price due to the first and second oil price (LeBlanc and Chinn, 2004). In 1990 the oil price doubled, but the inflation rate stayed relatively stable. As seen in the figure 2, the rise in the oil price has a smaller impact on the inflation. The same applies to the oil shock in 2000, which caused a small change in the inflation rate despite the doubling of the oil price. However, the sharp decline of the oil price in June 2014 did not lead to a fall in the inflation (Castro, Jerez and Barge-Gil, 2016). Castro et al. (2016) concluded that a deflationary scenario of oil prices would not lead to a negative
inflation rate in the Euro area for December 2016. The oil pass through to inflation is decreased over time. The correlation coefficient for the period 1970-1980 is 0.82, but the correlation coefficient for the period 1980-2014 is 0.1479 in the U.S. The
coefficient is lower in the second period. These correlation coefficients explain the decreasing in the oil price pass-through into inflation.
Figure 2: Source LeBlacn and Chinn (2004)
Since 1990, studies have found that the relationship is smaller since 1980 (Gregorio et al., 2007, Killian, 2009, LeBlanc, 2004, Hooker, 2002). There are a few factors that may explain this. The active monetary policy keeps the inflation around 2% (Chen, 2009). According to Taylor (2000) a lower inflation environment leads to a lower oil price through. According to Gregorio et al. (2007) the oil price pass-through is lower due to the fact that oil is substituted for other energy sources. The oil consumption has grown less much than the GDP per capita in the U.S, so the oil became less intensive in the industry of the U.S. The oil production and consumption is different for each country. An oil intensive country is more dependent on oil, so more sensitive for a change in the oil price than other countries (Égert et al. 2010).
The relation between the change in oil price and core inflation is something different than the relation between the change in oil price and inflation. To show the difference, the correlation coefficient is also calculated for the core inflation. The correlation coefficient is 0.2812 in the period 1970-2014. This is lower than the correlation coefficient of the change in oil price and the inflation (0.66). The same applies to the different periods of 1970-1980 and 1980-2014. The correlation
coefficient is lower in both periods than the correlation coefficient of the change in oil price and inflation, 0.55 and 0.04 respectively. The relation between the change in oil price and core inflation is thus lower for all the periods. The explanation is that the core inflation predicts the long run inflation better than the overall inflation itself (Cavallo, 2015).
2.3 Difference between emerging countries and developed countries
Many studies focus around the effect of a change in oil price upon inflation in an emerging country or a developed country. A few studies focus on the difference between emerging countries and developed countries. This part describes the relation between a change in oil price and the inflation in emerging countries and developed countries and the difference between these two groups of countries.
Since the 21th century, the oil demand is more driven by the emerging countries, like China, India and Brazil (Gregorio et al., 2007). This economy is more powerful to increase the oil price as result of the subsidy for fuel in emerging
countries (Kesichi, 2009). Another part is due to the changing structure of the U.S industry. The production in the U.S. is less energy intensive, thereby less dependent on oil than in the emerging countries (LeBlanc and Chinn, 2004). The effect of a change in oil price inflation is not equal across countries. In addition, the economic importance of oil has declined for developed countries, because developed countries move from an industrial sector to a more service oriented economic (LeBlanc and Chinn, 2004). However, not all countries have achieved an equal reduction in oil consumption.
The two graphs below show the development of the oil price in dollars and the inflation in industrial countries and developing countries from 1970:Q1 until 2006:Q2. The left scale is the oil price in dollars and the right scale is the average inflation rate. The inflation rate is the grey line and the black line is the oil price. Be aware of the different scale for each graph. The inflation is different between the two groups of countries. In the industrial countries, the inflation rate was high before 1980. With emphasis on the oil crises in 1973 and 1979, the inflation is higher. Oil was important for the industry, so a change in oil price affected the inflation.
However, since 1980 the inflation rate started to decrease to a level of 2% per year in 2006. This can be explained by the lower oil price pass-through into inflation since 1980. The inflation is around the 2%, because the monetary policy pays close attention to a low inflation rate to prevent deflation.
