The Impact of Remittances on the Trade Balance: Empirical Evidence from North Macedonia

Rilind Ademi, Zana Beqiri Luma, Bujamin Bela

University of “Mother Teresa”, Skopje, North Macedonia.

DOI: 10.4236/tel.[phone]                

Abstract

The impact of remittances on various economic indicators has been analyzed for many developing countries, however, there is little evidence about their impact on the trade balance. For North Macedonia, remittances play an important role because they completely compensate the trade deficit, but on the other hand these remittances are expected to affect the trade deficit, because a large part of these remittances is oriented for further consumption. This paper aims to examine the impact of remittances on North Macedonia’s trade balance using quarterly data for the period Q12003 - Q32019. Like many developing countries, North Macedonia’s economy relies heavily on remittance flows making it an interesting laboratory for the research question at hand. We differentiate between formal remittances, as measured by workers’ compensations and informal remittances, as measured by private current transfers. We find that private current transfers have a significant impact and improve the trade balance, whereas workers’ compensation significantly worsens the trade balance. In addition, our results show that inflation worsens the trade balance and past trade rates themselves affect current trade rates. These results point to the need for a more accurate measure of remittance flows in official statistics to be able to capture the full extent of their impact in various economic indicators of a country.

Keywords

Trade Balance, Remittances, North Macedonia, Current Transfers, Informal Flows

1. Introduction

The free movement of goods and people and its impact on sending and receiving countries has been constantly discussed in the context of European Union developments and beyond.

The economy of Northern Macedonia, as a small open economy where about 80% of trade services are realized with EU countries, over the years faced a negative trade balance. According to the State Statistical Office for the period 2014-2019, the trade balance is almost 1.5 billion euros (or 17% of GDP) over the last five years, with exports of goods and services between 5 billion euros and imports of 6.5 - 7 billion euros.

Against a negative trade balance of 1.5 billion euros (17% of GDP), the current account balance over the last five years marks a deficit of close to 200 million euros (or 2% - 3% of GDP). Inflows from secondary income that on average over the last five years were close to 1.7 billion euros (or 20% of GDP) neutralize the negative effect of trade balance in the current account balance. Secondary income is mainly formed by current transfers that represent remittances from outside without providing any compensation. Secondary income account shows current transfers between residents and nonresidents, which transfers are made in cash or other forms (IMF, 2009).

From this sequence of event, we can see that private transfers, respectively remittances, apart from reducing poverty, improving living standards, financing education and health services they also offset all the negative impact of the trade balance in the current account in particular and the balance of payments in general.

According to the World Bank’s Migration and Development Reports for 2019 worldwide, about 200 million migrant workers have sent about $715 billion to their homes (World Bank, 2019). Considering the high levels of informal transfers of remittances this figure appears to be much higher. Estimates of the importance of informal flows vary widely, ranging from 35 to 75 percent of the flows registered in developing countries (Freund & Spatafora, 2005). At the same time, various market observers suspect that informal flows could range from about 50 percent to 250 percent of recorded flows (Celent, 2002). For these reasons and not only, the role of remittances from migrants in developing countries continues to be an important issue for researchers and policymakers. Remittances represent an essential influx of financial resources, mainly from developed countries to developing countries. With the continued and strong support of these flows, private remittances are important to their home countries, especially in terms of their potential for development (Freund & Spatafora, 2005).

Remittances help the population in receiveing countries to cope with poverty alleviation, improving inefficient local labor markets and improving the social security system (Ari & Cergibozan, 2017). It is emphasized that remittances support families to survive by providing them with the means to buy food and other resources, to invest in their children’s education, and to improve the housing situation. In a macroeconomic context, it is argued that remittances drive consumption and economic growth, help finance the trade deficit and contribute to exchange rate stability (Dietz, 2010). Furthermore, remittances are supposed to strengthen the banking system and increase competition in local banks by simplifying bank transfers (Kireyev, 2006).

It should be noted that remittances are also associated with a number of potentially influential effects on government policy, monetary indicators, investment and labor supply. Because remittances create a safety net for host families, they can ease pressure on their governments to engage in economic and social welfare reform. Also, a high inflow of remittances is likely to fuel inflationary pressure and widen the trade deficit. Furthermore, remittances can increase household consumption but cannot be used to invest in productive assets (Taylor, 1992).

