Print Email Download Reference This Send to Kindle Reddit This
submit to reddit

Relationship between exchange rate and trade balance in Norway

1.1 Background of Norway economy

Norway is a modern Europe country with officially name is Kingdom of Norway. According to the World Bank, Norway Gross Domestic Product for year 2010(estimate) is worth $433.304 million dollars, so Norwegians enjoy the third largest GDP per capita in the world. Norway is a country richly endowed with natural resources-petroleum and gas, hydropower, fish, forest and minerals. As a result, Norway economic is mainly dependent on the petroleum sector, which accounts for nearly half of the exports and over 30% of state revenue.

Furthermore, Norway also becomes the top five exporters in seafood, crude oil and shipping industries among the world. The rich of resources in the countries contribute the large revenue to Norwegian. Norway is a less open economy and reported a trade balance surplus which is 25.4 million NOK in June of 2010. A positive of trade balance indicate that Norway current account surplus and exchange rate increase in the foreign exchange market.

1.2 History of Norwegian krone

Scandinavian Monetary Union had been established on May 5, 1873. This monetary union was formed by Sweden and Denmark to fix their currencies and stability their exchange rates. Norway currency which is krone (NOK) was introduced in 1875 after Norway join this union. In doing so, Norway krone have same level with Sweden and Denmark which is 2.48 kroner/kronor per gram of gold.

Many countries have remained their floating exchange rate after the collapse of Bretton Woods System in 1973. In this case, Norway authorities has stable the fix exchange rate on 1986 to 1992. However, this system has change during the period of unrest in European exchange market. As a result, the authorities have allowed the Norwegian krone to float on 10 December 1992. Moreover, the change of Norway exchange rate becomes more fluctuation after 1996 when it was unstable of oil price and petroleum in Norway. Later, the change in price of oil has influence the Norway future economy development.

1.3 Exchange rate behavior in Norway

Norway is the third largest of oil and gas exporter in the world after Saudi Arabia and Russia. The Norway currency change from one year to another compared to other currencies. During the mid 1980, Norway bank were strongly regulated the financial system in both the quantity of lending money and interest rate. However, when the oil price started to decline in 1985, many main industries in Norway had been influenced and damaged. To overcome this problem, Norway monetary policy has increase the interest rate to protect the kroner from devaluation. Besides that, monetary policy also reduces the disposal income of household and control the size of import from foreign country. At the same time, fiscal policy was implemented on December 1986 to fix the exchange rate. Moreover, Norway economy has strongly damage when the Asian currency crisis in 1997.Due to the price of oil increase in 2001, Norwegian Kroner grew to the high levels which is 8.99 NOK and against the US dollar and Euro. However, following by 2009, USD worth about 5.99 NOK. Here is the Norway exchange rate from year 1980 to year 2009.

Source: Federal Reserve System

1.4 The Concept of Exchange rate

Exchange rate can express as how much the nation currency worth respect to the foreign currency. In a slightly difference perspective, exchange rate defined as a currency price which can influence by the demand and supply of the currencies involved. The exchange rate concept can formulated as:

E = D / F

Where: E=prevailing exchange rate in the foreign exchange market;

F= foreign currency;

D= domestic currency trade in foreign exchange market.

There are two types of exchange rate which are nominal exchange rate and real exchange rate. Nominal exchange rate is the rate which the currency of a country is traded against for the currency of another. Central bank can fix and determinants the level of the nominal exchange rate because the quantity of money is printed by the central bank.

Real exchange rate is the nominal exchange rate adjusted for relative price levels. The real exchange rate is the nominal rate multiplied by the ratio of the foreign to domestic price levels. However, to measure the level of a country’s economic and assessing the impact of exchange rate on trade balance, real exchange rate will be apply rather than nominal exchange rate.

