Foreign banks and sovereign credit ratings: Reputational capital in sovereign debt markets

Abstract

Sovereign credit ratings importantly influence the borrowing costs of governments in international capital markets. Yet, there is limited understanding of how credit-rating agencies determine sovereign bond ratings. I provide theoretical justification and empirical evidence to support the proposition that a substantial presence of established global banks, acting as foreign direct investors, enhances the perceived creditworthiness of the host countries that have weak domestic institutions. Foreign banks can render the host countries’ commitments to make good on their debt obligations more credible by encouraging the transparency in the financial system, disciplining their fiscal policies, and mitigating the incentives for and impact of bank bailouts. Statistical evidence from countries in emerging Europe shows that countries with high levels of foreign bank ownership tend to be assigned better sovereign credit ratings and find it easier to obtain credit at lower interest rates in sovereign bond markets. My findings are robust to various estimation techniques, to extensive controls for alternative determinants of credit ratings, for the endogeneity of foreign bank entry, and for sample-selection bias. Interviews with bankers and senior analysts at credit-rating agencies were used to complement quantitative analyses. This article is the first attempt in the literature on sovereign borrowing and debt to examine whether private market agents, such as global banks, can enhance the government’s international creditworthiness.

Keywords

Crisis international finance global governance globalization economic interdependence political economy

Introduction

Sovereign credit ratings are assessments of a government’s willingness and ability to repay its sovereign debt. They are estimates of the probability of default. Ratings are a key determinant of the cost of government borrowing: the higher the perceived sovereign credit risk, the higher the risk premium in the form of higher interest rates a government faces in international credit markets. Why do some countries receive more favorable sovereign credit ratings than others? How can a government convince credit-rating agencies (CRAs) of its intentions to pay the obligations it assumes? In this article, I test the proposition that multinational banks, acting as foreign direct investors, positively influence the perceptions of the credit quality of the host country’s sovereign debt by CRAs. I argue that foreign banks may serve as credibility-building mechanisms. These banks can render the host country’s commitments to make good on its debt obligations more credible by disciplining its fiscal policies and mitigating the incentives for and impact of bank bailouts. Their presence thus creates the perception that the host country will honor its debt obligations. However, foreign banks can also generate pressures on the host country’s government to change its behavior by threatening to exit.

Less developed countries, in particular, face a serious challenge in signaling to sovereign bond markets that they are creditworthy borrowers (Mosley, 2003). Many of them have legacies of wars and conflicts, new or weak domestic institutions, and no credit history; hence, they are unable to signal a credible commitment of their intent and ability to repay sovereign debt. As these countries have a poor endogenously created reputation for honoring their promises, they face difficulties in raising money in sovereign bond markets. Furthermore, when a government lacks credibility in international financial markets, it will be charged a risk premium in the form of high interest rates to compensate for the risk of default (Gros, 2003: 2).¹ The costs of borrowing in sovereign bond markets matter a lot for capital-scarce transition and developing countries that rely on foreign funds to finance their general expenses and development projects.

This article contributes to several strands of the International Relations literature. First, it provides a significant extension of the literature on how countries accumulate reputational capital in international relations. Reputation is particularly important in sovereign borrowing, in which an enforcement mechanism is limited because collateral in the sense used in domestic contracts is “irrelevant” (Bulow and Rogoff, 1989).² In his influential study, Tomz (2007: 14) argues that reputational concerns in international lending “motivate countries to repay and inspire investors to lend”; hence, countries repay their debt to accumulate reputational capital. He develops a dynamic theory of reputation in which investors continually update their beliefs about the type of government they are dealing with. The prior literature has focused predominantly on the reputational costs of sovereign defaults, however. Furthermore, it has ignored the role of private market agents. Tomz (2007), for instance, sees lenders in sovereign debt markets as “atomized” actors. This article examines how governments can use foreign private banks to build their reputation in sovereign bond markets.³

Second, this article also fills the gap in the literature on financial globalization. One of the liveliest debates in this literature involves the implications of the growing foreign bank presence for less-developed countries. Before the 2008 global financial crisis, the general consensus in the literature on banking globalization was that the benefits of foreign bank presence were more significant than the costs in many respects.⁴ Scholars have shown that foreign banks enhance competition and improve the efficiency of financial intermediation in the local banking system, stimulate financial regulatory reforms, and improve access of the host country to international markets.⁵ Nonetheless, this literature remains largely silent on the implications of foreign banks for the creditworthiness of their host countries.

My empirical analysis of the influence of foreign banks on sovereign credit ratings focuses on the countries of Eastern Europe (EE). These countries are a particularly good test case. Having a thin track record of capitalist functioning, EE countries have faced a severe credibility problem in international markets after the collapse of communism.⁶ The development of their financial systems provides an ideal setting because all EE countries started the transition process in the early 1990s with a monobank system — in which money supply and credit allocation functions were located within a single institution — and went through a transformation from a state-dominated banking system to a market-oriented one. It is also worth noting that the banking systems in emerging European countries have witnessed the most significant growth of foreign ownership, with the share of assets held by Western banks increasing from an average of 17% in 1996 to 52% in 2005 (Bonin and Schnabel, 2011). This masks substantial differences across emerging European countries, however.

Using data on all EE countries rated by CRAs from 1995 to 2006, I demonstrate that foreign banks have a substantial positive effect on countries’ reputation for honoring debt commitments, and therefore on their sovereign credit ratings. This finding is robust to numerous controls, changes of model specification, and various estimation techniques. Controlling for economic reforms does not detract from the association between foreign bank ownership and more favorable bond ratings. The problem of selection is addressed by estimating a selection model to model, first, the reason why a country enters the sovereign credit market and, given that the country is rated, what determines its rating by the major CRAs. I also used instrumental variable regressions to deal with the potential endogeneity of foreign bank entry. Finally, I interviewed senior sovereign analysts at Standard & Poor’s and Moody’s to better understand how these CRAs assess the creditworthiness of sovereign borrowers, as well as bankers at large European multinational banks.

The economics and politics of sovereign credit ratings

Initially, CRAs issued ratings mainly for US railroads and municipalities, but their importance grew as the bond market expanded to include issues by utilities, manufacturers, and national governments (Sylla, 2002: 33–34).⁷ The sovereign rating was not a major focus of the CRAs until the 1980s when the debt crisis in Latin America led to the issue of Brady bonds (defaulted Latin American debt repackaged as bonds) (Sinclair, 2005: 137–38). The influence of CRAs has increased in recent decades as international credit markets have become the main source of finance (Sinclair, 2005: 10). Sovereign and corporate borrowers seek credit ratings to facilitate their market access because investors have a greater trust in rated securities than in unrated securities, even if they bear a similar credit risk (Cantor and Packer, 1996: 38).

