Next Article in Journal
Analysis of the Competitiveness, Complementarity, and Trade Combination of Kazakhstan and China in the Oil and Gas Trade
Previous Article in Journal
Granular Cities
Previous Article in Special Issue
The Tale of Two Economies: Inflationary Dynamics in the Euro Area and the US in the Context of Uncertainty
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Economies
Volume 12
Issue 7
10.3390/economies12070181
economies-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Economic and Political Determinants of Sovereign Default and IMF Credit Use: A Robustness Assessment Post 2010

by
Lina Maddah
,
Hassan Sherry
and
Hussein Zeaiter
*
Department of Economics, Lebanese American University, Beirut P.O. Box 13-5053, Lebanon
*
Author to whom correspondence should be addressed.
Economies 2024, 12(7), 181; https://doi.org/10.3390/economies12070181
Submission received: 9 May 2024 / Revised: 15 June 2024 / Accepted: 25 June 2024 / Published: 9 July 2024
(This article belongs to the Special Issue The Political Economy of Money)
Versions Notes

Abstract

:
According to the IMF, the current public debt makes up nearly 40 percent of the global debt, marking the highest share since the mid-1960s. Despite the vast research on alarming levels of sovereign default, the literature remains inconclusive. This paper investigates macroeconomic, financial, and political determinants of IMF credit use in the post-2010 era. The main contribution of our study lies in its temporal analysis as we investigate how the robustness of different factors has evolved. By utilizing an extensive dataset on 216 countries over the period of 2010–2021 and employing a variant of the Extreme Bounds Analysis (EBA) method, our study reveals that fluctuations in the IMF credit to external debt ratio can be attributed to changes in the total reserves to external debt ratio, where this relationship is statistically significant and reliable. However, high political risks seem to no longer affect the IMF’s decision, post 2010. Furthermore, our findings demonstrate that excluding countries with low debt arrears strengthens the results’ robustness. These findings contribute to a better understanding of the complexities surrounding IMF credit use in the contemporary global economic scene and offer new standpoints on the Fund’s lending choices.
Keywords:
sovereign debt default; extreme bound analysis; IMF; panel
JEL Classification:
F34; H63

1. Introduction

Rising indebtedness within a nation often precipitates financial instability, constraining resources available for investment purposes. In instances where governments face a balance of payments (BoP) crisis as well as difficulties in securing new loans from commercial creditors, recourse is frequently sought from the International Monetary Fund (IMF). This is because the IMF is recognized as an important source of foreign financing, which can significantly impact GDP and the Human Development Index (HDI) by enhancing the availability of financial resources that can be used for investment in vital sectors such as industry, services, and agriculture (Das 2023). Additionally, the IMF is regarded as a lender of last resort, particularly for countries facing financial crises or on the brink of such crises (IMF n.d.).
While the idea is shared that the primary objective of IMF credit is to restore economic stability by providing the necessary funds to restore BoP equilibrium, a substantial body of the literature suggests that reliance on IMF borrowing may carry adverse repercussions such as economic contraction, higher unemployment, and social unrest. This is attributed to the increase in a country’s debt burden, which exacerbates concerns about debt sustainability (El Ghonemy 1998; El-Said and Harrigan 2014). The repercussions associated with using IMF credit have prompted a scholarly1 interest in further investigating the role of IMF credit. Such an investigation aims to address existing gaps in the literature and provide insights into policymaking processes.
This study employs a comprehensive and temporal methodology, encompassing economic, financial, and political dimensions. Its focus is on the post-2010 period. Employing the Extreme Bounds Analysis (EBA) approach, the study examines the primary determinants of sovereign default, while acknowledging the inherent limitations associated with this analytical approach. This analytical framework is particularly attuned to concerns regarding the reliability and consistency of estimated correlations. More specifically, the analysis involves examining the robustness and credibility of prevailing empirical findings, as well as correcting for omitted variable bias. Moreover, the study incorporates a comprehensive dataset comprising annual observations spanning from 2010 to 2021, across 216 countries.
The findings in this study are instructive for policy; they hold relevance for policymakers and international financial institutions as they provide a better understanding of the implications of influencing decisions related to IMF loans. This can effectively offer valuable insights for the effective management of future debt crises. Section 2 presents a historical account of the changing landscape of IMF lending decisions, followed by a tabular presentation of the explanatory variables covered in an extensive list of preceding studies. Section 3 explains the data and methodology. Section 4 presents the findings, and Section 5 concludes.

2. Literature Review

The extant literature on IMF lending has significantly evolved over time, with various studies examining its relationship with debt sustainability or the likelihood of sovereign debt default. For instance, Jorra (2012) investigates the impact of IMF lending on the chances of sovereign default. The study finds that although IMF programs may offer short-term relief, they typically do not lower the long-term risk of default. Similarly, Goldstein (2003) explores debt sustainability and overall fiscal health in Brazil in the context of IMF lending. The study shows that IMF programs often fail to ensure long-term debt sustainability (https://euc-word-edit.officeapps.live.com/we/wordeditorframe.aspx?ui=en-US&rs=en-US&wopisrc=https%3A%2F%2Flauedu74602-my.sharepoint.com%2Fpersonal%2Fhassan_sherry_lau_edu_lb%2F_vti_bin%2Fwopi.ashx%2Ffiles%2Fdc1a4b417fa3452383f0959382a5ef24&wdenableroaming=1&mscc=1&wdodb=1&hid=E41630A1-5049-8000-FED0-AA640B05C0C5.0&uih=sharepointcom&wdlcid=en-US&jsapi=1&jsapiver=v2&corrid=766644dc-9451-14f1-20c0-f1833d237ce9&usid=766644dc-9451-14f1-20c0-f1833d237ce9&newsession=1&sftc=1&uihit=docaspx&muv=1&cac=1&sams=1&mtf=1&sfp=1&sdp=1&hch=1&hwfh=1&dchat=1&sc=%7B%22pmo%22%3A%22https%3A%2F%2Flauedu74602-my.sharepoint.com%22%2C%22pmshare%22%3Atrue%7D&ctp=LeastProtected&rct=Normal&wdorigin=Sharing.DirectLink.Copy&instantedit=1&wopicomplete=1&wdredirectionreason=Unified_SingleFlush#_ftn1 accessed on 11 April 2024), despite its success in short-term stabilization. On the other hand, Shadlen (2006) provides a historical account on the IMF’s role in managing debt crises in Latin America over three decades. The study highlights an unchanging pattern in the IMF’s approach to macroeconomic stabilization, namely the imposition of austerity measures, which offers mixed results in terms of resolving long-term debt problems.
Despite the large corpus of research on IMF lending, several gaps remain in the literature. One area which merits further investigation is the long-term impact of IMF programs on economic growth and development. While short-term stabilization is often achieved, the longer-term impact on debt sustainability, poverty reduction, and social welfare is less examined.
Moreover, many studies, such as Abosedra and Fakih (2017), Jorra (2012) and Goldstein (2003), focus on specific countries or regions, leaving a gap in comparative analyses that could provide more generalizable results. Last but not least, as outlined in papers by Shadlen (2006) and Guzman and Heymann (2016), the literature on IMF lending and its relationship with sovereign default requires further examination of the political economy of IMF lending—an area that could provide valuable insights for policymakers and scholars as well as allow for more effective IMF policies and interventions.
This study aims to contribute to bridging these gaps in the literature by providing insights into these underexplored areas.
Guzman and Heymann (2016) (https://euc-word-edit.officeapps.live.com/we/wordeditorframe.aspx?ui=en-US&rs=en-US&wopisrc=https%3A%2F%2Flauedu74602-my.sharepoint.com%2Fpersonal%2Fhassan_sherry_lau_edu_lb%2F_vti_bin%2Fwopi.ashx%2Ffiles%2Fdc1a4b417fa3452383f0959382a5ef24&wdenableroaming=1&mscc=1&wdodb=1&hid=E41630A1-5049-8000-FED0-AA640B05C0C5.0&uih=sharepointcom&wdlcid=en-US&jsapi=1&jsapiver=v2&corrid=766644dc-9451-14f1-20c0-f1833d237ce9&usid=766644dc-9451-14f1-20c0-f1833d237ce9&newsession=1&sftc=1&uihit=docaspx&muv=1&cac=1&sams=1&mtf=1&sfp=1&sdp=1&hch=1&hwfh=1&dchat=1&sc=%7B%22pmo%22%3A%22https%3A%2F%2Flauedu74602-my.sharepoint.com%22%2C%22pmshare%22%3Atrue%7D&ctp=LeastProtected&rct=Normal&wdorigin=Sharing.DirectLink.Copy&instantedit=1&wopicomplete=1&wdredirectionreason=Unified_SingleFlush#_ftnref1 accessed on 11 April 2024) offer a critical evaluation of the IMF’s Debt Sustainability Analysis (DSA). They argue that the DSA often underestimates the risks associated with debt and overestimates the effectiveness of IMF policies.
The following segment of the literature review offers a historical context on the International Monetary Fund’s (IMF) practice of providing loans to countries in debt. It investigates the origins, development, and consequences of this policy in relation to sovereign debt restructurings and crisis management. The segment after next, namely Table 1, provides a comprehensive overview of the existing body of research pertaining to the documented impact of different explanatory factors on the occurrence of debt default.

2.1. The IMF’s Policy of Lending: A Historical Perspective

The global financial crisis of 2007–2008, which had put at risk the stability of the global financial system, provoked governments to dispense large public guarantees and generous fiscal stimuli. What followed was a problem of looming debt crises, especially for developing countries. As Jorra (2012) and Dibeh et al. (2018) argue, politicians in advanced economies have turned to the International Monetary Fund (IMF) in an attempt to remedy the impending problem of global debt crises.
Nonetheless, the evolution of the role of the IMF as crisis manager, and, more specifically, as a lender to countries in default precedes the Great Recession. In fact, it has to be sought in the emergence of an uneven bargaining power between private international creditors, namely commercial banks, and the IMF, towards the end of the 1980s. As Díaz-Cassou et al. (2008) argue, the IMF had adhered to a policy of not lending to countries in default before 1989, meaning that any potential lending arrangement from the Fund to a member country was conditional on the country’s elimination of debt to private creditors during the period of the program. Such a long-standing policy was reversed in 1989 when the Fund started to lend to countries in default. Nonetheless, the seeds of transformation in the Fund’s policy to lend countries with accumulated debt to commercial creditors have sprung in the aftermath of the first oil crisis in the early 1970s.

2.1.1. The First Oil Crisis of 1973

The fourfold increase in the price of oil in 1973–74 allowed oil exporters to accumulate current account surpluses, which have driven multilateral institutions such as the IMF and the OECD, as well as governments of oil-importing countries, to explore optimal ways to ‘recycle’ those surpluses (Boughton 2000). What followed was a textbook example of financial intermediation, whereby the surplus economies—that is, the oil exporters—turned to advanced economies’ banking systems to invest the bulk of these surpluses, which in turn were partially on-lent to the oil-importing emerging economies. As indicated in Boughton (2000) and Herrera et al. (2019), between 1973 and 1978, international bank lending increased threefold, especially since, on the one hand, the loans came with no strings attached and, on the other hand, real interest rates during the said period were negative. As for developing country debt, it soared from USD 130 billion in 1973 to around USD 612 billion in 1982 (IMF 1984).

2.1.2. The World Debt Crisis of the 1980s

The early 1980s witnessed the onset of a global debt crisis, which weighed heavily on developing countries. The crisis was the outcome of a rapid and disproportionate increase in external debt relative to those countries’ debt service capacities, triggered to a large degree by their need to finance the deficits on the current accounts of their balance of payments—especially since the start of the oil crisis of 1973–1974—coupled with unfettered lending2 by international banks (Stambuli 1998). Their internal fiscal positions and balance of payments were further worsened in the wake of the collapse of commodity prices, rising market interest rates, and the ensuing appreciation of the US dollar. Consequently, many developing countries resorted to rolling over their debts or obtaining loans to re-finance maturing debt (Stefanidis 2023).
At this time, international banks suddenly stopped lending or even rolling over existing debt to a large number of crisis-ridden emerging economies. As Stambuli (1998) points out, twenty-one countries had defaulted on their debt by 1983. This was a turning point for the IMF, let alone for the international financial system; the IMF took on the role of crisis manager. It would lend to a country in default granted that international banks, or creditors, would provide written assurances to the Fund that they would raise their exposure to this country. This practice was referred to as ‘concerted lending’, and it was carried out on a case-by-case basis (Boughton 2000). The rationale for the role of crisis management by the IMF in the early 1980s was as follows (ibid.): first, many developing countries with a large debt burden would easily default on their debt if denied additional financing, and a default would pose a grave threat not only to the economic health of the defaulting countries but also to the international financial system; second, more borrowing could prove to be a good strategy if carried out in conjunction with macroeconomic reforms; and third, both large-scale funding and economic reforms could not be advanced without multilateral cooperation.
For most of the decade, until 1989, the practice of ‘concerted lending’ was in congruent with the IMF’s policy of non-default to private creditors. In other words, any financial arrangement between the Fund and the indebted country would require the elimination of arrears by the latter and the non-accumulation of new arrears during the program period (Díaz-Cassou et al. 2008).

2.1.3. The Fund’s Departure in 1989: Lending to Countries in Default

The IMF’s policy of not lending to countries in default in the 1980s was capable of being maintained mainly because international creditors viewed their cooperation in funding IMF-supported programs to be in their own interest. By the end of the 1980s, however, private creditors demonstrated increasing unwillingness to negotiate with their sovereign borrowers and thereby give the requisite financing assurances that the Fund required in the context of the standard ‘concerted lending’ approach. Such reluctance could be attributed, in large part, to the development of a secondary market for banks’ claims and to the bolstering of commercial banks’ balance sheets (IMF 2022). An important implication was that IMF support for structural adjustment programs in developing countries had been impeded. This meant that the basis for the Fund’s policy of concerted lending had slowly been undermined because the Fund’s long-standing policy of not lending had inadvertently conferred excessive clout on creditors at the expense of debtors. As Díaz-Cassou et al. (2008, p. 11) point out, the policy of not lending to countries in default (vis vis-à-vis private creditors) has increasingly come to be regarded as a ‘de facto veto power assigned to commercial banks over the Fund’s lending decisions.’
In 1989, the IMF departed from the concerted lending approach and introduced the policy of lending to countries in default. This was regarded as an overt move to curtail the leverage of international private banks over their lending decisions. The Fund’s modified policy allowed it to lend to countries in default—and/or tolerate temporary arrears to private creditors—contingent on the existence of a straightforward process of negotiations between private creditors and the sovereign. Pursuant to the terms under its 1989 policy, the Fund would lend to countries in default under the following conditions (IMF 2022): first, an IMF-supported program of a member country was deemed crucial for the country’s implementation of reforms; second, both creditors and debtors had embarked on a process of negotiations around restructuring the latter’s debt; and third, any financing arrangement with the debtor would be consistent with its external viability, and approved within a sensible time frame.
This new policy of lending to countries in default has cleared the way for the IMF to become an active player in debt-crisis resolutions and, more precisely, sovereign debt restructurings. Since 1989, the global economy has faced a handful of debt- and financial crises that weighed heavily on individual economies. Nonetheless, most elements of the role of the IMF as crisis manager have already been in place, and the countries3 in default—namely those that have not been able to navigate crises internally or via bilateral support—would swiftly resort to the Fund in an attempt to resolve their problems (Wells 1993; Boughton 2000; Haidar 2011, 2012).