1.4.1 Exchange rate Regimes

There are two types of exchange regimes, which are fix or flexible. Under a fixed exchange rate regime, the government intervenes in the foreign exchange market in order to stabilize the exchange rate at fixed level. When the inflation in country is very high, government can pegged the exchange rate to stabilize the currency. In floating exchange rate regime, the nominal exchange rate is determined by the market. Under this regime, exchange rate will merely change over time to adjust the inflation rate. However, choosing a best exchange rate regime is depends on social choice problem by understand social, economic, and policies of that country.

1.5 Determinant of exchange rate

In a free market, the equilibrium of exchange rate occurs at the point at which quantity of demand equal the quantity of supply of the foreign currency. The factors causes the supply and demand of exchange rate change is economies variables such as export, import, domestic and foreign income. Since the exchange rate under the floating system is determined by the market forces, market fundamental and market expectations also interact to influence the exchange rate in both long run and short run.

Interest rate plays an important role to influence the movement of exchange rate. The increase in the interest rate thus causes the home currency appreciates in the short run. When the home exchange rate is appreciates, residents will purchase more goods from abroad country because the foreign good is cheaper for them. As a result, the size of import will increase while the quantity of export will decrease.

Another factor which is often mentioned in explaining the movement of exchange rate is inflation rate. A high inflation means the relative price level was increase over a period, and this case usually accompanied by the depreciation of exchange rate. As a general rule, the lower inflation rate will raise the currency value, thus the purchasing power for residents increase relative to other foreign exchange rate.

The current account and the balance of payment is a term of trade to influence the exchange rate. When the demand of export for foreign countries is greater that import, the term of trade will become positive because current accounts is surplus. However, current accounts deficit when the price of exports raise smaller rate than import. As a result, the value of exchange rate decrease and demand for currency also reduce.

Finally, the increase of money supply in a country will affect the exchange rate. A country with large public deficit and debt will attractive less foreign investor. To overcome the debt obligation, government will sell the bond in market to increase the money supply. In other words, the increase of money supply cause the higher inflation exists in the economy. Thus, a large debt in country may influence the exchange rate and worsen the trade balance.

1.6 The Concept of Trade Balance

The trade balance, or sometimes called “net export” is a component of GDP when measured an economic growth. It is the relationship between nation’s export and import of merchandise over a period of time. In short, trade balance is export less import. A trade balance is known as trade surplus when the export is more than import; while the trade balance defined as trade deficit when import is more than export. There are three major factors influence the balance of trade, which are foreign exchange rate, domestic and foreign income and foreign price level.

According to the Marshall Lerner condition, when the exchange rate depreciation, the size of export in the country will increase more than import, thus trade balance will improve. On the other hand, when the exchange rate appreciation, it will causing the trade deficit to rise because the import is more than export.

Besides that, the change in domestic and foreign income also plays an important factor to influence the trade balance. If the Norway nation income rises, the residents will consume more foreign goods in abroad countries. As a result, trade balance will deficit because the increase of Norway income cause the import increase. At the same time, when the Norway income decrease, the demand for foreign goods will reduce, so this leads to trade surplus.

Furthermore, an increase in price level in country also will affect the trade balance. If Norway price level increase against foreign countries, it will cause the size of import increase more than export, thus the trade deficit become worsen. In general, the falls of price level in country will cause the foreign trade surplus to rise.

1.7 Marshall Lerner Condition vs. J-curve

Marshall Lerner condition explains the relationship between exchange rate and trade balance. When a country’s exchange rate devaluation relative to other countries, this will improve the balance of trade. At the same time, export is increase because the price of export is cheaper for foreign countries. Besides that, Marshall Lerner condition also examines the price of elasticity for export and import goods. If the demand elasticity for imports plus the demand elasticity for the export is exceeds one, then the Marshall Lerner condition holds which the depreciation of exchange rate will improve the trade balance.

However, Marshall Lerner condition does not meet when the price of goods tends to inelasticity. In the short run, depreciation of exchange rate initially worsens the trade balance. But with the time lag, the quantity adjustment effect becomes relevant, thus the trade balance will improve in the long term. This effect is called J-curve patterns. In conclusion, both Marshall Lerner condition and J-curve are two concept that explain the effect on trade balance when exchange rate depreciation or appreciation.