King and Sinclair (2003: 353) posit that CRAs are “hybrid forms of authority, operating between the state and the market, which have acquired public authority due to their professional expertise, specialist knowledge, reputation and acceptance by market actors.” The big three CRAs are Standard & Poor’s, Moody’s, and Fitch Ratings. CRAs’ assessments of a government’s credit risk are both quantitative and qualitative. The quantitative analysis assesses the importance of economic and financial performance. However, it is difficult to identify the relationship between the various criteria that CRAs enumerate in their statements because some of them are qualitative and non-quantifiable (Cantor and Packer, 1996: 39).⁸ This makes sovereign rating “one of the most subjective areas of rating” (Sinclair, 2005: 139).

Sovereign credit ratings are solicited by governments that want to attract investors and obtain credit to finance their fiscal expenditures. Once this process is started, the CRAs continue to monitor a country’s economic and political developments to assess its ability to service and repay its debt. Although CRAs claim that they simply provide an independent opinion on credit quality, their ratings are very influential. Investors pay close attention to sovereign ratings when investing capital, particularly in developing countries where the risk is greater and information is scarcer and less reliable. Sovereign ratings directly influence the ratings assigned to local municipalities, provincial governments, and domestic banks and firms (Cantor and Packer, 1996), as well as sovereign bond yield spreads. For example, a one-notch upgrade in a credit rating decreases the spread on average by 14% in emerging market economies (Sy, 2002: 382).

In the first systematic study of the determinants of the sovereign ratings assigned by Moody’s and Standard & Poor’s, Cantor and Packer (1996) found that the credit ratings were determined by a small number of economic factors that included per capita income (accounting for 80% of the variance), external debt, economic development, and a country’s history of default.⁹ Another line of research explored the relationship between political regime and credit ratings. According to the “democratic advantage” argument, representative legal and political institutions, such as constitutional checks and balances or strong property rights, enhance a nation’s market access to credit at low interest rates because they provide a commitment technology and increase the probability that the government will repay its debts. As a result, democracies can sell more sovereign bonds and are charged lower risk premiums than non-democracies (Biglaiser and Staats, 2012; Butler and Fauver, 2006; Schultz and Weingast, 2003; Stasavage, 2007). Other studies, however, cast doubt on the validity of the democratic advantage thesis. Saiegh (2005) found that democracies were more likely to reschedule their debt and paid interest rates that were similar to those of their authoritarian counterparts. Archer, Biglaiser, and DeRouen (2007) argue that democracies receive no more favorable sovereign credit ratings than otherwise similar autocracies. Beaulieu, Cox, and Saiegh (2012) criticize the previous literature for failing to account for selection effects: governments must seek credit ratings, and authoritarian governments are less likely to choose to enter international bond markets than are democracies.

Yet another strand of the literature concurs that national governments can improve their credibility by delegating powers internationally (Dreher and Voigt, 2011). Hauner, Jonas, and Kumar (2007) and Gray (2009) suggest that European Union (EU) accession improves the credibility of the economic reforms and policies of a country that joins the club, thus decreasing the perceived default risk by bond investors.

Although we have learned much from the scholarship examining the role of economic factors, domestic political institutions, and membership in international organizations in sovereign borrowing, this study represents a significant departure by exploring whether the presence of foreign banks can infuse government reputation, thus boosting a country’s standing with CRAs.

Foreign banks as an exogenous source of reputational capital

In developing my theoretical framework, I draw on the work of Hellman and Murdock (1998), which focuses on the role of governments in creating the “reputational capital” of (small) domestic banks in (underdeveloped) domestic financial markets. I extend this framework to explore reputation creation by governments in sovereign debt markets. A financial transaction is the exchange of an asset for a promise of the borrower to make a payment sometime in the future. The value of this promise, and thus the value of the transaction to the lender, depends on the quality of the promise. The latter depends on the reputation of the borrower, on the ability of the lender to monitor and control the borrower, and on the consequences of a default for the borrower. One of the key features of any financial transaction is the “reputational capital” possessed by agents, defined by Hellman and Murdock (1998: 11) as “the incremental profits that accrue to borrowers when they have a good reputation.”

This article tests the proposition that a substantial presence of well-established multinational banks, as foreign direct investors, in a local financial system is a recipe for building a host government’s reputational capital, which is reflected in better sovereign credit ratings.¹⁰ In low-information environments, CRAs and bond investors look for signals of a sovereign borrower’s intent and ability to repay its debts. The presence of large, internationally oriented banks with good reputations resulting from their long-term participation in the financial markets of developed countries signals a host government’s commitment to openness and economic reforms, thus rendering the government’s promise to honor its sovereign obligations credible. The opening of the domestic banking sector to foreign banks entails political costs because doing so erodes the rents of domestic financial institutions. Consequently, governments that allow the entry of foreign banks send a powerful signal that they value financial stability and reputation over the patronage of domestic interest groups.

The reasoning behind this is that rating methodologies may reflect “mental frameworks” shared between CRAs and international financial market participants on what constitutes appropriate policies on the part of borrowers.¹¹ This “signal” is credible because there is a commonality of interpretation of the implications of foreign bank entry for an emerging market country among CRAs and bond investors.¹² Foreign bank ownership thus has a signaling effect of financial trustworthiness and commitment to the fiscal responsibility by host countries. There is abundant empirical evidence in the literature on global banking showing the real benefits of foreign bank entry to host countries’ financial systems and economies at large.¹³ Prior studies (e.g. Hau et al., 2013) and my interview evidence also suggest that CRAs systematically assign more positive ratings to large global banks, citing substantial governmental guarantees or overlapping business interests as some of the reasons behind this rating privilege.¹⁴ CRAs and bond investors know that large multinational banks can secure certain economic policies in host countries since they can punish the defaulters by threatening to exit; thus, these banks may be able to affect both the capacity and willingness of host governments to pay their debts. Consequently, CRAs may rely on both signaling and the information about the real changes that foreign bank entry entails to draw inferences about the creditworthiness of host-country governments.