2.2. Explanatory Variables in Existing Literature

While sound literature includes models with varying sets of explanatory variables, there is an ongoing debate regarding which variables are crucial determinants of default risk. For instance, many studies have investigated the ‘real effective exchange rate’4 effect on debt default. The results are mixed with studies like Sargen (1977) and Solberg (1988) finding a positive effect; Ghulam and Saunby (2023) and Hajivassiliou (1994) finding a negative effect; and Edwards (1984) and Al Fawwaz (2016) finding an insignificant effect. Also, the variable ‘reserves/GDP’ has been studied by multiple authors such as Edwards (1984), Taffler and Abassi (1984), and Ghulam and Saunby (2023), among others. All these studies observed that the effect of ‘reserves/GDP’ on debt default is insignificant. In contrast, the variable ‘Total external debt/GDP (GNP)’ has studies that observed a positive, negative, and insignificant effect on debt default. Edwards (1984), Rivoli and Brewer (1997), and Taffler and Abassi (1984) are some authors who contributed to these observations. These studies underscore the fact that there is no agreement in the academic literature regarding the significance and magnitude of the impact of explanatory variables as key determinants of default.

3. Data and Methodology

Annual observations on 216 countries are included in this study over the period of 2010–2021.5 Depending on the data that are available for the variables used in specific regressions, the sample size differs among various EBA regressions. The political indices were obtained from ICRG tables of the PRS group, and the rest of the data were collected from the World Bank database. Table 2 describes the variables employed in this investigation. Political indices include the elements of corruption, political stability, democratic accountability, military involvement in politics, and ethnic tensions; the risk levels go from 0 (highest risk) through 6, 6, 12, 6, and 6, respectively.

3.1. Extreme Bound Analysis

In our study, we employ Extreme Bounds Analysis (EBA), a comprehensive sensitivity analysis technique, to assess the robustness of empirical findings and examine the sensitivity of results to potential sources of bias or omitted variables. In the context of sovereign debt, EBA helps identify the key determinants of defaults while accounting for the uncertainties and complexities inherent in the analysis. It provides a way to address concerns regarding the robustness and reliability of the estimated relationships. Many of the earlier analyses of sovereign default fall short of offering a complete and cohesive understanding of which variables should be included in the model as consistent determinants of sovereign risk. Because sovereign default rates vary across nations and across time, no particular independent variable, within prior studies, has been identified as a certain significant determinant with static influence. In the literature on sovereign debt, no single variable is employed in all the earlier studies, and no single variable produces consistent outcomes and indicators.
The early inspiration for the EBA method came from Cooley and LeRoy (1981) who argued that economic theory typically fails to provide a full description of which variables are to be held unchanged when statistical examinations are carried out. Thus, the EBA evaluates the robustness of the estimates’ coefficients, and its findings can provide convincing justifications. Leamer (1983) conceptualized the EBA technique, which was then utilized by Levine and Renelt (1992) to investigate the factors influencing cross-country economic growth. Soudis (2016) applied the EBA to approximately 30 factors proposed by the literature to assess their robustness in shaping sovereign bond ratings. The EBA is also applied as an important tool to test the robustness of conventional growth regression coefficients to changes in the set of conditioning variables (Haan 2000; Minier 2005). Additionally, Hwang and Wang (2004) applied it to examine the robustness of trade to the determination of productivity growth. Its primary purpose is to offer robustness and sensitivity analyses of the explanatory variables in an economic regression.
Our panel data set includes annual observations on 216 countries from 2010 through 2020. The sample size varies across the regressions run for the EBA depending on the availability of data on the specific variables included in a particular regression. Data sources include The World Bank and The International Country Risk Guide (ICRG) (n.d.) by the PRS Group. Careful attention is given to data quality, consistency, and comparability across countries and periods. The model specification is guided by economic theory, previous literature, and statistical considerations to identify the key factors affecting sovereign debt (see Table 1 in the literature).
To conduct the analysis, a comprehensive dataset is thus compiled, consisting of relevant macroeconomic, financial, political, and institutional variables that are expected to influence sovereign debt outcomes by countries borrowing from the IMF. All variables that have been identified in the literature as potential determinants of default risk are considered, and we categorize them as follows:
Macroeconomic indicators: inflation based on consumer price annual growth (Inf), GDP annual growth (GGDP), exports of goods and services annual growth (GExp), net barter terms of trade growth (ToT), standard deviation of the previous seven years of GNP (VGNP), external debt stocks of GNP (ED/GNP), total reserves in months of imports (Res/Imp), exports + imports to GDP ratio (Openness), total debt service of exports of goods (TDS/Exp), imports of goods and services of GDP (Imp/GDP), exports of goods and services of GDP (Exp/GDP), external debt stocks of exports of goods (ED/Exp), GNP per capita PPP current international currency (PCGNP), real effective exchange rate index (CPI 2000) (ExRate), current account balance of GDP (Capflow), fiscal balance to GDP (FisBal).
Financial indicators: reserves to external debt (Res/ED), interest and principal arrears to external debt (Arrears), rescheduled debt to external debt (ResDebt/ED), principal arrears to external debt (PrinArrears/ED), money and quasi money M2 to total reserves (Money), real lending interest rate (Intrate), long term debt service to reserves (LTServ/Res), long term debt service to GDP (LTServ/GDP), domestic credit to the private sector of GDP (Crdt/GDP), short term debt of total reserves (STD/Res), principal repayments to exports (PR/exp), total debt service of GNP (DebtServ), and central government debt total of GDP (GovDebt).
Institutional and political variables: the political risk variables are derived from ICRG tables (www.prsgroup.com) accessed on 10 November 2023; every index ranges from 0 to 6, Corruption (Corr), military involvement in politics (MilPol), ethnic tensions (EthTen), government stability (GovStab), and democratic accountability (Democ).
Where the value zero indicates high risk, these variables are expected to have a negative effect on IMF credit. A higher value of the index indicates increased political risk in the country, which increases the probability of default.

3.2. Sensitivity Analysis

The baseline econometric model is specified to estimate the relationship between the sovereign debt default and the selected explanatory variables. In the EBA, this model is systematically modified by introducing alternative specifications, such as different model structures, or variable inclusion/exclusion. Each modified model is iteratively estimated to assess the sensitivity to different modeling choices. This process identifies the robustness of the empirical findings and sheds light on the key factors driving credit from the IMF.
The general form of the baseline model can be expressed as
W = β 0 + β 1 X + β 2 Q + β 3 Z + ε
where
W is the dependent variable (IMFcredit/ED), the use of the IMF credit to the external debt, representing the phenomenon under investigation (sovereign debt default). β 0 , β 1 , β 2 , β 3 are the coefficients that measure the relationship between the dependent variables and the corresponding explanatory variables. X is the variable always included upfront in the regression due to its theoretical and empirical relevance (also called a free variable), Q is the focus variable or variable of primary interest, Z is a vector of doubtful variables (financial, macroeconomic, and political) that can potentially be correlated with W (based on existing literature), and ε represents the error term capturing unobserved factors or random variation that cannot be explained by the explanatory variables (Table 2).6
By iteratively estimating different model specifications, EBA allows for assessing the sensitivity of the estimated coefficients to changes in the model. Following Leamer (1983), Levine and Renelt (1992)7, and Chakrabarti and Zeaiter (2014), the empirical exercise of the EBA involves an order of steps as follows. First, the above model is estimated for each Q-variable, initially including X, the free variable, and Q, the focus variable. Second, the model is re-estimated N times, after adding all possible subset combinations of Z, doubtful variables, to produce the extreme upper and lower bounds. The extreme upper bound (EUB) represents the maximum value of the estimated coefficient, β 2 , of the Q-variable plus two standard deviations. The extreme lower bound (ELB) represents the minimum value of the estimated coefficient, β 2 , of the focus variable, Q, minus two standard deviations. For the Q-variable to be robust, β 2 should remain significant at the 90% level of confidence, and the coefficients of β 2 must hold the same sign for the lower and upper bounds in all regressions. This process helps identify the robustness of the empirical findings and provides insights into the factors that are consistently associated with W across different model specifications. The significance of the estimated coefficients and their stability across different model specifications are key considerations in interpreting the results of EBA. Robust findings, where the estimated coefficients remain consistently significant and exhibit similar magnitudes across various model specifications, provide stronger evidence for the relationships between the Q-variable and the W-variable. Otherwise, Q is judged to be fragile if it cannot withstand any changes in the Z-variable subset.8

4. Results

We first conducted an Extreme Bounds Analysis on the primary equation, utilizing all the variables delineated in Table 1. Initially, a regression analysis was executed, where the IMF credit to external debt ratio (IMFcredit/ED) was regressed against the ratio of total reserves to external debt (Res/ED). The t-statistics are presented in parentheses.
IMFcredit/ED = 0.081 + 0.014 Res/ED ***
(19.09)   (79.73)
(n = 2592; R2 = 0.99; F = 63.4)
According to the aforementioned results, approximately 99% of the fluctuations in the IMF credit to external debt ratio can be attributed to changes in the total reserves to external debt ratio. This finding is consistent with several studies listed in Table 1 (previous studies), which also indicate a significant positive coefficient at the 1% level.
As depicted in Table 3, the Extreme Bounds Analysis (EBA) for the independent variable Res/ED reveals that the coefficient is not only positive but also robust. Specifically, the coefficients of Res/ED at the upper and lower bounds are 0.0106 and 0.0533, respectively, with corresponding t-statistics of 11.45 and 6.53. According to Chakrabarti and Zeaiter (2014), Zeaiter and El-Khalil (2016), and Bitar et al. (2018), this robustness confirms that “there is enough independent variation of the Res/ED that explains variations in the dependent variable”. The variance inflation factor (VIF) determines whether the variable of interest may be thought of as a linear combination of Z-variables. 1/VIF is also utilized in numerous research to assess the degree of collinearity. Many studies suggest that a VIF greater than 10 indicates a substantial correlation between the independent variables. A VIF between 1 and 5 indicates no connection, whereas a VIF between 5 and 10 indicates considerable correlation. The X-variable’s VIF reveals no meaningful link with the other Z-variables.
Table 4 lists the EBA tests for the Q-variables: MIN and MAX in the second column denote the extreme lower bound, ELB and extreme upper bound, EUB, respectively. It is vital to notice that the same sample of Z-variables is utilized across all regressions in each test. As a result, the results of these tests are dependent on the combination of the Z-variables and the Q-variables, rather than on the sample size.
Openness, Imp/GDP, and PCGNP exhibit robustness, indicating that variations in these Q-variables consistently explain a significant portion of the variance in the use of IMF credit.
More specifically, the base coefficient for Imp/GDP is 22.75, with a t-value of 4.97 and a p-value of 0.00, which falls within the 95% confidence interval of 13.77 to 31.73. This suggests that an increase in Imp/GDP by one unit increases the IMFcredit/ED by 22.75 units, holding all other variables constant. This is consistent with the economic theory that more open economies and those with higher imports relative to GDP (relatively poorer countries with scarce natural resources) may require more external financing, including IMF credit, to manage their balance of payments and external debt. This is in line with the argument that “financial development is an equilibrium outcome that depends strongly on a country’s trade patterns” (Do and Levchenko 2007).
Conversely, ExRate, ED/GDP, Money, Exp/GDP, ED/Exp, LTServ/Res, STD/Res, Res/Imp, Inf, IntRate, CA/GDP, Crdt/GDP, GGDP, CapFlow, ToT, Gexp, MilExpn, VGN, TDS/Exp, LTServ/GDP, DebtServ/GNP, GovDebt/GDP, FisBal, LTServ/GDP, and PR/Exp do not demonstrate robustness, despite many of these variables showcasing predictive value in numerous studies documented in the literature.
Numerous studies have underscored the significance of political factors as crucial determinants of sovereign default, including works by Maddah et al. (2023), Kohlscheen (2010), Snider (1990), Feder and Uy (1985), McFadden et al. (1985), Berg and Sachs (1988), Thacker (1999), and Celasun and Harms (2010). These studies utilized political risk indices such as corruption, government stability, democratic accountability, military involvement in politics, and ethnic tensions. The present study also employs the same political indices, with the inclusion of identical Z-variables in the dataset. As shown in Table 5, the EBA tests on Corr, Democ, MilPol, GovStab, and EthTen are not robust. This confirms that in the post-2010 period, countries having high political risks did not significantly impact the IMF’s decision to use its credits. One can argue that post 2010, the IMF has been adopting a selective financing strategy that solely relies on debt risk indicators (accumulated debt arrears) without any consideration for political risks. Policymakers could interpret these findings as an indication that the IMF may need to reassess its lending policies.
Testimony to this is that in the aftermath of the Arab Uprisings, the IMF has still lent money to Arab countries that witnessed political instability (Sherry 2017). Since the uprisings, and as of 30 April 2017, the IMF’s financial arrangements with Arab countries totaled USD 57.43 billion, of which USD 42 billion was in 2016 alone (IMF 2017).
Moreover, all the political risk indices are not shown to have significant impact after they are added to the main dataset (see Table 5). On the other hand, the Z-variables used in this paper include all the Q-variables in addition to other less important variables and used in the literature. A number of z-variables are used more than once due to their inclusion in the literature, which creates multicollinearity (see correlation matrix, Table A1).
This is why we dropped the variables that show high correlation with other variables, PrinArrears/ED, PR/Exp, Imp/GDP, Exrate, ED/Exp, Exp/GDP, and DebtServ/GNP.
As shown in Table 6, ED/GNP, Res/Imp, TDS/Exp, PC/GNP, Openness, and CA/GDP are robust. On the other hand, Inf, Money, IntRate, GGDP and the political indices are not robust. Relative to the previous EBA results (Table 4 and Table 5), the results after correcting for multicollinearity show that the number of variables that are robust is six compared to three. This proves that excluding the variables that are highly correlated with each other gives more significance power. In addition, all the variables have VIF values less than 5. By removing highly correlated variables, the VIF values improves significantly (see Table 6).
Several studies have deemed arrears to be significant if the ratio of principal arrears to external debt exceeds 0.1 percent, and the ratio of total arrears to external debt is over 1 percent, as noted by Hajivassiliou (1989, 1994) and McFadden et al. (1985). Table A2 showed that 10 variables are robust compared to 3 in the previous section when all data are included. These results are in line with the aforementioned studies that small amounts of arrears do not seriously threaten the economy. Moreover, it increases the robustness of the variables. Compared to the previous sections discussed, the EBA models demonstrate improved robustness and significance when countries with a low number of principal arrears are omitted from the analysis.
Using total arrears to external debt above 5% and 1% as dependent variables in this section shows more Q-variables are robust (see Appendix A). This verifies the importance of the dependent variable that is used in this study, which, as explained in the previous section, is the use of the IMF credit. This result aligns with Zeaiter (2015a) who finds that debt arrears precede rescheduling agreements and can be considered a signal for sovereign defaults. On the other hand, numerous studies cited in the literature have employed a binary measure for sovereign default, where a value of 1 signifies receiving credit from the IMF. Even though this model shows better results (14 robust variables in Table A5), according to Zeaiter (2015b), “the use of a discrete measure of the dependent variable presents only a resort when a continuous measure of the dependent variable is absent since conversion of qualitative information to cardinal measures is necessary for any estimation. Using a binary dependent variable does not allow comparison in terms of the ‘degree’ of solvency and whether the default itself is complete or partial”.
Data from the World Bank corroborate this trend. An analysis of IMF credit in relation to Syria’s external debt reveals a progressive escalation from 2010 to 2021. Initially constituting 10% of the external debt in 2010, this proportion soared to a peak of 22% in 2021. This trajectory suggests a concurrent increase in both Syria’s arrears and the proportion of IMF credit to its total external debt. In a parallel examination, Yemen’s IMF credit, expressed as a percentage of its external debt, demonstrates notable fluctuations over the same period. Starting at 7% in 2010, there was a marginal decline in subsequent years, followed by an increase to 16% by 2021.
Finally, we test the appropriateness of the random effects estimator by including all the variables in one equation without distinguishing between X-, Q-, and Z-variables. As shown in Table A6, Model 1 includes all the variables of the original model, excluding the political risk indices. Several variables in fixed and random effects models do not show significant effects. According to the Hausman test, the null hypotheses cannot be accepted and shows that the fixed effect coefficients are consistent. In Model 2, all the political risk indices are added to the variables listed in Model 1. In this model, the random effect coefficients are consistent. In Model 3, variables that were omitted from Table 6 because of multicollinearity concerns have been excluded from the analysis. Thus, the model shows that the fixed effect model is consistent. Comparing Models 1, 2, and 3, variables in Model 3 show more significant and predictive power.

