Impact of Exports and Imports on Economic Growth in Rwanda: 1961-2018

The aim of this paper is to investigate the impact of trade on economic growth in Rwanda. This paper uses exports and imports for trade and gross domestic product for economic growth. Research questions were formulated as (1) Are exports, imports and economic growth cointegrated? (2) Is there a long or short run relationship between those Variables? (3) Are there any causal relationships between factors (4) what the direction of the causality is it? Annual time series data from World Development Indicators for the period from 1961 to 2018 have been used. The methods of linear regression for estimation of Vector Auto regressions models have been used. Our findings established that VAR was appropriate model, and GDP, Exports were stationary at first differences while Imports was stationary at second difference but not at levels. Hence the two series were integrated of order one and the third one was integrated of order two. Tests of cointegration indicates that the three variables were not cointegrated, implying there was no long run equilibrium relationship between the three series. The causality test indicated that exports and imports influenced GDP. On the other hand, we found that there was a strong evidence of unidirectional causality from exports to economic growth. However, there was bidirectional causality between GDP and imports. These results provide evidence that exports and imports, thus, were seen as the source of economic growth in Rwanda.


Introduction
Theoretically, it has been asserted that both exports and imports can play a vital role in economic development . The abstract and empirical research mainly concentrate on either the association between exports and imports on economic growth (Serhat Yuksel, 2016). Exports of goods and services are seen as an engine of economic and social development for a number of motivations, include exports that ask companies to alter and improve to maintain market share (Sahoo, 2014).
Exports guarantee higher revenues and profits, on the other side. Alternatively, they reduce dependency on local markets since; the market base will grow, leading only to a decrease of local clients, in the case of growth in overseas markets (Dickey D. &., 1979). If it is not, export are able to minimize the effect of the market volatility, business become more captive to financial change, altering client requirements and seasonal fluctuations in local economy (Dickey D. W., 1981). Finally, and with regard to the advantages of exports, it can be summarized that an increase in export leads to an increase in the access to the currencies, which increases the national income, the turnover and surpluses of the state (Khan, 2012). This lead to the improvement in the standard of living. Despite these export advantages, they sometimes do not produce profitable outcomes that do not add to the country's greater development, for several reasons, including: the existence of competition is greater than expected, the products is unpopular or popular in the markets other, instability in the target country as a result of wars or civil conflicts, weak media publicity and definition exported product or other similar reasons (Engle, 1987).
As far as imports are concerned, the weakness of the country in meeting its own requirements is usually expressed and made dependent and at the expense of foreign nations (Karamanaj, 2014). Imports lead, contrary to exports, to a local currency departure and weaken the equilibrium of trade so that economic growth will weaken (Serhat Yuksel, 2016). However, and in some cases it is considered the import source of economic growth, especially if it includes hardware and electronic equipment to help and contribute to the increase and improvement of the investment, or include products that require a production value of more than imported (Hatem Hatef Abdulkadhim Altaee, February 2016). Due to these reasons, the export and import remains a controversial topic their ability to influence the social and economic growth of the countries.
Rwanda is a developing country and geographically landlocked country and also is considered the fastest growing economy in East African Community with a GDP of 9.80 billion (2018) compared to 9.137 billion USD (2017), thus the per capita GDP is 826.30 USD per year, and after that was 780.80 USD last year, an increase over the previous reporting 4.2 percent (Planning, 2018(Planning, -2019. In 2018 Rwanda's total trade amounted to US$1,090.34 million, an increase of 9.16 percent over the fourth quarter of 2017, Exports totaled US$ 169.91 million, Imports totaled US$ 841.92 million and Re-exports were valued at US$ 78.5 million (NISR, 2018).
Researchers and economists continue to discuss and argue about the connection between economic growth and growth in exports and imports (Uddin M. K., 2010). Some believes that export leads to economic growth and thereby encourage Export Led growth Hypothesis, while others sees economics growth leads to growth in export and support Growth Led Export Hypothesis (Shirazi N. S, 2005).
These divergent views on the relationship between export, import and economic growth put many developing economies in dilemma of whether to open up their economies to promote international trade or whether they should concentrate on economic activities that will promote international trade. It is due to these contradicting evidences about the dynamic relationship between exports, imports and economic growth that the paper attempt to revisit these relationships in the case of Rwanda for the period 1960 to 2017.
The objectives of this research are to: 1. Examine the relationships between Exports, Imports and GDP. 2. Determine whether the relationships are long run or short run phenomena, or both. 3. Determine whether there is any causal relationship between Exports, Imports and GDP. 4. Examine the direction of the causality.

Literature review
From its very beginning, Bangladesh, a developing economy, has had a trade deficit This document strives to comprehend the conduct of the time series of Bangladesh's complete exports and imports. Root test units acknowledge the random walk in the complete time series of exports and imports. The Johansen co-integration test demonstrate that these two factors have a long-term balance connection. The research seeks to discover causal relationship through an error correction mechanism, if cointegration exists. The outcomes of the test of the test show long-term bidirectional causality and short-term unidirectional causality between Bangladesh import and export. The findings of the research shows that the global budget restrictions of Bangladesh do not violate them (Uddin J. , August 2009).. The prospective weapon of export, importation and economic growth in Bangladesh's economy (Akhter, 2015). Bangladesh is a developing country; from its very beginning, its economy has been hit by trade deficits. The aim of this investigation is to investigate the effect on the economy of Bangladesh of exports and imports. In all manner of products and services, no nation in the globe is self-sufficient. To satisfy its own requirement, each nation must import. This is no exception for Bangladesh. The manufacturing of needed equipment, medicine, cars and other products in Bangladesh are dedicatory. Therefore, it must import products of this kind from outside. However, after satisfying their own requirement, all nations can generate one or more products excessively. Once its request has been met, Bangladesh can excessively generate jute, tea, clothing and many other products. Therefore, it is recognized that by importation of products, Bangladesh can satisfy its deficit and export its surplus products once the demand is satisfied. This research aims to investigate the links between Bangladesh's exports, imports and economic growth. The study was based primarily on secondary information sources gathered from different books, newspapers, research works, publications etc. In this research, the latest export, import and economic growth literature in Bangladesh has been tested. The research shows the favorable and reverse situation with regard to imports for the effects of exports on economic growth.
(Md. Tareq ferdous Khan, March 2012) The document shows unthinkable development in the GDP, the exporting of goods and services (EXPORT) and the importation of goods and service (IMPORT) during the last three decades in the macro-economic indicators. Each EXPORT and IMPORT indicator has occupied a substantial proportion of GDP in the past, although its

Methodology
The analysis used in this study coat annual time series of 1960-2017 in Rwanda from World Bank. The data set entails of observation for GDP (current US$), exports of goods and services (current US$), and imports of goods and services (current US$). All data set are secondary brought from World Bank.
Total Exports (EXP) by Rwanda, complete imports (IMP) and economic growth (GDP) are the factors used in this research. As proxy for economic growth in Rwanda Gross Domestic Products is used. All the variables are taken in their raw data.
We will operate the further apt method, which is aimed to firstly set up the degree of integration of each variable. If the variables are all integrated in level, we will clench an estimate founded on a linear regression.
In contrast, if the variables are all integrated into the first difference, our estimation will be based on an estimate of the VAR model. Concerning the estimation of VAR models types and when the variables are impartial in the first difference we will look and plot the cointegration between the variables, if the cointegration test denotes the non-attendance of cointegration relation then we will involve the Unrestricted vector Auto-Regression (Unrestricted VAR) and the Granger causality Tests. If the cointegration test elects the presence of a cointegration relation between the different variable studied, Vector error correction model (VECM) will be employed.
Early empirical formulation well tested to track down and to assume the causal bond between imports, exports and GDP growth by mixing exports and imports into the aggregate production function, and which is used (Ramos, 2001). The augmented production function, including both export and import is expressed as: The function can also be represented in a log-linear econometric format thus: Where: 0 is the constant term, 1 coefficient of variable (exports), 2 Coefficient of variable (imports), is the time trend and ℇ is the error term.

Unit roots Tests
Unit root tests of Log (GDP), Log (Exp) and Log (Imp) have been conducted using Augmented Dickey Fuller (ADF) and Philip Peron (PP) tests.

Source Eviews 7
The results of the stationary tests ADF and PP shows that GDP and Exports variables are integrated in order (1), while the Imports variable is integrated in order (2),namely in first difference for both GDP and Exports and second difference for Imports and in all levels (1%, 5%, 10%).

VAR Lag Order selection Criteria
As soon as the order of integration of the studied variables was in first differentials, we have determined the cointegration between them. However, before this step, one must determine the number of delay existing in this estimate. To accomplish this, we have applied the VAR Lag order selection criteria method.

Source Eviews 7
The lag order selection was determined using five different criteria LR, FPE, AIC, SC, HQ.
The results of the VAR lag or selection criteria show that the number of delay chosen is equal to 1. Therefore, and at this moment, we are ready to process the existing cointegration number.

Cointegration Analysis: Johanson Test
To obtain the number of cointegration existing in our situation, we use the most effective and suitable tests, which is the Johanson test.

Source Eviews 7
The Johanson test indicates the absence of a cointegration relation between the variables studied. Therefore, we will use an estimate based on the VAR model and the Granger causality test.

VAR Estimation
The purpose of the model estimation is to identify and see if there are effects between the independent variable that are negative or positive on the dependent variable.  The results of the VAR model estimate show that the variable that designates exports has positive effects on GDP, but it does not have significant probability. On the other hand, the variable that designates imports has a negative effect on GDP, but it also has a probability that is not significant.

Granger Causality Test
The objective of applying the Granger Causality Test is to verify whether there is a causal relationship between the different variables existing in our empirical investigation. The causality test indicates that exports influence the GDP and also imports influence GDP. On the other hand, GDP influence imports but does not influences exports.

Residual Diagnostics Tests
To verify that our empirical work is acceptable and that our estimate is well treated. We use a set of tests called residual diagnostic tests. All residual diagnostic tests are satisfactory and assert that our model is acceptable and well treated (R 2 is greater than 60%, Fisher statistical probability is less than 5%, Breusch-Godfrey Serial Correlation LM T Tests are superior to 5%). Since the probability is not significant we have weak evidence against the null hypothesis then we fail to reject the null hypothesis, hence our residuals are normally distributed.

Summary of findings
The aim of this research was to determine the association between the international trades on economic growth of Rwanda in the period of 1960-2017. The cointegration test, VAR model and Granger causality tests are used here to look into the association between these three variables.  The unit root properties of the data were examined using the Augmented Dickey Fuller test (ADF) and Phillips Perron test (PP) after that the cointegration and causality tests were conducted. The cointegration test results show the absence of cointegration relation, which suggests using the VAR model. The estimation of the VAR model shows that exports influence economic growth while imports do influence Rwandan economic growth. Finally, and from the causality test, we find that exports cause economic growth and imports do cause economic growth. These results indicate that exports and imports are presented and seen as a source of economic growth in Rwanda. Therefore, our findings have led to the discovery that exports and imports cause economic growth; these results are expressed using the several advantages that characterize the social economic situation of Rwanda and its economic policy strategy.
We found that the coefficient of exports (Exp) and imports (Impo) have a positive impact on growth. It is statistically significant at 5 percent significance level. A one percent increase in exports will lead to an increase in real GDP by 4.67 percent approximately, and a one percent increase in imports will lead to an increase in real GDP by 11.42 percent approximately. This positive relationship between exports and real GDP and imports and GDP in Rwanda were consistent with a previous empirical study. This interesting result obtained from the empirical study confirms the positive sectors of Exports and imports dominance, which do generate direct growth impacts on the wider economy growth in Rwanda. Some conditions that are often associated with official exports and imports to developing countries, Rwanda inclusive, might be directly favorable to initiating higher levels of industrial performance as well as economic growth. For instance, substantial exports go to manufacturing sectors of the economy, particularly services sector for which reason exports and imports will make any significant impact on industrial performance and on economic growth (Adenutsi, 2008). The result obtained is consistent with several studies in developing countries.

Conclusion
The study examined the macroeconomic determinants of economic growth in Rwanda using cointegration approach. The empirical analysis was based on time Series econometrics. It is found in current study that all variables, growth in real GDP, exports and Imports, turned out to be non-stationary at their levels but became stationary at their first difference. The results of Johansen's cointegration test indicates that there exist a long run and short run relationship between growth in real GDP, exports and imports. All explanatory variables have a positive effect on growth in real GDP. As a result, increase in these variables lead to improve the real GDP. Hence, variables are significant determinants of growth in real GDP in Rwanda.
All coefficients are positive as expected based on theory. It is statistically significant at 5 percent significance level. Specifically, a one percent increase the explanatories had caused real economic growth in other words, gross domestic product (GDP) has been increased respectively, and approximately for each determinant enumerate above. It was inferred that these variables have the positive impacts on real GDP in Rwanda, in general. This means that Rwanda economic growth depend clearly of the above mentioned determinants. These results are in support of theory and concurs with the results obtained during analysis. Through the obtained results, the study has found that there is a positive relationship between exogenous variables mentioned and economic growth in Rwanda, all other things being equal. Results of the study indicate that exports and imports have positively and significantly contributed to economic growth. Cusum test of squares Normality test