The impact of exchange rate on trade balance between vietnam and china

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

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 EXCHANGE RATE ON

TRADE BALANCE BETWEEN VIETNAM AND

CHINA

BY

TRAN QUOC KHANH CUONG

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

HO CHI MINH CITY, January, 2015

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 EXCHANGE RATE ON

TRADE BALANCE BETWEEN VIETNAM AND

CHINA

A thesis submitted in partial fulfilment of the requirements for the degree of

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

By

TRAN QUOC KHANH CUONG

Academic Supervisor:

NGUYEN VAN NGAI

HO CHI MINH CITY, January, 2015

i

ACKNOWLEDGEMENT

First of all, I would like to show my deepest thanks to my supervisor, Associate Professor, Ph D Nguyen Van Ngai who gave scientific guidance, useful advice and instruct me to complete this thesis

I would like to thank Professor Chinn M D., Professor Bahmani-Oskooee, M and Dr Pham Khanh Nam who gave many useful comments during the process of doing my thesis

I am very grateful to the lecturers and classmates who help me to much during studying

Finally, I would like to give the special thanks my family who support me moral and financial during the time I study

ii

ABSTRACTS

The goal of this research is to investigate the short term and long term relationship between exchange rate and trade balance Differing test such as Johansen Cointegraion test, Vector Error Correction, Wald test etc., have been conducted The data was collected from International Financial Statistics (IFS), Organization for Economic Cooperation and Development (OECD) and Asian Regional Integration Center (ARIC) The conclusion is as follow:

• Long run relationship between real exchange rate and trade balance occurs, and

• The coefficient of real exchange rate is negative and significant

The results implies that devaluation of currency to cure trade deficit between Vietnam and China may not an appropriate method, for instance, the trade performanceof Vietnam to China could deteriorate

iii

Contents

CHAPTER 1 1

INTRODUCTION 1

1.1 Problem statement 1

1.2 Research objectives 3

1.3 Research methodology 3

1.4 Structure of this thesis 3

CHAPTER 2 5

LITERATURE REVIEW 5

2.1 Definition of exchange rate, real exchange rate and misalignment 5

2.2 Theoretical model of real exchange rate equilibrium 6

2.2.1 Model – based approach 6

2.2.2 The fundamental equilibrium exchange rate approach 6

2.2.3 The purchasing power parity approach 8

2.3 Theory of the impact of exchange rate on trade balance 10

2.3.1 Marshall – Lerner condition 10

2.3.2 The J-Curve 10

iv

2.4 Empirical studies 11

2.4.1 Empirical studies for the PPP approach 11

2.4.2 Empirical studies for the BEER and the FEER 14

2.4.3 Empirical studies of the impact of exchange rate on trade balance 15

2.5 Conceptual framework 20

CHAPTER 3 22

THE EXCHANGE RATE AND TRADE 22

BETWEEN VIETNAM AND CHINA 22

3.1 Tendency of the exchange rate between Vietnam and China 22

3.2 Trade balance between Vietnam and China 23

3.3 The main commodities of import and export between Vietnam and China in recent years 24

3.3.1 The main commodities of export to China in recent years 24

3.3.2 Imported commodities from China Error! Bookmark not defined CHAPTER 4 29

MODEL SPECITICATION AND DATA SOURCE 29

4.1 Model for purchasing power parity approach 29

4.2 Model for trade balance 33

v

4.3 Robustness check for VECM 37

4.3.1 Serial Correlation Test 37

4.3.2 Heteroskedasticity test: 37

4.3.3 Normality test: 38

4.4 Data source 38

4.4.1 Data for PPP approach 38

4.4.2 Data for trade balance 39

CHAPTER 5 41

THE IMPACT OF EXCHANGE RATE ON TRADE BALANCE BETWEEN VIETNAM AND CHINA 41

5.1 The estimation of real exchange rate 41

5.1.1 Unit root test 41

5.1.2 Optimal lag for VECM 42

5.1.3 Johansen (1988) procedure for cointegration test 45

5.1.4 The estimation of real exchange rate between Vietnam and China 48

5.1.5 Robustness check for PPP model 49

5.2 The exchange rate misalignment between Vietnamese currency (VND) and Chinese currency (RMB) 51

vi

5.3 The impact of the exchange rate on trade balance 53

5.3.1 Unit root test 53

5.3.2 Optimal lag for VECM 55

5.3.3 Diagnostic check for every lag 56

5.3.4 Inverse Roots of AR characteristic polynomial 57

5.3.5 Johansen procedure for cointegration test for the impact of exchange rate on trade balance 57

5.3.6 The impact of exchange rate on trade balance in long run 60

5.3.7 The impact of exchange rate on trade balance in short run 60

5.3.8 Robustness Check 61

CHAPTER 6 66

CONCLUSIONS AND RECOMMENDATIONS 66

6.1 Conclusions 66

6.2 Policy implications 67

6.3 Limitations and further researches 67

6.3.1 Limitations 67

6.3.2 Further researches 67

vii

LISTS OF FIGURES

Figure 2.1: J curve effect 11

Figure 2.2: Conceptual framework of this study 21

Figure 3.1 Nominal exchange rate between Vietnam and China, (VND/RMB) 22

Figure 3.2: Trade balance between Vietnam and China (million US dollar) 23

Figure 5.1: Inverse Roots of AR characteristic polynomial for PPP 44

Figure 5.2 Real exchange rate between Vietnam and China (VND/RMB) 48

Figure 5.3: Normality test 51

Figure 5.3: Misalignment between VND and RMB 52

Figure 5.4: Inverse Roots of AR characteristic polynomial for PPP 57

Figure 5.5: Normality test for VECM model 63

Figure 5.6: Cumulative sum of recursive 64

Figure 5.7: Cumulative sum of square of recursive 64

viii

LIST OF TABLES

Table 3.1: The main commodities of export to China during 2010-2013 (million USD) 24

Table 3.2: The main commodities of import from China during 2010-2013, million USD 26

Table 4.1 Data for PPP approach 39

Table 4.2: Data for trade balance approach 39

Table 5.1: Mackinon (1996) critical value 41

Table 5.2: Unit root test for PPP approach 41

Table 5.3: Lag criteria for PPP approach 43

Table 5.6: Wald test for Horvath – Watson procedure 47

Table 5.9: Unit root test on time series 54

Table 5.10: Lag selection of VECM 55

Table 5.11: Diagnostic check for every lag 56

Table 5.12: Johansen cointegration test with 18 lags 58

Table 5.13: Cointegrating equation 59

Table 5.13: The speed of adjustment coefficient of long run 60

Table 5.14: Wald test for the short run relationship 60

Table 5.16: Heteroskedasticity Test: Breusch-Pagan-Godfrey 62

ix

LISTS OF ABRREVIAION

ARDL: Autoregressive Distribution Lag

BEER: Behavior Equilibrium Exchange Rate

CPI: Consumer Price Index

FEER: Fundamental Equilibrium Exchange Rate

IIPCN: Index of Industrial Production of China

IIPVN: Index of Industrial Production of Vietnam

PPP: Purchasing Power Parity

RER: Real Exchange Rate

STAR: Smooth Transition Autoregressive

TB: Trade Balance

VECM: Vector Error Correction Model

In general, there was a substantial increase in both export and import throughout the period from 2000 to 2013 Nevertheless, this upward trend did not grow constantly In particular, as a result of global finance crisis, the proportion of import and export witnessed a marginal downward trend between 2008 and 2009 At that time, balance trade plunged into the lowest level for 11 years However, it recovered and rose considerably more than the previous stage

Although export increased over and over, it was still lower than import that leaded to trade deficit for Vietnam Until 2012, export already kept up with the import development, comprised nearly 115.5 billion USD that enhanced the effect and advantage of trade balancedue to trade surplus.The most striking feature was that the

of trade balance between Vietnam and Asian, especially China compared withUS or

EU In many years, VN always has trade surplus with US and EU; nevertheless, VN takes trade deficit with ASEAN, and especially China

In many years, economists have concerned about the trade deficit between Vietnam and China And exchange rate between Vietnam and China has been considering as a reason of trade deficit The question is that should we depreciate our currency? This problem is very important because exchange rate plays an important role in the

economy, especially in emergency market such as Vietnam (Stone, M.et al., 2009) It

will effect on inflation targeting whichis one of the hottest issue in Vietnam inrecent years

According to Marshall-Learner (ML) condition or J Curve, exchange rate plays an important role in trade balance Both of them state that if the currency is depreciation, with the condition that adding the elasticity of export and import more than one, the trade balance will increase in the long run

As VND is appreciated, Vietnamese finds imported goods from China cheaper and Chinese finds imported goods from Vietnam more expensive Therefore, Vietnam has trade deficit Nevertheless, if VND is depreciated, an opposite effect takes place and Vietnam takes trade surplus with China

3

Does VND over-valuated compare with China? Does exchange rate effect to trade balance between Vietnam and China? And does depreciation increase trade balance between Vietnam and China?

1.2 Research objectives

Based on the problem statement, this study aims to find out:

First, estimating of the real exchange rate between Vietnam and China

Second,evaluatingof the value of Vietnamese currency against Chinese currency

Third, examiningthe influenceof the exchange rate on trade balance between two countries

1.3 Research methodology

There are three procedures to solve the three objectives of this research Firstly, Purchasing Power Parity is used to estimate the real exchange rate between Vietnam

and China Secondly, time trend is used to evaluateof the value of Vietnamese currency

against Chinese currency And finally, cointegration and vector error correction is used

to estimate the impact of the exchange rate on trade balance between two countries

1.4 Structure of this thesis

The thesis consists of five chapters which are arranged as follows:

Chapter 1 provides an overview for the trade balance of Vietnam in recent years, and explains several reasons for choosing this topic and research objectives

Chapter 2 is assigned to literature review This chapter provides the theories and empirical studies for calculating real exchange rate, the impact of exchange rate on

Chapter 4 describes the methodology and data collection for this thesis This chapter containsempirical models that will apply for the next chapter as well as data sources

Following chapter 4, chapter 5 presents and discusses empirical results Econometric results are shown and discussed to determine the real exchange rate, misalignments between Vietnam and China Finally, the role of exchange rate in trade balance between Vietnam and China is covered

Finally, with all results and analyses from previous chapters, the chapter 6 provides recommendations for policy makers Additionally, the last part contains limitations and further researches as well

2.1 Definitionof exchange rate, real exchange rate and misalignment

GÄrtner (2009) stated that exchange rate is defined either as the price of one unit foreign currency in termsof domestic currencyor the ratio between the price of a bundle

of goods foreign and domestic

Definition of real exchange rate (RER) is the ratio price of traded goods to non-traded goods The equilibrium real exchange rate (ERER) is defined as the rate price of traded goods to non-traded goods that results from the simultaneous achievement of equilibrium in the external sector and internal sector of a country

The misalignment is the differencebetween real exchange rate and equilibrium exchange rate The misalignment implies the currency overvalued when it is positive, otherwise the currency undervalued

In order to calculate the misalignment, we have to calculate the equilibrium real exchange rate (ERER) There are some methods to calculate ERER such as PPP approach, model – based approach

6

2.2 Theoretical model of real exchange rate equilibrium

2.2.1 Model – based approach

Model – based approach (Clark and McDonald, 1998) measures RER misalignments based on the ERER This is familiar with “behavior equilibrium exchange rate” (BEER) approach and “fundamental equilibrium exchange rate” (FEER) approach These approaches are engines to estimate exchange rate

2.2.2 The fundamental equilibrium exchange rate approach

Clark and McDonald (1998) said that the FEER approach was developed by Williamson in 1994 This approach estimates real exchange rate balances with current account (CA) at no unemployment and low inflation with sustainable net capital flow

and current account equals (the adverse) of capital account (KA)

7

2.2.2.1 The behavior equilibrium exchange rate approach

Clark and McDonald (1998) built this model based on the fundamental macroeconomicssuch as:

Real effective exchange rateis the currency of home economy relative to foreign currency This variable is denoted in natural logs: log(q)

Term of trade representsthe domestic export ratio value divide into import value be relative to the equal effective foreign ratio, where the trade weight is calculated trade weighted This variable is denoted in natural logs: log(tot)

Relative price of nontrade to trade goods is defined as the ratio between home country

of consumer price index (CPI) and the producer price index (PPI) be relative to the equivalent foreign effective ratio This variable is denoted in natural logs: log(tnt)

Net foreign assetsequals the total foreign assets (minus gold holding) minus total liability to foreigners, express as the ratio to GNP This variable is denoted in nfa

Relative stock of government debt is defined as the ratio of domestic government net financial liabilities to nominal GDP relative to the effective ratio of partners

Real interest rateis defined as home interest rate (r) minus the foreign interest rate (r*)

r is defined as home interest rate.This is calculated inlong term (10 years) government bonds minus the change in the CPI from the previous year r* is the weighted average partners real interest rate with the same calculate style

Therefore the equation:

Xt = [(r - r*), ltntt, ltott, nfat, λt] (2.4)

8

Where:

x isa gross vector,q represents real effective exchange rate,tot stands for term of trade, tnt is relative price of nontrade to trade goods, nfa is defined net foreign assets, λ represents relative stock of government debt and real interest rate is r, r*

The equation (2.4) includes severalfundamentaleconomics that could affect to the bilateral exchange rate Therefore, the movement in the real effective exchange rate could be explained by BEER approach However, the fundamental economicscould explain exchange rate in a variety of variables such as technological process, control over capital flows, etc (Doroodian, Jung and Yucel, 2010) This reason implies thatthere is no normative formula Thus, it is difficult to put all variables in the equation with proper data As a consequence, when variables are limited or exceeded,

the coefficient would change certainly

2.2.3 The purchasing power parity approach

2.2.3.1 The law of one price

The Law of One Price proposed by Krugman, Obstfeld and Melitz (2012) expresses that the price of an identical good sold asthesame price in the world when express in termsof the same currencyin the competitive market without transportation cost and official barriers to trade, etc That means

 = st × ∗ (2.5)

Where  is theprice of goodsi in term of local currency, st is the nominal exchange

rate and ∗ is the price of goodsi in the foreign currency

However, in the reality, there is no evidence that goods and services have the same price in the international market because of transportation cost, tax, etc

9

2.2.3.2 Purchasing Power Parity

The PPP was first expressed by the Salamanca School inSpanishin 16thcentury At that time, PPP was basically that when we changed to the common currency, the price level

of every country should be the same (Rogoff, 1996)

Cassel (1918) introduced the term purchasing power parity (PPP) After that, PPP became the benchmark for acenter bank to build up the exchange rate and for scholars study exchange rate determinant The model of PPP of Cassel became the inspiration for Balassa (1964) and Samuelson (1964) set up their models They independently worked and gavethe final explanation why absolute PPP become the good theory of exchange rate (Asea, P and Corden, W 1994) The reason is that the relative price of each good in different countries should equal to the same price when changinginto the same currency

The PPP has two versions including absolute and relative PPP (Balassa, 1964)

According to the first version, Krugman et al., (2012), defined that absolute PPP

implies that exchange rate between two currencies of pair countries equal tothe ratio of the price level of these countries This mean:

st = pt/∗ (2.6)

Shapiro (1983) stated that the relative PPP implies the ratio of domestic to foreign prices would equalthe ratio change in the equilibrium exchange rate This states that there is a constant k which has the relationship between price level and equilibrium exchange rate,

10

2.3 Theory of the impact of exchange rate on trade balance

2.3.1 Marshall – Lerner condition

According to Bahmani-Oskooee (1991), Marshall – Lerner was that if the currency depreciated, the trade balance would improve in the long run with the condition that the sum of the absolute value of the elasticity of export and import is more than one Assuming that the condition is satisfied.The appreciation of home country makes imported goods from abroad cheaperthus, the consumers of Home country buy more Foreign consumer finds imported goods are more expensive and they buy less than before appreciation Consequence, Home country has trade deficit Nevertheless, if Home country is depreciated, an opposite effect would be taken place and Home country would enjoy trade surplus with the rest of the world

Furthermore, the condition states that devaluation of currency has positive effects on trade balance if the total of the absolute value of the elasticity of export and import is larger than one For that reason, the effects on trade balance rely upon elasticity of price

However, the disadvantage of the Marshall – Lerner condition is that it could not explain why trade deficit occurs in the short run after currency depreciation

2.3.2 The J-Curve

When currency depreciated, trade balance would become worse in the short run, after that they become better because of the lag of time This means that in the long run, the depreciation of currency makes trade balance increased The reason why trade deficit which happens in the short run, was explained by Akbostanci in 2004 Most exporters and importers have signed the contract before depreciation In the short run, the quantity of export and import does not change much;nevertheless, the depreciation

11

makes the imported goods costs more in domestic currency Therefore, the value of imported goods rises while exported products does not changealot As a result, trade balance becomes deficit

Figure 2.1: J curve effect

2.4 Empirical studies

2.4.1 Empirical studies for the PPP approach

The valuation of real exchange rate is very important for Vietnam Kaminsky et al., (1998) or Chinn (2000) stated that the appreciation of exchange rate maylead to the crisis of emerging economies The studies for PPP approach have two popular models, linear and nonlinear models Using linear model, almost papers use cointegration test,Vector Error Correction Model (VECM) or test unit root to check whether all variables move along together or mean reverting Contrary, using nonlinear model, almost papers apply STAR-family (Smooth Transition Auto Regressive) model and then testing the unit root of real exchange rate in nonlinear model framework

12

Tastan (2005) and Narayan (2005) tested the stationary of real exchange rate by using unit root test Tastan(2005) attempted to seek for the stationary of real exchange rate between Turkey and four main partners: US, England, German and Italian From 1982

to 2003, the empirical result stated non-stationary in the long run between Turkey and

US and Turkey and England as well While Tastan (2005) used single country, Narayan examined for 17 OECD countries The results of his research was mixed If he usescurrency based on US dollar, there were three countries France, Portugal and Denmark satisfied If the currency was used as Deutschmark, there were seven countries satisfied The authors used univariate technique to find out the equilibrium of

the real exchange rate Kremerset al., (1992) stated that this technique(univariate

approach) suffered low power against multivariate approach because of the deception

of improper common factor limitation implicit in the ADF test

After Johansen (1988) developed a method of conducting VECM, there were papers applied it to test PPP; nevertheless, testing PPP in terms of stationary by using Johansen (1988) procedure is not good asHorvath – Watson (1995)procedure (Edison

et at., 1997) Therefore, Chinn (2000) estimated the East Asian currencies overvalued

or undervalued with VECM Currencies of Hong Kong, Indonesian, Thai, Malaysian, Philippine and Singapore were overvalued However, calculation the misalignmentby the CPI deflator is broader than PPI deflator In this paper, time trend is used to calculate the equilibrium exchange rate Then misalignment was calculated follow the equation:

In case of using nonlinear regression model, many papers such as Baharumshahet

al.,(2010),Ahmad Y, & Glosser S, (2011) have been applied inrecent years However, Sarno (1999) stated that when using nonlinear Smooth transition autoregressive

13

model(STAR), the presumption of real exchange rate following STAR model could lead to wrong conclusion

Kapetanios G et al.,(2003)who developed a new test (KSS test), to test unit root for 11

OECD countries, with nonlinear Smooth Transition Auto Regressive model The authors used monthly data for 41 years from 1957 to 1998 and US dollar as a numeraire currency While KSS test did not accept unit root in some case, ADF test

provided reverse results For that reason KSS is superior to ADF test Liewet al.,

(2003) used KSS test to check whether RER of Asian is stationary or not In his research, data covered for 11 Asian countries with quarterly bilateral exchange rate from 1968 to 2001 with US dollar and Japanese Yen as a represented currencies They concluded that there was a conflict between using KSS test and ADF test for unit root

In the paper, the ADF test was accepted all case; however, KSS test was not accepted

in 8 countries with US dollar numeraire and 6 countries YEN as a numeraire The other kinds of unit root test for nonlinear model are Saikkonen and Lutkepol(2002) and

Lanneet al., (2002) which were used by Assaf(2006) to test real exchange rate (RER)

stationary or not for 8 EU countries There was no stationary of RER in the structural breaks after the post of Bretton Woods era from this paper The reason fornonstationary is that authorities may interfere with the exchange market to decide exchange rate undervalued or overvalued

Baharumshahet al.,(2010) attempted to test the nonlinear mean reverting of 6 Asian

countriesbased on nonlinear unit root test and Smooth Transition Auto Regressive model The authors used quarterly data for 40 years from 1965 to 2004 and US dollar

as a numeraire currency This was a new method to test the unit root of real exchange rate First, he proved that real exchange rate was nonlinear model, then he testedunit root of real exchange rate in nonlinear model The evidence proves that RER of these countries are nonlinear mean reverting Which means the calculation of the

14

misalignment of these currencies should be conducted with US dollar as a numeraire This evidence will leadto different resultswhen using ADF test for unit root

2.4.2 Empirical studiesfor the BEER and the FEER

Clark and Macdonald (2002) and Baak (2012) used BEER with four fundamental variables: real interest rate difference,Net Foreign Asset, Term of Trade While Baak measured the misalignment of Korean currency, Clark and Macdonaldestimated the equilibrium Exchange rate of 3 countries such as: United Stated, Canadian and England currency Johansen(1988) test was employed for cointegration and Vector Error Correction model for estimating long run relationship among variables Clark and Macdonald (2002)found that there wasa contrast between US and Canadian currencies with sterling pound US and Canadian dollar moved along together in BEER and PEER methods While Sterling pound has the adverse way of two methods The authors showed that these variables are not enough for the model as well That is the reason why the model can be put more variables whichdepend on authors For instance,Doroodian, Jung and Yucel (2010) seek for the equilibrium RER of Turkey based on 7 fundamental variables (fiscal and monetary policies) such as external term

of trade, ratio of investment over GDP, government consumption of non- tradable, capital control, exchange and trade control, the process of technical and capital accumulation

Moreover, the misalignment of real exchange rate is depended on statistical

econometric Such as Dağdevirenet al., (2012) attempted to measure the RER

misalignment of Turkey for each structural break The authors whofound out 2 break points of RER for Turkey, used S2S for the first break point and GLS for the second break point Saayman (2010) used panel data of South Africa from 1999Q1 to 2008Q4

to find out the equilibrium exchange rate of these countries He used BEER approach

15

with dynamic OLS and fully modify OLS The dynamic OLS has the disadvantage when variables and lags are added, it will reduce degree of freedom That is the reason

why fully modify OLS was performed to recheck dynamic OLS.Coudertet al.,(2013)

found out the misalignment of RER for EU countries They used GMM method with given VECM to estimate coefficient of model and long run relationship

Aflouk et al., (2011) and Ganet al.,(2013) attempted to estimate the long run

equilibrium for China and in Asian and Latin Americarespectively They used FEER model and tested unit root and then used Johansen (1988)procedure Unit root test of all-time series is I(1) and fundamentals are cointegrated together For that reason misalignments can be calculated based on FEER approach

Koske (2007) used both BEER and FEER approach to calculate equilibrium of Malaysia exchange rate by using quarterly data from 1980 to 2006 The conclusion for this paper is that Malaysian Ringgit wasovervalued from 1980 to 1984, after that it was undervalued That result wassuitable in the macroeconomic policy of Malaysia Last but not least, in the paper, the disadvantages of both BEER and FEER approach was stated in the context of emergency countries Because there is not an accurate model, the choice of variables aredepended on authors As a result, the coefficient of parameters will be changed by these choices

2.4.3 Empirical studies of the impact ofexchange rate on trade balance

Seeking for the relationship between trade balance and exchange rate is one of the most interested topics for international trade There have been many authors such as Rose, Yellen, Bahmani-Oskooee, Arize, Shahbaz have conducted researches for many countries from developing to developed one The common techniques, whichareapplied,areJohnsencointegration (1988)technique such as Bahmani-Oskooee

16

(1991), Onafowora (2003) etc., and Autoregressive Distributed Lags (ARDL) such as

Narayan (2006), Shahbazet al., (2012) etc

Rose and Yellen (1988) looked for J curve existent in U.S Theymostly used quarterly data from 1960 to 1985 and the bilateral trade between U.S and six her main partners (Japan, Canada, U.K, France, Germany, and Italy) The result indicated that J curve did not exist in the long run or short run Then, Rose (1990) used monthly date, seasonal adjustment, from 1974 to 1986 to estimate five major OECD countries (The United Kingdom, Canada, Germany, Japan and U.S) in the post Bretton Woods era By the same token as his study above, the results implied no evidence support Marshall – Lerner condition or exchange rate does not effect on trade balance of these countries

Shirvani and Wilbratte (1997) used monthly bilateral data among US and her trade partners to investigate the relationship between real exchange rate and trade balance with multivariate regression approach This report did not demonstrate the relationship between trade balance and exchange rate in the short run However,variables affected each other in the long run In addition, the disadvantage of this paper is that the authors did not find out the effect of exchange rate, real income of domestic and foreign on trade balance in a particular number

Matesanz andFugarolas (2009) examined the relationship between trade balance and exchange rate in Argentina from 1962 to 2005 In their paper, GDP Argentina, GDP

US, RER were the independent variables, and cointegration test and VECM were methods The results indicated that the most important thing is that trade balance of Argentina was supported by Argentina currency undervaluation As a results, M-L condition took place in Argentina Last but not least, in theirs paper, they argued that the coefficient of real exchange rate canbe positive or negative.If coefficient ispositive,

17

the M-L condition will hold, this means that when currency is devaluated, trade balance will increase, or vice versa

After Pesaranet al., (2001) developed Autoregressive Distributed Lagstechnique Many

papers have appliedthis method to prove the relationship between exchange rate and

trade balance such as: Arora et al., (2003), Narayan (2006), Shahbazet al., (2012).Arora et al., (2003) used this technique to study bilateral trade between India

and her trading partners They used quarterly data spans from 1977 to 1998 In the seven largest trading partners of India, there are four countries such as Australia, German, Italia and Japan had the long run effectwhen devaluating Rupee In case of USA, when devaluating of Rupee, it will make bilateral trade balance become worsen For instance, 1% of devaluation will make trade balance shorten 2.98%

In case oftrade balance relationship between China and US, Narayan (2006) employed cointegration test and ARDL model to make clear whether the relationship between exchange rate and trade balance exit?

where TB is trade balance, RER is real exchange rate

He used monthly data span from 1979 through 2002 Trade balance is independent variable and real exchange rate is dependent variable When using cointegration test, the empirical result indicated that there was a relationship between real exchange rate and trade balance ARDL suggested that the undervaluation of China’s currency would help trade balance increase with US Therefore, ML condition exited in case of China and US

Arize (1994) and Shahbazet al.,(2012) used the same above function of Narayan

(2006) toexamine the relationship between trade balance and real exchange rate Arize

18

(1994) attempted to explore the relationship between REER and TB for 9 Asian countries, Johansen (1988)cointegration test was employed Data conducted in his paper from and quarterly data from 1973 to 1991 The results fromsignificant statistic stated that there was a long run connection between RER and TB The undervalued of currencies supportedto increase trade balance This reason implied ML condition is hold This conclusion is suitable forthe paper of Bahmani-Oskooee (1991) Instead of

using cointegration test of Johansen(1988), Shahbazet al.,(2012) seek for ML condition

of Parkistan from 1980 to 2006 with ARDL The impact of the real exchange rate on trade balance was discovered; however, the coefficient is negative (0.7677) and significant This indicates that when Pakistan reduces the valuation of real exchange rate, it will make trade balance become worsen For instance, if currency is devaluated

at 1%, the trade balance will decrease at 0.7677% Last but not least, this finding is suitable for many papers ofPakistan priory

Inrecent years, the role of exchange rate in every commodity in the economy are hot

topics, therefore, Bahmani-Oskooee, Ardalani, Bolhasani (2010), Bahmani-Oskooeeet

al., (2013), Baek (2014)haveapplied exchange rate volatility method to examine the effect of exchange rate for every commodity of economy Such asBahmani-Oskooee, Ardalani, Bolhasani (2010) used exchange rate volatility to discover the influence of exchange rate on industrial commodities Monthly data was used from 1991 to 2007 Two models for import and export were employed separately

whereX is the valuation of export commodity, M represents the valuation of import commodity, REER is real effective exchange rate Yt is US income and VL stands for volatility of dollars

19

The result indicated that a little industrial commodity was influenced by volatility, 24 industries for importand 20 industries for export, or 36% and 30% respectively, in the long run And the misalignment of USD did not seem to influence on trade commodities in the long-term, except 13 commodities or 20% Therefore, ML condition only held in 20%of economic sector

The conclusions for J curve or M-L Condition are very different Many papers indicate that there is only the long run relationship between trade balance and exchange ratesuch as Bahmani-Oskooee (1991), Narayan (2006).However, Rose and Yellen (1989) and Rose (1990) noticed that trade balance and real exchange rate have no any relationship Shirvani and Wilbratte (1997), Rahman et al.,(1997) had the same consequence Gomes, Paz (2005) realized that the J-curve exist in theirs paper.Some papers indicate that if real exchange rate is depreciation, it will lead to trade balance

deterioration such as Arora et al., (2003) and Shahbazet al., (2012)

In short, the disadvantages of FEER approach which is no evidence associated with equilibrium exchange rate in the long run And BEER approach implies thatthere is not exit normative formula For these reasons, PPP approach is employed to calculate real exchange rate

When testing the relationship between trade balance and exchange rate, many papers calculate real exchange rate following formula:

20

authors did not prove PPP hold However, in this research, s, p and p* between Vietnam and China will be proved PPP approach in the long term before calculating the RER

2.5 Conceptual framework

The conceptual framework shall show the way how PPP approach hold in the long run and how to measure the relationship between exchange rate and trade balance Using PPP approach with Horvath – Watson(1995) procedurerequires nominal exchange rate, Consumer Price Index (CPI) of Vietnam and CPI China for data After proving PPP hold between Vietnam and China by PPP, the real exchange rate is calculated Then, time trend is applied to measure misalignment between two countries.Finally, Johansen (1988)cointegration and VECM are applied toinvestigate the impact of exchange rate

on trade balance inlong term and short term These last techniques require the real exchange rate between Vietnam and China, income of Vietnam and China

21

Figure 2.2: Conceptual framework of this study

Nominal exchange

CPI Vietnam

CPI China

Purchasing power parity

Real exchange rate

Calculating misalignment

Trade balance

3.1 Tendency of the exchange rate between Vietnam and China

Figure 3.1 Nominal exchange rate between Vietnam and China,(VND/RMB)

The first feature that should be mentioned between 2002 and 2005 is that the nominal exchange rate between two countries experienced slight changes year by year before showing a considerable growth until 2008

A similar trend can be observed in the next stage with a steady rise from 2008 to 2010 Most noticeably of all, it can be seen that the proportion of this rate only went up dramatically in 2011 in comparison with other years There was one noticeable exception, however, this rate suffered a slight fall in the middle of the final year before reaching a new peak at over 3 200 until the end of 2012

3.2 Trade balance between Vietnam and China

Figure 3.2: Trade balance between Vietnam and China (million US dollar)

24

Firstly, we can see that with regard to Viet Nam’s export and import over this 14 year periods, there was one basic trend: upward trend However, the line graph represents the trade balance has decreased gradually since year 2000 What was the reason for this phenomenon?

The most striking feature was that import grew rapidly, especially from 2006 to 2012, though it had increased steadily over the previous years Then this figure rocketed and comprised nearly 26.7%, approximately 36.9 billion USD compared with previous year

On the other hand, the biggest contract was that export almost made up a very low proportion during a few constant years Particularly, it was obviously described in two stages, before 2010 export was under 5 billion USD The most dramatic changes only occurred from 2010, thus this figure improved much more and get amount to 13.3 billion USD in 2013 Nevertheless, this proportion was still lower than import over 2.77 times, hence trade deficit was a certain result

3.3 The main commodities of import and export between Vietnam and China in recent years

3.3.1 The main commodities of export to China in recent years

Table 3.1: The main commodities of export to China during 2010-2013 (million USD)

Chemicals and chemichal products

Machinery, instrument, accessory and

Note: Author’s calculation Source: GSO (2014)

The table3.1gives information about the amount of valuation of main commodities exported to China from 2010 to 2013 in three groups The lowest group (group 1) is lower than 500 million USD, group 2 is between 500 million 1 billion USD and the largest group (group 3) isover 1 billion USD

The valuation of commodities has upward trend year after year.In 2010, there was only one commodity, rubber, has the valuation over 1 billion USD and two commodities belong to group 2, these were coal, computer and their part Then, in the year of 2011, there was four commodities over than 1 billion USD and four commodities over than

500 million USD The total value of 4 biggest commodities were approximate 5 billion USD and four commodities in the group 2 were approximate 2.8 billion USD In 2012, the biggest commodities remained with the same number of four, however, the valuation increased to 5.43 billion USD The second largest group still remain four and

26

the valuation rose approximately to 3.1 billion USD In the contrary, at the end ofthe year 2014, the biggest group had only three commodities and the valuation was4.3 billion USD; nevertheless, the second group increased to six commodities and the valuation was4.5 billion USD

Generally, most of commodities have increased over times;however, the low valuation goods are increasing slightly and the high valuation goods are unstable As the rich mineral and agriculture country, most of exported goods are mineral such as coal, crude oil or agriculture such as rice, coffee, rubber, cashew nutetc.For the reasons of low commodities, low technology or in other words, these are low profit and many competitors in comparison with high technology Especially, the volume of agriculture products will be reduced easily by the weather and it is the weakness point in Vietnam’s export structure to China as well

3.3.2 The main commodities of importfrom China in recent years

Table 3.2: The main commodities of import from Chinaduring 2010-2013 (million USD)

Animal fodder and materials

97,544

108,232

248,178

- Auxiliary materials for cigarettes

60,630

44,554

57,539

58,233 Auxiliary materials for

medicaments

84,068

79,190

141,467

- Auxiliary materials for textile,

garment, leather, footwear

3,130,127

3,957,241

4,379,237

5,556,882 Chemicals and chemical products

912,332

1,137,705

1,202,801

1,379,494 Computers, Electronically

products & parts

1,860,312

2,596,554

3,643,599

4,854,978 Fertilizers

603,400

878,770

848,779

853,467 Fishery products

10,222

18,716

16,201

23,187

27

Fresh and processed vegetables

and fruit

156,133

157,972

163,388

157,834 Insecticides and materials

226,934

269,424

318,534

385,732 Machinery, instrument, accessory

4,477,616

5,182,375

5,190,668

6,567,814 Means of transport and equipment

631,196

629,654

448,322

505,297 Metal products

2,361,114

2,504,933

3,025,870

3,961,547 Paper and paper products

173,686

200,475

243,787

309,981 Petroleum oil and petrol products

1,395,848

1,700,857

1,700,595

1,710,081 Pharmaceutical and

pharmaceutical products

30,424

32,194

43,944

204,906 Rubber and plastics

693,154

934,100

1,003,205

1,289,309 Telephones, cameras and theirs

accessory

-

1,744,262

3,425,515

6,391,679 Wood and wooden products

169,133

186,595

200,361

200,955

Note: Author’s calculation Source: GSO (2014)

The table3.2gives information about the amount of valuation of main commodities importedfrom China from 2010 to 2013 The table 3.2 illustratesthe structure of import goods is quite similar and tend to increase from year to year These are necessary goods for Vietnam to reproduce products that supply to domestic and export to the world as well Contrast to exported goods from Vietnam, the table 3.2indicates that the higher valuation of imported goods always increase year over year All of them are the signs that Vietnam are depending on China materials What the problems if the supply from China stop or isshortage, the quality is not good? Do we become a landfill siteof industrial of China?Therefore, Vietnam should have scenarios to escape from China supplyor find new suppliersfrom developed countries

In 2013, according to GSO, Vietnam exported to China approximate 13.1 billion USD (increased by6%) and imported around 36.9 billion USD (increased by28%), viceversa The most important problem is that Vietnam often exports unrefined, raw material and

28

imports finished or refined products For instance, we export crude oil and imports petroleum refined or gas, we export latex of rubber and import tires, rubber products, plastic material, we export coal, mineral to import iron, steel export wooden and import paper products, wood and wooden products…

In conclusion, nominal exchange rate has an upward trend; nevertheless, the trade deficit has increased gradually since year 2000 The reason is that import grows faster than export and that reason makes the trade balance becomes worsen, especially from

2007 to present

29

MODEL SPECITICATION AND DATA SOURCE

This chapter describes the methodology and data collection Data was collected from International Financial Statistics (IFS), Organization for Economic Cooperation and Development(OECD) and Asian Regional Integration Center (ARIC) indicators system This thesis aims to prove PPP approach hold and the role of exchange rate and trade balance between Vietnam and China

4.1 Model for purchasing power parity approach

Because the disadvantage of the FEER approach is that it cannot calculate the equilibrium in the long term and it ignores the debt shock in estimate the exchange rate And the BEER approach has the disadvantage is thatthe fundamentals economic can explain exchange rate is vary such as technological process, control over capital flows (Doroodian, Jung and Yucel, 2010) So that it is difficult for us to put all variables in the equation with the good data Last but not least, two approaches above can find out the misalignment with itself, cannot compare between two countries

PPP approach is used to calculate the misalignment of VN currency against CN currency Calculating the misalignment after using PPP approachispositive means that VND is overvalued than RMB (Chinese currency) and vice versa

Take log from the equation (2.6) we have:

Log(s t ) = log(p t

) – log(p∗) (4.1)

So when we run regression, the formula is:

st = c + α1pt + α2p∗+ Ɛt(4.2)