5. Concluding Remarks and Policy Implication

In summary, this study comprehensively analyzes the macroeconomic, financial, and political factors that contribute to sovereign default and the use of IMF financing post 2010. It particularly investigates the lending policy in situations when a country has outstanding debt payments. The study uncovers a noteworthy change in the factors and mechanisms influencing sovereign defaults and the use of IMF credit, which have not been sufficiently examined in the current body of scholarly research, using the EBA methodology to contribute to a better understanding of the processes that regulate global financial markets. The primary focus of this study is to test the resilience and importance of several economic and political factors that influence the use of IMF credit, specifically within the framework of the IMF’s policy of extending loans to countries with sovereign debts. The novelty of this study not only lies in its examination of the robustness of economic drivers of IMF loan use but on applying the EBA, which allows to include all the explanatory variables that have been used in the sovereign debt literature.
Also, the present analysis not only validates the resilience of certain economic indicators (e.g., the ratio of total reserves to foreign debt) as influential factors in the use of IMF loans but also importantly reveals those political concerns in the period after 2010. Our main findings suggest that the relationship between the total reserves to external debt ratio and the use of IMF loans in foreign debt are statistically significant and reliable. Furthermore, it is seen that political considerations, which have been recognized as crucial predictors of sovereign default in prior research, do not exhibit a substantial influence on the IMF’s choice to use its credit after the year 2010. It may be inferred that since 2010, the IMF has implemented a discerning approach to lending that only depends on debt risk indicators, namely cumulative debt arrears while disregarding political concerns.
Furthermore, the findings of this analysis demonstrate that by excluding countries with low debt arrears, the robustness of the results is strengthened. This offers new perspectives on the fund’s lending choices. The tendency to lend to countries with higher accumulated arrears may signal a breakup of the fund with its past practices of not lending to these countries. Still, this might serve as a window of opportunity for a more proactive role played by the IMF in devising debt relief strategies for developing countries based on economic and social reforms.
Policymakers may interpret these results as suggesting the need for the IMF to reevaluate its lending policies. It may consider investigating the creation of financial instruments that are specifically designed to meet the requirements of nations with varying degrees of outstanding debt. The inclination to grant loans to nations with more outstanding debts necessitates a more sophisticated strategy to guarantee the efficient allocation of cash and the successful provision of assistance to countries that would use it optimally for the purposes of economic stability and development. The IMF might consider implementing early intervention measures in countries with rising debt arrears to prevent them from escalating to levels where they significantly affect the economy’s stability. Countries that have avoided high arrears and successfully managed to prevent significant debt repayment defaults may be implementing effective economic policies and undertaking structural changes. The IMF could help these nations by acknowledging and strengthening their successful strategies, therefore acting as exemplars for other countries. Overall, credit use may require a more informed and proactive approach to ensure that funds are allocated effectively at an early stage and that assistance is provided to countries that will utilize it most successfully for economic stabilization and growth.
Finally, our study employs the Extreme Bounds Analysis (EBA) approach, which identifies the most important variables in a regression model by estimating the extreme bounds of the regression coefficients. The EBA approach does not differentiate between short- and long-term variables, as it is designed to identify the most robust variables regardless of time period. Future research could explore this distinction using alternative methodologies, such as a Panel Vector Error Correction Model (VECM), to analyze short-run dynamics and long-run equilibrium. Also, future research can address regional differences that can be attributed to various factors such as economic, political, and financial conditions. For instance, countries in Africa and Asia may have different economic structures, trade patterns, and institutional frameworks that influence their demand for IMF credit use.

Author Contributions

Conceptualization, L.M. and H.Z.; methodology, L.M. and H.S.; software, H.Z.; validation, L.M., H.S. and H.Z.; formal analysis, H.S.; investigation, L.M., H.S., and H.Z.; resources, L.M. and H.S..; data curation, H.Z..; writing—original draft preparation, L.M., H.S., and H.Z.; writing—review and editing, L.M., H.S., and H.Z.; visualization, L.M. and H.S.; supervision, H.Z.; project administration, H.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Conflicts of Interest

The authors declare no conflict of interest.