There is also historical support for the idea that international banks are relevant to countries lacking their own reputation in sovereign bond markets. Flandreau and Flores (2009: 679) argue that in international credit markets in the 19th-century, prestigious financial intermediaries (e.g. the Rothschilds and Barings), as underwriters of sovereign bonds, loaned their excellent reputations to those sovereign borrowers that faithfully implemented the “policy adjustments” required by these banks.¹⁵ Instead of relying on the reputation of sovereign borrowers, bond purchasers relied on the reputations of these financial intermediaries. Prestigious bankers were also able to prevent countries from defaulting due to their access to funds. Stasavage (2007) puts forward the idea that merchants, who were the primary source of credit for sovereign borrowers and a dominant force within the political elite, facilitated the access of small European city-states in the Middle Ages to credit at low interest rates. This is primarily because merchants “could ensure that decisions were taken to prioritize servicing debt rather than opting for opportunistic defaults” (Stasavage, 2007: 491).

The literature on multinational banking identifies three mechanisms through which the effect of foreign banks on the reputational capital of host governments in sovereign debt markets might operate. First, foreign bank presence is said to make the policy commitments of countries with underdeveloped and opaque financial systems more credible by increasing the transparency of the balance sheets of domestic financial institutions (Gros, 2003). Shen, Huang, and Hasan (2012) posit that there is a severe asymmetric information problem between CRAs and domestic banks in countries with weak institutional quality. The lack of transparency and poor quality of the financial accounts of such banks leads CRAs to question the credibility of their financial statements and assign these banks lower ratings. Furthermore, in many countries, regulatory authorities protect their “privileged data access” and are reluctant to share important bank data publicly or with other institutions (Hau et al., 2013: 26).

Foreign banks from environments of high institutional quality in developed countries tend to have more transparent accounts than most domestic banks in developing countries (Gros, 2003: 11). However, foreign banks do not have access to the same insider information as domestic banks; hence, they have incentives to push for improvements in the transparency and quality of information in host countries by encouraging better standards in auditing and accounting, as well as stricter disclosure requirements (Goldberg, 2009; Mishkin, 2006). This is well illustrated by the efforts of Belgian, Swedish, and Austrian banks that helped establish credit-evaluation systems in emerging European countries to facilitate the access of small firms to credit in their local subsidiaries (De Haas and Naaborg, 2006). Similarly, Levine (2001) shows that US banks encouraged the establishment of accounting and auditing firms and credit bureaus in Mexico. More available information on the business operations of local banks increases the transparency of their balance sheets, which prevents the accumulation of bad debts; hence, it should reduce the fiscal shocks emanating from the domestic financial system (Gros, 2003: 14). Greater transparency in the financial sector also decreases asymmetries of information, allowing CRAs and bond traders to monitor government policies and evaluate policy changes (Hauner et al., 2007).¹⁶

Second, foreign banks, headquartered in countries with better regulation and oversight, can prevent the fiscal irresponsibility of host governments by being less subject to manipulation by these governments (Goldberg, 2009). Sound fiscal policy means that market players expect that budgetary objectives will be attained and that a government will be able to service its debt without resorting to inflationary finance or default (Gros, 2003).¹⁷ Governments are able to use reserve and liquidity requirements to force-feed government bonds to domestic (often state-owned) banks as a way of financing budget deficits (Dinç, 2005). Forced purchases of government bonds and lending to favored political constituents by domestic banks lead to the accumulation of bad loans. Government bailouts of banks whose portfolios are burdened by large volumes of bad loans subsequently increase budget deficits, thus negatively affecting the government’s ability to service its debt. By contrast, it is politically difficult to convince the public that a foreign bank needs to be bailed out by the local government (Gros, 2003: 12).

Foreign banks reluctant to fund a host government by holding its debt can ask their home governments for support, thus making this an issue of international trade and investment (Tschoegl, 2005: 225–226). Alternatively, foreign banks may themselves engage in a voluntary acquisition of sovereign bonds issued by host governments, which enables them to monitor countries’ economic policies, collect information, and force them to pay their sovereign debt because banks can cut off a defaulting country from future loans or investments. Furthermore, foreign banks can mitigate problems with related lending to political cronies because these banks do not have domestic political ties (Giannetti and Ongena, 2009). Therefore, they follow an arm’s-length approach to lending rather than relationship lending (Corsetti et al., 2001: 35). Evidence from Bulgaria confirms the idea that foreign bank presence is associated with better financial governance. As Epstein (2017: 116) argues, foreign banks, which in 2008 owned 80% of the Bulgarian banking system in terms of assets, “reinforced” the independence of the Bulgarian central bank; thus, “in an otherwise corrupt country, bank governance in Bulgaria stood out for being professionalized, conservative, risk-averse, and independent.”¹⁸

Financial and fiscal stability are linked since banking crises wreak havoc on public finances. Takeovers of former government-owned banks by foreign banks should lead to reduced levels of bad loans and fewer banking crises (Tschoegl, 2005). If a country with a substantial foreign bank ownership experienced a banking crisis, this crisis was often limited to the need to bail out the depositors of a domestic (often state-owned) bank (Tschoegl, 2005: 213–214). The case of Parex Bank, the largest locally owned bank in Latvia, illustrates this well. Parex Bank, which engaged in excessive borrowing from abroad that dramatically increased the country’s foreign exchange exposure, was taken over by the Latvian government in November 2008 after a run on deposits (Epstein, 2014). According to the Financial Times, the bank “succumbed to rumours because it had no foreign parent.”¹⁹

Third, foreign banks can act as providers of guarantees: they strengthen financial stability in host countries by providing emergency liquidity in times of crisis (Tschoegl, 2005; Vogel and Winkler, 2010). Large international banks have easier access than weak domestic banks to funding in foreign currency, particularly in times of crisis (Tschoegl, 2005: 215). Parent banks provide liquidity to their foreign affiliates through internal capital markets (De Haas and Van Lelyveld, 2006).²⁰ Foreign banks tend to benefit from a “flight to quality” (depositors withdraw their savings from weak domestic banks and deposit them with foreign bank affiliates), which reduces the demand for foreign exchange due to capital flight with potentially destabilizing effects on the country’s exchange rate (Tschoegl, 2005: 215).

