The Determinants of Country Risk Analysis

An Empirical Approach
Madhu Vij
Abstract

The paper examines the effect of various economic and political factors on country risk ratings published
by Euromoney and Institutional Investor. As global competition drives corporations, managers frequently
rely on country risk analysis as a crucial aspect of strategic decision-making. The purpose of this paper
is to investigate the extent to which country risk measures can help in predicting country ratings. We
examine seven widely used measures of country risk across sixty-one countries. Results from the empirical
analysis indicate that country risk ratings can be replicated to a significant degree with a few available
political and economic indicators. Political risk was found to exert a significant influence on country
ratings. The results also confirmed that both Euromoney and Institutional ratings predicted similar
outcomes.

Keywords:Country Risk Analysis, Gross National Product, Gross Capital Formation

I. Introduction the 1990s. These ratings are an important

component of country risk management because
Recent years have witnessed an unprecedented
they provide a framework for establishing country
interest and research in identifying the
exposure limits that reflect the institution’s
determinants of country risk analysis. Institutions
tolerance for risk. Various empirical studies (for
engaged in managing global investment strategies
example, Feder and Uy 1985; Brewer and Rivoli,
are exposed to country risk — the risk that
1990) have developed quantitative models to
economic, social and political events in a foreign
replicate the country risk ratings published by
country would adversely affect an institution’s
banking magazines such as Euromoney and
financial interest. Managers frequently rely on
Institutional Investor. These ratings, which are
country risk analysis while formulating their
considered to provide a measure of the credit
strategies. Moreover, empirical researchers have
worthiness of sovereign borrowers, affect the cost
agreed that country risk is the result of political,
of capital flows to sovereign borrowers because
social and economic factors (Oetzel, et. al. 2001).
they seem to have been systematically linked to
Practitioners of country risk analysis face a
credit pricing in the Euro market (Feder & Ross,
daunting task in their selection of variables and
1982).
evaluation systems to represent and interpret the
various economic and social-political factors Feder & Uy (1985) were the first to identify the
(Burton and Inou, 1985). determinants of country risk ratings. They
replicated the creditworthiness ratings reported in
Research on the extent to which country risk
the 1979-1983 issues of the Institutional Investor.
measures predict country risk ratings flourished in
Feder and Uy were able to explain 70 per cent of
the variance of Institutional Investor ratings for a
Madhu Vij group of 55 developing countries.
Faculty of Management Studies Another study by Brewer & Rivoli (1990) focused
University of Delhi on the effects of political instability on perceived
Delhi - 110007
country creditworthiness of the 30 most heavily During the last decade, there has been a multitude
indebted developing countries by using the 1987 of economic and political crises — the Mexican
rating of Institutional Investor and Euromoney. peso crashed in 1994, and between 1997-1998 we
witnessed widespread default of financial
Saini and Bates (1984) in their survey of the institutions in Bangkok, the collapse of the
quantitative analysis of country risk analysis Indonesian government and the Malaysian
discuss the possibility of judgmental change. They economy, and the devaluation of the Brazilian
argue that given rapid transformation in the currency (Octzel, et. al. 2001).
economies of developing countries, and the
international environment, we could expect Thus, following the series of currency crashes in
structural shifts overtime in the influence of the 1990s-with the latest crisis in Turkey in
explanatory variables. February 2001, country ratings will continue to be
a crucial area of interest to business managers and
Another significant issue here is that rating researchers in strategic decision making. It is for
agencies have been accused of promoting financial this reason that managers are interested in
excesses. As discussed by Ferri, Lui and Stiglitz determining various measures to forecast political-
(1999), their pro-cyclical behavior, upgrading economic events in a country that may help in their
countries in good times and downgrading them in assessment of creditworthiness indicators.
bad times, may have contributed to magnifying the
boom-business pattern in stock markets. Even if This paper focusses on developing a country risk
rating agencies do not behave pro-cyclically, their appraisal model, to identify the various political
announcements may still trigger market filters. and economic factors that can help in predicting
Rating changes may unveil new (private) country risk ratings. To do this, we have examined
information about a country and they may fuel seven widely-used measures of country risk across
rallies or down turns. This effect is likely to be 61 countries. The paper employed logistic multiple
stronger in emerging markets where problems of regression analysis to develop a country risk model
asymmetric information and transparency are more to identify the determinants of country risk rating.
severe. Moreover, changes in rating might act as a The model focuses on country risk rating by
wake-up call, with rating changes for one country Euromoney and Institutional Investor for the year 2001
affecting other countries with similar economies. and the important economic and political variables
(Kaminsky, G. and Schmukler, S. 2001) for the year 2000. The primary purpose of the
study is to illustrate an approach to country risk
Given the growing importance of trade in the appraisal that can help to identify the important
world economy and also the increasing global factors that affect country risk perceptions. For
economic integration, which started around 1980 most of the empirical work in this area, multiple
and continues till today (World Development regression analysis has been employed.
indicators, 2002), managers are more concerned
about forecasting the various potential economic
and political indicators affecting their ratings. II. The Model
Private capital flows to developing economies have The Logistic Multiple Regression Analysis used in
increased dramatically from around $44 billion in this paper is a technique, which allows us to
1990 to $257 billion in 2000, while official flows incorporate multiple independent variables to
have decreased from $57 billion to $ 39 billion estimate the dependent variable. The model can be
during the same period. Foreign direct investment expressed as
has become the major form of international
finance for developing countries, accounting for Yi = α 0 + β 1 X1i + β 2 X2i + ….. + β n Xni + ∈i (1)
about 70 per cent of the private capital flows in The notation Xni indicates the values of the nth
2000 (World Development Indicators, 2002.) independent variable for case i. The beta terms are

Volume 5, Number 1 • April 2005 21

unknown parameters and the ∈i terms are creditworthiness rating reported by Institutional
independent random variables that are normally Investor or Euromoney magazine. Also, only seven
distributed with mean zero and constant variance independent variables are included in the multiple
σ2. regression model, as including a large number of
independent variable in a regression model is never
Burton & Inou (1985) express country risk as a good strategy, unless there are strong previous
CRit = f (ECi(t-1), POi(t-1) (2) reasons to suggest that they all should be included.
Including irrelevant variables increases the standard
Country risk in equation 2 is assumed to be error of all estimates without improving the
composed of economic and political related risk. prediction. To reduce the skewness in the
ECi(t-1) is economic-related risk and Po i(t-1) is the distribution, as also to make the independent
political-related risk for country i in the period t- 1 . variables comparable, logarithmic transformation is
applied to all the variables.
A one-year time lag characterizes the assumption
that in their assessment of new financial
commitment to specific countries, lenders are III. The Dependent and Explanatory
influenced principally by the most recent economic Variables
and political circumstances. This section discusses in detail the dependent and
The selection of various independent variables that explanatory variables that were used in the present
were used to estimate country risk ratings are as study.
follows:
The Dependent Variable
CRt = f (GNPt-1, GKFORMt-1, FDEXP t-1, RESIMP t-1,
CURGNPt-1, EXPGRTH t-1, POLRSKt-1) (3) 1. The dependent variable is the country credit
worthiness rating reported in the September
where
2001 issues of the Institutional Investor.
CRt = Country risk in period t Institutional Investor country credit ratings
are based on information provided by leading
GNP = Gross national income per capita, in
t-1
period t-1 international banks. Responses are weighted
by using a formula that gives more
GKFORM t-1 = Gross capital formation in period importance to responses from banks with
t-1 greater worldwide exposure and more
FDEXP t-1 = Net foreign debt/exports ratio in sophisticated country analysis systems.
period t-1 Countries are rated on a scale of 0 to 100
(highest risk to lowest), and ratings are
RESIMP = Reserves to imports ratio in period
t-1
t-1 updated every six months (World
Development Indicators, World Bank, 2002,
CURGNP t-1 = Current account balance on gross page 287).
national income in period t-1
2. Eurocurrency country creditworthiness
EXPGRTH = Exports growth rate in period t-1
t-1
ratings are based on nine weighted categories
POLRSK t-1 = Political Instability Indicator in (covering debt, economic performance,
period t-1 political risk and access to financial and
capital markets) that access country risk. The
The above mentioned political and economic risk
rating on a scale of 0 to 100 (highest risk to
indicators serve as the independent variables that
lowest risk) are based on polls of economists
are used to calculate the predictive power of the
and political analysts supplemented by
dependent variable, CRt that is the country
quantitative data such as debt ratios and

22 Journal of Management Research

 access to capital markets (World measures the burden of a country’s debt relative to
Development Indicators, World Bank, 2002, the major source of foreign exchange.
page 287).
Total external debt is debt owed to nonresidents
Thus, the dependent variable here is the country repayable in foreign currency goods or service. A
credit worthiness ratings from two major country with a high external debt to exports ratio
international rating services — The Institutional is more vulnerable to foreign exchange crises and
Investor and the Euromoney. However, risk rating may more likely to default (Frank and Cline 1971, Cline
be highly subjective, reflecting external perceptions 1984). Thus, the variable is negatively related to
that do not always capture the actual situation in a country credit ratings as a higher external debt to
country. But these subjective perceptions are the exports ratio is expected to lead to lower
reality that policy makers face. creditworthiness rating (Cosset etc.)

The Explanatory Variables 4. Gross International Reserves to Imports

Ratio
The set of explanatory variable used in the present
study is derived from previous empirical researches Reserves provide a short-term safeguard against
and from the suggestions of theoretical research fluctuations in foreign receipt. Feder, Ross and Just
on international borrowings. (1981), and Cline (1984), argue that larger the
reserves are relative to imports, the more reserves
1. Gross National Income (GNI) per capita are available to service debt and the lower is the
(or GNP) probability of default. Thus, this variable is
positively related to a country’s creditworthiness
This variable is used by World Bank to classify rating.
countries for analytical purposes and to determine
borrowing eligibility. The variable measures the
5. Current Account Balance on Gross
level of development of a country. Frank and
National Income
Cline (1971), and Feder and Just (1977) argue that
poorer countries may have less flexibility to reduce This variable is negatively related to the probability
consumption than richer countries. The variable is of default (Cline 1984), since the current account
positively related to country credit worthiness deficit broadly equals the amount of new
rating. Thus, countries with low gross national ‘financing required. Thus, countries with large
income per capita will be generally less current account deficits are less creditworthy.
creditworthy.
6. Exports Growth Rate
2. Gross Capital Formation
Countries with high export growth rates are more
This variable consists of outlays on additions to likely to service their debt and hence enjoy better
the fixed assets of the economy plus net changes creditworthiness ratings (Feder and Uy, 1985) as
in the level of inventories. The variable captures a exports are the main source of foreign exchange
country’s prospects for future growth and is earnings for most countries. This variable is
positively related to its risk rating. This variable is positively related to country creditworthiness
calculated as the ratio of Gross Domestic rating.
Investment to Gross Domestic Product and is also
known as the propensity to invest. 7. Political Risk Indicator
Political instability may indirectly accelerate debt
3. Total External Debt to Exports Ratio
service problems through a decline in long-term
The ratio of total external debt to exports

Volume 5, Number 1 • April 2005 23

capital flows and a consequent unwillingness of First Stage Analysis
lenders to roll over matured loans. Over a period
The two multiple regression analysis equations
of time, political instability may slow economic
estimated in the first stage were
growth, contribute to inflation, domestic
bottlenecks and production shortages and create Log CR1(euro) = 1.671 + .171 (GNP) + .304
foreign exchange shortage from an imbalance (GKFORM) – 0.70 (FDEXP) +
between exports and imports (Burton and Inoue, .117(RESIMP) – .019 (CURGNP)
+ .079 (EXPGRTH) (4)
1985).
Log CR2(Instit) = –.271 + .292 (GNP) +
Aliber (1980), and Brewer and Rivoli (1990) have .592 (GKFORM) – .074
argued that political instability can reduce a (FDEXP) + .167 (RESIMP) –
country’s willingness to service debt. .033 (CURGNP) + .104
(EXPGRTH) (5)
It has also been suggested in literature (Burton and
Inoue, 1985 and Citron and Nicklesburg, 1987) For the time period analyzed, equation 4 explains
that disruptive political events frequently precede about 84% (R square=.84) while equation 5
debt rescheduling. Thus, countries experiencing explains about 83% (R square=.83) of the variance
high political turmoil are more likely to default. in country risk.
Thus, the predictability of the equation does not
Data Sources change significantly whether we consider the
The data on the various independent and Euromoney or Institutional Investor ratings. This
dependent variables was obtained from the World is once again proved in stage two analysis. It can
Development Indicators, 2002, a publication of the therefore be concluded that the rating assigned by
World Bank. The Political risk indicator was the two institutions, Euromoney and Institutional
obtained from the Euromoney Journal (November Investor, are generally always in the same direction
2002). Sixty-one countries were selected as the and any of the two ratings can be used for analysis.
sample for this analysis. The names of the One of the important purposes of this paper was
countries selected along with their groupings are to evaluate any significant difference in ratings by
given in the Appendix. the two institutions. It can thus be safely concluded
that for the time period analyzed, both the ratings
will give the same result. Our results are also in
IV Results
agreement with the conclusions of Cosset and Roy
For the same set of independent variables, two (1991) who found that both ratings could be
equations have been computed – one with replicated to a significant degree with only a few
Euromoney ranking as the dependent variable and widely available economic statistics and both
the other with Institutional investor ratings. This models predicted similar outcomes.
has been done so as to find out if any significant
differences exist between the two ratings assigned Gross capital formation seems to be the most
to different countries for the same period. Also, important factor in evaluating country risk
the calculations have been done in two stages. The followed by Gross National Product or Income per
computation in stage 1 excludes the political risk capita. Both these variables are positively related to
indicator. In stage II all the variables are included country risk. The reserves to imports ratio and
to test the credence of arguments that political risk exports growth rate are positively related to
has an important bearing on country risk. country risk while the external debt to exports ratio
and the current account balance on GNI bear their
hypothesized negative sign

24 Journal of Management Research

Second Stage Analysis explaining the equation. The first variable is
Log CR = .735 + .180 (GNP) + .383
selected which best explains the dependent
(Instit) (GKFORM) – .075 (FDEXP) + variable. In the next step, in combination with this
.210 (RESIMP) – .039 (CURGNP) – .076 variable, the second variable is selected. This
(EXPGRTH) + .214 (POLTRIS) (3) procedure goes on till the remaining variables cease
to contribute in a significant manner towards the
Log CR = 2.270 + .105 (GNP) +.180 (GRFORM)
predictability of the equation.
(Euro)] – .071 (FDEXP) + .143 (RESIMP)
– .023 (CURGNP) – .061 (EXPGRTH)
+ .127 (POLTRIS) (4) Results
Tables 1 and 2 present the result of the stepwise
The effects of political instability indicator were
analysis when political risk was not considered as
examined by the addition of the political risk
an independent variable. Table 1 indicates that the
variable to the equations in the first stage. The
most powerful variable is the gross national
results are presented in equations 3 and 4.
product per capita. The importance of this variable
As discussed in stage one, the predictability of the is consistent with the theoretical literature as GNP
equation remains nearly the same whether we use per capita measures the level of development of a
Euro money ratings (R Square = .861) or country. The variable has the maximum correlation
Institutional Investor ratings (R Square = .862) with the dependent variable and the predictability
The two important variables in equations 3 and of the dependent variable with only GNP per
4 are gross capital formation and political risk. capita as the independent variable is 58%. Gross
Both are positively related to country risk ratings. capital formation is the second most important
The other variables with a positive sign are gross independent variable and when this variable, along
national product and reserves to imports ratio. As with GNP per capita is used, the predictability of
observed in stage one also, the foreign debt to the equation increases to approx. 74%. In both the
exports ratio and the current account balance on cases the F values are found to be significant ( 1%
GNP show a negative sign. and 5%, respectively).
Table 2 summarizes the result when the dependent
Stepwise Analysis variable is the Institutional Investor rating. The
The stepwise analysis technique is also employed results once again confirm the importance of the
in this paper. This technique helps to select two variables, GNP per capita and Gross capital
variables in the order of their importance in formation. The F values and t values are also
significant.

Table 1: Stepwise regression analysis (excluding political risk)

Model β Std. Error t Sig. R square Significance

1. (Constant) 2.255 .427 5.284 .000

LGNP .219 .059 3.713 .004 .580 .004

2 (Constant) .747 .718 1.040 .325

LGNP .258 .051 5.044 .001 .744 .040

LGKFORM .409 .170 2.406 .040

Dependent variable: Euromoney

Volume 5, Number 1 • April 2005 25

 Table 2: Stepwise Regression Analysis (Excluding Political Risk)

Model β Std. Error t Sig. R square Significance

1. (Constant) 1.338 .670 1.993 .074
LGNP .314 .093 3.388 .007 .534 .007
2 (Constant) -1.232 1.062 -1.160 .276
LGNP .381 .076 5.026 .001 .739 .022
LGKFORM .696 .251 2.770 .022
Dependent Variable: Institutional Investor

Table 3: Stepwise regression analysis (Including Political Risk)

Model β Std. Error t Sig. R square Significance

1. (Constant) 2.315 .299 7.748 .000

LPOLTRIS .515 .117 4.390 .001 .658 .001
Dependent variable: Institutional Investor

Table 4: Stepwise regression analysis (Including Political Risk)

Model β Std. Error t Sig. R square Significance

1. (Constant) 2.997 .210 14.249 .000
LPOLTRIS .336 .083 4.064 .002 .789 .002
Dependent variable: Euromoney

Table 5: Correlation between the Dependent and Independent Variables

LGNP LGKFORM LFDEXP LRES- LCUR- LEXP- LPOLTRIS LINSTIT LEURO-

IMP GNP GRTH MONEY

GNP 1.000 .279* -.384* -.109 .209 -.117 .810** .790** .906**
Pearson Correlation Sig. (2-tailed)
GKFORM .279* 1.000 -.139 -.206 .089 .133 .470** .357** .435**
Pearson Correlation Sig. (2-tailed)
LFDEXP .384* -.139 1.000 .001 -.364 .319 -.471** -.417** -.506**
Pearson Correlation Sig. (2-tailed)
LRESIMP .109 -.206 .001 1.000 .231 -.195 -.175 -.179 -.265*
Pearson Correlation Sig. (2-tailed)
LCURGNP .209 .089 -.364 .231 1.00 -.227 .196 .130 .144
Pearson Correlation Sig. (2-tailed)
LEXPGROT -.117 .133 .319 -.195 -.227 1.000 -.032 -.173 -.159
Pearson Correlation Sig. (2-tailed)
LPOLTRIS .810* .470** -.471** -.175 .196 -.032 1.000 .852** .877**
Pearson Correlation Sig. (2-tailed)
LINSTIT .790** .357** -.417** -.179 -.130 -.173 .852** 1.000 .829**
Pearson Correlation
LEUROM .906** .435** -.506** -.265* .144 -.159 .877** .829** 1.000
Pearson Correlation

* Correlation is significant at the .05 level (2 tailed)

** Correlation is significant of at the .01 level (2 tailed)

26 Journal of Management Research

Similarly, a stepwise analysis was performed by political risk and gross capital formation. Once
including political risk as one of the independent again the F values and t values are significant at
variables. Tables 3 and 4 present the results. In 1% level.
both the cases, political risk exerts a significant
influence on the ratings and is the single most Latin America and Caribbean
crucial factor driving country risk analysis. The F
values and the t values are also significant. The political risk factor seems to be the most
important variable here and the power of this
Finally, a correlation analysis was conducted variable lends support to empirical research (e.g.
between all the dependent and independent Ingram, 1974) that countries tend to experience
variables taken together. Table 5 presents the political instability before expropriation measures
results. are imposed. Most findings have been drawn from
experiences in Latin America.
GNP per capita, gross capital formation and
political risk show a high positive correlation in the
Institutional Investor and Euromoney ratings and are Europe and Central Asia
significant at 1% level of significance. Both net The influence of political risk once again seems to
foreign debt to exports and exports growth rate be the most important discriminating variable for
show a negative correlation with the two ratings countries in Europe and Central Asia and is highly
and net foreign debt to exports is significant at 1% significant. However, in the stepwise analysis, if
level of significance. the political risk factor was excluded from our
Comparison Across Countries analysis, then gross capital formation seems to
become more relevant (R square = .684, F value is
The principal purpose of country risk evaluation significant).
is also to set country exposure limits. In this part
of our study we divided the countries into seven Middle East and North Africa
groups based on the classification given by the
World Bank. Then, for each country group, a Political risk is the relevant variable in this group
stepwise analysis was carried out to find out if and is highly significant.
there are significant differences in factors across
the countries. Table 6 summarizes the results. Sub-Saharan Africa
We will briefly comment on the results of the A stepwise analysis here picks up political risk and
stepwise multiple regression analysis for each GNP per capita as the two most significant
country classification. variables that improve the predictability of the
equation. Once again both the F values and t
East Asia and the Pacific
values are significant.
GNP per capita seems to be the most important
factor in evaluating country risk. The coefficient of High Income
determination (R square) is .756 and both the F
values and t values of the equation are significant. For countries in this group political risk is the most
Political risk gets excluded in the stepwise analysis important variable (R square =.769) and the beta
and is not significant. values are significant at 1% level. GNP per capita
is the second most important variable and the two
South Asia variables together explain approximately 89% of
the variation (R square = .892) in the dependent
For South Asian countries the two factors, which
variable. In the next stage, reserves to imports ratio
significantly affect the rating of a country are,
is the relevant variable (R square = .960)

Volume 5, Number 1 • April 2005 27

 Table 6: Country Classification Results

B S.E. R square

East Asia and Pacific

Constant 2.410* .361
LGNP .219* .049 .756*

South Asia
Model 1 (Constant) 2.234* .209 .939*
LPOLTRSK .665* .098
Model 2 (Constant) 2.979* .087
LPOLTRSK .756* .020 .999*
LGKFORM -.301* .032

Latin America & Caribbean

Constant 1.973** .620
LPOLTRSK .756** .235 .775**

Europe and Central Asia

Constant 2.335* .160 .927**
LPOLTRSK .615* .071

Middle East & North Africa

Constant 2.126* .215
LPOLTRSK .718* .087 .945*

Sub Saharan Africa

Model 1 (Constant) 2.532* .132 .916*
POLTRSK .542* .067
Model 2 (Constant) 2.170 .083
LPOLTRSK .352 .043 .989*
LGNP .116 .021

High Income Income

Model 1 (Constant) 2.727* .244 .769*
LPOLTRSK .5666* .077
Model 2 (Constant) 2.452* .184
LPOLTRSK .473* .059 .892*
LGNP .056* .014
Model 3 (Constant) 2.259* .123
LPOLTRISK .521* .039 .960*
LGNP .062* .009
LRESIMP .011* .002
* significant at 1% level
** significant at 5% level
*** significant at 10% level

28 Journal of Management Research

V Conclusion Inou (1985) who found foreign exchange
reserves to imports to be the most critical
This paper examines the effect of various
factor among country risk economic related
economic and political factors on country risk
variables.
ratings published by Euromoney and Institutional
Investor. Some of the conclusions, which can be 4. In this paper, we focussed on identifying the
drawn from the analysis have been discussed effects of various economic and political
below. factors on changes in ratings of a country.
We also tested for any differences in ratings
1. Political risk exerts a significant influence on
given by Euromoney and Institutional
the country rankings. It is the single most
Investor. We found that no significant
crucial factor influencing country risk
differences exist and any one of the two
analysis when stepwise multiple regression
ratings can be used to explain the effects of
analysis is employed for all the countries
credit ratings on financial markets.
taken together.
5. The evidence suggests that country risk rating
2. When a stepwise analysis was performed
can be replicated to a significant degree with
excluding political risk, GNP per capita and
a few available political and economic
gross capital formation are the two economic
indicators. The results also indicate the
factors, which significantly explain country
dominating influence of GNP per capita on
risk ratings. The other variables for the period
country risk ratings. The other determinants
either contribute less to an explanation of the
of country risk rating are gross capital
variance, contradict the expected sign as a
formation, the ratio of net foreign debt to
consequence of multicollinearity or have a
exports, the ratio of reserves to imports, the
weak significance level.
ratio of current account balance on GNP and
3. The country group indicators also seem to the exports growth variable. All these
indicate the importance of political risk for variables show a high correlation with both
all regions except for East Asia and Pacific. the Euro Money and Institutional Investor
The influence of GNP per capita is ratings.
particularly important here. For high-income
6. The political risk approach is of greater
countries, the ratio of foreign exchange
relevance since most of the government
reserves to imports is the third most
decisions affect the economic factors directly
important variable (after political risk and
and it is difficult to define an accurate
GNP per capita). For other country groups
measure to predict cross border risks.
the ratio is not significant. This finding
contradicts the earlier finding of Burton &

References
Aliber, R.Z. (1980), A Conceptual Approach to the Analysis of External Debt of Developing Countries, World Bank Staff Working
paper no. 421, The World Bank, Washington, DC.
Brewer, T. and P. Rivoli (1990), Politics and Perceived Country Creditworthiness in International Banking, Journal of Money, Credit
and Banking 22 (August 1990): 357-369.
Burton, F. and H. Inoue (1985), An Appraisal of the Early Warning Indicators of Sovereign Loan Default in Country Risk
Evaluation System, Management International Review 25: 45-56.
Burton, F.N (1987), A country Risk Appraisal Model of Foreign Asset Expropriation in Developing Countries, Applied Economics
19(8): 1009-1048.

Volume 5, Number 1 • April 2005 29

Citron, J T and G Nicklesburg(1987), Country Risk and Political Instability, Journal of Development Economics 25 (April): 385-392.
Cosset, J.C. and J. Roy (1991), The Determinants of Country Risk Ratings, Journal of International Business Studies 22 (First Quarter):
135-142.
Cosset, J., Daouas, M., Kettani, O. and Oral, M. (1993), Replicating Country Risk Rating, Journal of Multinational Financial Management
Vol. 3.
Eaton, J. and M. Gersovitz, and J. Stiglitz (1986), The Pure Theory of Country Risk, European Economic Review 30: 481-513.
Erbs, C.B., Campbell, R.H. and Viskanta, T.E (1996), Political Risk, Economic Risk and Financial Risk, Financial Analysts Journal
52 (Nov./Dec): 28-46.
Feder, G. and R. Just (1977), A Study of Debt Servicing Capacity Applying Logit Analysis, Journal of Development Economics 4: 25-
38.
Feder, G.R. Just and K. Ross (1981), Projecting Debt-Servicing Capacity of Developing Countries, Journal of Financial and
Quantitative Analysis 16 (December): 651-669.
Feder, G. and K. Ross (1982), Risk Assessments and Risk Premiums in the Euro-Dollar Market, The Journal of Finance 37 (June):
670-691.
Feder, G. and L. Uy (1985), The Determinants of International Creditworthiness and their Policy Implications, Journal of Policy
Modeling 7: 133-156.
Frank, C. and W. Cline (1971), Measurement of Debt Servicing Capacity: An Application of Discriminant Analysis, Journal of
International Economics 1: 327-344.
Kaminsky, G. and Schmukler, S (2001), Emerging Markets Instability: Do Sovereign Ratings Affect Country Risk and Stock Returns ?, World
Bank, February 28, 2001, Accessed Online.
Oetzel, Jennifer M., Bettis, Richard A. (2001), Country Risk Measures: How risky are They ?, Journal of World Business 36(2).
Saini, K.G. and P.S. Bates (1984), A Survey of the Quantitative Approaches to Country Risk Analysis, Journal of Banking and Finance
8: 341-356.
Various Issues of Euromoney.
World Development Indicators, 2002.

30 Journal of Management Research

 APPENDIX
List of 61 Countries

Group Countries

East Asia and the Pacific China, Indonesia, Malaysia, Papua New Guinea, Philippines, Singapore,
Thailand, Vietnam

South Asia Bangladesh, India, Nepal, Pakistan, Sri Lanka

Latin America and Caribbean Argentina, Brazil, Chile, Colombia, Mexico

Europe & Central Asia Bulgaria, Hungary, Poland, Romania, Russian Federation, Turkey,
Uzbekistan, Yugoslavia

Middle East & North Africa Algeria, Egypt, Iran, Iraq, Jordan, Libya, Oman, Saudi Arabia, Tunisia

Sub-Saharan Africa Ethiopia, Ghana, Kenya, Mauritius, Nigeria, South Africa, Sudan,
Tanzania

High Income Australia, Austria, Canada, Hong Kong, (China) Denmark, France,
Germany, Ireland, Italy, Japan, Netherlands, New Zealand, Norway,
Portugal Sweden, Switzerland, United Kingdom, United States

Volume 5, Number 1 • April 2005 31