Appendix A

List of Countries:
AfghanistanDjiboutiLatviaSamoa
AlbaniaDoMinicaLebanonSan Marino
AlgeriaDoMinican RepublicLesothoSao Tome and Principe
American SamoaEcuadorLiberiaSaudi Arabia
AndorraEgypt, Arab Rep.LibyaSenegal
AngolaEl SalvadorLiechtensteinSerbia
Antigua and BarbudaEquatorial GuineaLithuaniaSeychelles
ArgentinaEritreaLuxembourgSierra Leone
ArmeniaEstoniaMacao SAR, ChinaSingapore
ArubaEswatiniMadagascarSint Maarten (Dutch part)
AustraliaEthiopiaMalawiSlovak Republic
AustriaFaroe IslandsMalaysiaSlovenia
AzerbaijanFijiMaldivesSolomon Islands
Bahamas, TheFinlandMaliSomalia
BahrainFranceMaltaSouth Africa
BangladeshFrench PolynesiaMarshall IslandsSouth Sudan
BarbadosGabonMauritaniaSpain
BelarusGambia, TheMauritiusSri Lanka
BelgiumGeorgiaMexicoSt. Kitts and Nevis
BelizeGermanyMicronesia, Fed. Sts.St. Lucia
BeninGhanaMoldovaSt. Martin (French part)
BermudaGibraltarMonacoSt. Vincent and the Grenadines
BhutanGreeceMongoliaSudan
BoliviaGreenlandMontenegroSuriname
Bosnia and HerzegovinaGrenadaMoroccoSweden
BotswanaGuamMozambiqueSwitzerland
BrazilGuatemalaMyanmarSyrian Arab Republic
British Virgin IslandsGuineanamibiaTajikistan
Brunei DarussalamGuinea-BissaunauruTanzania
BulgariaGuyanaNepalThailand
Burkina FasoHaitiNetherlandsTimor-Leste
BurundiHondurasNew CaledoniaTogo
Cabo VerdeHong Kong SAR, ChinaNew ZealandTonga
CambodiaHungaryNicaraguaTrinidad and Tobago
CameroonIcelandNigerTunisia
CanadaIndiaNigeriaTurkiye
Cayman IslandsIndonesiaNorth MacedoniaTurkmenistan
Central African RepublicIran, Islamic Rep.Northern Mariana IslandsTurks and Caicos Islands
ChadIraqNorwayTuvalu
Channel IslandsIrelandOmanUganda
ChileIsle of ManPakistanUkraine
ChinaItalyPalauUnited Arab Emirates
ColombiaJamaicaPanamaUnited Kingdom
ComorosJapanPapua New GuineaUnited States
Congo, Dem. Rep.JordanParaguayUruguay
Congo, Rep.KazakhstanPeruUzbekistan
Costa RicaKenyaPhilippinesVanuatu
Cote d’IvoireKiribatiPolandVenezuela, RB
CroatiaKorea, Dem. People’s Rep.PortugalVietnam
CubaKorea, Rep.Puerto RicoVirgin Islands (U.S.)
CuracaoKosovoQatarWest Bank and Gaza
CyprusKuwaitRomaniaYemen, Rep.
CzechiaKyrgyz RepublicRussian FederationZambia
DenmarkLao PDRRwandaZimbabwe
Summary Statistics:
VariableObservationsMeanStd. dev.MinMax
IMFcredit/ED14450.944647231.566120.002871200
Arrears14457.04346320.448120258.193
ExRate1123101.162224.1490653.7878741.702
ED/GNP142250.2854541.759631.15419429.738
Money 16974.96238210.845130.001312195.47
Exp/GDP 219843.242732.480090.459601340.303
PrinArears/ED14450.04683470.130764300.952
LTServ/GDP 14220.04550730.061745500.857
DebtServ14220.04688730.064155900.923
ED/Exp13301.4932161.1816760.000113
LTServ/Res12631.84 × 10¹⁰4.91 × 10¹¹4974.95 × 10¹¹
PR/exp13280.11510070.137567901.31
STD/Res12640.75061784.3296480122
Res/Imp 12635.1825023.6417430.06419735.0888
Oppeness21750.91419260.56282020.1215.27
Imp/GDP21750.49300160.284990.07542.62
Inf21735.53462520.50108−4.29487557.202
Intrate14898.5900965.69554056.5183
CA/GDP2175−2.57846814.35847−65.0289235.784
Crdt/GDP202254.6544744.586024.80 × 10⁻⁶267.934
GovDebt75161.079341.17149−1.17073252.523
PCGNP229919716.8320873.39590152630
GGDP24462.6846245.895908−54.235986.8268
Capflow 208124.613968.826149−3.9459279.4011
ToT 2209120.524644.3814142.8438458.574
GExp19894.73872331.73221−96.36441051.42
FisBal2198−6.00799718.53837−127.9759.8889
MilExpn17901.8638671.5204390.005415.4796
VGNP25224.18 × 10¹⁰1.74 × 10¹¹21000002.30 × 10¹²
TDS/Exp 136214.3888914.698760.003495133.177
Res/ED 126471.019692475.1640.0030488000
ResDebt/ED4351.2308345.339036052.1952
Corr 14172.7203351.1979760.3756
Democ 14174.2270351.49870906
Govtstab14177.2083921.0870374.45833311.5
MilPol14173.8380381.64912606
EthTen14173.9354561.19726716
Table A1. Pairwise correlation matrix.
Table A1. Pairwise correlation matrix.
CorrDemocGovStabMilPolEthTenArrearsResDebt/EDIMFcredit/EDPrinArrears/EDRes/EDLTServ/GDPDebtServED/GNPED/ExpTDS/ExpLTServ/GDPPR/expSTD/Res
Corr1.00
Democ0.571.00
0.00
GovStab0.07−0.231.00
0.010.00
MilPol0.700.600.021.00
0.000.000.53
EthTen0.310.070.120.391.00
0.000.010.000.00
Arrears−0.45−0.21−0.11−0.28−0.221.00
0.000.000.000.000.00
ResDebt/ED−0.060.000.05−0.12−0.040.231.00
0.360.990.410.050.510.00
IMFcredit/ED−0.020.03−0.02−0.12−0.13−0.010.181.00
0.580.350.550.000.000.760.00
PrinArrears/ED−0.45−0.21−0.10−0.29−0.230.96 ***0.22−0.011.00
0.000.000.000.000.000.000.000.76
Res/ED−0.01−0.04−0.02−0.04−0.03−0.01−0.010.99 ***−0.011.00
0.840.260.620.230.370.700.840.000.71
LTServ/GDP0.000.10−0.070.250.34−0.13−0.03−0.18−0.14−0.101.00
0.890.000.050.000.000.000.560.000.000.00
DebtServ0.000.09−0.060.250.34−0.13−0.03−0.18−0.14−0.100.99 ***1.00
0.940.010.090.000.000.000.580.000.000.000.00
ED/GNP−0.040.08−0.070.280.260.000.00−0.03−0.02−0.030.660.661.00
0.270.020.040.000.000.960.950.220.520.240.000.00
ED/Exp−0.100.10−0.090.090.070.100.00−0.040.10−0.040.340.340.721.00
0.000.000.010.010.050.001.000.190.000.210.000.000.00
TDS/Exp0.030.16−0.110.210.28−0.100.00−0.03−0.11−0.030.84 ***0.84 ***0.570.491.00
0.360.000.000.000.000.000.980.300.000.310.000.000.000.00
LTServ/GDP0.050.000.010.15−0.03−0.050.02−0.14−0.04−0.050.270.280.100.080.421.00
0.160.960.820.000.400.110.770.000.130.070.000.000.000.000.00
PR/exp0.010.12−0.110.190.27−0.090.01−0.21−0.09−0.110.89 ***0.88 ***0.510.420.91 ***0.371.00
0.840.000.000.000.000.000.800.000.000.000.000.000.000.000.000.00
STD/Res−0.26−0.16−0.09−0.15−0.060.45−0.020.000.430.000.000.000.020.070.020.150.031.00
0.000.000.010.000.090.000.740.870.000.870.880.870.460.020.450.000.36
Res/Imp0.110.11−0.010.06−0.02−0.18−0.030.01−0.170.01−0.02−0.02−0.11−0.070.01−0.020.01−0.15
0.000.000.770.130.640.000.600.740.000.730.590.520.000.020.790.590.620.00
Exp/GDP0.370.140.120.420.28−0.18−0.01−0.04−0.20−0.040.220.230.23−0.28−0.06−0.07−0.02−0.07
0.000.000.000.000.000.000.880.160.000.160.000.000.000.000.040.010.500.03
Oppeness0.320.130.100.380.27−0.12−0.030.06−0.120.060.240.250.33−0.15−0.04−0.180.01−0.09
0.000.000.000.000.000.000.540.030.000.040.000.000.000.000.150.000.780.00
Imp/GDP0.240.130.070.320.23−0.07−0.030.14−0.070.140.210.200.34−0.01−0.04−0.260.02−0.09
0.000.000.010.000.000.010.540.000.010.000.000.000.000.690.110.000.580.00
Inf−0.18−0.13−0.08−0.17−0.080.32−0.030.010.320.010.020.020.070.070.070.160.050.76
0.000.000.000.000.010.000.580.790.000.770.420.400.010.010.020.000.070.00
Intrate−0.30−0.070.00−0.25−0.150.150.110.000.140.00−0.10−0.10−0.030.05−0.06−0.07−0.04−0.06
0.000.050.900.000.000.000.050.970.000.900.000.000.280.120.060.040.180.09
ExRate−0.03−0.210.11−0.13−0.090.200.000.310.210.00−0.22−0.21−0.21−0.21−0.29−0.37−0.26−0.18
0.400.000.000.000.020.000.970.000.000.940.000.000.000.000.000.000.000.00
CA/GDP0.28−0.030.060.200.14−0.15−0.090.52−0.160.53−0.08−0.08−0.25−0.30−0.080.11−0.060.02
0.000.290.020.000.000.000.080.000.000.000.000.000.000.000.000.000.040.52
Crdt/GDP0.670.42−0.010.550.29−0.27−0.09−0.02−0.27−0.030.250.250.14−0.070.160.100.19−0.08
0.000.000.830.000.000.000.080.380.000.380.000.000.000.010.000.000.000.00
GovDebt0.290.29−0.060.310.36−0.05−0.02−0.13−0.05−0.340.190.170.440.480.31−0.110.200.01
0.000.000.180.000.000.370.810.010.340.000.000.000.000.000.000.040.000.80
PCGNP0.730.330.140.620.41−0.23−0.070.00−0.240.000.390.390.13−0.100.340.450.38−0.05
0.000.000.000.000.000.000.140.950.000.990.000.000.000.000.000.000.000.09
GGDP−0.10−0.110.14−0.10−0.10−0.04−0.050.01−0.050.01−0.11−0.10−0.12−0.15−0.18−0.07−0.15−0.08
0.000.000.000.000.000.120.320.780.060.770.000.000.000.000.000.010.000.00
Capflow−0.05−0.180.23−0.02−0.05−0.05−0.090.10−0.080.11−0.02−0.020.130.17−0.08−0.02−0.06−0.03
0.060.000.000.520.080.070.090.000.000.000.400.430.000.000.000.480.040.40
ToT−0.29−0.310.03−0.260.030.020.000.060.010.200.010.03−0.09−0.050.000.010.000.06
0.000.000.310.000.330.540.940.030.790.000.690.340.000.060.880.820.930.05
GExp−0.050.000.05−0.01−0.010.00−0.050.030.000.03−0.04−0.04−0.04−0.07−0.07−0.04−0.06−0.04
0.100.870.070.760.810.880.380.360.960.390.200.210.150.020.030.180.050.19
FisBal0.380.030.160.330.24−0.19−0.04−0.20−0.17−0.210.000.01−0.20−0.36−0.010.24−0.050.05
0.000.210.000.000.000.000.440.000.000.000.870.690.000.000.660.000.070.13
MilExpn−0.07−0.320.16−0.060.14−0.030.03−0.03−0.03−0.030.00−0.01−0.08−0.110.03−0.010.01−0.05
0.010.000.000.030.000.320.560.310.250.320.900.800.000.000.380.860.820.13
Money0.080.03−0.08−0.07−0.130.38−0.03−0.010.37−0.01−0.06−0.06−0.060.07−0.010.09−0.030.56
0.020.290.020.030.000.000.640.730.000.740.030.030.050.020.780.000.360.00
VGNP0.110.000.120.040.01−0.05−0.03−0.01−0.05−0.01−0.06−0.06−0.13−0.12−0.030.15−0.02−0.02
0.000.900.000.120.780.060.540.830.060.840.020.020.000.000.330.000.380.56
Res/ImpExp/GDPOppenessImp/GDPInfIntrateExRateCA/GDPCrdt/GDPGovDebtPCGNPGGDPCapflowToTGExpFisBalMilExpnMoneyVGNP
Res/Imp1.00
Exp/GDP−0.131.00
0.00
Oppeness−0.170.94 ***1.00
0.000.00
Imp/GDP−0.180.800.94 ***1.00
0.000.000.00
Inf−0.07−0.10−0.08−0.071.00
0.020.000.000.00
Intrate−0.10−0.21−0.17−0.09−0.031.00
0.000.000.000.000.28
ExRate0.030.010.050.040.71−0.121.00
0.510.660.130.240.000.00
CA/GDP0.080.170.120.04−0.01−0.170.071.00
0.000.000.000.100.680.000.02
Crdt/GDP0.230.330.280.21−0.14−0.380.010.121.00
0.000.000.000.000.000.000.780.00
GovDebt−0.020.070.050.01−0.20−0.14−0.200.020.431.00
0.750.070.180.710.000.000.000.520.00
PCGNP0.190.530.360.18−0.12−0.37−0.020.320.580.371.00
0.000.000.000.000.000.000.490.000.000.00
GGDP−0.090.030.040.03−0.080.100.020.04−0.14−0.23−0.081.00
0.000.130.090.230.000.000.440.040.000.000.00
Capflow0.13−0.040.040.12−0.05−0.010.04−0.10−0.04−0.18−0.060.151.00
0.000.070.070.000.030.660.260.000.100.000.010.00
ToT0.10−0.12−0.21−0.240.120.040.020.08−0.27−0.33−0.110.070.111.00
0.000.000.000.000.000.160.560.000.000.000.000.000.00
GExp−0.040.010.030.03−0.030.05−0.030.02−0.06−0.07−0.040.42−0.01−0.031.00
0.150.520.290.160.180.060.310.420.020.080.120.000.530.16
FisBal0.100.440.16−0.13−0.01−0.21−0.010.190.260.120.570.02−0.270.15−0.011.00
0.000.000.000.000.620.000.650.000.000.000.000.270.000.000.78
MilExpn0.32−0.02−0.05−0.10−0.01−0.130.000.010.030.020.14−0.120.070.10−0.060.161.00
0.000.370.030.000.550.000.990.690.220.610.000.000.000.000.020.00
Money−0.19−0.13−0.18−0.200.23−0.200.02−0.010.120.360.06−0.09−0.060.04−0.020.050.051.00
0.000.000.000.000.000.000.580.640.000.000.020.000.030.130.500.060.05
VGNP0.37−0.12−0.16−0.20−0.02−0.100.040.060.350.240.130.030.11−0.07−0.010.100.060.171.00
0.000.000.000.000.410.000.140.010.000.000.000.100.000.000.530.000.010.00
Significance codes: * p < 0.1; ** p < 0.05; *** p < 0.01
Table A2. Sensitivity analysis for the Q-variables, panel including all Z-variables, PrinArrears/ED > 0.1% (dependent variable: IMFcredit/ED, X-variable Res/ED).
Table A2. Sensitivity analysis for the Q-variables, panel including all Z-variables, PrinArrears/ED > 0.1% (dependent variable: IMFcredit/ED, X-variable Res/ED).
Q-Variables β tp-Value0.95 C.IVIFZ-VariablesRobust?
Min0.00−6.080.10−0.010.001.53ExRateED/Exp
ED/GNPBase0.00−1.310.190.000.00 No
Max0.003.160.200.000.001.35Res/ImpEthTen
Min−0.07−9.880.06−0.160.021.43MoneyGovDebt/GDP
Res/ImpBase0.013.970.000.000.01 Yes
Max−0.02−6.510.10−0.060.021.56CA/GDPTDS/Exp
Min−0.53−7.720.08−1.400.342.07PCGNPGovtstab
OpennessBase−0.04−2.830.01−0.08−0.01 Yes
Max−0.22−6.780.09−0.640.191.3ExRateCorr
Min0.00−4.320.14−0.020.011.37ExRateED/Exp
TDS/ExpBase0.00−4.610.00−0.010.00 No
Max0.014.370.14−0.020.031.68ExRateDebtServ/GDP
Min0.016.880.09−0.010.031.24OpennessExRate
InfBase0.000.600.550.000.00 Yes
Max0.017.190.09−0.010.041.32ExRateMoney
Min−0.04−4.690.13−0.150.071.34ExRateDemoc
MoneyBase0.00−1.840.070.000.00 No
Max0.00−3.340.18−0.010.001.23PrinArrears/ED ToT
Min0.017.500.08−0.010.041.3PrinArrears/ED GovDebt/GDP
IntRateBase0.00−0.260.800.000.00 No
Max0.017.500.08−0.010.041.3PrinArrears/ED GovDebt/GDP
Min−0.49−6.810.09−1.420.431.31ExRateFisBal
Imp/GDPBase−0.01−0.420.67−0.070.05 Yes
Max0.717.460.08−0.501.921.17OpennessED/Exp
Min−0.01−9.910.06−0.020.001.22ExRateSTD/Res
Exp/GDPBase0.00−2.970.000.000.00 Yes
Max0.00−6.680.090.000.001.16TDS/ExpCapFlow
Min0.00−2.580.24−0.030.021.35GExpMilExpn
GGDPBase0.00−0.320.75−0.010.00 No
Max0.012.070.29−0.030.041.28PrinArrears/ED GovDebt/GDP
Min−0.03−5.350.12−0.100.041.26CA/GDPMilExpn
ED/ExpBase−0.02−4.130.00−0.03−0.01 Yes
Max0.085.760.11−0.100.262.29Res/ImpExRate
Min0.00−11.520.060.000.001.1ExRateMilExpn
PCGNPBase−1.38−0.570.57−6.153.39 Yes
Max0.00−6.890.090.000.001.3CA/GDPCapFlow
Min0.006.430.100.000.011.21InfTDS/Exp
ExRateBase0.0210.680.000.010.02 Yes
Max0.017.130.09−0.010.021.17InfGexp
Min−0.01−6.860.09−0.030.012.1ED/ExpGovDebt/GDP
CA/GDPBase0.00−6.450.000.000.00 Yes
Max0.00−6.310.10−0.010.004.6TDS/ExpPR/Exp
Min−0.17−4.620.14−0.650.301.49GovDebt/GDPToT
CorrBase0.043.030.000.020.07 No
Max0.154.370.14−0.280.571.76ExRateIntRate
Min0.046.360.10−0.040.111.23Res/ImpCrdt/GDP
DemocBase0.033.860.000.010.04 Yes
Max0.116.430.10−0.110.332.07OpennessGovDebt/GDP
Min−0.02−3.400.18−0.120.071.26CorrPCGNP
GovtstabBase0.010.660.51−0.010.02 No
Max−0.02−3.140.20−0.110.071.23CA/GDPED/GNP
Min−0.02−3.430.18−0.100.061.59DemocLTServ/Res
MilPolBase−0.01−1.050.30−0.020.01 No
Max0.064.930.13−0.090.212.81Exp/GDPGovDebt/GDP
Min−0.02−2.060.29−0.110.081.59MilPolCrdt/GDP
EthTenBase−0.02−2.440.02−0.030.00 No
Max0.032.880.21−0.100.171.35ExRateToT
Table A3. Sensitivity analysis for the Q-variables, panel including all Z-variables, Arrears/ED > 5% (dependent variable: IMFcredit/ED, X-variable Res/ED).
Table A3. Sensitivity analysis for the Q-variables, panel including all Z-variables, Arrears/ED > 5% (dependent variable: IMFcredit/ED, X-variable Res/ED).
Q-Variables β tp-Value0.95 C.I.VIFZ-VariablesRobust?
Min0.00−3.960.16−0.010.011.54ExRateGexp
ED/GNPBase0.00−5.180.000.000.00 No
Max0.004.350.140.000.011.74Crdt/GDPMilPol
Min−0.08−6.320.10−0.240.081.79MoneyGovDebt/GDP
Res/ImpBase0.00−0.790.43−0.010.01 No
Max−0.03−6.460.10−0.080.021.36PCGNPEthTen
Min−0.65−6.660.09−1.880.599.05ED/ExpImp/GDP
OpennessBase−0.03−0.680.50−0.100.05 No
Max0.356.500.10−0.331.035.29Exp/GDPCrdt/GDP
