Dynamics of external debts among heavily indebted poor countries (HIPCs): a panel data approach.

Anoruo, Emmanuel ; Dimkpah, Young ; Ahmad, Yusuf 等

ABSTRACT

This study uses panel data for 29 Heavily Indebted Poor Countries (HIPCs) from 1984 to 2000 to examine the dynamic relationships between growth of external debts with other determinant variables (exchange rate, interest payment on debt, and non-interest current account balance) and control variables such as governance indicators. The fixed- and random-effect models were used to investigate these relationships. First, the results show that high interest payments have adverse effects on the growth of external debts. Second, real exchange rates have positive influence on growth of external debts. Third, corruption is found to distort economic growth and reduces the efficiency of the public sector. Finally, stability index contributes negatively to the growth of external debts. Therefore, given these dynamic relationships; this study suggests that there are strong correlation between growth of external debts and exchange rate policy, interest payments and some governance indicators. This evidence may partially explain the explosive external debt position of the HIPCs.

INTRODUCTION

The world witnessed unprecedented explosion in public debt throughout the 1980s and the 1990s in developing countries, especially those of the Heavily Indebted Poor Countries (HIPCs). Most of the public debt holdings of developing countries are external debts. In most of these countries the share of debt in gross domestic product grew over time (see Figures 1 and 2). In fact, most of these countries face external debt that is more than two times the size of their gross domestic product. Due to scarce foreign exchange in most of these countries, efforts to service the debt consumed large shares of their government revenues. As the debt burden increases, these countries must allocate a greater portion of their revenues to service external debts, resulting in higher taxes, more borrowing, and eventually debt default.

The severe difficulties that most HIPCs faced in servicing their external debts resulted in the persistent accumulation of arrears, which are unpaid debt service obligations. Despite several repeated attempts at rescheduling, many HIPCs have not been able to meet their debt service obligations fully and on time for several years.

The worldwide economic growth slowdown has resulted in an increase in the level of debt burden, especially in the 1980s and 1990s (Easterly, 2001). This is partly because these countries have low incomes and their economies tend to grow slower than those of the higher income countries. Slower economic growth poses problems in expanding exports and delays progress in debt restructuring. This, in turn, impedes flows of capital to HIPCs. The massive external debts of these countries have reduced the inflows of foreign direct investment, employment, and growth of their economies and therefore have become a stumbling block to sustainable development.

In light of the above, this paper attempts to address the primary questions including: What are the factors behind the growth of external debts of the HIPCs? What are the consequences of massive external debts for these countries? What are the correlating effects within the factors themselves, so as to determine future patterns?

LITERATURE REVIEW

Over the years, the issue of public debt has occupied primary importance in both local and international arenas. Claessen, et al (1997) argued that HIPCs are characterized not only by high debt relative to income, but also by relatively poor economic performance. The reason is the combination effect of the large inflows of concessional finance despite the emerging debt burden and low growth rates of output and exports. In addition, the poor economic performance in these countries could be attributed to adverse terms of trade development, civil and political unrest, weak macroeconomic management, and inefficient allocation of resources.

The most recent article by Easterly (2001) confirmed that the slowing down of economic growth in the past decades since 1975 can be attributed to the increases in the burden of public debts in most middle income countries and HIPCs. In the case of HIPCs, Easterly (2001) found that the public debt burden was worse than other lower income countries due to their slow economic growth after 1975 compared to their counterparts as a result of their weak policies. Elbadawi, Ndulu, and Ndungu (1997) found that while current debt inflows enhanced economic growth, past debt accumulation, which was viewed as a proxy for debt overhang had a negative impact on economic growth. They argued that if the accumulation of past external debts reached a certain critical level, it would actually discourage investment and retard economic growth. The authors also confirmed that the liquidity constraints caused by rising external debt servicing payments reduced exports and thus hampered economic growth. Ajayi (1997) analyzed the correlation between external debt and capital flight in the HIPCs and found that the building up of excessive external debts would encourage capital flight. Capital flight, he argued, resulted in increased demand for external debts to fill the domestic investment needs. In addition, capital flight reduces growth, erodes the tax base, worsens income distribution, and reduces debt-servicing capacity by diverting domestic savings from investment.

de Larosiere (1984) found that the growth of external debt in most developing countries are attributable to their fiscal imbalances. In order to finance the deficit, these countries needed to increase taxes while at the same time reduce the non-interest expenditures. The two policies, however, are uncommon for these countries both because they are politically difficult to implement.