For the developing countries, is the inflation rate lower than the industrialized countries between 1970-1980. However, since 1980 the inflation was higher than in the industrialized countries because these countries used more oil than before. Since 1996 the inflation rate began to fall to a level of 5% per year in the developing
countries in 2006. Overall, before 1980 the inflation in industrial countries was higher than in the emerging countries, but since 1980 the inflation in the developing
attracted energy-intensive industries from the industrial countries. But for the two groups of countries applies that the inflation is lower than it was in the period 1970- 1980. The low oil price pass-through is visible in the two graphs. Before 1980, the inflation is more volatile than after 1980. After 1980 the inflation is more constant despite the large oil price shocks.
Figure 3: source: LeBlanc and Chinn 2004
To give better insight in the relation between oil price and inflation in
developed countries and emerging countries, the oil price pass-through into inflation is calculated. In figure 4 below, Gregorio et al. (2007) has calculated the oil price pass-through into inflation. The oil price pass-through is defined as a ratio for the change in oil price and the inflation (Gregorio et al., 2007). Gregorio et al. (2007) shows in his figure the four important oil price shocks of 1973, 1979, 1999 and 2004. The first row gives the pass-trough coefficients in industrial countries and emerging countries. The coefficient is almost equal but differs in the late 1979s. At the time of the shock in 1999, the oil price pass-through had fallen in both countries (Gregorio et al., 2007). The oil price pass-through fell further during the second Gulf War in 2003, especially in the industrial countries. Through the years, the oil price pass-through decreased in both countries but the fall in the oil price pass-through is stronger in the industrial countries than in the emerging countries. Gregorio et al. (2007) points out that one of the main causes is the large economic growth in China. China is an important emerging country and the demand for was high. So the large oil price shock influenced China more than the industrial countries that use less oil for the production process.
The second row shows the inflation rate in the different countries. A higher oil price through is associated with higher inflation. In this table the oil price
pass-through is higher in emerging countries than in the industrial countries, so the inflation in percentage per year is higher in the emerging countries than in the industrial countries (Gregorio et al., 2007). The third row shows the change in inflation. The first two oil shocks, in 1973 and 1979 had a large impact on the inflation, but the last two oil shocks show a lower pass-through.
Developed countries and emerging countries show a strong reaction to the oil shock of 1970 and 1979. There is also a positive, but smaller reaction to the shock in 1990 (Gregorio et al., 2007).These results indicate that the oil price pass-through is lower for both countries. However, the effect of a change in oil price upon inflation is larger in emerging countries than in developed countries.
Figure 4: Source: Gregorio et al. (2007)
To study the change in oil price and the inflation, the relationship is still visible. It is clear that the effect of a change in oil price upon inflation is different between countries, especially emerging countries and developed countries. But since 2000, the oil price pass-through is lower for both countries, so there is still a debate about the effect of a change in oil price upon inflation for emerging countries and developed countries.
3. Literature review
This chapter describes a part of all the literature and researche on the relationship between the change in oil price and the inflation. There is a lot of literature that describes the relationship between the change in oil price and macroeconomic variables, in particular inflation and GDP growth. Many studies have investigated the relationship between the change in oil price and inflation in developed countries since the first oil crisis of 1973. Since 2000, more studies have focused on the
emerging countries, while only a few studies have looked into the difference between emerging countries and developed countries.
Hamilton (1983) and Darby (1982) are the first and most influential papers in this field focus on the oil crisis period, 1973-1980. Hamilton (1983) investigated the change in oil price upon inflation and GDP growth and focused only on the U.S. data. He found a positive significant relationship between the change in oil price upon GDP growth and inflation. Hamilton (1983) concluded that oil price shocks were an
important factor in almost all US recessions from 1949 to 1973. 7 Out of 8 recessions were caused by the large oil price shocks between 1949-1973. Darby (1982) focuses on United States, United Kingdom, Canada, France, Germany, Japan, Italy and the Netherlands, different industrialized countries. He took data from 1971-1974, the years of the first oil crisis. Darby (1982) found that the first oil crisis influenced the inflation and caused the recession in these years.
These empirical studies found a positive significant effect between oil price shocks and inflation during the two oil crises. But more recently, empirical studies found a smaller effect on the relationship between oil price shocks and the inflation. However, nowadays the oil price pass-through to the inflation is lower in comparison to the seventies. The overall reason is that oil is less important these days than in the seventies because oil can be substituted for other energy sources nowadays.