Min0.00−3.780.16−0.020.011.4ExRateGExp
TDS/ExpBase−0.01−3.110.00−0.010.00 No
Max0.00−3.090.20−0.010.001.26ToTFisBal
Min0.016.650.10−0.010.031.68Res/ImpExRate
InfBase0.00−0.260.800.000.00 No
Max0.018.420.08−0.010.031.36ExRatePR/Exp
Min−0.05−4.380.14−0.180.091.95ExRateDemoc
MoneyBase0.00−2.060.040.000.00 No
Max0.004.230.150.000.011.48ToTEthTen
Min−0.01−4.550.14−0.020.011.47PCGNPDemoc
IntRateBase0.00−0.630.53−0.010.00 No
Max0.023.560.17−0.060.101.65Crdt/GDPGovDebt/GDP
Min−0.78−6.390.10−2.340.772.3ExRateFisBal
Imp/GDPBase0.111.690.09−0.020.23 Yes
Max1.136.820.09−0.973.239.05OpennessED/Exp
Min−0.01−9.450.07−0.030.001.58ExRateVGNP
Exp/GDPBase0.00−2.330.02−0.010.00 Yes
Max0.00−6.590.10−0.010.001.17MoneyVGNP
Min0.00−3.180.19−0.020.011.34Res/ImpCorr
GGDPBase0.000.450.66−0.010.01 No
Max0.00−3.120.20−0.020.011.18Res/ImpMilPol
Min0.076.490.10−0.070.221.67Res/ImpCorr
ED/ExpBase−0.04−3.580.00−0.06−0.02 Yes
Max0.076.490.10−0.070.221.67Res/ImpCorr
Min0.00−12.960.050.000.001.49ExRateMilExpn
PCGNPBase0.000.200.840.000.00 No
Max0.00−6.360.100.000.001.13Imp/GDPCapFlow
Min−0.01−3.630.17−0.020.013.84Exp/GDPGovDebt/GDP
ExRateBase0.028.770.000.020.03 No
Max0.014.190.15−0.020.032.28InfFisBal
Min−0.01−7.140.09−0.040.012.28Imp/GDPGovDebt/GDP
CA/GDPBase−0.01−7.130.00−0.010.00 Yes
Max0.00−6.360.10−0.010.001.63Res/ImpCapFlow
Min0.136.330.10−0.130.382.35PrinArrears/EDFisBal
CorrBase0.114.780.000.070.16 Yes
Max0.136.330.10−0.130.382.35PrinArrears/EDFisBal
Min0.066.840.09−0.050.171.93PCGNPPrinArrears/ED
DemocBase0.054.170.000.030.08 Yes
Max0.076.470.10−0.070.211.47PCGNPIntRate
Min−0.02−3.200.19−0.120.071.6DemocPCGNP
GovStabBase0.021.250.21−0.010.05 No
Max0.113.740.17−0.260.471.82MoneyGovDebt/GDP
Min−0.55−3.950.16−2.331.2224.6GovDebt/GDPEthTen
MilPolBase0.010.900.37−0.010.03 No
Max0.034.160.15−0.060.131.73PCGNPPrinArrears/ED
Min0.023.350.18−0.060.111.28Res/ImpPrinArrears/ED
EthTenBase−0.01−0.770.44−0.040.02 No
Max0.234.080.15−0.490.9424.6MilPolGovDebt/GDP
Table A4. Sensitivity analysis for the Q-variables, panel including all Z-variables, Arrears/ED > 1% (dependent variable: IMFcredit/ED, X-variable Res/ED).
Table A4. Sensitivity analysis for the Q-variables, panel including all Z-variables, Arrears/ED > 1% (dependent variable: IMFcredit/ED, X-variable Res/ED).
Q-Variables β tp-Value0.95 C.I.VIFZsRobust?
Min0.00−7.270.09−0.010.001.69ExRateED/Exp
ED/GNPBase0.00−5.360.000.000.00 Yes
Max0.00−7.270.09−0.010.001.69ExRateED/Exp
Min−0.07−8.900.07−0.180.031.54MoneyGovDebt/GDP
Res/ImpBase0.00−0.480.63−0.010.01 Yes
Max−0.02−6.440.10−0.070.021.24CA/GDPGovtstab
Min−0.56−7.370.09−1.530.415.08ExRateImp/GDP
OpennessBase−0.05−2.410.02−0.08−0.01 Yes
Max−0.25−7.060.09−0.690.201.24ExRateInf
Min−0.01−5.160.12−0.020.011.43ExRateED/Exp
TDS/ExpBase0.00−3.990.00−0.010.00 No
Max0.014.470.14−0.020.047.64ExRateDebtServ/GNP
Min0.017.320.09−0.010.031.23TDS/ExpExRate
InfBase0.000.280.780.000.00 No
Max0.017.690.08−0.010.031.16ExRateCA/GDP
Min−0.04−4.380.14−0.170.081.47ExRateEthTen
MoneyBase0.00−1.960.050.000.006 No
Max0.003.400.180.000.011.34ToTEthTen
Min0.00−3.610.17−0.020.011.34PCGNPDemoc
IntRateBase0.00−0.340.730.000.00 No
Max0.025.130.12−0.020.065.53ExRateGovDebt/GDP
Min−0.65−8.480.07−1.630.331.35ExRateFisBal
Imp/GDPBase−0.01−0.260.80−0.080.06 No
Max0.857.320.09−0.622.3212.16OpennessEXD/Exp
Min−0.01−10.550.06−0.020.001.33ExRateFisBal
Exp/GDPBase0.00−2.700.010.000.00 Yes
Max0.00−6.330.100.000.001.56Res/ImpVGNP
Min0.00−3.590.17−0.020.011.23Res/ImpCorr
GGDPBase0.00−0.150.88−0.010.00 No
Max0.00−3.220.19−0.020.011.21Res/ImpDemoc
Min−0.03−5.240.12−0.110.051.24CA/GDPMilExpn
ED/ExpBase−0.02−3.570.00−0.03−0.01 No
Max0.096.010.11−0.100.282.42Res/ImpExRate
Min0.00−9.500.070.000.002.86GovDebt/GDPDemoc
PCGNPBase−38.9−0.130.90−6.455.67 Yes
Max0.00−6.570.100.000.001.29CA/GDPCapFlow
Min0.016.750.09−0.010.021.22DebtServ/GDPInf
ExRateBase0.029.630.000.010.02 Yes
Max0.016.330.10−0.010.021.21InfGExp
Min−0.01−8.070.08−0.020.001.24ED/ExpMilExpn
CA/GDPBase0.00−6.110.00−0.010.00 Yes
Max0.00−6.450.10−0.010.001.32TDS/ExpLTServ/Res
Min−0.25−4.120.15−1.000.512.68GovDebt/GDPEthTen
CorrBase0.073.740.000.030.11 No
Max0.133.690.17−0.320.581.73IntRateExRate
Min−0.11−4.880.13−0.400.182.26GovDebt/GDPMilExpn
DemocBase0.033.570.000.010.05 No
Max0.063.660.17−0.160.292.86PCGNPGovDebt/GDP
Min−0.03−4.450.14−0.130.061.14CA/GDPPCGNP
GovStabBase0.010.830.41−0.010.03 No
Max0.093.240.19−0.260.451.7MoneyGovDebt/GDP
Min−0.04−3.270.19−0.220.131.42DemocExRate
MilPolBase0.00−0.690.49−0.020.01 No
Max0.085.780.11−0.100.263.71Exp/GDPGovDebt/GDP
Min0.033.950.16−0.060.111.18Res/ImpToT
EthTenBase−0.01−1.320.19−0.030.01 No
Max0.043.480.18−0.120.201.31IntRateToT
Table A5. Sensitivity analysis for the Q-variables, panel including all Z-variables, all observations (dependent variable: DummyIMFcredit/ED, X-variable Res/ED).
Table A5. Sensitivity analysis for the Q-variables, panel including all Z-variables, all observations (dependent variable: DummyIMFcredit/ED, X-variable Res/ED).
Q-Variables β tp-Value0.95 C.I.VIFZ-VariablesRobust?
Min0.00533.49250.1775−0.0140.02451.17ED/GNPIntRate
ExRateBase0.0210.680.000.010.02 no
Max0.00663.68690.1686−0.01630.02951.1IntRateMilPol
Min0.00256.48990.0973−0.00240.00731.46Crdt/GDPTDS/Exp
ED/GNPBase0.00−1.310.190.000.00 no
Max0.00626.85580.0922−0.00530.01761.44ExRatePR/Exp
Min−0.0961−4.4130.1419−0.3730.18071.3GovDebt/GDPMilExpn
MoneyBase0.00−1.840.070.000.00 no
Max0.01343.54180.1752−0.03460.06131.09LTServ/ResIntRate
Min0.0056.58730.0959−0.00460.01461.09PrinArrears/EDPCGNP
Exp/GDPBase0.00−2.970.000.000.00 yes
Max0.01589.00180.0704−0.00650.03811.21Crdt/GDPGovDebt/GDP
Min0.92236.53450.0967−0.87112.71581.15Crdt/GDPTOT
PrinArrears/EDBase0.76546.76650.0987−0.98765.876 yes
Max5.19626.98880.0905−4.250814.64311.22GovDebt/GDPFisBal
Min4.3576.55840.0963−4.084312.79844.2PR/ExpGovDebt/GDP
LTServ/ResBase0.00−5.020.000.000.00 yes
Max4.3576.55840.0963−4.084312.79844.2PR/ExpGovDebt/GDP
Min−2.7478−3.9090.1594−11.6786.18222.4ED/GNPExRate
DebtServ/GNPBase−0.77−7.030.00−0.98−0.55 No
Max18.19564.0950.1525−38.26374.6546348.3LTServ/GDPMilPol
Min−0.0758−4.3210.1448−0.29850.1472.34ED/GNPCrdt/GDP
ED/ExpBase−0.02−4.130.00−0.03−0.01 No
Max0.07783.6290.1712−0.19470.35041.5Exp/GDPGovDebt/GDP
Min0−7.4370.0851001.27ExRateVGNP
LTServ/ResBase0.00−5.020.000.000.00 yes
Max0−6.3730.0991001.06IntRateVGNP
Min−2.5579−6.5750.0961−7.50122.38543.42LTServ/GDPExRate
PR/ExpBase−0.25−7.830.00−0.31−0.19 yes
Max−2.5364−6.6270.0953−7.39962.32673.31DebtServ/GNPExRate
Min−0.0171−3.2950.1876−0.08320.0492.98PrinArrears/EDInf
STD/ResBase−0.03−0.150.88−0.460.40 No
Max0.26945.40970.1164−0.36330.90211.35IntRateGovDebt/GDP
Min−0.0834−9.5430.0665−0.19450.02771.59IntRateGovDebt/GDP
Res/ImpBase0.013.970.000.000.01 yes
Max−0.0278−6.6040.0957−0.08120.02571.19Crdt/GDPMilExpn
Min0.24836.56290.0963−0.23240.7291.11PrinArrears/EDCrdt/GDP
OpennessBase−0.04−2.830.01−0.08−0.01 yes
Max0.75657.3450.0861−0.55222.06511.29GovDebt/GDPEthTen
Min0.48686.76980.0934−0.42691.40051.32Crdt/GDPVGNP
Imp/GDPBase−0.01−0.420.67−0.070.05 yes
Max1.34997.57330.0836−0.91493.61471.21GovDebt/GDPEthTen
Min0.00223.08310.1997−0.0070.01151.01Imp/GDPGovStab
InfBase0.000.600.550.000.00 No
Max0.02173.79380.1641−0.05090.09421.21ExRateEthTen
Min0.01496.33710.0996−0.0150.04481.19Res/ImpVGNP
IntRateBase0.00−0.260.800.000.00 yes
Max0.0238.64610.0733−0.01080.05681.04OpennessGExp
Min−0.0157−4.01570.1554−0.06540.0341.02ExRateGExp
CA/GDPBase0.00−6.450.000.000.00 No
Max−0.0039−3.2780.1885−0.01890.01121.43Res/ImpPCGNP
Min−0.0078−14.690.0433−0.0145−0.0011.09Exp/GDPGExp
Crdt/GDPBase−0.03−0.810.42−0.090.04 yes
Max−0.0033−6.4860.0974−0.00970.00311.67Res/ImpTOT
Min−0.01−3.4460.1798−0.04670.02681.89ExRateDemoc
GovDebt/GDPBase0.00−1.980.050.000.00 No
Max−0.0091−3.2910.1878−0.04410.0261.93ExRateEthTen
Min0−7.3120.0865−0.000101.2IntRateEthTen
PCGNPBase−1−0.570.57−63 yes
Max0−6.4420.098001.09PrinArrears/EDED/GNP
Min0.00642.28890.2622−0.02930.04211.03InfED/GNP
GGDPBase0.00−0.320.75−0.010.00 No
Max0.01272.26010.2652−0.05860.0841.05ExRateMilPol
Min−0.0053−3.080.1999−0.0270.01651.68CA/GDPGExp
CapFlowBase0.00−0.320.75−0.010.00 No
Max0.01754.56350.1373−0.03120.06611.36ExRateCorr
Min0.0013.17760.1941−0.00310.00521.07PrinArrears/EDED/Exp
TOTBase0.002.120.030.000.00 No
Max0.0023.40340.1819−0.00560.00961.38ExRateFisBal
Min−0.0007−1.9650.2997−0.00540.0041.03Crdt/GDPRes/Imp
GExpBase0.030.970.33−0.030.08 No
Max−0.0007−1.9650.2997−0.00540.0041.03Crdt/GDPRes/Imp
Min−0.0115−3.9620.1574−0.04830.025314.58Imp/GDPExp/GDP
FisBalBase−0.44−7.070.00−0.56−0.31 No
Max0.00544.98340.1261−0.00830.0191.27PCGNPED/GNP
Min−0.1585−7.0670.0895−0.44340.12651.39ExRateEthTen
MilExpnBase−0.86−0.960.34−2.610.89 No
Max−0.0907−6.3190.0999−0.27310.09171.17PrinArrears/EDDemoc
Min0−6.7700.0934001.12FisBalGExp
VGNPBase0.00−0.240.810.000.00 yes
Max08.4110.0753001.32Crdt/GDPRes/Imp
Min−0.0188−5.5940.1126−0.06160.02393.06DebtServ/GNPExRate
TDS/ExpBase0.00−4.610.00−0.010.00 No
Max−0.0031−3.0840.1996−0.01580.00961.06TOTSTD/Res
Min−0.1536−4.5960.1364−0.5780.27091.03GexpExp/GDP
CorrBase0.043.030.000.020.07 No
Max−0.0993−3.0890.1993−0.50760.30911.05LTServ/ResED/GNP
Min−0.1249−7.6710.0825−0.33190.0821.05FisBalIntRate
DemocBase0.033.860.000.010.04 yes
Max−0.086−6.3410.0996−0.25820.08631.22PCGNPLTServ/Res
Min−0.1453−3.4450.1799−0.6810.39052.01GovDebt/GDPExRate
GovStabBase0.010.660.51−0.010.02 No
Max0.07363.85350.1616−0.16910.31631.08Crdt/GDPIntRate
Min0.28846.83210.0925−0.2480.82492.01GovDebt/GDPExRate
MilPolBase−0.01−1.050.30−0.020.01 yes
Max0.28846.83210.0925−0.2480.82492.01GovDebt/GDPExRate
Min−0.1461−5.210.1207−0.50220.21011.29GovDebt/GDPOpenness
EthTenBase−0.02−2.440.02−0.030.00 No
Max0.06473.72790.1668−0.15570.2851.2PCGNPIntRate
Table A6. Panel regressions (dependent variable: IMFcredit/ED).
Table A6. Panel regressions (dependent variable: IMFcredit/ED).
Model 1Model 2Model 3
FEREFEREFERE
Res/ED0.004070.0328560.032196−0.019580.007845 ***0.007901 ***
(0.03212)(0.02938)(0.048114)(0.040441)(0.00067)(0.000653)
ExRate−0.00046−0.000480.0008810.001025 **−0.00101 **−0.00109 **
(0.000552)(0.000525)(0.000529)(0.00048)(0.000556)(0.000549)
ED/GNP−0.00053−0.00036−0.00111−0.0013−0.00231 ***−0.00185 ***
(0.000596)(0.000547)(0.000952)(0.000656)(0.000531)(0.000491)
Money−0.00459−0.00875−0.02646−0.028 ***−0.00185−0.00228
(0.007571)(0.007177)(0.01585)(0.007861)(0.004933)(0.004858)
TDS/Exp−0.00354−0.00248−0.0041−0.0039−0.00525−0.00586 *
(0.002335)(0.00221)(0.00331)(0.002491)(0.003265)(0.003182)
Exp/GDP−0.00594−0.002120.010129 *0.007382 *0.01368 ***0.013286 ***
(0.003464)(0.003207)(0.00508)(0.004196)(0.003854)(0.003824
PrinArrears/ED0.7247 ***0.7247 ***0.7254 ***0.7073 ***0.589229 ***0.70562 ***
(0.137799)(0.136368)(0.188794)(0.147397)(0.162205)(0.155197)
LTServ/Res−5.82 × 10⁻¹⁴−1.31 × 10⁻¹³6.58 × 10⁻¹³ **7.06 × 10⁻¹³ ***−4.66 × 10⁻¹³ **−4.30 × 10⁻¹³
(1.53 × 10⁻¹³) (1.48 × 10⁻¹³) (3.01 × 10⁻¹³)(2.26 × 10⁻¹³)(1.92 × 10⁻¹³)(1.88 × 10⁻¹³)
PR/Exp0.55393 **0.5115 **0.1836920.1066990.4695260.55428 *
(0.222011)(0.210111)(0.333349)(0.234692)(0.346043)(0.337352)
STD/Res0.0040040.0132590.0026250.0120620.0639560.049873
(0.019958)(0.019616)(0.045264)(0.029428)(0.026833)(0.025499)
Openness−0.13458−0.45822−0.44158−0.29481−0.28744−0.25037
(0.318055)(0.299172)(0.384995)(0.301222)(0.34785)(0.338773)
Imp/GDP0.7687091.003588 *0.1324090.097066−0.59103−0.69179
(0.573326)(0.552981)(0.750223)(0.614203)(0.505987)(0.485108)
Inf0.0000250.0004020.000229−0.000230.0025020.001488
(0.001073)(0.000932)(0.00103)(0.000851)(0.001477)(0.001413)
IntRate0.0001890.0011150.00237*0.00247 ***0.000216−0.00058
(0.001108)(0.001054)(0.001289)(0.000853)(0.001292)(0.001146)
CA/GDP0.000624−0.00028−0.00397−0.00066−0.00737 ***−0.00716 ***
(0.001993)(0.001855)(0.002784)(0.002206)(0.002034)(0.002002)
Crdt/GDP0.0005220.0009540.0033 ***0.00336 ***−0.00105 ***−0.00093 **
(0.000695)(0.000609)(0.000752)(0.000711)(0.000377)(0.000369)
GGDP0.000738−0.000750.0018030.001192−0.00632***−0.00823 ***
(0.002011)(0.001464)(0.001688)(0.00132)(0.001994)(0.001701)
Capflow−0.00194−0.00234−0.00248−0.00040.0012370.001732
(0.001626)(0.001554)(0.003089)(0.002429)(0.001439)(0.00139)
ToT−0.000120.0001130.000160.000213−0.00011−0.00019
(0.000209)(0.000195)(0.000545)(0.000368)(0.000172)(0.000162)
FisBal0.0043080.004063−0.00492−0.00583−0.00957 ***−0.00937 ***
(0.003142)(0.003004)(0.004785)(0.003744)(0.002378)(0.002323)
MilExpn−0.00698 *−0.006750.0145380.0074710.029181 ***0.027927 ***
(0.004163)(0.004132)(0.010363)(0.009003)(0.006169)(0.006034)
VGNP−5.69 × 10⁻¹⁴−6.41 × 10⁻¹⁴−2.53 × 10⁻¹³ *−1.95E-13 **−7.98 × 10⁻¹⁵ −2.47 × 10⁻¹⁴
(7.37 × 10⁻¹⁴)(7.30 × 10⁻¹⁴)(1.23 × 10⁻¹³) (8.30 × 10⁻¹⁴)(3.35 × 10⁻¹⁴)(3.21 × 10⁻¹⁴)
Res/Imp−0.00507−0.00762−0.00452−0.00057
(0.004279)(0.00386)(0.005416)(0.004011)
ED/Exp−0.0051−0.024630.0554540.038044
(0.021857)(0.021942)(0.031346)(0.025637)
LTServ/GDP−1.25062−1.0154−0.87144−1.15904
(1.340687)(1.254754)(1.242649)(1.103773)
GovDebt/GDP−0.00078 *−0.0011 ***0.0003530.000577
(0.000454)(0.000405)(0.000587)(0.000469)
DebtServ/GNP1.3544760.8835921.1696621.500618
(1.344129)(1.217801)(1.287311)(1.1296)
PCGNP1.11 × 10⁻⁶−9.51 × 10⁻⁷−6.21 × 10⁻⁶−9.04 × 10⁻⁶ ***
(1.51 × 10⁻⁶)(1.34 × 10⁻⁶)(3.66 × 10⁻⁶)(1.95 × 10⁻⁶)
GExp−0.00022−0.00024−0.00071−0.00087
(0.000515)(0.000504)(0.000611)(0.000555)
Corr 0.079505 **0.0871 ***−0.01010.00342
(0.028471)(0.020705)(0.020046)(0.018862)
Democ 0.0037820.006688−0.01022−0.01271
(0.035375)(0.02217)(0.010535)(0.010305)
GovStab −0.00441−0.00472−0.000550.001141
(0.004489)(0.003901)(0.006582)(0.006406)
MilPol −0.02573−0.018880.0164370.010876
(0.020401)(0.018143)(0.011622)(0.011341)
EthTen −0.0525 ***−0.03452 **−0.01536 *−0.01954 **
(0.017399)(0.015495)(0.008519)(0.008333)
Constant0.2228690.261323−0.03684−0.139270.2550520.290645
(0.105432)(0.101237)(0.213568)(0.15359)(0.111614)(0.106625)
R20.820.870.950.970.850.89
Chi-Sq15.26 11.41 13.51
Prob. N Chi-sq.0.123 0.3264 0.226
Significance codes: * p < 0.1; ** p < 0.05; *** p < 0.01.

