“The relationship between external debt and economic growth: empirical evidence from Ukraine and other emerging economies”

The article examines the relationship between external debt and economic growth in emerging economies for the period 2006‒2016. The authors used different economet-ric tools, e.g., ADL model and correlation analysis. The regression results showed that the original values had no significant impact on the estimation of the parameters. Thus, there was made an assumption that emerging economies have a non-linear impact on macroeconomic parameters, including external debt that has a non-linear type of influence on economic growth. The authors established that high level of external debt, in conjunction with macroeconomic instability, impedes economic growth in such countries. The regression model also showed that there is a critical level of debt burden for emerging economies, where the marginal impact of external debt on economic growth becomes negative. The results of the study highlighted the significance of the problem of effective public debt management strategy implementation in Ukraine. This issue is predetermined by the appropriate organizational support. The study recommends improving a public external debt management model. In this paper, the authors proposed a new structure with the participation of new element – independent agencies. The unified external debt management system should integrate all state institutions and executive power structures in this area.


INTRODUCTION
The issue of external borrowing as a policy to promote economic growth creates serious debate among policy makers, researchers and economists.Currently, foreign (external) debt is a global problem for the present and future development of most countries in the world.In particular, this applies to developing economies and emerging economies, including Ukraine.
The main concern is whether or not external borrowing leads to economic growth in emerging economies.Ensuring of the economic development of the country is one of the most important national priorities, being a guarantee of the country's independence, a condition for stability and effective life of the society, achievement of success.According to the main economic theories, external debt should have an influence on the economy through investment and productivity of the labor force.Since the main reason for attracting external loans is a lack of own funds to finance investment projects, an increase in gross external debt should stimulate growth in investment and capital formation, which leads to an increase in the potential volume of GDP.However, often external loans can be used not for investment projects, but for short-term Investment Management and Financial Innovations, Volume 15, Issue 1, 2018 goals to cover emerging deficits of the economy.Thus, in economic theory, there are different approaches in evaluating the relationship between external debt and economic growth.On the one hand, the neoclassical growth models advocated that there is a positive influence of the external debt on economic growth.Moreover, they emphasized that borrowings are one of the sources for financing capital formation, and they could promote economic growth.On the other hand, there are followers of the opposite theory who aware that due to debt overhang, there is always a decrease in economic growth of the countries.
The main aim of this empirical investigation is to determine the impact of the external debt on economic growth in countries of the estimated group.Economic growth is measured by the dynamics of real GDP.The object of the study is the relationship between external debt and economic growth in countries with emerging economies.

LITERATURE REVIEW
The relationship between external debt and economic growth is analyzed in different studies, which are aimed at finding out the effect of influence of debt burden on economic growth in countries with different types of economies.
A brief review of the literature examining such an impact allows to highlight several types of relationship between external debt and economic growth.Thus, it is established that the influence of external debt on economic growth can have a positive, negative and non-linear relationship.There are three groups of theoretical models describing this interaction.
The first group offered supporting evidence for the positive relationship between economic growth and foreign debt.These views are based on the provisions of the Keynesian and neoclassical theory of growth, where external debt contributes to more intensive economic growth, provided the productive use of borrowed funds.This approach was observed by Warner (1992), Easterly (2003) and Cline (1985).Warner (1992) found a positive relationship between external debt and investment.Easterly (2003) also found certain empirical evidence of such an impact.According to Cline (1985), external debt contributes positively to growth.He argued that if marginal productivity of each available external debt is greater than or equal with the principal and the interest payment, external debt will have a positive impact on the economy of the borrowing country.
It is worth mentioning that the majority of existing empirical literature report that external debt adversely affects economic growth.The second group of models is devoted to the analysis of the reasons for the negative impact of a high level of external debt on economic growth.This group of models is based on the theory of "debt overhang" developed by Krugman (1988) in the late 1980s.A debt overhang occurs when the present value of the expected income of a country is less than the accumulated debt.Siddique and Selvanathan (2015) found a negative effect of the external debt level on economic growth of 40 poor countries over the period 1970-2007.Empirical evidence revealed the poor internal and external economic policies as the main causes for the debt overhang problem.In another single country study, Rifaqat and Usman (2012) examined the long-run and short-run impact of external debt on economic growth in Pakistan from 1970 to 2010, considering GNP as a function of annual education expenditures (proxy of human capital), capital, labor force and the external debt.They found that external debt exerts a negative impact on economic growth; clearly indicated that higher external debt discourages economic growth.
The third group of models, the most popular nowadays, considers the non-linear impact of external debt on economic growth, and is represented by eclectic concepts that combine the provisions of the previous two groups of models.This group of models is based on the principles of the Laffer curve.According to Sachs (1986), the central point in this concept is the assumption of the existence of a critical level of external debt after the achievement of which further attraction of foreign loans is impractical.The non-linear form of the relationship is explained by the reduction in the absolute size and efficiency of investments, caused by the tendency to reduce the marginal efficiency of capital and the impact of unfavorable macroeconomic situation.Emerenini and Nnanna (2015) used a Solow-type neoclassical growth model.The results showed the presence of non-linear effect of debt on economic growth in Nigeria.Soldatova (2006) in her study investigated that external debt contributes positively to growth up to a certain point, after which its contribution becomes negative.
Despite the theoretical study of the problem of public debt and its relationship with the main macroeconomic indicators of the economic growth, there is still a lack of fundamental research on external public debt, in particular, assessment of the ratio of public debt to GDP.