Although the impact of remittances has been extensively analyzed, their impact on various economic and social indicators remains unclear. This is mainly due to the fact that accurately collecting data is extremely difficult. A large part of these remittances are not channeled through the formal payment system and consequently do not appear in official statistics. To better understand the inaccuracy of official flows recorded in the current account of the balance of payments we can use the results of a study which was conducted in Macedonia in 2007, where the survey included 1046 households that receive remittances from abroad (Roberts, Markiewitz, Nikolov, & Stojkov, 2008). According to this study, only 38.6% of remittances were sent home by official money transfers (for example through Western Union) or bank transfers.

The purpose of this paper is to find out the extent of the impact of remittances on the trade balance. As mentioned above, they play an important role because they completely compensate the trade deficit, but on the other hand these remittances are expected to deepen the trade deficit, because a large part of these remittances is oriented for further consumption. To evaluate the impact of remittances on the trade balance we define remittances in the narrow sense (when taking into account only formal remittances of workers from abroad) and in the broad sense (taking into account private current transfers, which includes informal remittances).

The paper is organized as follows: Section Two briefly reviews the existing literature on the impact of remittances on the trade balance; Section Three elabo­rates the underlying methodology; Section Four presents the results and offers a discussion; and Section Five concludes.

2. Literature Review

A number of studies have argued that the inflow of remittances in various forms—from bank transfers to in-kind gifts—plays an important role in reducing poverty and contributing to economic growth (Chami, Hakura, & Montiel, 2009; Buch & Kuckulenz, 2004; World Bank, 2006).

The micro- and macro-economic impact of remittances has been widely documented. At the macro-level, remittances stabilize the balance of payments, hence contributing to closing the large and persistent trade gaps in many countries and preserve macro-economic stability (World Bank, 2006). At the micro-level, the development effects of remittances, with a certain degree of variety, have been documented for poverty alleviation, improving education and health outcomes, improving income distribution, steering entrepreneurial spirit, etc (Adams & Page, 2005).

Studies about the impact of remittances on the trade balance are limited. In fact, most of these studies focus on developing countries where remittance flows are an important phenomenon. Based on these studies, the resulting economic impact of remittances is sometimes found to be positive and sometimes negative. Furthermore, some studies take a bilateral perspective while others focus on the aggregate trade balance (Bahmani-Oskooee & Ratha, 2004).

A study measuring the impact of remittance inflows on the trade balance has been done for the developing countries of the Asia-Pacific region for the period 1980-2015. Results through OLS, 2-SLS and PGMM show that remittance inflows worsen the trade balance. GDP growth also affects the deterioration of the trade balance, while the devaluation of the currency has a positive impact on the trade balance (Tung, 2018).

Another similar study assesses the impact of remittances on the trade balance of Malaysia for the period 1990-2015. Regression results show that an increase in remittance flows improves the trade balance. In addition to remittances, currency devaluation affects the improvement of the trade balance, while an inflow of capital from abroad worsens the trade balance (Nguyen, 2017).

A study for North Macedonia was conducted to analyze an empirical link between the current account deficit and a range of macroeconomic variables for the period 1998Q1 - 2013Q4. There is a strong link between developments in the current account balance on the one hand and fiscal balance, financial development, market conditions and trade openness on the other. Also the current developments of the current account balance are influenced by the past dynamics of the current account. Delayed current account balance, monetary aggregate M2 and market conditions affect the improvement of the current account balance. On the other hand, trade opening, budget deficits affect the deterioration of the trade balance. While GDP growth and the effective real exchange rate (REER) are not statistically significant (Sadiku, Fetahi-Vehapi, Sadiku, & Berisha, 2015).

Another study was done for Poland, where through the VAR model the results show that national income and money supply has played a crucial role as a determining factor of the trade balance for the period 1997-2008. A weak relationship is found between the trade balance and the effective real exchange rate (Misztal, 2010).