1.8 Problem Statement

Exchange rates play an important role in the economic growth. Today, many economists have been studied the relationship between exchange rate and trade balance both in short run and long run. Under the rules of Breton Wood system, the exchange rate is fixed. However, after the collapse of the fix exchange rate in 1973, many nations change their currency to the flexible exchange rate. Most studies in the field of exchange rate have argued that large fluctuation of the exchange rate will impact the trade balance. Likewise, some studies point out their result that the change of exchange rate is determine by the domestic income, foreign income and interest rate. As a result, these variables have correlation relationship with the trade balance.

Indeed, large fluctuation of exchange rate in the world is a major issues concern by economists. Uncertainty in exchange rate could impair the smoothing of economic growth, which affects the trade and capital movements. Furthermore, when exporter and importer unable to overcome the risk they face, trade balance will result as deficit. This will follow by the negative effects in current account and balance of payment for developing countries. Besides that, instability of exchange rate could cause many issues and problems happen in economic growth.

The variability of currency affects the size of foreign exchange reserve. The nation whose loss the value of currency is more worsen when exchanges the goods in foreign exchange market. According to Marshall Lerner condition, when currency devaluation against another foreign currency, it will improve the economic trade balance. This condition states that devaluation of exchange rate will cause the price of exports cheaper for foreign countries, afterward export performance will increase. At the same time, demand for import will reduce because the foreign goods are expensive for the nation citizens. On the other hand, the price of import and export are determined by the price of elasticity, so the J-curve phenomena will exist.

Moreover, another problem that has happened under the floating of exchange rate is foreign debt obligation. Since the movement of exchange rate can cause the current account surplus or deficit, it tends to affect the transaction of currencies in commercial bank. Central bank and other monetary policies are hard to buy and sell the foreign exchange because the change in exchange rate is unpredictability. This operation might also cause the cost of debt service increase. In short, fluctuation of exchange rate tends to deterioration the economic growth and international trade in one country.

Thus, this paper tends to study Norway economic growth and the relationship between exchange rate and trade balance. Norway exchange rate change from one year to another compared with foreign currencies. Since the Norway is the third largest of gas and oil exporter in the world, Norwegian kroner were fluctuating with the price of oil, which may have affected investor expectation in Norway future development economy. So, one question needs to be asking, however, whether the Norway trade balance determined by the fluctuation of exchange rate ?

1.9 Research Objective

The general objective of this paper is to examine the relationship between exchange rate and trade balance in Norway. According to Marshall Lerner condition, price elasticity of demand for exports and imports must be greater than one. A depreciate of exchange rate will improve economic growth, thus it has positive relationship with trade balance. At the same time, this paper will find out whether the M-L condition satisfied and evidence of J-curve patterns exists in the Norway trade balance. More specific, this study tends to identify what factors determine the Norway exchange rate and trade balance in the long run economic growth.

1.10 Significant of Study

Exchange rate is an important variable for developing countries in the foreign exchange market. After the Bretton Wood System breakdown in 1973, many developing countries have moved their pegged exchange rate to the floating exchange rate. This study will focus the determinant of the exchange rate on trade balance in Norway. With the appropriate model, we can explain the exchange rate behavior in Norway and the impact of fluctuation exchange rate on economy growth.

In general, the Ordinary Least Squares (OLS) model seldom applied by researchers to examine the relationship between exchange rate and trade balance. Thus, with available and latest information, this study will use unit root test and OLS model to testing the long run relationship between these variable. Besides that, by using time series data from 2000 to 2009 and other useful variables, this study will present the empirical result whether there is a significant correlation exists among these variables.

Furthermore, the unstable of exchange rate has effect the economic performance and international trade balance. So this study will highlight the movement of exchange rate in order to maintain the trade balance performance in Norway. More specific, this study will allow the researchers to realize how much the impact size effect on the economy growth when the exchange rate devaluation or appreciation.