LeBlanc and Chinn (2004) examined the effect of a change in oil price upon inflation and focus on the G7 countries1. The results show that the change in oil price upon inflation has a positive significant effect in the U.S., Japan and Europe. But the effect is different between these developed countries. The effect of change in oil price upon inflation is larger in Europe than in Japan and in the U.S. LeBlanc and Chinn (2004) give two reasons for this phenomenon. The first reason, the labor unions are more powerful in Europe than in the U.S. and in Japan. The labor unions claim a higher wage in response to higher consumption prices for oil. The second
reason is that the competition is less intense in Europe than in U.S. Producers pass along the higher wage costs to consumers resulting in higher consumer prices in Europe. This is not possible in the U.S because of more competition. Hence the wage-price spiral is more active in Europe, so the oil price pass-through into inflation would be higher in Europe than in U.S. and in Japan. Despite the positive significant effect in these countries, LeBlanc and Chinn (2004) describe that the oil price has a lower pass through in comparison to the seventies, because the production in the U.S. is less dependent on oil, the economy is more deregulated and flexible in Europe and the world economy is more capable of absorbing the oil price shocks.
Before 2000 the studies focus mainly on developed countries, because the usage of oil was limited in emerging countries. After 2000, more studies focused also on emerging countries.
Du, He and Wei (2010) focus on the emerging country China, the second largest oil consumer after the U.S. They examined the relationship between the world oil price and China’s inflation. Du et al. (2010) used the method of multivariate vector auto regression (VAR) with monthly time series from 1995:1 to 2008:12. The
conclusion was that the change in oil price does not affect the inflation in China for the years 1995:1-2001:12. However, the sample 2002:1-2008:12 shows that the oil price change affects the inflation. The results show that the change in oil price affects the inflation in China since 2002. Du et al. (2010) expect that the effect of a change in oil price becomes higher significant in contrast to developed countries. China’s dependence on oil has increased during the past decades. This sharp increase in oil is due the upcoming position of China.
Also Adam, Rianse, Harafah, Cahyono and Rafiy (2016) focus on an emerging country, Indonesia. This study investigates the effect of a change in oil price upon inflation in the period between January 2004 and September 2015. Indonesia is one of the developing countries that has oil refinery facilities to produce oil, but Indonesia became an oil importing country in 2003. Indonesia became an oil importing country because the domestic demand exceeds the production. Since 1999 the oil production has decreased (EIA, 2015). The results show that the oil price has a significant effect upon inflation. If the oil price increases with 1% the inflation rate increases by 0.33%.
Gregorio et al. (2007) is one of the few studies that describes the effect of a change in oil price upon inflation for emerging countries and industrialized countries. They did research in thirty-three countries, twenty industrialized and thirteen
emerging countries. Before 1980 a doubling of the oil price was passed-through as an increase of 15 percentage points in inflation. But after 1980 a doubling in oil price
leads to a 3 percentage point higher inflation rate for developed countries. Gregorio et al. (2007) found a lower oil price pass-trough to inflation in developed countries but to a lesser degree for emerging countries. Because the emerging country increased the energy intensity, maybe due to fact that the emerging countries take over the energy-intensive industries from the developed countries.
Hooker (2002) investigates the effect of a change in oil price upon core inflation in the U.S. Core inflation is the inflation minus energy prices and food prices. This is a good indicator of the underlying long-term inflation, because the oil and food prices are volatile over time. Hooker (2002) estimated a model for the U.S. with core inflation as the dependent variable over the period 1962-1981. He researched two periods. In the first period, 1962-1981, he found a positive significant effect on the core inflation. However, in the second period, 1981-2000, he found that oil price has a positive insignificant effect on the core inflation in the U.S. This is not in line with the other studies that focus on the inflation. An explanation can be that the core inflation is more stabile during oil shocks in comparison to the inflation because the core inflation is excluded the oil price and food price. By definition, the impact upon core inflation will be less than upon total inflation.
Cavalla (2008) also also research on the change in oil price upon core inflation but over the period 1987-2008. The results show that the change in oil price upon core inflation has a positive insignificant effect for the U.S., U.K., and Canda, but a positive significant for Europe. An explanation is the lesser degree of
competitions in Europe as LeBlanc and Chinn (2004) also concluded. This only counts for the inflation, and not for the core inflation.
This paper further broadens the views of the existing literature in two ways. First way is to use the core inflation instead of the normal inflation rate according to Hooker (2002). In comparison to Hooker (2002), this paper contains a more recent dataset from 2000 until 2014, in comparison to Hookers dataset from 1962-2000.