The ability of multinational banks to provide emergency liquidity assistance stabilizes the host-country banking system and lowers expectations of bailouts by a struggling local government with negative fiscal consequences. Foreign parent banks can thus replace local central banks in their role as a lender of last resort in times of sudden liquidity needs (Corsetti et al., 2001). Foreign financiers can also reduce the risk of capital flow reversals by providing a financial buffer that can effectively mitigate the negative consequences of crises on a government’s ability to service its external debt (Moreno and Villar, 2005).

Recent history abounds with examples of foreign banks acting as a lender of last resort to their affiliates. In the aftermath of the 1998 Russian financial crisis, when the brokerage subsidiaries of foreign-owned banks in Hungary experienced financial difficulties, parent banks recapitalized them (Cardenas et al., 2003). When foreign subsidiaries of West European banks suffered large losses in the aftermath of the 2008 global credit crisis, head offices quickly injected capital. Parent banks of Nordic banks provided the necessary capital and liquidity to their affiliates in the Baltic countries because they understood that “it was necessary for the parent banks to continue transferring capital to their subsidiaries to maintain their creditworthiness.”²¹ Furthermore, the Swedish and Danish central banks showed their commitment by providing a swap line of €500 million to Latvia, followed by bridging loans and provisions of liquidity to the financial system in order to prevent a run on the Latvian currency in December 2008 (Åslund and Dombrovskis, 2011; Epstein, 2017). The Swedish central bank also provided an emergency swap line to Estonia, intended to enable local authorities to provide liquidity to Swedish subsidiaries in the local currency (Epstein, 2017). In sum, my testable prediction is that conditional on entering the sovereign debt market, countries with a substantial foreign bank presence should receive higher sovereign credit ratings than otherwise comparable countries.

Interviews with sovereign analysts

To assess the plausibility of the expectations presented earlier, I conducted interviews with senior analysts at Standard & Poor’s and Moody’s in Paris and London in December 2013 and January 2014. The interviews consisted of open-ended questions regarding the relative importance of factors used by CRAs in assessing the creditworthiness of sovereign borrowers.

All credit analysts that I interviewed highlighted the importance of foreign direct investments in the banking sector in their rating decisions and mentioned some of the channels identified earlier through which foreign bank presence would positively affect their assessments of countries’ creditworthiness. Moody’s considers “banking sector risks” when assigning credit ratings and includes “the upward adjustment factor for foreign bank ownership” in its sovereign rating model because of the expectation of lower government liabilities in countries with a substantial presence of foreign bank affiliates.²² Sovereign risk analysts in both agencies suggested that foreign direct investments exert a positive effect on ratings because they represent stable sources of finance that facilitate the reduction of the current account deficit; hence, they increase the likelihood that a government will uphold its financial commitments. In my interviews, it became clear that credit analysts expected and perceived positively the commitment of parent banks to stand behind their subsidiaries and branches and provide external liquidity support during times of crisis by transferring funds within the banking group, instead of requesting a bailout from the host-country government.²³

Adams, Mathieson, and Schinasi (1999) argue that rating analysts consider political factors and processes as less important in their agencies’ ratings because it is harder to assess them than the risks associated with financial instability and crises. Sovereign analysts that I interviewed recognized the importance of the political environment, particularly the rule of law, government effectiveness, and commitment to reforms. Perhaps surprisingly, their notion of politics was not tied to regime type. Rather, bond raters look favorably at the stability of the political regime and how it translates into the ability and willingness of a sovereign to uphold its financial commitments. As a senior analyst from Standard & Poor’s stated: “The label ‘democracy’ is a too vague concept: some democracies are able to implement brave reforms, while others are completely paralyzed.”²⁴ Analysts at Standard & Poor’s and Moody’s also named economic size, wealth, growth, and the overall soundness of economic and fiscal policies as important factors in their rating decisions.

Credit analysts in both agencies recognized that EU membership exerts a positive influence on ratings. However, they also suggested that the effect of the EU is “indirect and not automatic,” but rather through its impact on institutional and fiscal factors.²⁵ The positive impact of the integration process, as bond raters remarked, reflects the ability of the EU to put pressures on accession governments to implement reforms and provide external checks and balances, whereas European funds represent an important source of capital.

Empirical analysis

Measuring ratings

A country’s credit rating is the main indirect measure of its borrowing costs. All three leading CRAs make announcements of various types, including rating changes (upgrades and downgrades), revisions of outlook (positive and negative), and reviews for future rating changes. The highest rating awarded by Standard & Poor’s is AAA, and the lowest is D/SD (default/selective default). The ratings are published with ratings outlooks ranging from Credit Watch–Positive to Credit Watch–Negative, indicating that closer attention is given for potential upgrading or downgrading in the next few months. The investment-grade ratings are BBB and above. The ratings BB+ and below are considered non-investment grade (speculative).

I use the data for foreign currency long-term credit ratings for all EE countries rated by Moody’s (16 countries), Standard & Poor’s (16 countries), and Fitch Ratings (18 countries).²⁶ As it is standard in the literature, I group the ratings into 17 categories,²⁷ in which the best ratings — AAA for Standard & Poor’s and Fitch, and Aaa for Moody’s — are assigned the value of 17 and the ratings below B– for Standard & Poor’s and Fitch, and B3 for Moody’s, are assigned the value of 1. The sample period extends from 1995 to 2006.²⁸ This time frame reflects foreign bank investment activity, which was particularly high in the late 1990s and early 2000s and again in 2006−2007 before it slowed down after the start of the global credit crisis (Claessens and Van Horen, 2012).²⁹

Foreign banks

A bank is considered as foreign owned if 50% or more of its shares are owned by foreigners (Claessens and Van Horen, 2012: 8). The primary proxy measure for the degree of foreign bank influence across host countries is the percentage of foreign bank assets among total bank assets (by country). The data on foreign banks come from Claessens and Van Horen (2008, 2012), and 1995 is the first year of data reporting.

The local presence of foreign banks has sharply increased over the past two decades. At the end of 2007, foreign banks held, on average, close to 50% of the market share in terms of loans, deposits, and profits in emerging and developing countries (Claessens and Van Horen, 2012: 5). The rate of growth in foreign bank ownership is particularly striking in emerging Europe, but foreign bank ownership is also high in Latin America and sub-Saharan Africa (reflecting past colonial linkages), with a 34% and close to 50% share of assets, respectively (Cull and Martinez Peria, 2010: 3–4). This trend has not been mirrored in East Asia and the Middle East, which have embraced foreign bank participation less enthusiastically. In China and India, state-owned banks still play a dominant role in domestic financial intermediation (Goldberg, 2009).