Lastly, the contribution for this study will enhance future researchers to compare and investigate the exchange rate model for other developing countries.

2.0 Introduction

Large fluctuation of exchange rate is a major issue faced by many developing countries. This is because uncertainly of the exchange rate will affect the economic growth both in short run and long run. Many economists have been examined the relationship between exchange rate and trade balance in economic. Due to their empirical studies, they found that J-curve pattern exist in the long run where trade balance surplus or deficit depends on the prices elasticity. In the short run, devaluation initially worsens the trade balance, but then the trade balance start to increase in the long run and higher than the initially volume. However, this result disagree by some economists, Rose and Yellen (1989) suggest that no evidence of J- curve response in America data .Their major finding showed that only little relationship between US exchange rate and trade balance.

2.1 Literature Review

Yusof (2007) examined the relationship between real exchange rate and trade balance both in short run and long run in Malaysia. They used co- integration model and causality test to indicate whether the J-curve effect got exists in the long run. In this regards, they were covering the quarterly data from 1997: 1 to 1998: 2. Their final results suggested that devaluation of exchange rate improve the trade balance in the long run. They also examined that the J-curve pattern exists and effect in the long run. In the short run, other variables such as domestic income and foreign income also determined the trade balance.

Narayan (2004) used the Granger causality test and co-integration test to present the relationship between trade balance, exchange rate, domestic income and foreign income in New Zealand’s trade balance. However, the empirical result showed that there is no significant relationship exists between trade balance, exchange rate and other variables. This result stated that these three variables are no co- integration in the long run. Besides that, he also used impulse response analysis and found that J-curve behavior exists in the New Zealand’s trade balance during the period 1970 to 2000.

Similarly, Narayan and Seema (2004) also using co-integration test to studies the Fijian economy which covers the period from 1970 to 2002. The variables use for their study is real effective exchange rate, domestic income and foreign income. By using co-integration analysis, they found that there is a significant relationship between trade balance and these variables. However, the empirical result from ADRL, DOLS, and FM-OLS model show that domestic income is the main factor determine the Fijian trade balance and follow by the foreign income. Finally, they also found that the evidence of J-curve hypothesis exist in the Fijian economy.

Gomes and Paz (2005) studies this case for Brazil economic and using the data from Jan 1990 to December 1998. In their studies, they focus on the Marshall-Lerner condition and the J-curve phenomenon to explain the change in exchange rate on trade balance in Brazil economy. Basically, they are used VEC-M model and VAR model to construct the ML model. Their final result fount that Brazil economic meet the ML condition. In long run, when Brazil exchanges rate devaluation, trade balance improves, thus Brazil current account is surplus and balance of payment is positive. However, the Brazil trade balance worsen when exchange rate devaluation in the short run. This condition made the J-curve behavior present in the Brazil.

Petrovic and Gligoric (2010) apply granger causality test to explore whether exchange rate depreciation increase trade balance in Serbia or worsen it. The times series data for their research is from Jan 2002 until September 2007 which take monthly data. Moreover, the methodology used in their paper is Johansen’s co-integration test and Autoregressive Distribute Lag (ADRL) approach. From their findings, they conducted that real exchange rate devaluation has a significant relationship with the trade balance in Serbia. Besides that, they also point out that there is a J-curve pattern exists in the Serbia trade balance.

However, some empirical studies found that J-curve pattern does not necessary exists in all country and affect the trade balance. Unlike other countries, China has no J-curve pattern both in the short run and long run. Brada et al. (1993) used quarterly data from 1980:1 to 1989:4 to address how China’s trade balance affected by the fluctuation of the real exchange rate, domestic income and foreign income. They argued that with no J-curve effect exists, devaluation of China’s real exchange rate has improved balance of trade in economy. By apply a Johansen’s co-integration test; they provide strong evidence that there is a stable long run relationship exists between exchange rate and trade balance.