The second way is to analyze the effect of oil price upon core inflation in emerging and developed countries instead of the U.S. Due to the low oil price pass-through since 1980 is there as significant difference between emerging and
4. Empirical analysis
This thesis examines the effect of a change in oil price upon the core inflation in emerging countries and developed countries. This chapter describes the empirical model and the results. First the dataset and sample are described that is used in this model. This is followed by an explanation of the dependent and independent
variables. The third paragraph gives the hypothesis, followed by the explanations of the model. Finally the results will be discussed and the internal validity of this model will be discussed.
4.1 Data and sample
The baseline regression uses data of the variables core inflation, GDP growth and the change in oil price. The dataset is obtained from the datasets World
Development Indicators, OECD and trading economics.
The sample time period is taken from 2000 until 2014 and each variable is measured in annual data. The dataset consists of 13 developed countries and 13 emerging countries. There exists a difference between the definition of emerging countries and developed counties. This thesis determines wheather a country is developed or an emerging country on the income criteria. If the gross net income per capita is less than $12735 the country is emerging. An income higher than $12735 the country is developed. Based on this income criteria, the emerging countries are: Brazil, China, Colombia, Egypt, Hungary, Poland, Russia, Turkey, Mexico, South Africa, South Arabia, India and Indonesia. The developed countries in the dataset are: Australia, Canada, Denmark, Finland, Iceland, Germany, Japan, Netherlands, Norway, Singapore, Sweden, United States and United Kingdom. These countries are pooled to two groups: developed countries and emerging countries. Each group contains three variables with each 15 observations.
4.2 Variables
This paragraph describes all the variables that are included in the model. As dependent variable the core inflation is used and the independent variables are a change in oil price and the GDP growth. How the independent variable is correlated with the dependent variable is explained below.
The core inflation is the dependent variable in this model. The core inflation is the inflation excluded from energy prices and food prices. The change in oil price has an indirect effect on the core inflation. The increase in oil price leads also to higher other energy prices, because this energy sources are substitutes (Cavallo, 2008). Workers claim a higher wage to compensate for the higher energy prices. The higher wage can be passed-though into the production prices whereby the core inflation will increase. This paper used core inflation, because core inflation is a better prediction for the long run inflation rate. The exclusion of food and energy price makes it more reliable indicator for the underlying trend in the inflation (Cavallo, 2008). Other measurements for the inflation are the consumer price index and the GDP deflator. The difference between these two is that the CPI contains the prices bought by the consumer including foreign goods, but the GDP deflator contains the price of all goods and services produced domestically. The key difference between the core inflation is that the CPI and GDP deflator are include with the food and energy prices. The data for the core inflation is obtained from OECD and trading economics. For the country Singapore the core inflation is obtained from Monetary Authority Singapore. The core inflation is measured in percentage per year. For the emerging country India is missing the data of core inflation from 2000-2003. The dataset has enough data points for the core inflation to give a good prediction for the weighted average over the difference countries each year. Thereby India is not a large oil producer or consumer.
The most important independent variable in this model is the oil price. There are different types of oils with each another price. Most of the oil prices depend on the World Texas Intermediate (WTI), Brent, Dubai and OPEC. This paper uses the WTI oil benchmark. WTI oil references two-thirds of all the crude oil contracts around the world. WTI is a light, sweet crude oil. Light and sweet oil is the most common oil in the industry for consumption and production.
The oil price is included in the model to test if the oil price changes have an effect on the core inflation. Annual oil prices are calculated by EIA from daily data by taking an unweighted average of the daily closing spot prices over the specified time period. The oil price is constant over all countries, but different over time, because it is complex to calculate the oil price over different countries due the taxes, price controls and the exchange rates.
Gross domestic product growth is the control variable in this model. GDP growth is correlated with the core inflation, because it is possible that a higher economic growth causes higher inflation. Higher output leads to a higher GDP growth following with higher wages. Due to the wage price spiral, the costs are
passed through the consumption prices and leads to a higher inflation (Hamilton, 2003). The GDP growth is extracted from WTI and measured in annual percentage growth rate based on local currency. GDP is defined as the total market value of the goods and services and some nonmarket production, education and defense.