In 2011, the top foreign bank investors in the EE region included the Italian UniCredit and Intesa Sanpaolo, the Austrian Erste and Raiffeisen Bank, the French Société Générale, the Belgian KBC Bank, the Spanish Santander, and the Swedish Swedbank.³⁰ Austrian and Italian banks are the key players, with diversified exposures across EE countries, while Swedish banks are mainly present in the Baltic countries. Most foreign banks operating in the EE region acquired their subsidiaries via takeovers involving the privatization of previously government-owned banks.³¹ The long-term business model of these multinational banks involves “making a commitment to a transition economy host country, so as to build the requisite reputational capital necessary for further expansion in the region” (Bonin, 2010: 467).

The theoretical framework presented earlier suggests that countries with a substantial foreign ownership of their banking markets receive better credit ratings than otherwise comparable countries. Figure 1 shows the cross-country bivariate relationship between the share of banking assets held by foreign banks and the credit ratings assigned by all three CRAs in 2001. The scatter plot illustrates that EE countries with higher levels of foreign bank ownership tend to receive better credit ratings. A large foreign bank presence is usually associated with an investment-grade rating, whereas low foreign bank ownership is associated with a speculative grade. For instance, Estonia and the Czech Republic, with the highest average share of foreign banks among total banks, amounting to 77.93% and 73.4%, received on average A–/BBB+ and A– ratings from Standard & Poor’s (low credit risk), respectively. In contrast, Russia and Ukraine, with a low average share of foreign banks among total banks, amounting to 13.53% and 22.71%, respectively, were on average assigned BB–/BB and B/B+ ratings by Standard & Poor’s (high credit risk).

Figure 1.

Foreign bank ownership and sovereign credit ratings in 2001 by (a) Fitch, (b) Standard & Poor’s, and (c) Moody’s.

Control variables

In selecting the control variables, I include those variables that are most common in the scholarship on sovereign credit ratings. Per capita income (GDP at purchasing power parity per capita of internationally comparable USD) is a proxy for the stability and quality of institutions that prevent excessive government borrowing. A higher GDP growth rate improves the government’s ability to service its foreign debt obligations. Inflation signals structural problems in government finances. Alternatively, high inflation reduces existing government debt in domestic currency; thus, it eases the government’s ability to service its foreign debt obligations. The central government balance (scaled by GDP) is an important predictor of ratings because large and persistent fiscal deficits absorb domestic savings, thus increasing the bond default risk. A high current account deficit, associated with overconsumption, could undermine long-term debt sustainability. However, it may also reflect a rapid accumulation of investment, higher growth, and thus the greater ability of a government to repay its debt. The size of external debt over exports determines a country’s solvency. Higher indebtedness is associated with a higher risk of default. Macroeconomic data were obtained from the World Bank’s World Development Indicators and supplemented with data reported by the European Bank for Reconstruction and Development (EBRD).

Past sovereign defaults increase the perceived credit risk of a country.³² The data on defaults on foreign currency bond debt were obtained from Standard & Poor’s publications.³³ The bond default variable is coded as 1 beginning in the year of default and thereafter. Finally, I also test for the “democratic advantage” thesis, according to which democracies are more creditworthy than their authoritarian counterparts, by using a continuous aggregate democracy measure from the Polity IV data set, varying from −10 (full democracy) to +10 (full autocracy). Summary statistics for all variables are presented in Table 7 in the Appendix. I employ a panel data model with random effects on a linear transformation of the ratings and ordered probit models.³⁴

Main findings

The main results from the baseline regressions with foreign currency ratings from each rating agency as the dependent variable are presented in Table 1. In estimating models 1–9, I employ the share of foreign bank assets in total banking assets as a proxy for foreign bank influence. The explanatory power of the models is high, with R-square values between 0.76 and 0.78. The results are quite consistent across all regression specifications. Consistent with my theory, foreign banks positively influence sovereign credit ratings by all CRAs. An increase in the foreign bank ownership is associated with more favorable credit ratings. The coefficient on foreign banks variable is highly statistically significant (at least at 5%) across all estimations.

Table 1.

Foreign banks and sovereign credit ratings.

	Standard & Poor’s			Moody’s			Fitch
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8	Model 9
FOREIGN BANKS	0.009** (0.004)	0.009** (0.004)	0.009** (0.004)	0.024*** (0.005)	0.024*** (0.005)	0.024*** (0.005)	0.016*** (0.004)	0.016*** (0.004)	0.015*** (0.004)
PER CAPITA INCOME (log)	5.420*** (0.462)	5.369*** (0.439)	5.768*** (0.518)	5.559*** (0.619)	5.542*** (0.620)	5.472*** (0.613)	4.366*** (0.480)	4.132*** (0.334)	4.263*** (0.396)
GDP GROWTH	0.078*** (0.029)	0.079*** (0.029)	0.077*** (0.029)	0.001 (0.030)	0.002 (0.030)	−0.000 (0.031)	0.038 (0.026)	0.056** (0.028)	0.049* (0.027)
INFLATION (log)	0.008 (0.096)	0.007 (0.096)	0.027 (0.096)	−0.170 (0.109)	−0.166 (0.109)	−0.178 (0.113)	0.018 (0.094)	−0.012 (0.099)	−0.006 (0.098)
GOVERNMENT BALANCE	0.086*** (0.033)	0.076** (0.034)	0.093*** (0.033)	0.063* (0.035)	0.058 (0.037)	0.075** (0.036)	0.097*** (0.032)	0.095*** (0.035)	0.100*** (0.035)
CURRENT ACCOUNT	−0.035 (0.026)	−0.043* (0.025)	−0.044* (0.027)	−0.025 (0.028)	−0.028 (0.029)	−0.038 (0.029)	−0.020 (0.020)	−0.037* (0.020)	−0.039* (0.021)
EXTERNAL DEBT	−0.014*** (0.003)	−0.014*** (0.003)	−0.014*** (0.003)	−0.011*** (0.003)	−0.012*** (0.003)	−0.010*** (0.003)	−0.011*** (0.003)	−0.009*** (0.003)	−0.009*** (0.003)
BOND DEFAULT	−1.335** (0.558)	−1.559*** (0.587)	−1.275** (0.560)	−0.884** (0.436)	−0.940** (0.450)	−0.954** (0.447)	−1.396*** (0.445)	−1.473*** (0.505)	−1.502*** (0.486)
DEMOCRACY	0.003 (0.041)	−0.016 (0.043)	0.014 (0.046)	−0.092* (0.051)	−0.103* (0.055)	−0.041 (0.056)	−0.045 (0.044)	−0.038 (0.039)	−0.016 (0.044)
BANK DEFAULT		−0.753(0.612)			−0.348(0.669)			0.094(0.642)
EU APPLY			−0.298(0.452)			−0.946(0.602)			−0.519(0.494)
EU MEMBERSHIP			−0.432(0.309)			0.147(0.344)			0.112(0.324)
Constant	−40.881*** (4.106)	−40.368*** (3.908)	−43.965*** (4.599)	−41.425*** (5.587)	−41.217*** (5.602)	−40.523*** (5.501)	−31.211*** (4.235)	−29.485*** (2.961)	−30.388*** (3.468)
R-squared	0.770	0.781	0.761	0.773	0.774	0.780	0.763	0.777	0.773
Observations	138	138	138	135	135	135	148	148	148