Buluswar et al. (1996) carried out that there is no co-integration relationship between trade balance and exchange rate in India during period 1960-1990. By using the Johansen model, they points out that India government policies didn’t have effective monetary approach to control the trade deficit. Devaluation the exchange rate in the long run significant no affects the India trade balance. At the same time, Granger test conclusive the other variables such as money supply and price index, including exchange rate no significant relationship with the trade balance. Finally, this study does not show J-curve effect in the India Trade balance.

Similarly, this result was support by Singh (2004) who also studies the J-curve hypothesis in India economic growth. By using the GARCH model, the result showed that volatility of exchange rate does no significant with the trade balance. However, the error correction model carried out that real exchange rate and domestic income plays a major role in determinants on India trade balance. This model is covering the quarterly data from 1975:02 to 1996:03. His final result does not support the exits of J-curve effects in India trade balance and there is no significant relationship between exchange rate and trade balance.

Narayan (2004) previous studies point out that there have existence the of J-curve pattern in New Zealand and Fiji. However, Narayan (2006) later studies argued that there is no evidence of J-curve behavior in the China. This result were based upon monthly data, cover the period from November 1976 to September 2002. The main purpose of his studies is to investigate the relationship between China’s trade balance and the real exchange rate with the USA. Furthermore, Narayan (2006) applied impulse responses function to estimate the relationship between China balance of and exchange rate. Although his final result point out that Marshall Lerner condition was hold but the J-curve affects does no deal with the case in China.

Baharumshah (2001) carried out a similar result when he studies the effect of exchange rate on bilateral trade balance. In his major study, Baharumshah (2001) using the quarterly data from 1980:1 to 1996:4. He attempts to identify what factor determinants the trade balance of Malaysia and Thailand with US and Japan. By apply the Johansen co integration test, he found that real effective exchange rate, domestic incomes and foreign incomes are important variables to influence the trade balance. Besides that, his paper has reported that devaluation of exchange rate improves the trade balance, but no evidence of J-curve pattern to fit the data. Wilson and Kua (2001) also study the bilateral trade for US and Singapore. The present of partial reduced form model resulted that the real exchange rate does not influence the bilateral trade balance. They covering the quarterly data from 1970 to 1996 and found that only little evidence of J-curve phenomena exists when the exchange rate devaluation.

Furthermore, Baharumshah and Yol (2007) study the fluctuation of exchange rate on the bilateral trade balance between 10 African countries and US over period 1997 to 2002. They use Johansen co integration analysis that the trade balance is co integrated with the real exchange rate. Besides that, the Marshall Lerner condition does not satisfy in all 10 countries because the elasticity of exchange rate is less than 1. The most important finding shows that all 10 African countries trade balance has significant determinant by the fluctuation of exchange rate.

Most studies in the field of exchange rate resulted that Marshall Lerner condition was hold when the home currency devaluation. Brahmasrene (2002) empirical results conclude that Marshall Lerner condition was hold when he studies the impact of real exchange rate for Thailand trade balance with its major trading partners. Besides that, quarterly data from 1990 to 2000 was collected for his empirical study. The result from co integration test showed that real exchange rate has significant relationship with the Thailand bilateral trade balance. Furthermore, his result suggests that real income does not have significant when determinant the Thailand trade balance.

Similarly, Tochitskaya (2007) using quarterly data from 1995:4 to 2004:4 to study Belarus economic and trade balance. The OLS regression model result that depreciation of exchange rate has a significant impact on Belarus trade balance. Therefore, Marshall Lerner condition meets. Moreover, unit root test carried out the dependent variables and other variables does not show the relationship in the long run. However, Rose (1991) examines that Marshall Lerner condition does not satisfied when he studies the role of exchange rate in 5 major OECD countries which are UK, Canada, Germany, Japan and US. The data are monthly and seasonally adjusted from 1973 to 1986. By using imperfect substitute model, the findings show that exchange rate is not an important factor determinant the trade balance.