4.3 Model
This paragraph describes the baseline model and the implementation of the regression. The baseline model is included the change in oil price, GDP growth as independent variable and core inflation as dependent variable
Model 1: 𝜋 = 𝛽!+ 𝛽!𝑜𝑖𝑙𝑝𝑟𝑖𝑐𝑒 + 𝜖
Model 2: 𝜋 = 𝛽!+ 𝛽!𝑜𝑖𝑙𝑝𝑟𝑖𝑐𝑒 + 𝛽!𝐺𝐷𝑃 𝑔𝑟𝑜𝑤𝑡ℎ + 𝜖
The first model is the relationship between core inflation and the change in oil price. Where 𝜋 is the core inflation measured in annual percentage and oil price is the change in oil price (Appendix I). The second model is included with the control variable GDP growth. This thesis examines if 𝛽! has a positive significant effect on core inflation for the emerging countries but is equal to zero for developed countries.
To test whether the change in oil price has a positive significant effect on core inflation for emerging countries and developed countries, two separated regressions be done. One regression for emerging countries and one for developed countries to test if the change in oil price has a positive significant effect on core inflation. This model includes 13 emerging countries and 13 developed countries. The average is calculated for each variable for each year from 2000-2014 over the two groups of countries. This resulted in 15 observations per variable for the two groups developed and emerging countries.
4.4 Hypothesis
Before doing the regression and interpret the results, I expect that the effect of a change in oil price upon core inflation has a positive significant effect among
emerging countries, but an insignificant effect for developed countries. Specifically in U.S. and Europe, in the recent oil price shock in 2014, the decreasing in oil price does not reduce the inflation.
From 1945-1980 the developed countries were more oil dependent, but after 1980 the energy intensity reduced. The usage of oil decreased and increased in the emerging countries since 1980 (Gregorio et al. 2007). Before 1980 the developed countries used more oil for the oil-intensive industry and the use of cars. The
production was mainly manual labor in emerging countries. The effect of a change in the oil price had more effect on the core inflation in developed countries than in the emerging countries. But after 1980, the developed countries used more electricity and other energy sources instead of oil. Also the structure of the industry changed in developed countries. The industry became less oil dependent because other energy sources served as substituted. Another reason is the outsourcing of industry
production to the emerging countries due to the more globalization around the world. This was accompanied by the fact that in the emerging countries the development from agriculture sector to industrialized sector began (Kesichi, 2009). In both sectors the amount of oil increase because industry needed more oil and in the agriculture sector more modern mechanize tools are developed that increased the usage of oil. So the emerging countries were more oil dependent than before. Since the 21th century the oil demand is more driven by the emerging countries (Gregorio et al., 2007). The oil intensive firms are more powerful to increase the oil price in emerging countries than in the developed countries. Because in the emerging countries subsidy is given for oil. (Kesichi, 2009). Hence the expectation is that the effect of a change in oil price is larger in emerging countries than in developed countries.
In formula:
𝐻!"#"$%&"! !"#$%&!"#: 𝛽!"# !"#$% = 0
A change in oil prices has no significant effect on the inflation 𝐻!"!#$%&$ !"#$%&'(): 𝛽!"# !"#$% > 0
4.5 Empirical results
This paragraph describes the results of the regression model. Four regressions are performed to explain if a change in oil price has an effect on the core inflation in developed countries and emerging countries.
Table 1- Regression results Dep. variable Core inflation Developed (1) Emerging (2) Developed (3) Emerging (4) GDP growth -0,1821 (-3,54)* -1.8514 (-3.63)* Oil price 0.000147 (0.03) 0.069 (2.01)* 0.0013824 (0.19) 0.1273 (3.49)* Constant 1.8867 (14.33)* 6.0982 (6.63)* 1.9958 (11.56)* 13.5290 (6.29)* R-squared 0.0467 0.1345 0.123 0.5493 N 15 15 15 15 ∗ 5% significant, … t − value
The first two regressions give the results of the effect of a change in oil price upon core inflation without the control variable GDP growth. The first one (1) gives the results for the developed country and the second model for emerging countries. The change in oil price has insignificant effect on the core inflation. If the oil price
increases with 1%, the core inflation increases with 0.000147%. In contrast to the emerging countries, the change of oil price has a positive significant effect at the 𝛼 = 5%. If the oil price increases by 1%, the core inflation increases by 0.069%. The R-squared is 0.0467 for developed countries and 0.1345 for the emerging countries. This means that 4.67% of the variance in the core inflation is explained by the oil
price shocks in developed countries but 13.45% of the variance in the core inflation is explained by the oil price shocks. The variable change in oil price has a larger effect on the core inflation in emerging countries than in the developed countries. The results are consistent with the expectations.