Also, studies find that private transfers are factors of economic development in Macedonia, where they have a positive impact on GDP according to a study for the period 2005-2015 (Ferati, Sallai, & Deari, 2018).

For the Turkish economy, a study was conducted using the Johansen procedure and the VEC model for a long-term and short-term relationship between the trade balance and other variables. Their results show that the real effective exchange rate, real national and foreign income are the main determinants that explain the variation of Turkey’s trade balance. Thus, currency devaluation improves the trade balance in the long run, rising national income improves the trade balance and rising foreign income worsens the trade balance (Ari & Cergibozan, 2017).

A comprehensive study attempting to measure the real rate of remittances and its structure for Macedonia was made in 2008. Results show that 63.8% of remittances go for consumption, 15.8% for capital investments and 20% for savings (Roberts, Markiewitz, Nikolov, & Stojkov, 2008). This structure of remittance expenditures also creates the logic that a large part of consumption goes to foreign goods, implying that remittances on the one hand are offsetting the negative contribution of the trade deficit on the current account deficit but on the other hand they are also contributing in increasing this trade deficit. There are also concerns that the large increase in remittances is causing an increase in the trade deficit through the phenomenon of “Dutch disease”. The increase in remittances may affect an appreciation of the exchange rate, which causes the price of exports to increase and the price of imports to decrease, which may lead to an increase in the trade deficit (Hien, 2017). The findings in the fixed exchange rate regimes also stand in the same line where positive and significant coefficient on this interacted variable suggests that a fixed exchange rate regime allows for an even more pronounced spending effect of remittances and thus higher real appreciation (Lartey & Mandelman, 2008).

Also a more comprehensive study for the new EU member states is done to assess the determinants of the current account balance. According to the results, if remittances are equal to or exceed 5% of GDP then they improve the current account balance by 0.02% (Rahman, 2008).

3. Data and Methodology

Previous empirical and theoretical work suggests that the trade balance is influenced by a number of important factors (see Table 1 below). In the following, we advance a number of hypotheses regarding the relationship between the trade balance and the potential factors affecting the trade balance, which we will continue to evaluate in the following empirical work.

The basic trade balance equation can be specified as:

$\begin{array}{l}{\text{TDGDP}}_t=\alpha +{\alpha }_{\text{1}}{\text{CPI}}_t+{\alpha }_{\text{2}}{\text{RemittGDP}}_t+{\alpha }_{\text{3}}{\text{CurrTransGDP}}_t+{\alpha }_{\text{4}}{\text{BBGDP}}_t\\ +{\alpha }_{\text{5}}{\text{FDIGDP}}_t+{\alpha }_{\text{6}}{\text{TD4GDP}}_{t-4}+{\alpha }_{\text{7}}{\text{GDPchan}}_t+{\alpha }_{\text{8}}{\text{REER}}_{\text{t}}+{\epsilon }_t\end{array}$ (1)

where t represents the time period t = Q12003 - Q32019. TDGDP as a dependent variable represents the difference between exports and imports of goods and

Table 1. Variables, definitions and sources.

services in relation to nominal GDP. While no trade surplus periods were recorded for Macedonia, this dependent variable was termed trade deficit relative to GDP, so a positive coefficient of any independent variable indicates an increase in the trade deficit and a negative sign indicates a decrease in the trade deficit. The independent variable CPI, represents the consumer price index, changes in percentage compared to the same period last year, an indicator which measures the impact of inflation on the dependent variable. The flow of money from emigrants is measured by two variables: RemittGDP and CurrTransGDP. RemittGDP represents money transfers through formal channels through persons—residents and immigrants who have non-resident status due to residence abroad for one or more years. This variable represents a narrow perspective of the real impact of transfers which is put in relation to GDP. The CurrTransGDP variable, on the other hand, includes private current transfers, a broader component of transfers, which is believed to include informal transfers because it includes: employee remittances, rents, pensions, invalidity coming from abroad and activities of the purchase of foreign net effect in the foreign exchange market. This represents the purchase of foreign cash from exchange offices by banks. It is believed that a significant part of the net effective is drawn from private immigration transfers (Нацевска, Крстевска, Петровска, & Уневска, 2006). BBGDP represents the budget balance in relation to GDP. When revenues exceed budget expenditures it represents a surplus and vice versa budget deficit. FDIGDP represents inflows in the form of foreign direct investment relative to GDP. The variable TD4GDP represents the time delay of the dependent variable (four-quarter lags), because it is believed that previous commercial rates affect actual trading rates. GDPchan presents the percentage changes of the most important economic indicator, gross domestic product, measured as a percentage change for the quarter compared to the same period last year, according to the expenditure method. The REER variable represents the real effective exchange rate which is presented in the form of indices where the base year is 2015 = 100. An increase in the index means an appreciation of the national currency and a decrease in competitiveness, and vice versa when we have a decrease in this index, which represents currency devaluation and an increase in international competitiveness. The purpose of effective exchange rates is to provide substantial and comparable measures of price and cost competition between countries. ε represents the common error term.

The created linear regression (using least square method) is made in the form of finding the best model. This method lets us choose the best model from amongst all the models which can handle a number of variables varying from “Min variables” to “Max variables”. Furthermore, we choose several “criteria” to determine the best model: Adjusted R², Mean Square of Errors (MSE), Mallows Cp, Akaike’s AIC, Schwarz’s SBC, Amemiya’s PC (see TableA1 in Appendix).

Using this approach then the best model is with six variables excluding from the model GDPchan and REER. Also out of curiosity, these two variables if included in the model are also statistically insignificant. The linearity assumption we can see by scatterplots between residuals and predicted values and also by observed dependent variable and predicted values. As we can see by scatter plots we conclude that we have positive linear relations between dependent and independents variables (see FigureA1 in Appendix). The normality of the residuals we have checked by running a Shapiro-Wilk test on the residuals, so the Ho was accepted (the residuals follow a normal distribution, see TableA2 in appendix). Testing for homoscedasticity based on the Breusch-Pagan and Abridged-White test, we can see that the significance F > 0.1 which means that Ho will be accepted where there is homoscedasticity of term errors and Ha is rejected in the presence of heteroskedasticity (see TableA3 in Appendix). This means that the error terms are correctly specified. Variance inflation factor (VIF) measure is used to see if we have problems with multicollinearity, and according to the results we see that the VIF indicator is below 2, which shows that we have no problems with the multicollinearity of the variables (see TableA4 in Appendix). Testing for auto-correlation of residuals we use Durbin-Watson test which a rule of thumb is that test statistics values in range of 1.5 - 2.5 are relatively normal, so in our dataset Durbin-Watson test is 2.2 (see TableA5 in Appendix) which means that we don’t have problems with auto-correlation.

4. Estimation Results

Table2 contains descriptive statistics (also TableA6 in Appendix). Across all years, the trade deficit as a percentage of GDP is 18.29%, with a variation 7.64% - 30.04% of GDP over time. In addition to these indicators in Table2 you can find other statistical data for all variables included in the model. The period analyzed on the basis of the quarter Q12003 - Q32019 creates a dataset with 67 data and it is important to highlight that there is no missing data in this time period included in the model.

To measure the relationship and the direction of the relationship between the two variables we use Pearson correlation. As can be seen from Table3 (also TableA7 in Appendix) where the matrix of correlations between the dependent variable and the independent variables is presented, as well as the relation between the independent variables, we can also read the relation but also the

Table 2. Descriptive statistics.

Source: Author’s calculations.

Table 3. Correlation matrix.

Values in bold are different from 0 with a significance level alpha = 0.05.

direction of this relation. Values in bold are different from 0 with a significance level alpha = 0.05.

As can be seen strong positive relationship between the two variables we have between the dependent variable (TDGDP) and TD4GDP, which is expected and logical. We have a positive average correlation between the dependent variable and CPI as well as RemittGDP. We also have a negative average correlation with the dependent variable and CurrTransfGDP. We have a low positive correlation between the dependent variable and the FDIGDP variable. We have no correlation between the dependent variable and GDPchan, REER and BBGDP. Also for the reader can be seen the relations and directions of these relations even between the explanatory variables among themselves. Also through graphs (see FigureA1 in Appendix) you can see the relationship between the dependent variable and the explanatory variables.

From Table4 below (also TableA8 in the Appendix) we can see the parameters of the model, respectively the significance of the independent variables and their impact on the dependent variable (TDGDP). As can be seen, two models are presented, the one with six variables which is also the best model as a result of the consultation of several “criteria” to determine the best model: Adjusted R², Mean Square of Errors (MSE), Mallows Cp, Akaike’s AIC, Schwarz’s SBC, Amemiya’s PC (see TableA9 in Appendix).

For the purpose of explanation, we add to the model the other two variables (GDPchan; REER) that are excluded from the first model where we can see that the two variables excluded from the first model are not statistically significant and the significance of the other variables remains the same as in the base model with six variables.

R-squared is 0.63 (see TableA1 and TableA5 in the Appendix) which means that the dependent variable (TDGDP) is explained up to 63% of the independent variables presented in the model. Following the results from Table4, we can see

Table 4. Model parameters (dependent variable: TDGDP).

Notes: the dependent variable is the trade deficits as a percentage of GDP. ***, ** and * denote significance at the 1 percent, 5 percent and 10 percent level.

that the independent variables: CPI, RemittGDP, CurrtransfGDP, TD4GDP are variables that statistically affect and explain the dependent variable (TDGDP) and all variables are significant at 1 percent level.

With the increase of inflation by 1 percent, the trade deficit increases by 0.76% in relation to GDP, ceteris paribus. The presence of inflation causes the prices of domestic products to rise faster than the prices of products of trading partners, thus making domestic products less competitive and consequently reducing the demand for those products and increasing the demand for imported goods. Also, the inflation rate increases the cost of borrowing due to a higher interest rate. This in turn discourages potential investors from undertaking exports activities (Aye Mengistu & Lee, 2014). The results of other authors are also in line, underlining the fact that inflation is harmful to the economic growth and this has effects on the trade balance (Ademe, 2016).

To measure the impact of cash flows from emigrants as we have said two variables have been introduced. Initially, the dilemma arises whether these two variables explain the same thing and we can have the presence of multicollinearity, but as can be seen from Table 3, the correlation coefficient is 0.169, which means that these two indicators are explaining distinct phenomena from each other. Also from Table 2 you can see the average of the RemittGDP variable which is 2.28% of GDP and the average of the CurrTransfGDP variable which is 16.2% of GDP. As we presented earlier the first variable explains in a narrow concept remittances taking only formal flows from workers who work one or more years abroad. While the second variable represents a broader concept which includes the category of net effective operations that are believed to be from informal channels of remittances. Both of these variables have emerged statistically significant at 1 percent level and in the opposite direction, respectively with the increase of the RemittGDP variable by 1% in relation to GDP increases the trade deficit of 3.79% in relation to GDP and with the increase of the CurrTransfGDP variable for 1% of GDP reduces the trade deficit of 0.35% of GDP, in ceteris paribus. The most accessible and realistic indicator is definitely the second variable and therefore we think it is more relevant in the real explanation of these cash flows coming from abroad. Also studies of other authors that have been done for different countries and for different periods of time show that the results are not in line, respectively in some authors it is seen that remittances worsen the trade balance (Tung, 2018) and some results show that remittances improve the trade balance (Nguyen, 2017; Rahman, 2008).

It is believed that opening trade routes, finding trading partners affect trade rates. Therefore positive trading rates that have been realized in finding new trading partners or routes in the past definitely affect today’s trading positions. We have therefore introduced the lagged dependent variable into a regression model, implying the fact that past trading rates definitely affect current trading rates. The result finds this variable as statistically significant at 1 percent level and an increase in the trade deficit relative to GDP before the four quarters affects the increase of 0.415% of the current trade deficit relative to GDP.

The budget balance does not affect the trade balance. The theory of “twin deficits” is not proven in our model, although as can be seen it has a statistical relevance (p-value) very close to the level of acceptability. The expressed causality is very logical, because with the improvement of the budget balance of 1% in relation to GDP, the trade deficit decreases by 0.25% in relation to GDP. In fact, there is room for a possible new work on the budget balance and the effects on the trade deficit, but we cannot dwell only on the traditional view of linking the impact of the budget balance on the trade balance. There is evidence of a budget-to-trade deficit causality but there is also strong evidence of trade-to-budget deficit causality. Also the evidence that there are periods when there is no relationship between these variables. Foreign direct investment is not proven to be statistically significant and has a positive coefficient.

A standardized beta coefficient (see TableA9 and FigureA2 in Appendix) compares the strength of the effect of each individual independent variable to the dependent variable. The higher the absolute value of the beta coefficient, the stronger the effect.

5. Conclusion

The purpose of this paper is to find out the extent of the impact of remittances on the trade balance. Remittances play an important role because they completely compensate the trade deficit, but on the other hand these remittances are expected to deepen the trade deficit, because a large part of these remittances is oriented for further consumption. To evaluate the impact of remittances on the trade balance we define remittances in the narrow sense (when taking into account only formal remittances of workers from abroad) and in the broad sense (taking into account private current transfers, which includes informal remittances). In addition to remittances, the model examines other variables that may have an impact on the trade balance.

The created linear regression (using least square method) is made in the form of finding the best model. This method lets us choose the best model from amongst all the models which can handle a number of variables varying from “Min variables” to “Max variables”. Furthermore, we choose several “criteria” to determine the best model: Adjusted R², Mean Square of Errors (MSE), Mallows Cp, Akaike’s AIC, Schwarz’s SBC, Amemiya’s PC where the best model is with six variables excluding from the model GDPchan and REER.

R-squared is 0.63 which means that the dependent variable (TDGDP) is explained up to 63% by the independent variables included in the model. Following the results from the table we can see that the independent variables: CPI, RemittGDP, CurrtransfGDP, TD4GDP are variables that statistically affect and explain the dependent variable (TDGDP) and all variables are significant at 1 percent level. Emigrant remittances measured through the private current transfers’ indicator are presented as statistically significant and the increase of this indicator improves the trade balance. On the other hand, the other variable, workers’ compensation is statistically significant and its increase worsens the trade balance. Looking at the qualitative aspect, we believe that the variable current transfers better represent the real flows of emigrants, because the variable workers’ compensation only narrowly and formally records these flows. On the other hand, rising inflation worsens the trade balance, and also past trade rates positively and significantly affect current trade rates.

Appendix

Figure A1. Scatter plots.

Figure A2. TDGDP/Standardized coefficients.

Table A1. Summary of the variables selection TDGDP.

The best model for the selected selection criterion is displayed in blue.

Table A2. Test on the normality of the residuals (Shapiro-Wilk) (TDGDP).

H0: The residuals follow a Normal distribution. Ha: The residuals do not follow a Normal distribution. As the computed p-value is greater than the significance level alpha = 0.05, one cannot reject the null hypothesis H0.

Table A3. Analysis of variance (TDGDP).

Table A4. Multicolinearity statistics.

Table A5. Goodness of fit statistics (TDGDP).

Table A6. Descriptive statistics.

Table A7. Correlation matrix.

Table A8. Model parameters.

Table A9. Standardized coefficients (TDGDP).

Conflicts of Interest

The authors declare no conflicts of interest regarding the publication of this paper.

References

[1]	Adams, R., & Page, J. (2005). Do International Migration and Remittances Reduce Poverty in Developing Countries? World Development, 33, 1645-1669. https://doi.org/10.1016/j.worlddev.2005.05.004
[2]	Ademe, A. (2016). Determinants of Ethiopian Trade Balance: Vector Error Correction Model (VECM) Approach. Journal of Research in Business, Economics and Management, 6, 858-867.
[3]	Ari, A., & Cergibozan, R. (2017). Determinants of the Trade Balance in the Turkish Economy. In EBEEC Conference Proceedings, the Economies of Balkan and Eastern Europe Countries in the Changed World (pp. 160-169). KnE Social Sciences.
[4]	Aye Mengistu, A., & Lee, J. (2014). Examining the Effects of Currency Depreciation on Trade Balance in Selected Asian Economies. International Journal of Global Business, 7, 59-76.
[5]	Bahmani-Oskooee, M., & Ratha, A. (2004). The J-Curve: A Literature Review. Applied Economics, 36, 1377-1398. https://doi.org/10.1080/0003684042000201794
[6]	Buch, C., & Kuckulenz, A. (2004). Worker Remittances and Capital Flows to Developing Countries. ZEW Discussion Paper No. 04-31. https://doi.org/10.2139/ssrn.535802
[7]	Celent (2002). Global Money Transfers.
[8]	Chami, R., Hakura, D., & Montiel, P. (2009). Remittances: An Automatic Output Stabilizer? IMF Working Papers 09/91. https://doi.org/10.5089/9781451872385.001
[9]	Dietz, B. (2010). Migration and Remittances in Macedonia: A Review. Osteuropa Institut. https://doi.org/10.2139/ssrn.1713747
[10]	Ferati, R., Sallai, N., & Deari, H. (2018). Impact of Remittances on Economic Growth and Development in Republic of Macedonia. Global Scientific Journals, 6, 432-441.
[11]	Freund, C., & Spatafora, N. (2005). Remittances: Transaction Costs, Determinants, and Informal Flows. World Bank Policy Research Working Paper 3704. https://doi.org/10.1596/1813-9450-3704
[12]	Hien, N. (2017). The Impact of Remittance on Trade Balance: The Case of Malaysia. Journal of Economics and Public Finance, 3, 531-540. https://doi.org/10.22158/jepf.v3n4p531
[13]	IMF (2009). Balance of Payments and International Investment Position (6th ed.). International Monetary Fund.
[14]	Kireyev, A. (2006). Macroeconomics of Remittances: The Case of Tajikistan. IMF Working Paper 06/02. https://doi.org/10.5089/9781451862621.001
[15]	Lartey, E., & Mandelman, F. (2008). Remittances, Exchange Rate Regimes and the Dutch Disease: A Panel Data Analysis. Federal Reserve Bank of Atlanta, Working Paper 2008-12. https://doi.org/10.2139/ssrn.1109206
[16]	Misztal, P. (2010). Main Determinants of the Trade Balance in Poland. Ekonomista, No. 1, 59-75.
[17]	Nguyen, P. (2017). The Impact of Remittance on Trade Balance: The Case of Malaysia. Journal of Economics and Public Finance, 3, 531-540. https://doi.org/10.22158/jepf.v3n4p531
[18]	Rahman, J. (2008). Current Account Developments in New Member States of the European Union: Equilibrium, Excess, and EU-Phoria. IMF Working Paper, WP/08/92. https://doi.org/10.5089/9781451869538.001
[19]	Roberts, B., Markiewitz, M., Nikolov, M., & Stojkov, A. (2008). A Study on Determinants and Trends in Remittances Flows in Macedonia. Center for Economic Analysis.
[20]	Sadiku, L., Fetahi-Vehapi, M., Sadiku, M., & Berisha, N. (2015). The Persistence and Determinants of Current Account Deficit of FYROM: An Empirical Analysis. 7th International Conference, the Economies of Balkan and Eastern Europe Countries in the Changed World, EBEEC (pp. 90-102). https://doi.org/10.1016/S2212-5671(15)01696-2
[21]	Taylor, J. (1992). Remittances and Inequality Reconsidered: Direct, Indirect, and Intertemporal Effects. World Bank Staff Working Paper 481. https://doi.org/10.1016/0161-8938(92)90008-Z
[22]	Tung, L. (2018). Impact of Remittance Inflows on Trade Balance in Developing Countries. Economics and Sociology, 11, 80-94. https://doi.org/10.14254/2071-789X.2018/11-4/5
[23]	World Bank (2006). Global Economic Prospects. Economic Impacts of Remittances and Migration. The International Bank for Reconstruction and Development/The World Bank.
[24]	World Bank (2019). Migration and Remittances. World Bank Group.
[25]	Нацевска, Е., Крстевска, А., Петровска, М., & Уневска, Д. (2006). Анализа на приватните трансфери во билансот на плаќања на Република Македониjа [Analysis of Private Transfers in the Balance of Payments of the Republic of Macedonia]. Народна банка на Република Македониjа [Cenral Bank of North Macedonia].