Notes: Standard errors in parentheses. *p < 0.10; **p < 0.05; ***p < 0.01.

Furthermore, I find that per capita income is a strong, positive determinant of sovereign credit ratings from all three agencies: as GDP per capita increases, sovereign ratings are upgraded. Lower external indebtedness and lower fiscal deficits are also consistently related to better ratings from all CRAs. By contrast, bond defaults negatively affect CRA assessments. My statistical results further suggest that although all three CRAs broadly employ similar rating criteria, they weight some factors differently. I find that Standard & Poor’s and Fitch place more weight on economic growth and the current account than Moody’s. Political regime type is significant in the Moody’s model, but it does not have the expected sign. Inflation does not appear to influence credit ratings.

To illustrate the magnitude of the foreign bank advantage, the difference between the 10th and 90th percentile of the average share of foreign banks translates into an improvement of ratings by three notches for Moody’s and by over two notches for Standard & Poor’s and Fitch. This is a substantial improvement given a rather static rating trend as CRAs attempt to assess trends instead of “discrete events” (Archer et al., 2007: 347).

Beyond the set of core variables, the next specification incorporates a measure for default on foreign currency bank debt (bank default), which was not significant. I also test for EU accession by including dummy variables for the stages at which countries applied for EU membership and attained full membership. Following Gray (2009), the EU apply and EU membership variables are coded 1 for each country and year when a particular stage was initiated and for years thereafter. Surprisingly, EU accession does not appear to significantly influence countries’ credit ratings.

It has been argued that CRAs often react to movements in bond spreads (De Grauwe and Ji, 2012: 870); thus, they lag behind the market in adjusting their credit ratings. I examine the direct impact of foreign banks on government borrowing costs, measured by market sovereign bond yield spreads. I employ the yearly averages of sovereign bond spread indices included in J.P. Morgan’s Emerging Markets Bond Index-Global (the index of dollar-denominated sovereign bonds issued by a selection of emerging market countries) as an alternative dependent variable. These indices provide market participants with a performance benchmark for assessing sovereign risks and returns.

Looking at Table 2, one can observe that the coefficient on foreign banks is negative and highly significant at the 1% level, providing support for the idea that countries with higher levels of foreign bank ownership enjoy lower bond yield spreads (see model 10).³⁵ I also use the yearly averages of Moody’s and Standard & Poor’s credit ratings as an additional alternative dependent variable and obtain results that are consistent with my hypothesis (see model 11). Furthermore, I use the percentage of foreign banks among total banks as an alternative indicator of foreign bank influence. As models 12–14 (see Table 2) demonstrate, regardless of how I measure foreign bank ownership, I find that countries with a more significant foreign bank ownership are more likely to receive better ratings.

Table 2.

Alternative dependent and independent variables.

Panel A: Alternative dependent variables			Panel B: Foreign banks as the percentage among total banks (alternative independent variable)
	EMBI Spreads (log)	Standard & Poor’s and Moody’s average rating	Standard & Poor’s	Moody’s	Fitch
	Model 10	Model 11	Model 12	Model 13	Model 14
FOREIGN BANKS	−0.010*** (0.003)	0.017*** (0.004)	0.015** (0.007)	0.036*** (0.009)	0.027*** (0.007)
PER CAPITA INCOME (log)	−1.985** (0.886)	5.106*** (0.458)	5.336*** (0.473)	5.420*** (0.639)	4.214*** (0.498)
GDP GROWTH	−0.016(0.014)	0.027(0.025)	0.077*** (0.028)	0.000(0.031)	0.038(0.026)
INFLATION (log)	−0.113** (0.051)	−0.103(0.091)	−0.004(0.091)	−0.210* (0.111)	−0.009(0.092)
GOVERNMENT BALANCE	−0.065*** (0.018)	0.093*** (0.029)	0.082*** (0.031)	0.057(0.035)	0.087*** (0.032)
CURRENT ACCOUNT	−0.017* (0.009)	−0.048** (0.023)	−0.033(0.025)	−0.039(0.028)	−0.027(0.019)
EXTERNAL DEBT	0.006*** (0.002)	−0.012*** (0.003)	−0.015*** (0.003)	−0.011*** (0.003)	−0.011*** (0.003)
BOND DEFAULT	0.374* (0.196)	−0.785** (0.381)	−1.256** (0.539)	−0.676(0.442)	−1.184*** (0.439)
DEMOCRACY	0.057(0.057)	−0.042(0.043)	0.014(0.039)	−0.033(0.049)	−0.022(0.041)
Constant	23.202*** (7.879)	−37.694*** (4.079)	−40.365*** (4.161)	−41.026*** (5.658)	−30.373*** (4.324)
R-squared	0.954	0.806	0.782	0.772	0.762
Observations	65	152	147	144	157

Notes: Standard errors in parentheses. *p < 0.10; **p < 0.05; ***p < 0.01.

I also test the robustness of my results to alternative estimation methods. I estimate the ordered probit model, which accounts for the ordinal nature of the dependent variable. By contrast, the linear model assumes that ratings are divided into equally spaced rating categories. In other words, the linear models treat the difference between Aaa and Aa1 the same as that between Baa3 and Ba1 (or other numerical scores) (Mora, 2006: 2054). The results from the ordered probit estimations reported in Table 3 validate the findings presented earlier: the foreign banks variable (measured in terms of assets) has a positive sign and is statically significant at least at the 10% level in all cases.³⁶

Table 3.