The last two regressions are included with the control variable GDP growth. The regression is again done for developed countries and emerging countries.
The third regression (3) gives the results for the developed countries included the control variable GDP growth. The change in oil price stays not significant for developed countries despite the added variable GDP growth. This outcome corresponds to the expectations that are made in 4.4 Nowadays the oil price pass-through is lower than before 2000. Another reason for this appearance is the less oil intensity in developed countries. Developed countries are moved from industrialized sector to a more service sector. The industry is outsourced to the emerging
countries, so the emerging countries are more oil intensive than developed countries nowadays. According to Hooker (2002) and Cavallo (2008) there is also an
insignificant effect of a change in oil price upon core inflation.
The control variable GDP growth is significant at the α = 5% level. If the GDP grows by 1% the core inflation decreases by 0.1821. The R-squared has increased to 12.30%. The R-square is higher than the model without GDP growth.
The fourth regression gives the results for emerging countries including the control variable GDP growth. The oil price change is significant at the 𝛼 = 5% level. A 1% increase in the oil price leads to a 0.1253% higher core inflation. The GDP growth is still significant. The R-square is higher for the model with the variable oil price and GDP growth. The R-squared rose from 0.1345 to 0.5493. 54.93% of the variables explain the core inflation, so the variable GDP growth and change in oil price have a large effect on the core inflation in emerging countries.
4.6 Internal validity
An empirical analyze is intern validity when the statistical inferences about the causal effects is valid for the sample (Stock and Watson, 2012).
The baseline model omits variable that explain a part of the dependent variable core inflation, like exchange rate, unemployment and interest rates. If these variables are included, maybe the effect of a change in oil on core inflation for the different countries is more unbiased.
Another treat of internal validity is the simultaneous causality between the independent variables and the dependent variable, this leads to a correlation
between the variable in the model and the error term. GDP has an effect on the core inflation, but core inflation has also an effect on the GDP growth, because if the output increases may increase the core inflation.
Last internal validity is the form of misspecification, because this regression is a linear regression between the oil price change and the core inflation, but the
population model is maybe a non-linear model. The oil price change variable contains nonlinearity. Many studies suggested that the relation between oil prices and economic variables is nonlinear (Hamilton 2003). If there is a change in the oil price, a linear approximation to the relation between oil price and economic variables will be unstable over time (Hamilton 2003).
5. Conclusion
The changes in energy markets and the world economy around 1980 explain the difference between emerging countries and developed countries. Nowadays, the oil price pass-through is lower than prior to 1980. Since 1980, the oil price pass-through is decreasing in emerging countries and developed countries. However, the oil price pass-through decreased less in developed countries than in emerging countries.
This paper estimates the effect of a change in oil price upon core inflation in developed countries and emerging countries. The sample time period is taken from 2000-2014 and the dataset contains 13 emerging countries and 13 developed countries. The variable change in oil price is constant over all countries, but different over time, because it is hard to calculate the oil price over different countries due to the taxes, price controls and the exchange rates. Two conclusions that can be drawn based upon the empirical findings of this paper.
The first conclusion is that the change in oil price upon core inflation has a significant effect in emerging countries. The second conclusion is that the change in oil price does not have a significant impact on the core inflation in developed
countries. The results are consistent with the hypothesis. The effect of a change in oil price on core inflation has a larger effect in the emerging countries than in the
developed countries. Several reasons are given for this result. One reason is that in the emerging countries, the energy intensity increased, while it decreased in the developed countries. This is because of the developed countries becoming less dependent of oil, since they started using many other energy sources as a substitute.
The emerging countries increased the energy intensity, maybe due fact that the emerging countries take over the energy-intensive industry and production from the developed countries (Gregorio et al., 2007). The industry is outsourced to the emerging countries due to more globalization around the world. This was
accompanied by the fact that in the emerging countries the development from the agricultural sector to the industrialized sector began (Kesichi, 2009). Since the 21th, century the oil demand is more driven by the emerging countries (Gregorio et al., 2007). This economy is more powerful to increasing the oil price, as a result of the subsidy for fuel in emerging countries (Kesichi, 2009). The developed countries are moved from an industrialized sector to a service sector, that is less oil dependent than the industrialized sector.
The overall conclusion is that the change in oil price has a significant impact on core inflation in emerging countries but an insignificant effect in developed countries for the years 2000-2014.
5.2 Future research
Future researches could be pay more attention on the short and long run effects of the change in oil price upon core inflation, this paper doesn’t describe the difference between these two effects. A few studies examined the short and long run effects of the change in oil price, but not on the core inflation. Also the difference between oil exporting countries and oil importing countries is not taken into account in this paper, but the effect of a change in oil price upon the core inflation can be different between oil exporting countries and oil importing countries. In addition to measure the
asymmetric effects of the change in oil price is helpful for further investigate the difference between countries in their oil dependence.
6. References
Adam P., Rianse U., Harafah L. M., Cahyono E. and Rafiy M. (2016). A Model of the Dynamics of the Effect of World Crude Oil Price and World Rice Price on Indonesia’s Inflation Rate. Agris on-line Papers in Economics and Informatics.
8(1), pp. 3-12
Castro C., Jerez M. and Barge-Gil A. (2016). The deflationary effect of oil prices in the euro area. Energy Economics. 56(1), pp. 389-397
Cavallo M., (2008) Oil price and inflation. FRBSF Economic Letter. Chen, Sheng S. (2009). Oil price pass-through into inflation. Energy
Economics. 31(2). pp. 126-133.
Darby, Michael R. (1982) The price of oil and world inflation and recession.
American Economic Review 72, pp. 738–751.
Égert B., Jimenez-Rodriguez R. and Morales-Zumaquero, A. (2010) The effect of foreign shocks in Central and Eastern Europe. Journal of Policy Modeling.
32(4). pp.461-477.
E.L (2014, 8 June). The Economist explains: Why the oil price is falling? Retrieed from
http://www.economist.com/blogs/economist-explains/2014/12/economist-explains-4 on 3 June 2016.
Gregorio J., Landerretche O., Neilson C., Broda C., and Rigobon R. (2007). Another Pass-through Bites the Dust? Oil Prices and Inflation. Economia, Vol. 7, No. 2, pp. 155-208.
Hamilton, James D. (1983). Oil and the Macroeconomy since World War II.
Journal of Political Economy. 91(2). pp. 228-248.
Hamilton, James D. (2003). What is an oil shock? Journal of econometrics.
113, pp. 363–398.
Hem C. Basnet and Kamal P. Upadhyaya. (2015). Impact of oil price shocks on output, inflation and the real exchange rate: evidence from selected ASEAN countries. Applied Economics, 47(29), pp. 3078-3091
Hooker, Mark A. (2002). Are Oil Shocks Inflationary? Asymmetric and Nonlinear Specifications versus Changes in Regime. Journal of Money, Credit and
Banking. 34(2). pp. 540- 561.
Kesichi F. (2009). The third oil price surge – What’s different this time?
Du L., He Y. and Wei C. (2010). The relationship between oil price shocks and China’s macro-economy: An empirical analysis. Energy Policy. 38, pp. 4142– 4151.
Kilian L. (2009). Not All Oil Price Shocks Are Alike: Disentangling Demand and Supply Shocks in the Crude Oil Market. American Economic Review. 99(3). pp. 1053-1069.
LeBlanc M., Chinn Menzie D., (2004). Do High Oil Prices Presage Inflation?,
Business Economics. 39(2). pp. 38-48.
Stock James H. and Watson Mark M. (2012) Introduction to Econometrics. Essex: Pearson Education Ltd.
Taylor John B. (2000) Low inflation, pass-through, and the pricing power of firms. European Economic Review. 44(2). pp. 1389-1408.
Yan L. (2012) Analysis of the International Oil Price Fluctuations and Its Influencing Factors. American Journal of Industrial and Business Management. 2 pp. 39-46.
7. Appendix Appendix I
Year Price Change in oil price (%)
2000 30,38 0,570837642 2001 25,98 -0,144832126 2002 26,18 0,007698229 2003 31,08 0,187165775 2004 41,51 0,335585586 2005 56,64 0,364490484 2006 66,05 0,166137006 2007 72,34 0,095230886 2008 99,67 0,377799281 2009 61,95 -0,378448881 2010 79,48 0,282970137 2011 94,88 0,193759436 2012 94,05 -0,008747892 2013 97,98 0,041786284 2014 93,17 -0,049091651