Notes

1
The aftermath of the 2008 financial crisis has spurred a proliferation of empirical studies on this subject, particularly in light of the significant shifts observed in global external debt dynamics (Chakrabarti and Zeaiter 2014).
2
No thorough management of the risks involved with the lending ‘obsession’ among international banks in advanced economies until Mexico’s unilateral default on its external debt in August 1982 (Stambuli 1998).
3
As Christofides et al. (2005) point out, the Fund has, in cases such as the Mexican crisis of 1994–1995, helped countries avert a costly default on their sovereign debt.
4
The weighted average of a country’s currency in relation to an index or basket of other major currencies.
5
The appendix includes Descriptive statistics and list of countries.
6
For detailed analysis on free and doubtful variable selection, see Chakrabarti and Zeaiter (2014).
7
Also, we restrict the number of explanatory variables included in any regression to be 8 or less, as implied by Levine and Renelt (1992).
8
See Levine and Renelt (1992) for information on potential econometric issues with the specification relating to the selection of Z-variables, such as multicollinearity. We follow their approach, before implementing the EBA: we do not include more than 3 free variables (X), we only include one; we cap the total number of regressors by the size of the pool of doubtful variables, Z, from which a subset is selected for each regression run; and we drop any Q-variable that is found to actually measure the same attribute as a Z-variable.