Kyereme (2002) main objective is to explore the determinants of trade balance between US and Australia in the long run. By using co integration test and sample data 1965 to 1998, he found that there is a positive relationship between exchange rate and trade balance. Besides that, the regression result conducted that price ratio is a major factor determinant the trade balance. Other variables such as GDP ratio, money supply, real exchange rate and lending ration also have significant to influence the trade balance.

Bahmani-Oskooee (1991) also applies co integration analysis to determine whether there is a long run relationship exists between trade balance and real effective exchange rate. He study cover quarterly data from 1973 to 1988 for eight less developed countries (LDC), which are Argentina, Bahamas, Bangladesh, India, Korea, Philippines, Greece and Thailand. His final finding concludes that Marshall Lerner condition satisfied in the LDC trade balance. Further analysis point out the relationship between trade balance and exchange rate were co integrated in the long run.

A large and numerous studies have focusing their study on the relationship between trade balance and exchange rate. Similarly, Kale (2001) used co integration method to examine the relationship for these two variables in the Turkish economy. This study used the quarterly data from 1984:1 to 1996:2 for his framework model. Furthermore, Bickerdike Robinson Metzler condition was holds when the devaluation of exchange rate in the long run improves Turkey trade balance.

In another major study, Rose (1990) unable to found a significant relationship between exchange rate and trade balance. In this paper, he used non structural techniques to examine the impact of the real exchange rate on the developing countries. Rose (1990) used two set of data which is annual data spans 1970 to 1988 and quarterly data which cover from 1977 through 1987 for 30 developing countries. The result for his finding is difference with other studies because real exchange rate has insignificant impact on the multilateral trade balance for developing countries.

Furthermore, Wilson (2001) also examines the same result with Rose (1990) when studies the relationship between exchange rate and trade balance for bilateral trade balance. He chooses Singapore, Malaysia and Korea with USA and Japan on quarterly period which cover the period from 1970 to 1996. By using the Johansen and Granger test, he suggests that real exchange rate does not have a significant relationship with the trade balance. Besides that, he found that no evidence of J-curve effect in Singapore and Malaysia economy. However, J-curve patterns exists in Korea trade balance when its trade with Japan and USA.

Tai and Lin (1998) use granger causality model to study the Taiwan-Japan trade balance from 1981:1 to 1003: 6. The ADF unit root test show that the variables which are exchange rate, money supply, output and price index are integrated. However, Johansen results examine that trade balance does not have co integration relationship with these four variables. In both short run and medium term, the impulse response analysis point out that the impact of exchange rate on trade balance are lower than the impact of money supply and income. At the same time, Hui (2002) also study the relationship between Taiwan exchange rate and trade balance with Japan and US. The sample size is cover quarterly data from 1981: to 1998: 4. By using co integration test and error correction model, he suggests that the relationship between exchange rate, export and trade balance are exists in the long run.

Amano and Norden (1995) also apply the same econometric model with Tai and Lin (1998) which is Granger causality test to run the relationship between exchange rate and the term of trade for Canada-US bilateral trade balance. The causality result shows that the shocks of term of trade have a significant impact on exchange rate. The sample data use for their study is over the 1973M1 to 1992M2. Finally, their studies conclude that the term of trade plays an important role to Canada-US exchange rate, while the monetary policy is a second factor determinant the change of exchange rate.

Luis et al. (2008) main purpose of their study is to examine the role of exchange rate in the US and UK trade balance. The sample data for exchange rate and trade balance are from 1988: 01 to 2005:12. By apply the unit root test and co integration model, the result were different. The null hypothesis for the unit root test is rejected for the balance of trade while the hypothesis for the exchange rate does not reject. Thus, in regressive model, exchange rate treats as weakly exogenous variable in order to solve this problem. Their final finding show that the trade balance is non stationary variable and the impact of exchange rate take a long period to disappear.

In order to measure the overvaluation of rupiah during the Asian crisis in 1997, Saxena (2002) study the exchange rate dynamics in Indonesia over 1980:1 to 1997:4. This study using the co integration method, unobserved component model and structural vector auto regression (SVAR) model to analysis the exchange rate policy in Indonesia. Saxena (2002) empirical results show that exchange rate is positively related to the term of trade and government expenditure. Lastly, the unobserved component model carry out that Indonesia trade balance increase when exchange rate devaluation.

Arize (1996) main objective is to study the long run relationship between trade balance and the terms of trade in 16 countries. The quarterly data cover from 1973:2 to 1992:4. Johansen co integration test show that the term of trade and the trade balance are co integrated for all 16 countries. In the long run, devaluation of exchange rate improves the trade balance. Marshall Lerner condition holds, thus the price of export and import elasticity is greater than 1.

Peree and Steinherr (1989) examine the uncertainty of exchange rate and its effects on foreign trade in US country during the period 1960 to 1985. They use two measures to test the uncertainty of exchange rate. One is weighted function and the second is depends on the duration and amplitude of the exchange rate. Their empirical results suggest that the uncertainty of exchange rate has negative effects on the trade balance, therefore resources allocation also affects.

Yazici (2008) investigate and compare the change of exchange rate on Turkish three main economic sectors, which are agriculture, manufacturing, and mining sector. The empirical work uses the quarterly data from 1986:1 to 1988:3. By using the Bahmani-Oskooee (1985) model which is J-curve terms, Yazici (2008) found that J-curve pattern exists in these three sectors. When exchange rate devaluation, the trade balance for these sectors initially increase then worsen and then increase again. In the short run, all three sectors reaction to the change of exchange rate, while in the long run, these three sectors do not exhibit the similar way with exchange rate.

Alam (2010) investigate the volatility of exchange rate on Pakistan aggregate exports demand during 1979 Q3 – 2005 Q4.The ADRL analysis show that the real export are co integrated with the foreign economic activity, exchange rate and the volatility of exchange rate. Besides that, the short run dynamic causality did not found from the volatility of exchange rate on Pakistan aggregate export demand.

3.0 Introduction

This chapter discusses the properties of times series and some econometric model to investigate the relationship between exchange rate and trade balance in Norway. In this study, I will use the Unit root test and Ordinary Least Square (OLS) model to measure the empirical results. Besides that, Multiple Linear Regression applies in this chapter because two independent variables are used.

3.1 Data Sources

In this study, times series data are cover from year 2000 to year 2009. This annual data were obtained from various sources. For the export, import and trade balance, the data sources are come from U.S. Census Bureau, Foreign Trade Division, Data Dissemination Branch, Washington, D.C. 20233. The data for Norway exchange rate obtain from Federal Reserve System while the data for inflation rate collected from CIA World Factbook.

3.2 Hypothesis Formulation

In order to test the relationship between these variables, we can state an assumption in hypothesis. However, this assumption may or may not be true. There are two types of statistical hypothesis, which are null hypothesis and alternative hypothesis. In this study, these hypotheses would be expressed as:

Ho: β1=β2=0 (insignificant)

H1: β1≠ β2≠ 0 (significant)

3.3 The model Specification

In this study, the multiple regression model used is as below:

Y=α+β1X1+β2X2+µt

Where Y= trade balance

α=the Y intercept

β=regression of coefficients

X1=exchange rate

X2= inflation rate

µt=error term

The relationship between dependent and independent variables are below:

Trade balance- Dependent variable which is being predicted. Trade balance is export less import. Trade balance surplus exists when export more than import, while trade balance deficit occur when import more than export.

Exchange rate- Independent variable and use to determinant whether the exchange rate volatility can affect the trade balance. When exchange rate devaluation, trade balance will improve. However, balance of trade will deficit when exchange rate appreciation.

Inflation Rate- Another independent variable use to test the co integration relationship with the trade balance. The increase of inflation rate in national will increase the price level at country. Thus, it causes the import for country more than export, and presents the trade deficit.