Alternative estimation techniques.

Ordered probit	Standard & Poor’s	Moody’s	Fitch
Ordered probit	Model 15	Model 16	Model 17
FOREIGN BANKS	0.009** (0.004)	0.011** (0.004)	0.013*** (0.004)
PER CAPITA INCOME (log)	3.673*** (0.376)	3.305*** (0.333)	3.042*** (0.291)
GDP GROWTH	0.075*** (0.028)	0.044(0.028)	0.072*** (0.025)
INFLATION (log)	−0.140(0.086)	−0.240** (0.098)	−0.074(0.084)
GOVERNMENT BALANCE	0.041(0.031)	0.032(0.031)	0.055*(0.030)
CURRENT ACCOUNT	−0.059*** (0.020)	−0.046** (0.019)	−0.045*** (0.016)
EXTERNAL DEBT	−0.003(0.002)	−0.003(0.002)	−0.003(0.002)
BOND DEFAULT	−1.218* (0.624)	−0.528(0.474)	−1.193** (0.510)
DEMOCRACY	−0.013(0.025)	−0.064** (0.025)	−0.033(0.023)
Pseudo R-squared	0.312	0.333	0.319
Observations	141	136	149

Notes: Up to 13 rating cut-offs (not reported) are estimated for the ordered probit models with ratings from 17 to 1 for all three CRAs. Standard errors in parentheses. *p < 0.10; **p < 0.05; ***p < 0.01.

Sample selection: The decision to access the sovereign bond market

Beaulieu, Cox, and Saiegh (2012) argue that countries seek credit ratings only when they expect to receive “acceptable” ratings and favorable interest rates. For them, one of the reasons why some countries might enter the sovereign bond market is selection by interested third parties, most likely on the advice of investment or development banks providing funding. In this section, I examine the potential selection bias in estimating the foreign bank advantage: selection bias can occur if the presence of foreign banks influences not only the credit ratings of host countries, but also their decision to request a rating from one of the CRAs. I use the Heckman selection model to address this source of potential bias, which involves specifying equations for the determinants of sovereign credit ratings (outcome equation) and whether a country received a rating from a CRA (selection equation).³⁷

I estimate the parameters of the selection model with exclusion restrictions; therefore, I include an extra (exogenous) explanatory variable that influences the selection (i.e. receiving ratings) but not the outcome of interest (i.e. actual ratings).³⁸ I use MERCHANDISE EXPORTS to high-income countries as a proxy for friendliness and political ties with rich countries. It is conceivable that a country’s political connections may influence its decision to seek a credit rating. Data come from the World Bank’s World Development Indicators.

In addition to the selection effect, Beaulieu, Cox, and Saiegh (2012) also criticize previous studies for treating economic development and democracy as exogenous to each other. They highlight the arguments made by Acemoglu, Johnson, and Robinson (2001) that political institutions are a fundamental cause of development; hence, the development should be endogenized. Moreover, as argued earlier, foreign banks headquartered in countries with well-regulated banking systems can exert a long-term and cumulative influence on financial development, and subsequently on economic growth, in a host country, following the logic of arguments by Goldberg (2009) and Mishkin (2006). Therefore, including proxies for both foreign bank influence and GDP per capita in analyses of credit access and sovereign credit ratings may not provide an adequate test of the effect of foreign banks. To address this problem, I include GDP (instead of per capita income) from the World Bank’s World Development Indicators in my estimations. GDP is an equally important indicator of the ability of a country to repay its debt.

Table 4 displays the results of the selection models, showing that countries with higher levels of foreign bank ownership were more likely to be rated and more likely to receive more favorable ratings, conditional on being rated. The estimated correlation between the errors in the selection and outcome equations is generally not significantly different from zero (the coefficient of the inverse Mills ratio-denoted lambda is mostly insignificant). Therefore, one cannot reject the hypothesis that the selection and rating equations are independent. Across all specifications, the coefficient on the foreign banks variable remains highly significant at the 1% level, which indicates that even when controlling for the selection process, a substantial foreign bank presence yields higher credit ratings from all three CRAs. This should reassure us that the impact of foreign bank presence is not mistakenly attributed to any underlying factors that drive a country’s ability to access credit and receive credit ratings. In sum, countries with a substantial presence of foreign banks tend to enjoy an advantage in both accessing credit and sovereign credit ratings.

Table 4.

Sample selection: Access to credit markets and credit ratings.

Heckman selection model	Standard & Poor’s	Moody’s	Fitch
Heckman selection model	Model 18	Model 19	Model 20
Sovereign credit rating
FOREIGN BANKS	0.020*** (0.008)	0.030*** (0.010)	0.025*** (0.007)
GDP (log)	0.879*** (0.236)	1.604*** (0.348)	0.813*** (0.266)
GDP GROWTH	0.117** (0.059)	0.149** (0.069)	0.118*** (0.043)
INFLATION (log)	−0.845*** (0.181)	−1.075*** (0.226)	−0.573*** (0.171)
GOVERNMENT BALANCE	0.017(0.057)	−0.023(0.070)	0.016(0.051)
CURRENT ACCOUNT	−0.186*** (0.042)	−0.201*** (0.050)	−0.123*** (0.032)
EXTERNAL DEBT	−0.010* (0.005)	0.001(0.006)	−0.002(0.004)
BOND DEFAULT	−2.664** (1.340)	−2.350** (1.114)	−3.788*** (1.076)
DEMOCRACY	0.062(0.051)	0.041(0.057)	0.092** (0.043)
Constant	−13.366** (5.927)	−32.708*** (8.785)	−13.336** (6.752)
Obtaining rating
FOREIGN BANKS	0.017** (0.007)	0.001(0.005)	0.013** (0.006)
GDP (log)	1.938*** (0.262)	1.534*** (0.180)	1.521*** (0.196)
GDP GROWTH	0.031(0.033)	−0.002(0.031)	0.039(0.028)
INFLATION (log)	−0.323** (0.133)	−0.307*** (0.110)	−0.367*** (0.116)
GOVERNMENT BALANCE	0.054(0.053)	0.047(0.041)	0.070(0.045)
CURRENT ACCOUNT	−0.051* (0.029)	−0.020(0.026)	−0.072*** (0.025)
EXTERNAL DEBT	0.007* (0.004)	0.011*** (0.003)	0.004(0.003)
BOND DEFAULT	−3.205** (1.375)	1.048(1.060)	−0.195(0.657)
DEMOCRACY	0.090*** (0.032)	0.038(0.027)	0.005(0.027)
MERCHANDISE EXPORTS	0.002(0.013)	−0.004(0.012)	0.014(0.011)
Constant	−45.867*** (6.263)	−36.055*** (4.249)	−36.510*** (4.735)
Lambda (inverse Mills ratio)	−0.071(0.677)	1.691** (0.854)	−0.039(0.705)
Observations	242	242	242

Notes: Standard errors in parentheses. *p < 0.10; **p < 0.05; ***p < 0.01.