References

  1. Abosedra, Salah, and Ali Fakih. 2017. Assessing the Role of Remittances and Financial Deepening in Growth: The Experience of Lebanon. Global Economy Journal 17: 1–17. [Google Scholar] [CrossRef]
  2. Afonso, Antonio, Pedro Gomes, and Philip Rother. 2010. What Hides Behind Sovereign Debt Rating? European Central Bank Working Paper. Frankfurt: European Central Bank. [Google Scholar]
  3. Al Fawwaz, Torki Mejhim. 2016. Determinants of External Debt in Jordan: An Empirical study (1990–2014). International Business Research 9: 116. [Google Scholar] [CrossRef]
  4. Aylward, Lynn, and Rupert Thorne. 1998. Countries’ repayment performance vis-à-vis IMF. IMF Staff Papers 45: 595–619. [Google Scholar] [CrossRef]
  5. Bai, Yan, Seon Tae Kim, and Gabriel Mihalache. 2015. The Maturity and Payment Schedule of Sovereign Debt. New York: National Bureau of Economic Research. [Google Scholar] [CrossRef]
  6. Balkan, Erol. 1992. Political instability, country risk and probability of default. Applied Economics 24: 999–1008. [Google Scholar] [CrossRef]
  7. Bandiera, Luca, Jesus Crespo Cuaresma, and Gallina Andronova Vincelette. 2010. Unpleasant Surprises: Sovereign Default Determinants and Prospects. Washington, DC: The World Bank, Poverty Reduction and Economic Management Network, Economic Policy and Debt Department. [Google Scholar]
  8. Berg, Andrew, and Jeffrey Sachs. 1988. The debt crisis: Structural explanations of country performance. Journal of Development Economics 29: 271–306. [Google Scholar] [CrossRef]
  9. Berg, Tobias, and Daniel Streitz. 2016. Determinants of the size of the sovereign Credit default swap market. The Journal of Fixed Income 25: 58–73. [Google Scholar] [CrossRef]
  10. Bird, Graham, and Dane Rowlands. 2001. IMF lending: How is it affected by economic, political, and institutional factors? Journal of Policy Reform 4: 243–70. [Google Scholar] [CrossRef]
  11. Bitar, Nicholas, Hussein Zeaiter, and Avik Chakrabarti. 2018. Were Reinhart and Rogoff Right? International Review of Economics and Finance 58: 614–20. [Google Scholar] [CrossRef]
  12. Boughton, James M. 2000. From Suez to Tequila: The IMF as crisis manager. The Economic Journal 110: 273–91. [Google Scholar] [CrossRef]
  13. Callier, Philippe. 1985. Further results on countries debt service performance: The relevance of structural factors. Weltwirtschaftlishes Archive 121: 385–92. [Google Scholar]
  14. Carlos, Santos. 2011. The Euro Sovereign Debt Crisis, Determinants of Default Probabilities and Implied Ratings in the CDS Market: An Econometric Analysis Munich Personal RePEc Archive. Available online: https://mpra.ub.uni-muenchen.de/31341/ (accessed on 1 December 2023).
  15. Catao, Luis, and Bennett Sutton. 2002. Sovereign Defaults: The Role of Volatility. IMF Working Papers, No. 149. Washington, DC: IMF. [Google Scholar]
  16. Celasun, Oya, and Philipp Harms. 2010. Financial deregulation, private foreign borrowing and the risk of sovereign default: A political–economic analysis. Frontiers in Finance and Economic 7: 82–100. [Google Scholar]
  17. Chakrabarti, Avik, and Hussein Zeaiter. 2014. The determinants of sovereign default: A sensitivity analysis. International Review of Economics & Finance 33: 300–18. [Google Scholar]
  18. Christofides, Charalambos, Atish R. Ghosh, Uma Ramakrishnan, Alun H. Thomas, Laura Papi, Juan Zalduendo, and Juan I. Kim. 2005. The design of IMF-supported programs. In The Design of IMF-Supported Programs. Washington, DC: International Monetary Fund. [Google Scholar]
  19. Chronopoulos, Dimitris K., George Dotsis, and Nikolaos T. Milonas. 2019. International evidence on the determinants of domestic Sovereign Debt Bank holdings. Journal of Financial Services Research 58: 143–60. [Google Scholar] [CrossRef]
  20. Citron, Joel-Tomas, and Gerald Nickelsburg. 1987. Country risk and political instability. Journal of Development Economics 25: 385–92. [Google Scholar] [CrossRef]
  21. Cline, William R. 1984. International Debt: Systematic Risk and Policy Response. Washington, DC: Institute of International Economics. [Google Scholar]
  22. Cooley, Thomas F., and Stephen F. LeRoy. 1981. Identification and estimation of money demand. American Economic Review 71: 825–44. [Google Scholar]
  23. Das, Ramesh Chandra. 2023. Growth and Developmental Aspects of Credit Allocation: An Inquiry for Leading Countries and the Indian States. Bingley: Emerald Publishing Limited. [Google Scholar]
  24. Das, Ramesh Chandra, and Jagabandhu Mandal. 2022. Long-Run Associations and Short-Run Dynamics Between Corruption and Inflation: Examining Social Unsustainability for the Panel of Countries. International Journal of Social Ecology and Sustainable Development (IJSESD) 13: 1–16. [Google Scholar] [CrossRef]
  25. Detragiache, Enrica, and Antonio Spilimbergo. 2001. Crisis and Liquidity: Evidence and Interpretation. IMF Working Paper, No. 02. Washington, DC: IMF. [Google Scholar]
  26. Dibeh, Ghassan, Ali Fakih, and Walid Marrouch. 2018. Decision to Emigrate Amongst the Youth in Lebanon. International Migration 56: 5–22. [Google Scholar] [CrossRef]
  27. Díaz-Cassou, Javier, Aitor Erce-Domínguez, and Juan J. V. Zamora. 2008. The Role of the IMF in Recent Sovereign Debt Restructurings: Implications for the Policy of Lending into Arrears. Documentos Ocasionales-Banco de España. Washington, DC: IMF, pp. 1–39. [Google Scholar]
  28. Do, Quy-Toan, and Andrei A. Levchenko. 2007. Comparative Advantage, Demand for External Finance, and Financial Development. Journal of Financial Economics 86: 796–834. [Google Scholar] [CrossRef]
  29. Edwards, Sebastian. 1984. LDC foreign borrowing and default risk: An empirical investigation. American Economic Review 74: 726–34. [Google Scholar]
  30. El Ghonemy, M. Riad. 1998. Affluence and Poverty in the Middle East. London: Psychology Press. [Google Scholar]
  31. Elmore, C., and G. McKenzie. 1992. Predicting LDC Debt Arrears. Southampton: University of Southampton. [Google Scholar]
  32. El-Said, Hamed, and Jane Harrigan. 2014. Economic reform, social welfare, and instability: Jordan, Egypt, Morocco, and Tunisia, 1983–2004. The Middle East Journal 68: 99–121. [Google Scholar] [CrossRef]
  33. Euh, Yoon-Dae. 1979. Commercial Banks and the Credit Worthiness of Less Developed Countries. Ann Arbor: UMI Research Press, p. 11. [Google Scholar]
  34. Feder, Greshon, and Lily Uy. 1985. The determinants of international creditworthiness and their implications. Journal of Policy Modeling 7: 133–56. [Google Scholar] [CrossRef]
  35. Feder, Gershon, Richard Just, and Knud Ross. 1981. Projecting debt-servicing capacity of developing countries. Journal of Quantitative and Financial Analysis 16: 651–69. [Google Scholar] [CrossRef]
  36. Frank, Charles R., and William R. Cline. 1971. Measurement of debt servicing capacity: An application of discriminant analysis. Journal of International Economics 1: 327–44. [Google Scholar] [CrossRef]
  37. Gartner, Manfred, Bjorn Griesbach, and Florian Jung. 2011. Pigs or lambs? The European sovereign debt crisis and the role of rating agencies. International Advances in Economic Research 17: 288–99. [Google Scholar] [CrossRef]
  38. Georgievska, A., L. Georgievska, A. Stojanovi, and Natasa Todorovic. 2008. Sovereign rescheduling probabilities in emerging markets: A comparison with credit rating agencies’ ratings. Journal of Applied Statistics 35: 1031–51. [Google Scholar] [CrossRef]
  39. Ghulam, Yaseen, and Benedict Saunby. 2023. Does increase in defence spending lead to more risk of sovereign debt defaults? Defence and Peace Economics 1: 1–19. [Google Scholar] [CrossRef]
  40. Gokmenoglu, Korhan K., and Rabiatul Adawiyah Mohamed Rafik. 2018. Determinants of external Debt: The case of Malaysia. In Springer Proceedings in Business and Economics. Berlin and Heidelberg: Springer, pp. 16–33. [Google Scholar]
  41. Goldstein, Morris. 2003. Debt Sustainability, Brazil, and the IMF. Available online: https://ssrn.com/abstract=384022 (accessed on 1 January 2024). [CrossRef]
  42. González-Fernández, Marcos, and Carmen González-Velasco. 2018a. Bond Yields, Sovereign Risk and Maturity Structure. Open Access Journal of Risk Management 6: 109. [Google Scholar] [CrossRef]
  43. González-Fernández, Marcos, and Carmen González-Velasco. 2018b. What drives sovereign debt maturity in European countries? Panoeconomicus 65: 137–61. [Google Scholar] [CrossRef]
  44. Guzman, Martin, and D. Heymann. 2016. The IMF debt sustainability analysis: Issues and problems. Journal of Globalization and Development 6: 387–404. [Google Scholar] [CrossRef]
  45. Haan, Sturm. 2000. No Need to Run Millions of Regressions. SSRN Journal. [Google Scholar]
  46. Haidar, Jamal. 2011. Haidar, Jamal. 2011. Currency Valuation and Purchasing Power Parity. World Economics 12: 1–12. [Google Scholar]
  47. Haidar, Jamal. 2012. Sovereign Credit Risk in the Euro Zone. World Economics 13: 123–36. [Google Scholar] [CrossRef]
  48. Hajivassiliou, Vassilis. 1987. The external debt repayments problems of LDC’s. An Econometric Model Based on Panel Data 36: 205–30. [Google Scholar] [CrossRef]
  49. Hajivassiliou, Vassilis. 1989. Do the Secondary Markets Believe in Life after Debt? Policy Research Working Paper Series, No. 252; Washington, DC: World Bank. [Google Scholar]
  50. Hajivassiliou, Vassilis. 1994. A simulation estimation analysis of the external debt crises of developing countries. Journal of Applied Econometrics 9: 109–31. [Google Scholar] [CrossRef]
  51. Haque, Nadeem, Manmohan Kumar, Nelson Mark, and Donald Mathiesonm. 1996. The economic content of indicators of developing country creditworthiness. IMF Staff Papers 43: 688–724. [Google Scholar] [CrossRef]
  52. Herrera, Ana Maria, M. B. Karaki, and Sandeep Kumar Rangaraju. 2019. Oil price shocks and US economic activity. Energy Policy 129: 89–99. [Google Scholar] [CrossRef]
  53. Hilscher, Jens, and Yves Nosbusch. 2010. Determinants of sovereign risk: Macroeconomic fundamentals and the pricing of sovereign debt. Review of Finance 14: 235–62. [Google Scholar] [CrossRef]
  54. Hwang, Insang, and Eric C. Wang. 2004. Does openness to trade affect total factor productivity growth? Evidence from 45 Japanese manufacturing industries. Journal of Economic Research 9: 147–73. [Google Scholar]
  55. IMF. 2017. IMF Annual Report 2017: Promoting Inclusive Growth. Available online: https://www.imf.org/external/pubs/ft/ar/2017/eng/index.htm (accessed on 10 January 2024).
  56. IMF. 1984. International Financial Statistics. Yearbook. Washington, DC: IMF. [Google Scholar]
  57. IMF. 2022. Fund Policy on Lending into Arrears to Private Creditors—Further Consideration of the Good Faith Criterion. Washington, DC: International Monetary Fund. [Google Scholar]
  58. IMF. n.d. IMF at a Glance. Retrieved 3 June 2024. Available online: https://www.imf.org/en/About/Factsheets/IMF-at-a-Glance (accessed on 10 November 2023).
  59. International Country Risk Guide (ICRG). n.d. Political Risk Indices. Available online: http://www.prsgroup.com/icrg.aspx (accessed on 10 November 2023).
  60. Jorra, Markus. 2012. The effect of IMF lending on the probability of sovereign debt crises. Journal of International Money and Finance 31: 709–25. [Google Scholar] [CrossRef]
  61. Kadiric, Samir. 2022. The determinants of sovereign risk premiums in the UK and the European government bond market: The impact of Brexit. International Economics and Economic Policy 19: 267–98. [Google Scholar] [CrossRef]
  62. Karagol, Erdal, and Selami Sezgin. 2004. Do defense expenditures increase debt rescheduling in Turkey? probit model approach. Defense and Peace Economics 15: 471–80. [Google Scholar] [CrossRef]
  63. Karayalcin, Ayse, and Tugrul Temel. 1988. Debt Services Capacity of Developing Countries and the Case of Turkey. The Central Bank of Turkey (Discussion Paper No. 8810). Ankara: The Central Bank of Turkey. [Google Scholar]
  64. Kışla, Gul H., Muradoglu Gulnur, and A. Özlem Önder. 2022. Spillovers from one country’s sovereign debt to CDS (credit default swap) spreads of others during the European crisis: A spatial approach. Journal of Asset Management 23: 277. [Google Scholar] [CrossRef]
  65. Kohlscheen, Emanuel. 2010. Sovereign risk: Constitutions rule. Oxford Economic Papers 62: 62–85. [Google Scholar] [CrossRef]
  66. Kruger, Mark, and Miguel Messmacher. 2004. Sovereign Debt Defaults and Financing Needs. IMF Economic Papers, No. 53. Washington, DC: IMF. [Google Scholar]
  67. Lau, Evan, and Alvina Lee. 2016. Determinants of external debt in Thailand and the Philippines. Available online: https://econpapers.repec.org/RePEc:eco:journ1:2016-04-89 (accessed on 5 November 2023).
  68. Leamer, Edward E. 1983. Let’s take the con out of econometrics. American Economic Review 73: 31–43. [Google Scholar]
  69. Lee, Seund Jung. 1991. Ability and willingness to service debt as explanation for commercial and official rescheduling cases. Journal of Banking and Finance 15: 5–27. [Google Scholar] [CrossRef]
  70. Lee, Seung Jung. 2009. How information quality of macro aggregates affects sovereign risk: An empirical investigation. Review of International Economics 17: 510–32. [Google Scholar] [CrossRef]
  71. Levine, Ross, and David Renelt. 1992. A sensitivity analysis of cross-country growth regressions. American Economic Review 82: 942–63. [Google Scholar]
  72. Li, Carmen. 1992. Latin America: A case for debt relief and foreign aid? Journal of International Development 4: 217–31. [Google Scholar] [CrossRef]
  73. Maddah, Lina, Josep-Maria Arauzo-Carod, and Fernando A. López. 2023. Detection of geographical clustering: Cultural and creative industries in Barcelona. European Planning Studies 31: 554–75. [Google Scholar] [CrossRef]
  74. Maltritz, Dominic, and Alexander Molchanov. 2014. Country credit risk determinants with model uncertainty. International Review of Economics and Finance 29: 224–34. [Google Scholar] [CrossRef]
  75. Marashaden, Omar. 1997. A logit model to predict debt rescheduling by less developed countries. Asian Economies 26: 25–34. [Google Scholar]
  76. McFadden, Daniel, R. Eckaus, G. Feder, A. V. Hajivassiliou, and S. O’Connell. 1985. Is there life after debt? An econometric analysis of the creditworthiness of developing countries. In International Debt and the Developing Countries. Edited by G. W. Smith and J. T. Cuddington. Washington, DC: World Bank. [Google Scholar]
  77. Minier. 2005. Nonlinearities and Robustness in Growth Regressions. SSRN Journal. [Google Scholar] [CrossRef]
  78. Oatley, Thomas. 2002. Commercial Banks and the International Monetary Fund: An Empirical Analysis. Chapel Hill: University of North Carolina at Chapel Hill. [Google Scholar]
  79. Oatley, Thomas, and Jason Yackee. 2000. Political Determinants of IMF Balance of Payments Lending: The Curse of Carabosse? Chapel Hill: University of North Carolina at Chapel Hill. [Google Scholar]
  80. Ojeda-Joya, Jair, and Jose E. Gómez-González. 2012. The Term-Structure of Sovereign Default Risk in Colombia and Its Determinants. Bogotá: Banco de la Republica de Colombia. [Google Scholar]
  81. Panizza, Ugo, Federico Strurzenegger, and Jeromin Zettelmeyer. 2009. The economic and law of sovereign debt and default. Journal of Economic Literature 3: 651–98. [Google Scholar] [CrossRef]
  82. Peter, Marcel. 2002. Estimating Default Probabilities of Emerging Market Sovereigns: A New Look at-So-New Literature. HEI Working Paper, No. 6. Norwalk: HEI. [Google Scholar]
  83. Qian, Zongxin, and Qian Luo. 2015. Regime-Dependent determinants of China’s sovereign credit default swap spread. Emerging Markets Finance and Trade 52: 10–21. [Google Scholar] [CrossRef]
  84. Rahnama-Moghadam, Mashaalah. 1995. Debt rescheduling in less developed countries: World or regional crisis. Journal of Developing Areas 30: 11–22. [Google Scholar]
  85. Rivoli, Pietra, and Thomas Brewer. 1997. Political instability and country risk. Global Finance Journal 8: 309–21. [Google Scholar] [CrossRef]
  86. Saini, Krishan G., and Philip S. Bates. 1984. A survey of the quantitative approaches to country risk analysis. Journal of Banking and Finance 8: 341–56. [Google Scholar] [CrossRef]
  87. Sargen, Nicholas. 1977. Economic Indicators and Country Risk Appraisal. Economic Review of the Federal Reserve Bank of San Francisco. San Francisco: Federal Reserve Bank of San Francisco, pp. 19–35. [Google Scholar]
  88. Semik, Sofia, and Lilli Zimmermann. 2021. Determinants of substantial public debt reductions in Central and Eastern European Countries. Empirica 49: 53–70. [Google Scholar] [CrossRef]
  89. Shadlen, Ken. 2006. Debt, Finance, and the IMF: Three Decades of Debt Crises in Latin America. South America, Central America and the Caribbean, 8–12. [Google Scholar]
  90. Sherry, Hassan. 2017. Debunking the Myth of a Changing IMF: Unpacking Conditionality in the Arab Region Post-Uprisings. Beirut: Arab NGO Network for Development (ANND). [Google Scholar]