MODEL AND METHODOLOGY

In order to determine the factors that cause the growth of external debts, consider the following expression:

GEDBT = [alpha] + [beta]1NICA + [beta]2IPED + [beta]3RER + [mu] (1)

where,

GEDBT = growth of external debt to GDP

NICA = non-interest current accounts

IPED = interest payments on external debts

RER = real exchange rates

[mu] = the error term

In equation (1) the growth of external debts is regressed on non-interest current accounts, interest payments on external debts, exchange rates. The expected signs of the explanatory variables in relation to their effects on growth of external debts are presented in Table 1.

Control Variables

In addition to corruption index, this study uses governance indicators such as, internal conflict index, government instability index, and bureaucracy quality index as control variables. These control variables are included in this study because they are likely to affect the growth of external debts. For example, corruption has implications for the composition of government expenditures. Corrupt government officials try to channel public funds to finance their personal ventures. Such diversion of resources public funds to personal use negates economic growth. The coefficients of corruption index and bureaucracy quality is expected to be positive, while the coefficients of internal conflict and government stability are predicted to be negative. Equation (1) can be rewritten to include the governance indicators as follows:

GEDBT = [alpha] + [beta]1NICA + [beta]2IPED + [beta]3RER + [beta]4COR + [mu] (2)

GEDBT = [alpha] + [beta]1NICA + [beta]2IPED + [beta]3RER + [beta]4COR + [beta]5GI + [mu] (2a)

where,

GI = Governance indicators

COR = Corruption index

In equation (2) the growth of external debts is regressed on non-interest current accounts, interest payments on external debts, exchange rates and corruption index. In equation (2a) growth in external debts is regressed on non-interest current accounts, interest payments on external debts, real exchange rate, corruption index, and governance indicators. The expected effects of these variables on the growth of external debts are as follows:

Equations (1) through (2a) are estimated via the fixed- and random-effect. Fixed-effects model is used in order to allow the countries to have different intercepts that may be correlated with the regressors. The models are based on the following equation:

Yit = [chi]'it [gamma]it + [mu]it] (3)

where Y represents the dependent variable (Growth of External Debt), [chi]' is a vector of explanatory variables, i stands for the countries in the sample (i= 1, 2, 3, 4,...., 29), t is the period under investigation (t = 1984, 1985, 1986, 1987, 1988, 1989, 1990, 1991,...... 2000) and [[mu].sub.it] is the error term. From equation (3) we derive the fixed effects model in terms of the notations used in the study as follows:

GEDBTit = [[beta].sub.1]NICASit + [[beta].sub.2]IPEDit + [[beta].sub.3][RER.sub.it] + [[beta].sub.4][GI.sub.it] + [alpha]i + [delta]i + [mu]it (4)

where GEDBT represents growth of external debt, NICA stands for non-interest current account balance, IPED, RER, and GI as explained, while [mu] is the error term. In equation (4), [[alpha].sub.i] captures unobserved country-specific effects assumed fixed over time. The year-effects represented by [[delta].sub.i] are included to account for shocks that are common to all countries in the sample, such as rapid population, slow economic growth, and imperfect capital markets.

From equation (3), we again generate the random effects model as follows:

GEDBTit = [[beta].sub.1]NICAit [gamma]i + [[beta].sub.2][IPED.sub.it][gamma]I] + [[beta].sub.3][RER.sub.it][gamma]i + [[beta].sub.4][GI.sub.it][gamma]i + [delta]i + [mu]it, [gamma]I = [bar.[gamma]] + [??]i (5)

where [mu] is the error term, [??]i stands for random country effect while [bar.[gamma]] represents the mean of the coefficient vector. Under the random effects model, the slope coefficients are allowed to vary randomly across countries

Most of the previous country-studies applied the standard OLS procedure to examine the determinant of the growth of external public debts. These studies assumed that the omitted variables are independent of the explanatory variables and are independently, identically distributed. This assumption however leads to biased inferences especially when country-specific features such as policy changes. Hsiao (1986) points out that the OLS procedure yields biased and inconsistent estimates when the omitted country-specific variables are correlated with the explanatory variables.

The panel data approach provides avenues through which the country-specific characteristics (whether observed or unobserved) can be incorporated into cross-country studies to avoid biases resulting from the omission of relevant variables. The fixed-effect procedure yields unbiased and consistent estimates when the omitted country-specific variables are correlated with the explanatory variables. One of the shortcomings of the fixed-effects framework is that it assumes that differences across countries represent shift in the regression equation. This assumption implies that the fixed-effects model is appropriate when the entire population rather than the sample is investigated. However the random-effects model is applied when a sample rather the population is considered. The random-effects model is not without flaws. It yields biased regression estimates if the omitted country-specific variables are correlated with the explanatory variables. This study considers both the fixed-effect and random-effect procedures given the weaknesses associated with each of the models. Furthermore, our sample (25 countries) is large enough to warrant the application of both approaches.

DATA AND EMPIRICAL RESULTS

The data on external debts, non-interest current account, and interest payments on external debts were taken from the Global Development Finance published by the World Bank. The governance indicators were obtained from the International Country Risk Guide. The exchange rate data were obtained from the International Financial Statistics 2003 CD Rom version published by International Monetary Fund (IMF). The list of HIPCs consists of 29 countries including Angola, Burkina Faso, Cameroon, Congo, Democratic Republic, Congo Republic, Cote D Ivoire, Gambia, Ghana, Guinea, Guinea-Bissau, Guyana, Honduras, Kenya, Liberia, Madagascar, Malawi, Mali, Mozambique, Nicaragua, Niger, Senegal, Sierra Leone, Sudan, Tanzania, Togo, Uganda, Vietnam, and Zambia. The range of the study is from 1984 to 2000.

Table 2 displays the summary statistics for bureaucracy quality index (BQI), corruption index (COR), growth of external debt (GEDBT), real exchange rate (RER), government stability index GSI), internal conflict index (ICI), interest payment ratio of GDP (IPED), and current account deficits (NICA). The mean values for BQI, COR, GEDBT, RER, GSI, ICI, IPED, and NICA are 1.16, 2.58, 144.73, 1201.13, 6.22, 6.68, 0.10, and -0.32, respectively. The maximum and minimum values show cross-country variability among the variables used in the study. The standard deviations indicate that exchange rates fluctuated the most for the period under investigation.

Table 3 presents the bivariate correlations between growth of external debts, exchange rates, interest payments on external debts, non-interest current account, and governance indicators. From Table 3, it can be seen that growth of external debts are negatively correlated with non-interest current account, interest payments, corruption index, internal conflict index, government stability index, and bureaucracy quality index. However, growth of external debts and real exchange are positively correlated. Furthermore, the results show that non-interest current account, interest payments, corruption index, internal conflict index, government stability index, and bureaucracy index are positively correlated with each other.

Table 4 displays the results from the fixed and random effect models in conjunction with the associated test statistics. The explanatory variables are exchange rates, non-interest current accounts, interest payments on external debts, several governance indicators, and corruption index. All the t-statistics are given in parenthesis. The results from the fixed and random effect models for without governance indicators and corruption index are presented in column A of Table 4. The results from both the fixed and random effect models reveal that interest payments have significantly negative effect on the growth of external debts. Exchange rates are found to have significant influence on growth of external debts. Column B of Table 4 presents the results from both the fixed and random effect models with corruption index included as an additional explanatory variable. Again, the results reveal that interest payments have negative influence on growth of external debts. Exchange rates are found to affect growth of external debts positively. Interestingly, the results reveal that corruption engenders growth of external debts for the sample countries for the time period under investigation. The regression coefficient on corruption index is statistically significant at the 10 percent level. The regression coefficients on exchange rates and interest payments on external debts in Columns A and B are statistically significant at the 1 and 5 percent levels, respectively. The exchange rates and interest payments on external debts have the expected signs. The magnitudes of all significant variables are reasonable. The finding that interest payments have negative influence on growth of external debts suggests that high interest payments discourage foreign loans. However, it is important to point out that contrary to the conventional wisdom, some of the HIPCs have been able to increase their external debts through debt rescheduling and restructuring, irrespective of the level of interest payments.

The results from the fixed- and random-effect models with all governance indicators are presented in Column C of Table 4. The results from both procedures indicate that interest payments have significant adverse effects on the growth of external debts at the 5 percent significance level. However, non-interest current account deficits have no significant effects on growth of external debt for 29 HIPCs from 1984 to 2000. Interest payments, exchange rates and government stability index have significant effects on growth of external debts. Government stability index contributed negatively to the growth of external debt and is statistically significant at the 10 percent level. This result indicates that instability is associated with external debt problems. In other words, the less stable a country is, the more it encounters debt problems.

What is striking about these results is that when only corruption index was included in the model, it turned out the corruption played a positive role in determining the growth of external debts. It is important to point out that our prediction relative to effect of corruption on the growth of external debts was confirmed since the regression coefficient on corruption index has the expected sign. It is logical to say that corruption distorts economic growth and reduces the efficiency of government. Inefficiency in the public sector spurred by corruption leads to increase in demand for foreign loans at higher interest rates. In this particular study, we can conclude that as a country experiences less and less corruption, fewer funds are needed to supplement the loss of money as result of corruptions. On the contrary, one can also argue that activities associated with corruption can result in lack of trust in the eyes of international community. Lack of trust will reduce the inflow of foreign capital. Therefore, the less corrupt a country is, the easier it is for it to obtain external funds. The acquisition of additional foreign loans will increase the country's external debts to GDP ratio. Both arguments are quite valid depending on the credibility of the particular country under study.

It is interesting to note that the contributions of interest payment and exchange rate variables and signs remained relatively the same for with and without governance indicators as can be seen in Columns A through C of Table 4. Only non-interest current accounts balance really did not contribute to the growth of external debts. The regression sign on non-interest current accounts did not change in all three estimations and it is statistically insignificant in all of the cases. We have to be mindful of the fact that there has been an upward trend in the current account deficits for the HIPCs as a result of fluctuations in commodity prices. Fluctuations in commodity prices adversely affect the extent to which the HIPCs depend on foreign capital. However, given that the regression coefficient on non-interest current accounts is statistical insignificant in all cases, its contribution to the growth of external debts can be described at best, as marginal.

The regression coefficients on exchange rates and interest payments on external debts turned out as expected and statistically significant in all of the cases. This relationship is quite robust and obvious since most of the HIPCs followed aggressive exchange rate policies, which led to increases in devaluation of local currencies in terms of purchasing power parity. Inefficient exchange rate policies pursued by most of the HIPCs caused the ratio of debt to GDP to increase as a result of capital loss.

The results from the fixed- and random-effect models with all governance indicators are presented in Column C of Table 4. The results reveal that most of the governance indicators have insignificant effects on growth of external debts. In short, only government stability index significantly contributed to the growth of external debts. As expected, the regression coefficient on government stability index is negative. This result implies that the more stable a country is, the less it relies on foreign loans. Interestingly, all of the regression coefficients on the governance indicators exhibited the expected signs. Internal conflict and bureaucracy quality indexes appear not to have implications for growth of external debts, as they are statistically insignificant at the conventional levels. However, it is important to point out that most of the countries in this classification (HIPCs) are plagued with internal conflicts. Most of these countries divert substantial resources and political attention from economic, financial, and social programs to internal conflicts. The reallocation of resources in favor of internal conflicts has a negative effect on economic development, as important parts of the productive infrastructure are either neglected or destroyed due to internal strife.

CONCLUSIONS AND POLICY RECOMMENDATIONS

High external debts can erode confidence in economic reforms and thus diminish the sustainability of what might be an otherwise sound economic reform strategy. Massive external debts can have indirect negative consequences on governments in terms of public support insofar as debts are perceived to contribute to poor growth and poor policies. This paper attempted to ascertain the determinants of the growth of external debts. The study uses panel data for 29 HIPCs from 1984 to 2000. The fixed- and random-effect models were used to investigate the relationships between growth of external debts, exchange rate, interest payments on external debts, and non-interest current account balance. The governance indicators including internal conflict index, government instability index, and bureaucracy quality index were used as control variables. The main results of the paper are interesting and intuitive.

Interest payments on debts and real exchange rates have significant effects on growth of external debts. Exchange rates exhibited the lowest regression coefficient but has the highest significant level amongst all explanatory variables. Surprisingly, non-interest current account balance, which was expected to be significant along with interest payments and exchange rates, proved otherwise. In terms of control variables, only one (i.e. government stability index) out of the four governance indicators has implications for growth of external debts.

In all, the results suggest that there are strong correlation between growth of external debts to GDP ratio, exchange rates, interest payments on external debts and some of the governance indicators namely--corruption and government stability indexes. The results have important implications for the HIPCs, especially as they struggle to map out strategies to curtail their reliance on foreign capital and to avoid further debt-overhangs. The HIPCs should formulate strategies that will enable them to curtail their external debt burdens. The results of this study show that the increases in foreign debt burdens can be attributed to high debt service costs. The inability of the HIPCs to curb their external debt burdens can be blamed on exchange rate misalignments, rather than, the size of their current account deficits. The aggressive exchange rate policies pursued by the HIPCs actually weakened their currencies in terms of purchasing power parity. The implied capital loss due to worsening purchasing power parity exacerbates the external debt burdens of these countries. The HIPCs should confront the issue pertaining to corruption in their economies. They should develop policies designed to curtail wide spread corruptions in these countries. Above all, stable institutions and governance measures should be strongly encouraged, as these will enable the HIPCs to alleviate their external debt burdens.

It should be noted, however, that the economic and political situations of these countries make them a non-typical sub-sample. Further research is therefore necessary to provide a more definitive assessment of the relationship between growth of external debts and some of the variables used in this study.

ACKNOWLEDGMENTS

The authors wish to express their gratitude towards two anonymous referees of this journal and the participants at the Eastern Economic Association conference (Washington D.C.) 2004, for their valuable comments and suggestions. The usual disclaimers apply.

REFERENCES

Ajayi, S. I. (1997). An analysis of external debt and capital flight in the heavily indebted poor countries of Sub-Saharan Africa. In External Finance for Low-Income Countries, Zubair Iqbal and Ravi Kanbur, (eds.) IMF, 77-117.

Anand, R & Sweder Van Wijnbergen, (1989). Inflation and the financing of government expenditure: An introductory analysis with an application to Turkey. The World Bank Economic Review. 3(1), 17-38.

Bacha, E. (1991). Debt crisis, net transfers, and the GDP growth rate of developing countries", UNDP/UNCTAD, INT/88/021.

Boote, A. R. & Kamau Thugge, (1997). Debt relief for low-income countries and the HIPC initiative, IMF working Paper, No. WP/97/24.

Claessens, S., Detragiache, E., Ravi Kanbur & Peter Wickham (1997). Analytical aspects of the debt problem of HIPC, in External Finance for Low-Income Countries, Zubair Iqbal and Ravi Kanbur, (eds.) IMF, 21-48.

De Larosiere, J (1984). The growth of public debt and the need for fiscal discipline. IMF Paper.

Dickey, D. A & W. A. Fuller (1979). Distribution of estimated of autoregressive time series with unit root. Journal of American Statistical Association. 427-31.

Easterly, W. R. (2001). Growth implosions and debt explosions: Do growth slowdowns cause public debt crisis? Contributions To Macroeconomics, 1(1).

Elbadawi, I. A., Benno J. Ndulu, and Njuguna Ndungu (1997). Debt overhang and economic growth in Sub-Saharan Africa. In External Finance for Low-Income Countries, Zubair Iqbal and Ravi Kanbur, (eds.) IMF, 49-76.

Hsiao, C. (1986). Analysis of panel data. Econometric Society Monographs No. 11, Cambridge University Press: Cambridge, United Kingdom.

Humphreys, C & John Underwood (1989). The external debt difficulties of low-income Africa. IMF Policy, Planning, and Research Working Paper, WPS No. 255.

Husain, I. & Saumya Mitra (1989). Future financing needs of the highly indebted countries IMF Policy, Planning, and Research Working Paper, WPS No. 254.

Nallari, R & Sona Varma (1995). Deficit financing, external debt, and economic growth: An application to the Indian economy. Journal of Quantitative Economics, 11(2), 85-94.

Strauss, T. (1998). Growth and government: Is there a difference between developed and developing countries. Working Paper Series in Economics and Finance, No. 275.

Van Wijnbergen, S. (1990). External debt, inflation, and the public sector: Toward fiscal policy and sustainable growth. The World Bank Economic Review, 3(3), 297-320.

Van Wijnberger, S.(1989). Growth, external debt, and real exchange rate in Mexico. IMF Policy, Planning, and Research Working Paper, WPS No. 257.

Emmanuel Anoruo, Coppin State University Young Dimkpah, Virginia State University Yusuf Ahmad, The World Bank, Washington, D.C.

Table 1: Expected Signs of the Explanatory Variables

 BQI COR RER GSI ICI IPED NICA

Equation (1) + - -
Equation (2) + + - -
Equation (2a) - + + - + - -

BQI = Bureaucracy quality index, COR = Corruption index, GEDBT
= Growth of external debts/GDP, RER = Real exchange rates, GSI
= Government stability index, ICI = Internal conflict index,
IP = Interest payments on external debts/GDP, and NICA = Non-interest
current account balance.

Table 2: Summary Statistics

 BQI COR GEDBT RER

Mean 1.16 2.58 144.73 1201.13
Median 1.00 3.00 113.00 364.84
Maximum 3.00 5.00 1064.00 22332.50
Minimum 0.00 0.00 0.00 0.00
Std. Dev. 0.89 1.13 115.48 3244.07
Skewness 0.50 -0.38 3.26 4.30
Kurtosis 2.59 2.87 19.23 21.62
Jarque-Bera 23.54 11.98 6285.07 8638.11
Probability 0.00 0.00 0.00 0.00
Observations 493.00 493.00 493.00 493.00

 GSI ICI IPED NICA

Mean 6.22 6.68 0.10 -0.32
Median 6.00 7.00 0.05 -0.23
Maximum 11.00 12.00 0.79 3.27
Minimum 1.00 0.00 0.00 -2.59
Std. Dev. 2.44 2.57 0.13 0.40
Skewness 0.31 -0.22 2.70 -0.11
Kurtosis 2.41 2.59 11.91 20.28
Jarque-Bera 15.08 7.46 2229.43 6137.60
Probability 0.00 0.02 0.00 0.00
Observations 49300.00 493.00 493.00 493.00

BQI = Bureaucracy quality index, COR = Corruption index,
GEDBT = Growth of external debts/GDP, RER = Real exchange
rates, GSI = Government stability index, ICI = Internal
conflict index, IP = Interest payments on external debts/GDP, and
NICA = Non-interest current account balance

Table 3: Correlation Matrix

 BQI COR GEDBT RER

BQI 1.00
COR 0.33 1.00
GED BT -0.07 0.13 1.00
RER 0.08 0.15 0.05 1.00
GSI 0.15 0.09 -0.12 0.20
ICI 0.27 0.20 -0.10 0.25
IPED 0.38 0.06 -0.08 0.06
NICA -0.01 -0.03 -0.07 -0.17

 GSI ICI IPED NICA

BQI
COR
GED BT
RER
GSI 1.00
ICI 0.37 1.00
IPED 0.07 0.27 1.00
NICA -0.03 0.06 -0.09 1.00

BQI = Bureaucracy quality index, COR = Corruption index,
GEDBT = Growth of external debts/GDP, RER = Real exchange
rates, GSI = Government stability index, ICI = Internal
conflict index, IP = Interest payments on external debts/GDP,
and NICA = Non-interest current account balance

Table 4: Dependent Variable: Growth of External Debts

 Without Governance
 Indicators

Independent Variables Fixed Random
 Effects Effects

Constant 149.85 *** 148.81 ***
 (19.66) (7.67)

Interest Payments on -139.17 ** -127.56 **
External Debts
 (2.29) (2.25)

Exchange Rates 0.01 *** 0.01 ***
 (4.30) (4.16)

Non-Interest Current -3.07 -3.96
Account Balance
 (0.31) (0.40)

Internal Conflict Index -- --

Corruption Index -- --

Government Stability Index -- --

Bureaucracy Quality Index -- --

Adjusted R-Square 0.043 0.043

Number of Observations 493 493

 With Corruption only

Independent Variables Fixed Random
 Effects Effects

Constant 128.00 *** 126.57 ***
 (9.09) (5.61)

Interest Payments on -132.60 ** -122.30 **
External Debts
 (2.18) (2.16)

Exchange Rates 0.01 *** 0.01 ***
 (4.30) (4.15)

Non-Interest Current -2.83 -3.66
Account Balance
 (0.29) (0.38)

Internal Conflict Index -- --

Corruption Index 8.24 * 8.48 *
 (1.84) (1.94)

Government Stability Index -- --
 (2.61) (2.62)

Bureaucracy Quality Index -- --

Adjusted R-Square 0.050 0.050

Number of Observations 493 493

 With All Governance Indicators

Independent Variables Fixed Random
 Effects Effects

Constant 171.37 *** 168.99 ***
 (9.02) (6.51)

Interest Payments on -146.53 ** -130.93 **
External Debts
 (2.18) (2.10)

Exchange Rates 0.01 *** 0.01 ***
 (5.02) (4.87)

Non-Interest Current -3.37 -4.16
Account Balance
 (0.34) (0.43)

Internal Conflict Index -2.51 -2.52
 (1.29) (1.32)

Corruption Index 5.22 * 5.82 *
 (1.90) (1.98)

Government Stability Index -4.02 ** -4.00 **

Bureaucracy Quality Index 4.98 4.08
 (0.88) (0.73)

Adjusted R-Square 0.08 0.08

Number of Observations 493 493

Note: Absolute value of robust t -statistics are in parentheses;
* significant at 10%; ** significant at 5%; and *** significant at 1%