3.4 Method of Analysis

The Multiple Regression Models is an econometric technique use to examine the relationship between dependent and independent variables. More specifically, Multiple Regression Analysis enables us to understand how the dependent value will change when independent variables is varied. The form of the multiple regression models as below

Y=α+β1X1+β2X2+β3X3+…… βn Xn+ µt

to explain the behavior of the dependent variable. The µ is known as stochastic error term or simply an unobservable random variable taking positive or negative. We shall examine the nature of stochastic disturbance term, but for the movements assume that it is a proxy for all the omitted variables that may affect Y but are not included in the regression model.

3.4.1 The method of Ordinary Least squares (OLS)

In regression analysis, OLS is a popularly method used by estimators because it has very strong theoretical properties, which are known as the Gauss-Markov Theorem. Under the assumption of classical linear regression model (CLRM), the OLS have minimum variance, so they are BLUE (Best Linear Unbiased estimators).

3.4.2 Fit of the Regression Model

The goodness of fit in the regression model can be assessed by the coefficient of multiple determinations, denoted by the symbol R2. Now, let us define

R2=ESS/TSS

=1-RSS/TSS

=1-(∑ Ûi2 / ∑ yi2)

Now ∑ yi2 is the independent of the number of X variables in the model. The RSS, ∑ Ûi2, however, depends on the number of regressors present in the model. To compare two R2 terms, one must taken into account the number of X variables present in the model. Thus, we consider the adjusted R2, denoted by Ṝ2. The Ṝ2 is a modification of R2 that adjusts for the number of term in a model.

3.4.3 F-test

We use F-test to test the significant of a multiple regression because the independents variables are more than one. We reject Ho if calculated F > critical value.

F= (ESS/k-1) / (RSS/n-k)

Set up the appropriate null and alternative hypothesis:

Ho: β1=β2=0 (invalid)

H1: Not all the slope coefficients are simultaneously zero (valid)

If F > Fa (k-1, n-k), we reject Ho. The fitness of model is good because the higher F, the higher R2 will present.

3.5 Unit Root Test

Unit root test is a statistical method to tests whether a time series data is non stationary or stationary. We start with

Yt = pYt-1 + ut -1≤ p≤ 1

If p = 1, that we have a unit root, meaning the time series under consideration is non stationary. However, we cannot estimate and test the hypothesis that p=1 by T-test because the test is biased in the case of a unit root. Therefore, we regress Yt and lagged one period to Yt-1 and find out whether the estimated p is statistically equal to 1? Now we subtract Yt-1 from both sides to obtain:

Yt -Yt-1= pYt-1 - Yt-1+ ut

= (p-1) Yt-1+ ut

This can be alternatively written as

Δ Yt = δ Yt-1+ ut

If δ =0, then p = 1. So, the time series data is non stationary (unit root). The most famous unit root tests are Augmented Dickey Fuller (ADF) test and Phillips-Perron test. Both of these tests use the existence of a unit root as the null hypothesis.

3.5.1 Augmented Dickey Fuller (ADF) test

In ADF, we still test whether δ =0 and the ADF test follows the same asymptotic distribution as the DF statistic, so the same critical value can be used. The ADF model can be applied by as:

Δ Yt = α +βt + γ Yt-1 +δ1 Δ Yt-1 +…… + δp Δ Yt-p + ɛt

Where α is constant, β is the coefficient and the p is the lag order for the autoregressive process. The ADF test is only valid for AR (1) .To test the higher order autoregressive, the lag of the order p has to apply into the test. Another alternative approach to examine the information criteria is Akaike and Schwarz.

3.5.2 Phillips and Perron test

Phillips and Perron test also address the same result as the ADF test. They use nonparametric statistical methods to take care of the serial correlation in the error terms without adding lagged difference terms. To test a unit root, we reject Ho when the statistical value less than critical value.

Print Email Download Reference This Send to Kindle Reddit This

Share This Essay

To share this essay on Reddit, Facebook, Twitter, or Google+ just click on the buttons below:

Request Removal

If you are the original writer of this essay and no longer wish to have the essay published on the UK Essays website then please click on the link below to request removal:

Request the removal of this essay.


More from UK Essays

Doing your resits? We can help!