  91. Snider, Lewis. 1990. The political performance of third world governments and debt crisis. The American Political Science Review 84: 1263–80. [Google Scholar] [CrossRef]
  92. Solberg, Ronald. 1988. Sovereign Rescheduling: Risk and Portfolio Management. London: Unwin Hyman, pp. 9–21. [Google Scholar]
  93. Soudis, Dimitrios. 2016. Determinants of Sovereign Bonds Ratings: A Robustness Analysis. Bulletin of Economic Research 2: 164–77. [Google Scholar] [CrossRef]
  94. Stamatopoulos, Theodoros V., Stavros Arvanitis, and Dimitris Terzakis. 2016. The risk of the sovereign debt default: The Eurozone crisis 2008–2013. Applied Economics 49: 3782–96. [Google Scholar] [CrossRef]
  95. Stambuli, P. Kalogna. 1998. Causes and consequences of the 1982 world debt crisis. Predoctoral Research Paper Department of Economics University of Surrey Guildford 3: 1–35. [Google Scholar]
  96. Stefanidis, Georgios. 2023. IMF lending in Sovereign Default. Macroeconomic Dynamics 28: 1–35. [Google Scholar] [CrossRef]
  97. Taffler, R. J., and B. Abassi. 1984. A model for predicting debt-servicing problems in developing countries. Royal Statistical Society 147: 541–68. [Google Scholar] [CrossRef]
  98. Thacker, Strom. 1999. The high politics of IMF lending. World Politics 52: 38–75. [Google Scholar] [CrossRef]
  99. Wells, Robin. 1993. Tolerance of arrearages: How IMF loan policy can affect debt reduction. The American Economic Review 83: 621–33. [Google Scholar]
  100. Zeaiter, Hussein. 2015a. Debt Arrears as a Signal of Sovereign Default. The Journal of Developing Areas 2: 127–36. [Google Scholar] [CrossRef]
  101. Zeaiter, Hussein. 2015b. Sovereign Debt Defaults: Evidence from Developing Countries. Journal of Economic Research 20: 199–230. [Google Scholar]
  102. Zeaiter, Hussein, and Raed El-Khalil. 2016. Extreme bounds of sovereign defaults: Evidence from the MENA region. International Review of Economics & Finance 41: 391–410. [Google Scholar]
Table 1. The literature on explanatory variables.
Table 1. The literature on explanatory variables.
Explanatory VariablesObserved Effect on Debt Default
Significant, PositiveSignificant, NegativeInsignificant
Reserves/GDP Edwards (1984)
Taffler and Abassi (1984)
Rahnama-Moghadam (1995)
Thacker (1999)
Bird and Rowlands (2001)
Catao and Sutton (2002)
Georgievska et al. (2008)
Afonso et al. (2010)
Hilscher and Nosbusch (2010)
Kışla et al. (2022)
Ghulam and Saunby (2023)
Total external debt/GDP (GNP)Edwards (1984)Rivoli and Brewer (1997)Taffler and Abassi (1984)
Callier (1985)Lau and Lee (2016)Karayalcin and Temel (1988)
Feder and Uy (1985) Thacker (1999)
Solberg (1988) Oatley and Yackee (2000)
Lee (1991) Bird and Rowlands (2001)
Balkan (1992) Oatley (2002)
Li (1992) Georgievska et al. (2008)
Haque et al. (1996) Lau and Lee (2016)
Aylward and Thorne (1998) Gokmenoglu and Rafik (2018)
Detragiache and Spilimbergo (2001)
Peter (2002)
Kruger and Messmacher (2004)
Kohlscheen (2010)
Hilscher and Nosbusch (2010)
Celasun and Harms (2010)
Carlos (2011)
Ghulam and Saunby (2023)
Total external debt/exportsEuh (1979) Berg and Sachs (1988)
McFadden et al. (1985) Karayalcin and Temel (1988)
Hajivassiliou (1987) Oatley and Yackee (2000)
Hajivassiliou (1989) Oatley (2002)
Elmore and McKenzie (1992) Lee (2009)
Hajivassiliou (1994)
Marashaden (1997)
Peter (2002)
Kruger and Messmacher (2004)
Interest service (due)/exportsHajivassiliou (1989) Hajivassiliou (1987)
Li (1992)
Hajivassiliou (1994)
Marashaden (1997)
Total debt service (due)/exportsFrank and Cline (1971)Elmore and McKenzie (1992)Saini and Bates (1984)
Sargen (1977) Taffler and Abassi (1984)
Euh (1979)
Feder et al. (1981) Rahnama-Moghadam (1995)
Cline (1984) Aylward and Thorne (1998)
Edwards (1984) Oatley (2002)
McFadden et al. (1985)
Hajivassiliou (1987)
Berg and Sachs (1988)
Karayalcin and Temel (1988)
Solberg (1988)
Snider (1990)
Balkan (1992)
Marashaden (1997)
Rivoli and Brewer (1997)
Thacker (1999)
Detragiache and Spilimbergo (2001)
Bird and Rowlands (2001)
Catao and Sutton (2002)
Karagol and Sezgin (2004)
Georgievska et al. (2008)
Kohlscheen (2010)
Real effective exchange rateSargen (1977)Ghulam and Saunby (2023)Edwards (1984)
Solberg (1988) Hajivassiliou (1994)
Bird and Rowlands (2001) Haque et al. (1996)
Catao and Sutton (2002) Al Fawwaz (2016)
Peter (2002) Gokmenoglu and Rafik (2018)
Afonso et al. (2010)
Reserves/importsHajivassiliou (1987)Frank and Cline (1971)Sargen (1977)
Rivoli and Brewer (1997)Feder et al. (1981)Euh (1979)
Edwards (1984)Saini and Bates (1984)
Feder and Uy (1985)Hajivassiliou (1989)
Karayalcin and Temel (1988)Balkan (1992)
McFadden et al. (1985)Hajivassiliou (1994)
Elmore and McKenzie (1992)Aylward and Thorne (1998)
Li (1992)Detragiache and Spilimbergo (2001)
Haque et al. (1996)Celasun and Harms (2010)
Marashaden (1997)
Oatley and Yackee (2000)
Bird and Rowlands (2001)
Kohlscheen (2010)
Exports/GDPGhulam and Saunby (2023)Feder et al. (1981)Cline (1984)
Karagol and Sezgin (2004)McFadden et al. (1985)
Citron and Nickelsburg (1987)
Hajivassiliou (1987)
Balkan (1992)
Elmore and McKenzie (1992)
Rivoli and Brewer (1997)
Aylward and Thorne (1998)
Imports/GDPMcFadden et al. (1985)Solberg (1988)Frank and Cline (1971)
Elmore and McKenzie (1992) Hajivassiliou (1987)
Aylward and Thorne (1998) Edwards (1984)
Saini and Bates (1984)
Karayalcin and Temel (1988)
Current account balance/GDP (exports)Edwards (1984)Bird and Rowlands (2001)Callier (1985)
Hajivassiliou (1989)Peter (2002)Karayalcin and Temel (1988)
Oatley (2002)Kruger and Messmacher (2004)Hajivassiliou (1994)
Haque et al. (1996)
Marashaden (1997)
Thacker (1999)
Detragiache and Spilimbergo (2001)
Oatley and Yackee (2000)
Lee (2009)
Panizza et al. (2009)
Money and quasi money/international reserves (GDP)Sargen (1977) Cline (1984)
Saini and Bates (1984) Citron and Nickelsburg (1987)
Callier (1985) Balkan (1992)
Hajivassiliou (1994) Thacker (1999)
Peter (2002)
GDP (GNP) per capitaRahnama-Moghadam (1995)Feder and Uy (1985)Frank and Cline (1971)
Oatley and Yackee (2000)McFadden et al. (1985)Sargen (1977)
Celasun and Harms (2010)Berg and Sachs (1988)Euh (1979)
Berg and Streitz (2016)Aylward and Thorne (1998)Edwards (1984)
Thacker (1999)Saini and Bates (1984)
Bandiera et al. (2010)Hajivassiliou (1987)
Kışla et al. (2022)Karayalcin and Temel (1988)
Balkan (1992)
Elmore and McKenzie (1992)
Hajivassiliou (1994)
Bird and Rowlands (2001)
Panizza et al. (2009)
Gartner et al. (2011)
Al Fawwaz (2016)
Lau and Lee (2016)
González-Fernández and González-Velasco (2018a)
Gokmenoglu and Rafik (2018)
Chronopoulos et al. (2019)
GDP growth rate Sargen (1977)Edwards (1984)
Euh (1979)McFadden et al. (1985)
Feder and Uy (1985)Hajivassiliou (1987)
Lee (1991)Karayalcin and Temel (1988)
Li (1992)Bird and Rowlands (2001)
Haque et al. (1996)Lee (2009)
Detragiache and Spilimbergo (2001)Gartner et al. (2011)
Catao and Sutton (2002)
Kruger and Messmacher (2004)
Afonso et al. (2010)
Hilscher and Nosbusch (2010)
Kohlscheen (2010)
Celasun and Harms (2010)
Ojeda-Joya and Gómez-González (2012)
Inflation rateSargen (1977)Lau and Lee (2016)Hajivassiliou (1994)
Saini and Bates (1984)Ghulam and Saunby (2023)Balkan (1992)
Aylward and Thorne (1998) Celasun and Harms (2010)
Peter (2002) Panizza et al. (2009)
Bandiera et al. (2010) Edwards (1984)
Gartner et al. (2011) Stamatopoulos et al. (2016)
Kışla et al. (2022)
Das and Mandal (2022)
Das (2023)
Outward orientation (openness)Ojeda-Joya and Gómez-González (2012)Bandiera et al. (2010)Callier (1985)
Maltritz and Molchanov (2014) McFadden et al. (1985)
Thacker (1999)
Celasun and Harms (2010)
Al Fawwaz (2016)
Real interest rate on international lendingLee (1991)Bandiera et al. (2010)Callier (1985)
Detragiache and Spilimbergo (2001) McFadden et al. (1985)
Peter (2002) Hajivassiliou (1994)
Semik and Zimmermann (2021) Bird and Rowlands (2001)
Lau and Lee (2016)
Government expenditure Gokmenoglu and Rafik (2018)
González-Fernández and González-Velasco (2018b)
Short-term debt to reserves ratioBandiera et al. (2010)
Governmental debt/GDPStamatopoulos et al. (2016) Bai et al. (2015)
González-Fernández and González-Velasco (2018b)
Total debt/GDPOjeda-Joya and Gómez-González (2012) Berg and Streitz (2016)
Chronopoulos et al. (2019)
Kışla et al. (2022)
Terms of tradeAl Fawwaz (2016)
Regional risk aversionGhulam and Saunby (2023)
CDSQian and Luo (2015) Kadiric (2022)
Stamatopoulos et al. (2016)
Fiscal space Al Fawwaz (2016)
Stamatopoulos et al. (2016)
Table 2. List of variables: W, X, Q, and Z.
Table 2. List of variables: W, X, Q, and Z.
Variable NameDescription
Dependent Variable
IMFcredit/EDUse of IMF credit to the external debt
X-variable
Res/EDReserves to external debt
Q-Variables
ArrearsInterest and principal arrears to external debt
ResDebt/EDrescheduled debt to external debt
PrinArears/EDPrincipal arrears to external debt
ED/GNPExternal debt stocks of GNP
Res/ImpTotal reserves in months of imports
OpenessExports + imports to GDP ratio
TDS/ExpTotal debt service of exports of goods
InfInflation based on consumer prices annual growth
MoneyMoney and quasi money M2 to total reserves
IntrateReal lending interest rate
Imp/GDPImports of goods and services of GDP
Exp/GDPExports of goods and services of GDP
GGDPGDP annual growth
ED/ExpExternal debt stocks of exports of goods
PCGNPGNP per capita PPP current international currency
ExRateReal effective exchange rate index (CPI 2000)
CA/GDPCurrent account balance of GDP
EthTenEthnic tensions (ICRG)
MilPolMilitary involvement in politics (ICRG)
CorrCorruption (ICRG)
DemocDemocratic accountability (ICRG)
GovtStabGovernment stability (ICRG)
LTServ/ResLong-term debt service to reserves
Z-variables
LTServ/GDPLong-term debt service to GDP
Crdt/GDPDomestic credit to private sector of GDP
CapflowPrivate capital flows total of GDP
ToTNet barter terms of trade growth
DebtServTotal debt service of GNP
STD/ResShort-term debt of total reserves
PR/expPrincipal repayments to exports
GExpExports of goods and services annual growth
FisBalFiscal balance to GDP
VGNPStandard deviation of the previous seven years of GNP
MilExpnMilitary expenditure of central government
GovDebtCentral government debt total of GDP
Table 3. Sensitivity analysis for the X-variables (Res/ED), all Z-variables, all observations (dependent variable: IMFcredit/ED).
Table 3. Sensitivity analysis for the X-variables (Res/ED), all Z-variables, all observations (dependent variable: IMFcredit/ED).
X-Variable β tp-Value0.95 C.I.VIFZ-VariablesRobust?
Res/EDMin0.010611.45990.0554−0.00120.02231.28ExRate, MilExpn
Base0.01367962.7300.01360.01364 Yes
Max0.05336.53010.0967−0.05040.15691.6Res/Imp, GovDebt/GDP
Table 4. Sensitivity analysis for the Q-variables, panel including all Z-variable except political indices, all observations (dependent variable: IMFcredit/ED, X-variable Res/ED).
Table 4. Sensitivity analysis for the Q-variables, panel including all Z-variable except political indices, all observations (dependent variable: IMFcredit/ED, X-variable Res/ED).
Q-Variables β Tp-Value0.95 C.I.VIFZ-VariablesRobust?
Min0.00−9.800.060.000.001.12CA/GDPTOT
ED/GNPBase−0.02−1.040.30−0.060.02 No
Max0.00−6.820.090.000.001.11Res/ImpFisBal
Min−0.02−8.420.08−0.050.011.59IntRateGovDebt/GDP
Res/ImpBase0.120.450.65−0.400.64 No
Max−0.01−6.330.10−0.020.011.25ED/ExpFisBal
Min−0.47−6.870.09−1.350.407.61ExRatePR/Exp
OpennessBase5.362.060.040.2610.45 Yes
Max−0.21−7.430.09−0.570.158.53PR/ExpPCGNP
Min0.00−6.610.10−0.010.001.07MoneyTOT
TDS/ExpBase−0.05−0.830.41−0.170.07 No
Max0.00−6.370.100.000.001.11CapFlowTOT
Min0.016.870.09−0.010.031.14ExRateTOT
InfBase0.010.240.81−0.070.09 No
Max0.016.340.10−0.010.031.10ExRateCA/GDP
Min0.00−7.650.08−0.010.001.44OpennessPrinArrears/ED
MoneyBase−0.03−0.320.75−0.200.14 No
Max0.00−6.800.09−0.010.001.17PrinArrears/EDTOT
Min0.003.590.17−0.010.011.07Exp/GDPTOT
IntRateBase−0.08−0.410.68−0.490.32 No
Max0.014.660.13−0.010.021.22GovDebt/GDPVGNP
Min0.377.140.09−0.291.028.53OpennessCapFlow
Imp/GDPBase22.754.970.0013.7731.73 Yes
Max0.776.920.09−0.642.187.61OpennessExRate
Min−0.01−6.840.09−0.020.014.50OpennessExRate
Exp/GDPBase−0.08−1.460.15−0.180.03 No
Max0.00−6.320.100.000.001.13ED/ExpTOT
Min0.00−2.230.27−0.020.021.02ExRateCrdt/GDP
GGDPBase0.040.210.84−0.310.39 No
Max0.002.610.23−0.020.031.77ExRateGovDebt/GDP
Min−0.03−7.660.08−0.070.028.56Imp/GDPOpenness
ED/ExpBase−0.85−1.070.28−2.390.70 No
Max−0.02−6.410.10−0.070.021.58CA/GDPPCGNP
Min0.00−9.260.070.000.001.34ExRateMilExpn
PCGNPBase0.000.200.840.000.00 Yes
Max0.00−6.370.100.000.001.20Exp/GDPGovDebt/GDP
Min0.00−3.560.17−0.010.001.87LTServ/GDPGovDebt/GDP
ExRateBase−0.02−1.040.30−0.060.02 No
Max0.004.160.15−0.010.011.29GexpMilExpn
Min−0.01−6.440.10−0.030.011.02ExRateGEXP
CA/GDPBase1.3622.420.001.241.48 No
Max0.00−6.380.10−0.010.001.40TDS/ExpSTD/Res
Table 5. Sensitivity analysis for the Q-variables, panel including all Z-variables, political indices (dependent variable: IMFcredit/ED, X-variable Res/ED).
Table 5. Sensitivity analysis for the Q-variables, panel including all Z-variables, political indices (dependent variable: IMFcredit/ED, X-variable Res/ED).
Q-Variable β Tp-Value0.95 C.I.VIFZ-VariablesRobust?
Min−0.03−4.140.15−0.110.061.15GExpTOT
CorrBase−0.01−0.710.48−0.030.01 No
Max0.043.420.18−0.120.201.36PrinArrears/EDmilpol
Min0.013.120.20−0.040.071.09TOTED/GNP
DemocBase0.000.940.35−0.010.02 No
Max0.023.340.19−0.070.121.71GovDebt/GDPRes/Imp
Min−0.01−1.600.36−0.130.101.02PrinArrears/EDExRate
GovStabBase−0.01−0.960.34−0.020.01 No
Max−0.01−1.390.40−0.060.051.01LTServ/ResExp/GDP
Min−0.02−4.540.14−0.080.041.23DemocImp/GDP
MilPolBase−0.02−3.680.00−0.03−0.01 No
Max−0.01−3.460.18−0.040.021.09TOTExp/GDP
Min−0.03−4.710.13−0.100.051.18InfImp/GDP
EthTenBase−0.02−3.940.00−0.03−0.01 No
Max−0.01−3.150.20−0.070.041.1DemocTOT
Table 6. Sensitivity analysis for the Q-variables, panel including all Z-variables (dependent variable: IMFcredit/ED, X-variable Res/ED).
Table 6. Sensitivity analysis for the Q-variables, panel including all Z-variables (dependent variable: IMFcredit/ED, X-variable Res/ED).
Q-Variables β Tp-Value0.95 C.IVIFZ-VariablesRobust?
Min0.00−9.800.060.000.001.12CA/GDPTOT
ED/GNPBase−0.02−1.040.30−0.060.02 yes
Max0.00−6.820.090.000.001.11Res/ImpFisBal
Min−0.02−6.830.09−0.060.021.71GovDebt/GDPDemoc
Res/ImpBase0.120.450.65−0.400.64 yes
Max−0.01−6.460.10−0.020.011.05Res/ImpCapFlow
Min−0.06−3.400.18−0.270.161.1Res/ImpIntRate
OpennessBase5.362.060.040.2610.45 yes
Max0.064.950.13−0.090.201.16GovDebt/GDPCapFlow
Min0.00−6.610.10−0.010.001.07MoneyTOT
TDS/ExpBase−0.05−0.830.41−0.170.07 yes
Max0.00−6.370.100.000.001.11CapFlowTOT
Min0.003.120.200.000.012.41TDS/ExpSTD/Res
InfBase0.010.240.81−0.070.09 no
Max0.014.150.15−0.010.021.19GovDebt/GDPDemoc
Min−0.01−4.780.13−0.050.021.27GovDebt/GDPCapFlow
MoneyBase−0.03−0.320.75−0.200.14 no
Max0.00−3.330.19−0.010.001.57Res/ImpTOT
Min0.003.770.16−0.010.011.08TOTGExp
IntRateBase−0.08−0.410.68−0.490.32 no
Max0.014.660.13−0.010.021.22GovDebt/GDPVGNP
Min0.00−3.040.20−0.010.011.05PCGNPTOT
GGDPBase0.040.210.84−0.310.39 no
Max0.002.070.29−0.020.031.32GovDebt/GDPCorr
Min0.00−6.430.100.000.001.14IntRateCorr
PCGNPBase0.000.200.840.000.00 yes
Max0.00−6.510.100.000.001.2GovDebt/GDPCapFlow
Min−0.01−10.850.06−0.020.001.32GovDebt/GDPCorr
CA/GDPBase1.3622.420.001.241.48 yes
Max0.00−6.380.10−0.010.001.4TDS/ExpSTD/Res
Min−0.03−4.140.15−0.110.061.15TOTGexp
CorrBase−0.01−0.710.48−0.030.01 no
Max−0.03−4.140.15−0.110.061.15TOTGexp
Min0.013.120.20−0.040.071.09ED/GNPTOT
DemocBase0.000.940.35−0.010.02 no
Max0.023.340.19−0.070.121.71Res/ImpGovDebt/GDP
Min−0.01−1.510.37−0.060.051.18PCGNPFisBal
GovtStabBase−0.01−0.960.34−0.020.01 no
Max−0.01−1.450.38−0.060.051.06TDS/ExpFisBal
Min−0.02−4.290.15−0.080.041.21DemocMoney
MilPolBase−0.02−3.680.00−0.03−0.01 no
Max−0.01−3.910.16−0.040.021.15TOTFisBal
Min−0.02−4.040.15−0.090.041.74MoneyRes/Imp
EthTenBase−0.02−3.940.00−0.03−0.01 no
Max−0.01−3.150.20−0.070.041.1DemocTOT
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Maddah, L.; Sherry, H.; Zeaiter, H. Economic and Political Determinants of Sovereign Default and IMF Credit Use: A Robustness Assessment Post 2010. Economies 2024, 12, 181. https://doi.org/10.3390/economies12070181

AMA Style

Maddah L, Sherry H, Zeaiter H. Economic and Political Determinants of Sovereign Default and IMF Credit Use: A Robustness Assessment Post 2010. Economies. 2024; 12(7):181. https://doi.org/10.3390/economies12070181

Chicago/Turabian Style

Maddah, Lina, Hassan Sherry, and Hussein Zeaiter. 2024. "Economic and Political Determinants of Sovereign Default and IMF Credit Use: A Robustness Assessment Post 2010" Economies 12, no. 7: 181. https://doi.org/10.3390/economies12070181

APA Style

Maddah, L., Sherry, H., & Zeaiter, H. (2024). Economic and Political Determinants of Sovereign Default and IMF Credit Use: A Robustness Assessment Post 2010. Economies, 12(7), 181. https://doi.org/10.3390/economies12070181

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop