The Impact Of Oil Price On Inflation - The Case Of Vietnam.pdf

UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES HO CHI MINH CITY THE HAGUE VIETNAM THE NETHERLANDS VIETNAM NETHERLANDS PROGRAMME FOR M A IN DEVELOPMENT ECONOMICS THE IMPACT OF OIL PRICE ON INFLATI[.]

UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES HO CHI MINH CITY THE HAGUE VIETNAM THE NETHERLANDS VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS THE IMPACT OF OIL PRICE ON INFLATION THE CASE OF VIET NAM A thesis submitted in partial fulfilment of the requirements for the degree of MASTER OF ARTS IN DEVELOPMENT ECONOMICS By TRUONG NGO TRONG NGHIA Academic Supervisor: DR NGUYEN VAN NGAI HO CHI MINH CITY, NOVEMBER 2012 Acknowledgement Over two years not so long but it is one of the most interesting periods of my life with many impressive memories I would like to take this opportunity to express my deep gratitude to the Vietnam-Netherlands Master Program for Economics of Development for organizing many helpful and exciting curriculums For the completion of this thesis, I have indebted to many people who have given me their continued support, advice and guidance First of all, I would like to express my sincere gratitude to my supervisor Dr Nguyen Van Ngai who gave me valuable guidelines, comments, suggestions, and inspiration for the successful completion of this study Besides, his friendly and thoughtful instructions have given me a great deal of encouragements to overcome difficulties in the whole research process I am also thankful to Dr Nguyen Trong Hoai, Dr Nguyen Hoang Nam, Dr Tu Van Binh, MBA Ly Thi Minh Chau, Tutor-Mr Phung Thanh Binh and all lecturers and program administrators in the Vietnam – The Netherlands Program for M.A in Development Economics They gave me advanced knowledge and help me kindly during the course I would like to express my heartfelt thank to all my classmates in MDE15, MDE16, especially Tran Tuyet Hanh, Tran Van Long, Le Trong Binh, Vo Thi Ngoc Trinh, Le Anh Khang, Nguyen Van Dung, Nguyen Xuan Phap, Nguyen Le Thao Nguyen and other classmates for their continuous support and encouragement Last but not least, I would like to express my deepest thanks to my parents, my brother, my close friends who have always given the most favorable environment and kept encouraging me that made me feel more confident during my study ii Certification “I certify that the content in this thesis has not already been submitted for any degree and has not submitted for any other degree until now I certify that this thesis is done from the best of my knowledge All the aids that I received during the time in preparing the thesis as same as all sources used have been acknowledged in my thesis.” Signature Truong Ngo Trong Nghia Date: iii / / 2012 Abstract A steep upward trend in the price of crude oil in recent years, reaching a spike record in middle of 2008, has led to increasing concern about its impacts on macroeconomic, both abroad and in Vietnam In this study, using the vector auto regression approach (VAR) with monthly dataset from 2001M1 to 2010M10, I attempt to empirically investigate the dynamic effects of oil price and Vietnam macroeconomics Focusing on the reduced-form of causal relationships between world oil price (expressed in Vietnamese price) and macroeconomic variables, I have used both linear and non linear form of oil prices to get the results of their impact on price level and output In empirical analysis, I find consistent evidence that oil price shocks have a significant effect on output and price level in short term In detail, my research finds a weak and positive statistically significant association between oil price shocks and price level The output is more highly sensitive and I find an empirical evidence about the negative impact of oil price shocks on economic growth although it’s not straight forward I also assert the existence of asymmetric impact of oil price proxies’ changes on economic growth rate Key words: inflation, GDP, oil price shock, VAR models, Cointegration, Granger causality, Phillips curve iv TABLE OF CONTENTS Acknowledgement ii List of Tables List of Figures List of Acronyms Chapter 1: Introduction 1.1 Problem statement 1.2 Research Objectives 1.3 Research Questions 1.4 Scope and methodology of the Study 1.5 Thesis structure 10 Chapter 2: Literature Review 11 2.1 Review oil shocks in history 11 2.1.1 Suez Crisis in period 1956-1957 14 2.1.2 OPEC Embargo in 1973-1974: 14 2.1.3 Iranian revolution and oil price fluctuation in 1979 15 2.1.4 Iran-Iraq War in 1980-1981 15 2.1.5 The great price collapse in 1981-1986 15 2.1.6 First Persian Gulf War in 1990-1991 16 2.1.7 The downtrend of oil price in 2001 16 2.1.8 Growing demand and stagnant supply 17 2.2 How higher oil prices affect the economy 18 2.2.1 Why oil shock seems to be the big economic headache 18 2.2.2 The scenario of oil price and inflation in Vietnam 21 2.3 Review price level and inflation theories 22 2.4 The transmission mechanism of oil price shocks 25 2.5 Approaches to estimate oil impact on macro economy variables 30 2.6 Empirical studies about the oil price-macroeconomic relationship 32 2.7 Conceptual Framework 44 Chapter 3: Model Specification and Data 45 3.1 Analytical Framework 45 3.2 Model Specifications 46 3.3 Steps of Estimation 50 3.3.1 Descriptive statistics 50 3.3.2 Unit root testing 50 3.3.3 Granger Causality Test 51 3.3.4 Impulse response functions 53 3.3.5 Variance decomposition 53 3.4 Data Sources 53 Chapter 4: The impact of oil price on inflation-the case of Vietnam 56 4.1 Descriptive Statistics 56 4.2 Unit Root Test 57 4.3 Var Granger Causality Test 58 4.4 Impulse Responses and Variance Decompositions 61 4.5 Asymmetric Impacts 65 4.6 Result comparisons 70 Chapter 5: Conclusion and Policy Implication 73 5.1 Conclusions 73 5.2 Policy Implication 75 5.3 Limitation and Further Studies 78 5.3.1 Limitation 78 5.3.2 Further Studies 79 References 81 Appendix 88 List of Tables Table 2.1: Summary results of empirical studies 40 Table 3.1 The definition of variables in the model 55 Table 4.1: Description statistic of key variables 56 Table 4.2: Augment Dickey-Fuller test 57 Table 4.3: Philips-Perron test 58 Table 4.4: Optimal lag length 59 Table 4.5: VAR Granger- CausalityTest 60 Table 4.6: Variance Decompositions for Var Model 62 Table 4.7: Granger causality test of proxies of oil price shocks 65 Table 4.8: Accumulated Response of PC_IP_SA to non-linear oil price shocks 68 List of Figures Figure 2.1: Price of oil in 2009 dollars, 1973:M1-2010:M10 Price of West Texas Intermediate deflated by CPI (Hamilton, 2011) 12 Figure 2.2: The natural logarithm of the real price of oil, 1861-2009, in 2009 U.S dollars 12 Figure 2.3: When oil prices head up, the US turns grey: the oil market and US recessions 18 Figure 2.4: Illustration of oil price and macroeconomic variables in level 21 Figure 2.5: Factor contributions of price level 23 Figure 2.6: Cost-push inflation in the AS-AD model 24 Figure 2.7: Demand-pull inflation in the AS-AD model 25 Figure 2.8: Two round effects of oil price increases 26 Figure 2.9: Mix transmission channels of oil price shocks 27 Figure 2.10: Conceptual framework 44 Figure 4.1: The relationships of variables 60 Figure 4.2: The impulse response functions for basic model 62 Figure 4.3: The variance decompositions of variables 64 Figure 4.4: The impulse response functions of output to negative oil price changes 66 Figure 4.5: The impulse response functions of price level to net oil price increase 69 List of Acronyms AIC: Akaike Information Criterion AR: Autoregressive ARCH: Autoregressive Conditional Heteroskedasticity CPI: Consumer Price Index GARCH: Generalized Autoregressive Conditional Heteroskedasticity INF: Inflation OLS: Ordinary least squares OPEC: Organization of Petroleum Exporting Countries PPI: Producer Price Index SBV: State Bank of Vietnam SC: Schwartz criterion UK: the United Kingdom US: the United States of America VAR: Vector Auto Regressive WTI: West Texas Intermediate 2.6 Empirical studies about the oil price-macroeconomic relationship The early empirical studies about the link between inflation and macroeconomic were investigated by Hamilton (2003), one of the most famous economists in energy field He argued that except one time in 1960, most of the events of post-World War II (WWII) U.S recessions had at least partial impacts of oil-price increases Following Hamilton’s paper series, there are lots of the researches on many economic aspects In which, the relationship between the oil-price shocks and the whole performance of economic aggregate has been explored during the four decades in many nations by Burbidge and Harrison(1984); Gisser and Goodwin (1986); Mork (1989, 1994); Mork et al (1994); Lee et al (1995); Cologni and Manera (2008) In recent years, Tang et al (2010) suggested to broadly classify most of those studies into three categories The first category is the large group of gathering researches on the theoretical mechanisms and channels in which the oil-price increase is investigated as a major factor that can retard the economic activities (Hamilton, 1983, 1996, 2000; Mork ,1989; Mory ,1993,1994; Lee, Ni and Ratti, 1995; Hooker, 1996; Lee et al., 2002; Brown and Yücel, 2002; Cunado and Perez de Gracia, 2003, 2005; Jimenez and Sanchez, 2005; Grounder and Bartleet, 2007; Katsuya, 2008; Jin, 2008; Aliyu, 2009; Du et al., 2010…) The second category focuses mainly on the empirical investigation on the relationship between oil-price change and national aggregate economic activity Either symmetric or asymmetric, either linear or nonlinear, the mathematical relationship were verified most of the developed countries during the 1970s up to now (Burbidge and Harrison, 1994; Gisser and Goodwin, 1986; Chaudhuri, 2000; Cunado and Perez de Gracia, 2003, 2005; Gounder and Bartleet, 2007; Mohammad and Gunther, 2007; Cologni and Manera, 2008; Kiptui, 2009; Weiqi et al., 2009; Limin et al., 2010…) The remaining studies focus on analyzing the role of macroeconomic policies to be able to cope with the impact of oil price shocks All the elements can weaken the negative impact of oil price fluctuations in aggregate economic activities have also been studied and carefully analyzed over time And other factors can impact economy beside oil price effects have been considered (Ferder, 1996; Bernanke et al., 1997; Hamilton and Herrera, 2001; Hooker, 32 2002; Leduc and Sill, 2004; Huang et al., 2005; Blanchard and Gali, 2007…) Indeed, the majority of studies are concentrating to propose appropriate monetary policies for coping with the oil supply shock Meanwhile, the slowdown of total output and inflation are widely considered as the two inevitable impacts of oil-price fluctuations in the world wide context First category On the empirical side, Hamilton (1983, 1996) had lots of contributions in which most of U.S recessions were preceded by increases in the price of oil That suggested oil price increases had an essential role as one of the main cause of recessions Mork (1989), Lee et al (1995) and Hamilton (1996) introduced non-linear transformations of oil prices to examine the negative relationship between increasing oil prices and economic downturns as well as to re-confirm the existence of Granger causality between both variables Mork (1989) examined whether Hamilton’s outcomes were still accurate in the case of the 1980s’ oil market collapse, also considered real oil price Furthermore, that allowed an asymmetric response of the US economic activity to oil price changes in which the real prices of oil were separately specified increases and decreases by different variables In details, the impacts of oil price increases on economic activities were different from those of decreases, and even the latter’s were not statistically significant in most cases Furthermore, an asymmetric relationship for the US also found by Mory (1993) A little bit different from the confirmations of Loungani (1986) and Hamilton (1988) about the oil-induced dislocations, Mory argued that the macroeconomics would be recessionary whether initial triggered by price increases or decreases Mork et al (1994) observed Japan, Germany, USA, Norway and Canada They found that the USA was negatively impacted by both oil price increases and decreases, while Canadan and German were less affected; the outcomes of all the rest were impacted unclearly 33 After a long time researching, Lee, Ni and Ratti (1995) focused on volatility, maintaining that “an oil shock is likely to have greater impact in an environment where oil prices have been stable than in an environment where oil price movement has been frequent and erratic” because price changes in a volatile environment are likely to be soon reversed Economists used the Garch model to study the conditional variance through real price changes In all sample periods, they got the evidence of an asymmetry between the positive effects and negative normalized shocks Positive nonnormalized shocks in real oil price were strongly related to positive unemployment and negative real growth Meanwhile, negative oil price shock impacts were insignificant Jimenez-Rodriguez and Sanchez (2005) assessed empirically the effects of oil price shocks on the real economic activity of the main industrialized countries using multivariate VAR analysis, combined with both linear and non-linear models They had the evidence of a non-linear impact of oil prices on real GDP In details, their results analyzed that oil price increases intervened with more impact on GDP growth than that of oil price declines Specially, among oil importing countries, oil price increases had a negative impact on economic activity in all cases excluding Norway and Japan A negative impact of oil price shock of 1.1% on European Union countries after eight quarters following the shock and a positive impact to the tune of 0.89 % and 1.1 % in the case of Norway and Japan respectively were recorded Surprisingly, one phenomenon was exhibited: while it was expected an oil price shock which had positive effects on the GDP growth in a net oil exporting country as England; but in reality, the oil price increase 100% actually led to loss of British GDP growth rate more than 1% after the first year in all specifications The explanation was because of an extensive literature highlighted that the unexpected result of oil price hikes has done many facts In which, that led to a large real exchange rate appreciation of the pound, which was being described as the Dutch disease in the literature3 “The Dutch Disease is the standard example of the Paradox of Plenty In the 1970s large revenues to the Dutch state from the extraction of natural gas led to the temptation to build a welfare state that was unsustainable in the long run The competitive ability of the private sector was reduced and the industrial sector experienced a setback from which it took many years to recover In the case of oil exporting countries, this is even more likely 34 Katsuya (2008), he researched the relation between oil prices and Russian economy from 1997Q1 to 2007Q4 used the VAR model But in his data, the variables were non-stationary, so a vector error correction (VEC) model was suitable for the research The analysis led to the finding that a 1% increase in oil prices contributed to real GDP growth by 0.25% over the next 12 quarters, whereas that to inflation by 0.36% over the corresponding periods Jin (2008), in a recent research into the impact of oil price shocks and exchange rate volatility on economic growth, he showed that the oil price increases caused negative effect on economic growth in Japan and China On the other hand, an appreciation of the real exchange rate led to a positive GDP growth in Russia Aliyu (2009) showed that there was a unidirectional relationship in which the oil price shock Granger cause real GDP of the Nigerian economy In additions, the oil price shock and the appreciation of real exchange rate had both positive impacts on real economic growth through a long-run vector error correction model applied Moreover, based on the Hamilton’s (1983) linear specification, Jin (2008) and Aliyu (2009) were also confirmed the existence of a symmetric oil-real GDP relationship By applying VAR and Structure VAR, Du et al (2010) found out the break point of relationship between the world oil price and China’s macro-economy Before the break date (2002:M1), China had policy for the oil price regulation, so that the correlation between the world oil price and domestic oil price of China was not strong enough It caused the impact of the world oil price on China’s macro-economy was not significant After the break date, due to the oil price reforms, the relationship between the world oil price and China’s macro-economy became much more significant Furthermore, Granger causality tests showed that China’s macro-economy could not affect the world oil price while China’s GDP and CPI were both positively correlated with the world oil price And the impulse-response functions of the linear impact model showed the largest impact reached in the second month and disappeared completely after about 12 months Unlike most of the developed countries investigated in the existing literature, a 100% increase of the world oil price because abundant petroleum revenues change the calculations of even the prudent rulers, thus making learning more difficult, not only between countries but also within them” 35 cumulatively increased GDP and CPI of China by about 9% and 2%, respectively The results of non-linear models also informed there was an asymmetric impact of the world oil price on China’s GDP Second category By VAR model, Burbidge and Harrison (1984) found that consumer price index (CPI) and oil price have positive relationship in Germany and Japan, and larger effect in the UK Chaudhuri (2000) concluded that although oil does not exist in products but in does influence goods price Oil price fluctuation can affect the necessary goods price Cunado and Perez de Gracia (2003) researched the effect of oil prices on production and inflation of European countries VAR model was used with quarterly data for the duration from 1960 to 1999, and the result was that oil price had a constant impact on inflation and had inverse effects on production growth rate in the short term With regarding to Cunado and Perez de Gracia (2005), both economic activity and price indexes were significantly affected by oil prices in the short run, especially these impacts were more significant when oil price shocks were defined in local currencies in all analyzed countries Moreover, they found evidence of asymmetries in the oil prices–macro economy relationship Gounder and Bartleet (2007) had the proof in which the higher inflation and the higher unemployment resulting from energy price shock or energy crisis on the demand-side In fact, at the same time, the ‘oil crises’ of the 1970s and early 1980s led to rise both inflation and unemployment as the ‘stagflation’ phenomenon The nonlinear transformations of oil prices were also applied to re-establish the negative relationship between increases in oil prices and economic downturns, as well as to expose Granger causality between both variables Using VAR model in the case of Iran, Mohammad and Gunther (2007) conducted the result that if the oil prices changed either negatively or positively, the inflation would be seriously increased Moreover, there was a close relationship between industrial growth and positive oil price changes 36 Kiptui (2009) estimated a generalized Phillips curve Oil prices had positive significant effects on inflation in Kenya, that 10 per cent increasing in oil prices conducted 0.5 per cent inflation up in the short- run and per cent up in the long-run In the context of China, according to Wiki, Lebo, and Zhongziang (2009), increasing oil price affected inflation and interest rate positively while Chinese investment and output were seriously affected by higher oil prices In addition, by Limin, Yanan, and Chu (2010), that there were close relationships among the world oil price, economic growth and inflation although the impact was non-linear Others Beside the evidence that oil market disruptions affected the U.S economy through the several shocks and uncertainty channels over the 1970 to 1990 sample period, the Research of Ferder (1996) induced some important findings as followings: The tightening monetary policy was applied to response to oil price increases, this evidence could explain a part of the oil price-output correlation In particular, the non borrowed reserve growth was fallen and the Federal funds rate was raised following oil price increases These two tools of monetary policy all affected the output growth However, the monetary variables had a weaker and less significant impact than the oil price variables It suggested the monetary channel provided a partial explanation for reason why oil price increases adversely affected the economy The Federal funds rate was raised by the Federal Reserve in response to oil price increases as well as it hoped to be low as much as in response to oil price decreases Moreover, the inclusion of monetary variables into output equations could not effect on the coefficients of oil price increases and decreases to converge in value Therefore, the monetary policy channel was hard to explain the asymmetry problem as a puzzle In contrast, when oil price volatility was introduced into the output equations oil price volatility rose both positively and negatively and the coefficients of oil price increases and decreases that became much closer in value Thus a partial solution to answer the asymmetry puzzle was revealed by the spectral shocks and uncertainty channels 37 By Bernanke, Gertler and Watson (1997), the role of monetary policy was considered as the central factor rather than a factor contributing to discontinuity in the relationship between oil price and GDP Indeed, a positive increase in oil price was followed by a rise in the federal fund rate and the reduction in U.S output was accounted for about two-thirds to three-quarters in the reaction of this kind of monetary policy tightening subsequent to an oil shock Unfortunately, they found that the role of monetary policy in the 1970s was more stable than the recent periods Hamilton and Herrera (2001) re-examined Bernanke et al (1997) and arrived at some contrary conclusions There were the relative contributions of monetary policy in which oil price shocks related to the following recessions in 1973, 1979-80, and 1990 Through the impulse response functions, they argued that the potential of monetary policy fighting against the contraction cause of consequences of an oil price shock was not as great as proposed by the analysis of Bernanke, Gertler, and Watson In other words, they said that oil shocks caused a bigger effect on the economy rather than as Bernanke et al suggested in the VAR By Hooker (2002), the changing weight of oil prices as an explanatory variable in a traditional Phillips curve specification for the U.S economy was analyzed empirically His results showed that the pass-through from oil prices to price levels has become negligible since the early eighties, but the evidence to certify a significant role of the decline in energy intensity, the deregulation of energy industries, or changes in monetary policy as the factor behind this lower pass-through could not be found In addition, he opined that the energy price shocks might be a trigger for an external inflation spike as the immediate result had been recorded in the literature When the inflation was not caused by an increase in domestic money supply, just was the results from oil price movements, it could drive the negative consequences for real balances According to Cunado and Gracia (2005), the asymmetric relation was convinced European countries of a truth In addition, asymmetric relation of oil change impact couldn’t be explained by supply theory Therefore, in order to explore that relation, other theories such as monetary policies, asymmetry in petroleum product prices, and adjustment costs were employed Through which, monetary policy 38 theory could explain the asymmetric response of GDP to oil changes Rather, when nominal income would fall down by sudden inflation in case there was an oil price increase, monetary policy hardly couldn’t remain the nominal GDP as the same before Furthermore, from the oil price changes–inflation rate relationship in the cases of Thailand, Malaysia, South Korea, and Japan and from the oil price changes– economic growth rate relationship only in the case of South Korea, the evidences of these asymmetric relations were found The same matter on a group of industrialized nations also was researched and was compared the results with those of the 1970s (Blanchard and Gali, 2007) There was an argued that due to the smaller energy intensity, the more flexible labor market, and the better changes in monetary policies, the effects of oil price shocks on the economies nowadays is weaker The following table summarized the above-reviewed empirical studies in sequence of time as below: 39 Table 2.1: Summary results of eempirical studies Authors Data Models- Main Results Methodology Hamilton USA; (1983) Quarterly 1949-1972 VAR Oil price upswing caused slower (Y, OP, MP, IP, UN, productivity after 3-4 quarters later by W, INF) slower output growth with a recovery beginning after 6-7 quarters Burbidge and US, Japan, Harrison (1984) Germany, UK, Canada; VAR Consumer price index (CPI) and oil price (Y, OP, MP, IP, R, had positive relationship in Germany and W, INF) Japan, and larger effect in the UK Monthly 19611982 Gisser and USA; OLS They found the impact of oil price shocks is Goodwin Quarterly (Y, OP, MP, FP, little or no support for the form of cost-push (1986) 1961-1982 UN, I, INF) Mork (1989) USA; Quarterly 1949-1988 Mory (1993) USA; Annual 1952-1990 VAR inflation They presented an even stronger negative (Y, OP, MP, IP, UN, correlation between oil price increase and W, INF) OLS (Y, OP, MP, GOV) output growth than Hamilton’s results Some evidences of an asymmetric effect of oil price spikes on the US economy were presented Lee, Ni and USA; VAR The negative sum suggested that there was a Ratti, (1995) Quarterly (Y, OPV, MP, IP, decline in the level of GNP over 24 horizons 1949-1992 UN, W, INF) following an oil shock An asymmetry in effects was found Darrat et USA; al.,(1996) Quarterly (Y, OP, MP, FP, W, 1960-1993 R) Ferderer (1996) USA; Monthly 1970-1990 VAR Multivariate VAR-causality model suggested that oil prices were not a major cause of U.S business cycles VAR The Federal funds rate and oil price (Y, OPV, OPV MP) volatility explained 27% and 22% of the forecast error variance for industrial production at the 24-month horizon 40 Hooker (1996a) Hooker (1996b) USA; VAR 10% increase in oil price caused GDP to Quarterly (Y, OP, MP, IP, slow down by 0.6% in the third and fourth 1947-1974 INF) quarter after the shock USA; VAR They could not foresee the unemployment Quarterly (Y, OP, MP, IP, 1974-1994 INF) Hamilton USA; OLS (1996) Quarterly (Y, OP, MP, INF, 1948-1994 IP) Hooker (1999) USA; Quarterly rate or GDP growth by oil price levels The relation between GDP increase and net oil price increase was statistically meaningful ECM Suggested that different monetary policy (Y, TB, UN, OP, IP) (particularly prior to the oil shocks) could 1979-1998 have substantial impact on inflation Hamilton Quarterly OLS with lags Price increases caused more impact than the (2000) 1949:1999 variables decreases did From 1949 to 1980 a 10% (Y, Yt-1, Yt-2, Yt-3 ,Yt- increase in oil prices resulted four quarters 4; OP t-1, OPt-2, OPt-3, later in a level of GDP growth that was 1.4% OPt-4; OP+t-1, OP+t-2, lower OP+t-3, OP+t-4) Nagi Elthony et Kuwait; VAR, VECM The empirical evidence indicated that oil al., (2000) Quarterly data (OILR, OPLP, price shocks and hence oil revenues had a 1984-1998 EXDEV, EXCON, CPI, M2, notable impact on government expenditure, both development and current IMPORTS) Lee, Ni and Japan; Ratti., (2002) Monthly 1960-1996 Cunado and 15 European Perez Countries; de Gracia Quarterly (2003) 1960-1999 VAR In response to a shock in oil price, output (Y, OPV, MP, INF, decline occurred after a 10-month delay and R, CP, GOV) the decline was short-lived VAR The result was that oil price had constant (Y, OP, INF) impact on inflation in short term but had inverse effects on production growth rate 41 Cunado and Asian Bivariate VARs with The impact was higher when oil prices were Perez Countries; lags variables (Y,IF, measured in local currency, which could be de Gracia Quarterly constructed proxies due to the role of exchange rates or national (2005) 1975Q1– of oil shocks: OP, price 2002Q2 OP+ , OP- , SOPI, variations on macroeconomic variables NOPI4, NOPI12) Jimenez- OECD Rodriguez and Countries; Sanchez(2005) Quarterly VAR The conclusion was drawn that price (Y, OPV, INF, R, W, increases had greater effect on GDP growth EX) than the decreases 1972-2001 Blanchard and developed Multivariate VARs, They discovered that the weak reaction of Gali (2007) countries; Rolling bivariate the economies in recent period was due to Quarterly VARs; 6-variable smaller energy intensity, a more flexible 1960:1-2005:4 VAR: OP, GDP- labor DEF, INF,W, GDP, market, and better changes in monetary policies EM Gounder and New Zealand; VAR They found that the largest negative impact Bartleet (2007) Quarterly (Y, OPV, W, EX of NOPI innovations to GDP occurred in the 1989-2006 INF) third quarter and remains negative over years in new Zealand They reported a cumulative effect of 0.7 percent of GDP growth for New Zealand Katsuya (2008) Russia; Quarterly VECM (UOP, GDP, IF) The result showed that real GDP growth by 0, 25 percent was caused by percent 1997Q1 - increase in oil prices, whereas that to 2007Q4 inflation by 0, 36 percent over the next twelve quarters Jin (2008) Russia, Japan and China Aliyu (2009) VECM (Y, EX, OP) Specifically, a 10% permanent increase in international oil prices was associated with a Quarterly; 5.16% growth in Russian GDP and a 1.07% 1999-2007 decrease in Japanese GDP Nigeria; VECM Quarterly (Y, EX, OP) 42 The results tally were very well with the coefficient of oil price shock of 0.77 1986-2007 reported using a linear oil price model for the Nigerian economy Du et al., China; VAR, Structure The impulse-response functions of the linear (2010) Monthly VAR impact model showed that China’s GDP and 1995:1- (OP, GDP, CPI,M1, CPI were both positively correlated with the 2008:12 R) world oil price Source: adapted from Gounder and Bartleet (2007), Aliyu (2009) and author improved Notes: VAR is Vector Auto regression, VECM is Vector Error Correction Model, Y is economic growth, MP is Monetary Policy, OP is oil prices, UOP is converted from US dollars per barrel to the Russian rubles per barrel, IP is import prices, UN is unemployment, W is wages, INF is inflation, R is interest rate, I is investment, OPV is oil price volatility, CP is commodity prices, GDP is Gross domestic product, GOV is Government expenditures, EX is exchange rate, TB is treasury Bill rate, OILP is Oil Price of Kuwaiti Blend Crude, OILR is Oil Revenue, EXDEV is Government Development Expenditure, EXCON is Government Current Expenditure, CPI is Consumer Price Index, M1 is Money Supply, M2 is Money Demand (M2 Definition), IMPORTS is Value of Imports of Goods & Services; Proxies of oil prices: OP+ (Oil price increase), OP- (Oil price decrease), SOPI (Scaled oil price increases), NOPI4 (Net oil price increases from the past quarters), NOPI12 (Net oil price increases from the past 12 quarters) Seeing the table above, it’s obvious that VAR approach is also inspired by the existence of a large empirical literature using VARs Why? The most advantage of VAR models is that they can simultaneously estimates the interrelationship between more than one endogenous variable So in this study I employ vector auto regression (VAR) models to examine oil shocks transmission mechanism, which focuses primarily on reduced-form relationships between oil price and macroeconomic using a small number of variables We will concern more in section of chapter 43 2.7 Conceptual Framework In this section, we will test the validity of transmission mechanisms of oil price shock in Vietnam, by checking the statistical relationships between key variables on transmission chains shown in Fig 2.9 step by step We start checking the relationship between oil price and domestic consumer price index CPI Based on Fig 2.9, this relationship is corresponding to the arrows and Then, we will test the direct impact of oil price change to economic output with proxy IP (Industrial Production Index), i.e the short-term impact, indicated by arrow For more concise, let’s modify and improve the Fig.2.9 become a simple framework as below: Tải FULL (93 trang): https://bit.ly/3zCKkd4 Dự phòng: fb.com/TaiHo123doc.net Figure 2.10: Conceptual framework Short tem effects: OIL PRICE Hypothesis testing:  Granger causality tests (exist short term Econometric relation ships?!) technique of  VAR models GDP/IP CPI Sources: Author’s calculation 44 Asymmetric tests Chapter 3: Model Specification and Data After here, the analytical framework and model specifications are introduced Then, the steps of estimation and data sources are mentioned 3.1 Analytical Framework Estimation strategy First, stationary is check in each variable Second, the short-run dynamic behavior between oil prices and macroeconomic variables is studies and Granger causality Third, Impulse Response Functions, Variance Decompositions are to be used Finally, asymmetries in the proxies of oil price changes and macroeconomic relationships are tested Tải FULL (93 trang): https://bit.ly/3zCKkd4 Dự phòng: fb.com/TaiHo123doc.net Unit root testing (ADL and PP tests) Granger causality tests – short run relationships Impulse Response Functions, Variance Decompositions tests Asymmetric testing procedure Conclusion 45 3.2 Model Specifications The VAR system is based on empirical regularities embedded in the data As we have concerned before, one of the most advantage of VAR models is that they can simultaneously estimates the interrelationship between more than one endogenous variable So the VAR technique is suitable for this study because of its ability to expose the dynamic structure of the model as well as it helps us to avoid imposing excessive identifying restrictions associated with many different economic theories There are some kinds of VAR models One of the typical VAR models can be expressed as a system of reduced form equations in which each of the endogenous variables is regressed on its own lagged values and the lagged values of all other variables A reduced-form VAR model can be expressed in matrix form as, yt A (L)*yt-1 B*xt µt (3.9) Where yt is the vector of endogenous variables, is the vector of constants, xt is the vector of exogenous variables and μt is the vector of serially uncorrelated disturbances that have zero mean and a time invariant covariance matrix (generalization of a white noise process) A and B are coefficient matrices, L is the lag-operator In this study, the vector of endogenous variables yt consisted of three variables: percentage changing of consumer price index (PC_CPI_SA), percentage changing of industrial production (PC_IP_SA) and percentage changing of oil price (PC_VOP) yt = [PC_CPI_SA, PC_IP_SA, PC_VOP] The vector of exogenous variables xt for model contained percentage changing of money supply (PC_M2) in other to examine the effect of monetary policy; in light of the global crisis, dummy variable (DBP) was used for break point of month of peak oil price in middle of 2008; trend was used for time varying When these variables were treated as exogenous variables, their contemporaneous impact on the endogenous variables could be accepted, but not used for a feedback xt = (PC_M2, DBP, TREND, PC_M2*TREND, DBP*TREND) 46 6674596 ... the context of Vietnam over the period of 200 1-2 010? Sub-questions: - Which are the directions of the causality relationships among oil price, price level and output? - How are the impacts of oil. .. of high oil prices on world economics, the scenario of oil price and inflation in Vietnam It also includes the review inflation theories, the transmission mechanism of oil price shocks, the approaches... impacts of oil price on the macroeconomics? - Are there any different impacts of proxies of oil price on inflation rate and economic growth rate of Vietnam when oil price is in trend of increasing