RESEARCH METHODOLOGY
The following case study is based on the analysis of data of Ukraine and 10 other emerging economies, which had approximately the same economic mechanism, and, at the same time, began to transform their own political and economic systems, namely Armenia, Azerbaijan, Belarus, Bulgaria, Georgia, Kazakhstan, Kyrgyz Republic, Moldova, Romania and Tajikistan.
The reason for including those countries to the estimation group was the availability and the adequacy of comparable reliable data, which were necessary for analysis, and the stability of the socio-political situation in countries during the investigated period.The study period selected was from 2006 to 2016.Data were collected from World Bank, National Bank of Ukraine, OECD.The study used both descriptive and econometric tools.Parameters are estimated using the nonlinear regression and correlation analysis.The validity of the regression analysis was assessed and Durbin-Watson test was included to check for autocorrelation and included collinearity diagnostics.Pearson's Chi-square test 2 χ was applied to sets of categorical data to evaluate how likely it is that any observed difference between the sets arose by chance.

TRENDS AND PATTERNS OF THE EXTERNAL DEBT IN EMERGING ECONOMIES
The experience of Central European countries shows that macroeconomic stability is a necessary, but insufficient prerequisite for improving the well-being of the population and reducing poverty.
If the economy is sensitive to external and internal shock phenomena, in particular, due to excessive external borrowings, the poor and unprotected population is adversely affected.
The market mechanism is not able to provide socially equitable distribution of income, which is generated by the gross product.Unfair income distribution is one of the "defects" of the market system, along with unemployment and inflation.
In order to assess how the standard of living in emerging economies has changed over the past 10 years, the authors investigated such an indicator as the Gini index (Figure 1).The Gini coefficient measures the inequality among values of a frequency distribution (for exlample, levels of income).A Gini coefficient of zero expresses perfect equality, where all values are the same (where everyone has the same income).A Gini coefficient of 100% expresses maximal inequality among values.
The results from Figure 1 show that over the past 10 years, in general, the index had a tendency to decrease, except for Bulgaria and Tajikistan.Thus, in emerging economies, there has been an improvement of the fair distribution of income generated by the gross product in recent years.
It should be noted that in international practice, a special place is devoted to the analysis of GDP growth, which is considered as the major indicator of economic growth of the country and plays an extremely important role in determining the development of society.Annual growth rates of GDP and total debt service on external debt in emerging economies are presented in Figure 2.
Figure 2 plots that generally annual growth rates of GDP were almost always higher in the investigated countries than in Ukraine.Moreover, from 2013 to 2016, generally, this indicator has sharply decreased in the group of countries, Ukraine is not exception -the decline by 32% in 2016 compared with the previous year, whereas, in general, there was a decrease by 18%.
This situation indicates a decline of economic growth in recent years, among the factors -general deterioration of the macroeconomic situation, political problems, military conflict, as well as instability in the international market.
Regarding the dynamics of the ratio of debt servicing to the export of goods and services, the level of payments generally increased, especially in 2016 with a particularly significant level -by 37% in Ukraine, while in the group -only by 5%.
The matrix of the relationship between external public debt per capita and GDP per capita, depict- Also it should be mentioned that there are always many risks of using borrowed resources, especially threat for a financial security is the biggest issue for each country.Figure 4 shows that in Ukraine there is a problem of the service of external debt.During the period 2006-2016, the essential exceed of sum of payments on the public external debt over the sum of payments on public domestic debt is observed.
Besides, Figure 4 depicts that in 2015 in Ukraine, there was a rapid and sharp growth in payments on debt in 2015 and a decline in 2016.There is no evidence to suppose that there is a tendency of this parameter.The ratio of payments on public debt to GDP, % This analysis gives the possibility to confirm that Ukraine is the state with high dependence from external financing.In order to optimize the national debt level, it is essential to make its restructuring and concentration on internal credit resources and long-term loans with fixed interest rates.
To analyze the depth of the debt problem, there should be conducted an empirical research that will determine the impact of the external debt on economic growth in countries of the estimated group.

MODEL SPECIFICATION
In this study, there were estimated both linear and non-linear regressions using a set of variables common in the growth literature.The initial regression analysis, which requires the setting of initial values for each parameter, showed that, in practice, the original values had no significant impact on the estimation of the parameters.Furthermore, the validation of the regression estimates proved that there is no gain of such investigation and no evidence of the adequacy of regression model.Thus, we made a conclusion that emerging economies belong to the third group, where the impact of macroeconomic parameters, including external debt, has a non-linear influence on economic growth.
Therefore, the authors used the non-linear regression to estimate the relationship between external debt of emerging economies and economic growth.
There is no well-defined theoretical framework that sheds light on the relationship between Solow growth model and external debt.But some empirical works on external debt used Solow growth model as a basis to investigate its impact on economic growth.And the model is developed based on Cobb-Douglas production function given by the form: Since the investigation of a linear function has undeniable advantages over other classes of functions, the authors reduced non-linear functions to linear ones.
Thus, the power function becomes logical after logarithm of Cobb-Douglas production function:  RESGdp -log of ratio of total reserves to GDP; ln it EDGdp -log of ratio of external debt to GDP; it u − error term.
In this study, the authors used as dependent parameter a log of GDP growth and as the main independent variables: log of population growth, log of growth rate of investment rate (mainly from Solow's growth model), log of growth rate of GNI per capita; log of official exchange rate, log of ratio of unemployment to total labor force, log of ratio of the half-sum of exports and imports to GDP, log of ratio of total reserves to GDP, and the debt burden measuring variables: the ratio of external debt to GDP, debt service export ratio.
Before inferring conclusion from estimation results, it is important and vital to undertake some statistical tests.This paper tries to take the autocollinearity and multicollinearity tests.
Multicollinearity: Chi-square test was implemented for the detection of the existence and severity of multicollinearity.
Autocollinearity: another test was used for AR serial correlation -the Durbin-Watson test, based on the OLS residuals.

EMPIRICAL RESULT
This section shows the estimation results of the model (for all countries in general and each country of the assessed group) and the corresponding test statistics employed in the estimation.Analysis of variance (ANOVA), reported in Table 2, uses the F-test to determine whether the variability between group means is larger than the variability of the observations within the groups.The null hypothesis for an ANOVA always assumes the population means are equal.We reject the null hypothesis if F (observed value) > F (critical value).
The F-test statistics value is 0.411, which confirms the presence of at least one independent variable significant for the model.Empirical findings indicated that population growth rate has the biggest positive effect on economic growth.This means that 1 percent increase in population growth rate leads to increase of GDP by 9.99%.This relationship was significant at 5 percent level of significance.The results also indicated a significant positive relationship between growth rate of GNI per capita and economic growth.This is in line with the general assertion that the capital is a key factor of production, hence, it is positively associated to economic growth.The estimated results showed that 1 percent increase in GNI per capita leads to increase of GDP by 1.01 percent.The relationship was consistent with economic theory.
The result from the general model (for all emerging economies) shows that all the variables in the model are statistically significant.As we hypothesized based on Solow's growth model, growth rate of ratio of the external debt to GDP indicates that 1% increase in results in the long run increases by 0.18% of economic growth, but this change is statistically insignificant at p-value of 0.977, which is greater than 5%.
However, the regression analysis of the model for each country of the estimated group depicts different influence of the external debt (see Table A in Appendix A).For example, debt service export ratio almost in all countries has a positive relationship with economic growth, only in Azerbaijan, Kyrgyz Republic and Moldova, it has negative influence, i.e., 1% increase in debt service export ratio leads to decline of GDP by 0.27% in Armenia (but p-value was less than 5% -0.287), in Kyrgyz Republic, there was higher level of influence (-0.85%) and the greatest effect was in Moldova (-0.68 %).
The ratio of external debt to GDP was significant at 5 percent and negatively related to GDP in Armenia, Azerbaijan, Belarus, Kazakhstan, Moldova and Ukraine.Indicating that an increase in the external borrowing of those countries affects the economy negatively, i.e., in Ukraine 1%, increase in ratio of external debt to GDP leads to decrease of GDP by 0.20%.This relationship was significant at 5 percent level of significance (p-value is 0.563).Next step of the analysis was based on the abovementioned multiple regression model.The authors tried to establish the critical level of external debt, which negatively affects the economic growth of Ukraine.Based on the results obtained from the present empirical investigation, the following conclusions can be drawn: the influence of external debt on the dynamics of GDP is non-linear, the marginal impact of external debt on GDP dynamics becomes negative with the level of external debt 62%, and the GDP growth rate increases twice with 8% of the level of external debt to GDP.
Note: Significance levels are at 5%.
Obviously, the orientation of most countries towards the generally accepted (normative) level of external debt accumulation is inaccurate.This structure is not effective, moreover, in many emerging economies, the issue of centralizing public debt management has recently become quite acute.For example, in Kazakhstan and Belarus, there are assumptions that consider the possibility of creating debt agencies.

RECOMMENDATIONS
In general, there are three approaches of constructing the institutional models of the public external debt management adaptable to specific needs, as follows in Table 5.The first and the second approaches are united by one common principle feature -the management of external debt is built on the basis of traditional institutions of government administration.
Regarding the third approach, it should be noted that, as practice has shown, independent agencies alone do not guarantee the automatic creation of an effective management system.Meanwhile, in countries where such a structure was organized, it was possible to solve, on the one hand, the problem of the growing volume of external debt, and on the other hand, there appeared the problem of attracting highly qualified personnel to the civil service in such agencies.
The authors have developed own approximate structures of the institutional model in the field of public external debt management, which is presented in Figure 5. This

CONCLUSION
According to the author's research results, external resources have become an important source of the additional investment.However, the improper policy of external debt management and the inefficient use of external resources have a negative impact on the dynamics of economic development.
According to data analysis, the following conclusions can be obtained: external debt has a non-linear relationship on the dynamics of GDP in emerging economies; there is a marginal impact of external debt on GDP dynamics; the efficient use of external resources can be achieved by reforming the existing institutional structure of external debt management.
The regression model can be practically used as one of the tools in creating the state's macroeconomic policy.The use of the model will allow to analyze the impact of foreign economic debt on economic growth.The developed regression model allows to choose one or another model tool, in accordance with the current situation in the world economic system, managing such important factors as investments, population, reserves, openness of the economy, etc.
The application of the regression model partially will allow to solve the problem of "debt overhang".Based on the economic and statistical analysis of the interaction of model parameters, it can be concluded that the function of the dependence of external debt on GDP can be used as a decision-making tool in the economic policy of the country.

Figure 2 .%
Figure 2. Annual growth rates of GDP and total debt service on external debt in emerging economies, 2005−2016

Figure 3 .Figure 4 .
Figure 3. Matrix of correlation of present value of external debt per capita and GDP per capita in emerging economies, 2016 Thus, the conducted empirical testing of the influence of external debt on economic growth confirmed the theoretical assumptions concerning negative effects of the external debt service on the level of the development of the emerging economies.Consequently, it can be assumed that efficient use of external resources can be achieved by reforming the existing institutional structure of external debt management.That means that there is a significant problem of public debt management strategy in Ukraine.This issue is predetermined by the appropriate organizational institutional support.The direct participants in the management of the public debt in Ukraine are the Verkhovna Rada of Ukraine, the Ministry of Finance of Ukraine, one of the main tasks of which is the development of programs for state borrowing and their implementation on behalf of Ukraine, as well as management of public internal and external debt, National Bank of Ukraine (directly deals with the servicing and payment of external debt obligations of the government), the State Treasury Service of Ukraine (implementation of management of internal and external public debt and its servicing together with the National Bank of Ukraine and the Ministry of Finance), Ministry of Economic Development and Trade of Ukraine, Department for Regulation of Foreign Economic Activities (former National Agency of Ukraine for Reconstruction and Development), State Export-Import Bank of Ukraine (carries out servicing of foreign loans granted by international organizations under government guarantee), State Financial Inspection and Accounting Chamber of Ukraine (regularly informing the Verkhovna Rada of Ukraine and its committees on the progress of the implementation of the State Budget of Ukraine and the state of repayment of Ukraine's internal and external debt).

Figure 5 .
Figure 5. Institutional model of the public external debt management

Table 1
depicts the regression analysis of the general model.The coefficient of multiple determinations (R 2 ) with the value of 0.966 indicates that the total variation in the GDP growth was accounted for by 96.6% of the independent variables included in the model.Standard error isn't significant.

Table 1 .
Overview of model

Table 2 .
ANOVA of model

Table 3 .
T-test and estimation of regression coefficients for model

Table 5 .
Institutional models of the external public debt management model is simplified, therefore, more effective, since the functions of the bodies are not duplicated, and are concentrated mainly in such a link as the debt agency.It must function with the relevant Government and Parliament approval and realize the goals through debt service bodies (National Bank of Ukraine, State Treasury Service of Ukraine and State Export-Import Bank of Ukraine).

Table A .
Fixed effects panel regression for each variable in each country, 2006−2016

Table A .
(cont).Fixed effects panel regression for each variable in each country, 2006−2016