Causality: Reforms and foreign banks

One might argue that foreign bank presence and other policies leading to better credit ratings are the outcomes of financial and economic reforms that countries undertake prior to the entry of foreign banks. These reform policies might then cause the observed effect of foreign bank presence on ratings. To provide causal evidence on the effect of foreign bank ownership on the creditworthiness of their host countries, I follow Rose (2004) and include two leads and two lags of the foreign bank variable jointly (see Table 5). The rationale behind this approach is that if the government undertook policy reforms two years prior to the entry of foreign banks and these reforms improved the country’s credit ratings, the coefficient on the second lead of foreign banks variable should be significant. However, neither the coefficients on leads nor the coefficients on lags are significant at conventional levels, implying that the effect of foreign banks on a country’s credit rating materializes after rather than prior to their entry (see models 21, 23, and 25).

Table 5.

Foreign banks, reforms, and sovereign credit ratings.

	Standard & Poor’s		Moody’s		Fitch
	Model 21	Model 22	Model 23	Model 24	Model 25	Model 26
FOREIGN BANKS (t – 1)	0.005(0.010)	0.008** (0.004)	0.017* (0.010)	0.023*** (0.005)	0.009(0.007)	0.015*** (0.004)
FOREIGN BANKS (t – 2)	0.002(0.008)		0.006(0.008)		0.006(0.006)
FOREIGN BANKS (t + 1)	0.012(0.011)		0.014(0.011)		0.003(0.008)
FOREIGN BANKS (t + 2)	−0.011(0.011)		−0.018(0.012)		0.007(0.008)
PER CAPITA INCOME (log)	4.864*** (0.388)	5.335*** (0.536)	4.914*** (0.465)	5.006*** (0.546)	4.064*** (0.488)	4.098*** (0.510)
GDP GROWTH	0.174*** (0.041)	0.077*** (0.029)	0.042(0.038)	0.012(0.032)	0.029(0.028)	0.041(0.026)
INFLATION (log)	−0.207* (0.109)	0.015(0.098)	−0.288** (0.130)	−0.181(0.119)	−0.050(0.096)	0.029(0.097)
GOVERNMENT BALANCE	0.060(0.041)	0.086*** (0.033)	0.059(0.043)	0.068* (0.037)	0.136*** (0.035)	0.099*** (0.033)
CURRENT ACCOUNT	−0.089*** (0.027)	−0.033(0.026)	−0.056* (0.030)	−0.036(0.029)	−0.017(0.020)	−0.021(0.020)
EXTERNAL DEBT	−0.008** (0.003)	−0.014*** (0.003)	−0.009** (0.004)	−0.010*** (0.004)	−0.012*** (0.003)	−0.011*** (0.003)
BOND DEFAULT	−1.586** (0.737)	−1.310** (0.563)	−0.808(0.504)	0.913*(0.472)	−1.146*** (0.442)	−1.358*** (0.460)
DEMOCRACY	−0.005(0.033)	0.001(0.042)	−0.063(0.051)	−0.077(0.050)	−0.051(0.044)	−0.050(0.043)
REFORMS (t – 1)		0.230(0.687)		0.349(0.791)		0.543(0.709)
Constant	−36.779*** (3.466)	−40.876*** (4.137)	−35.871*** (4.174)	−37.919*** (4.431)	−28.556*** (4.278)	−30.656*** (3.822)
R-squared	0.816	0.773	0.809	0.791	0.771	0.775
Observations	130	138	128	135	140	148

Notes: Standard errors in parentheses. *p < 0.10; **p < 0.05; ***p < 0.01.

Furthermore, I explore more explicitly whether economic reforms affect the influence of foreign bank presence on country risk by employing the EBRD indicators that have been used to track economic reforms in EE countries in six areas, including privatization (large and small scale), governance and enterprise restructuring, price liberalization, trade and foreign exchange system, and competition policy.³⁹ If the influence of foreign banks is manifested through their impact on economic and institutional reforms after their entry, the effect should no longer be possible once I directly control for reforms in regressions. As models 22, 24, and 26 in Table 5 show, the coefficients on the lagged EBRD reform index are not significant at conventional levels, while the coefficients on foreign banks retain their strong statistical significance at least at the 5% level.

Endogeneity

It is plausible to assume that foreign banks are attracted to countries with responsible governments and better credit ratings. To reduce concerns about endogeneity with regards to foreign bank entry, I lag all regressors in all my estimations one period. In this section, I also apply an instrumental variable approach using two sets of instruments explaining foreign bank entry. These instruments have been used by scholars of multinational banking and were identified by bankers at major multinational banking groups whom I interviewed. Following the identification strategy similar to Detragiache, Tressel, and Gupta (2008), I use population as a measure of potential market size. The assumption behind this instrument is that international banks operating in many markets can better diversify country-specific risk than domestic banks can, which is particularly valuable in smaller countries. Furthermore, foreign banks can acquire a “monopolistic advantage” in small markets by making a relatively small initial investment.⁴⁰

The second instrument — the long-run political history of a host country — captures cultural and economic ties and previous history with Western Europe, the area where the majority of foreign banks that have penetrated the region of emerging Europe are headquartered. The cultural (and physical) distance between home and host countries is an important determinant of foreign bank entry (Focarelli and Pozzolo, 2001). Interview evidence also supports the assertion that historical links, including imperial legacies, influence the investment decisions of banks. For instance, Erste’s chief executive officer, Andreas Treichl, expressed this view: “Look at our map and then look at the map of Austria in 1914. Every country we’re in, we’d opened up saving banks 175 year ago.”⁴¹ To measure European cultural integration, I rely on Grosjean (2013), who used the Periodical Atlas of Europe to code the political history of EE states from 1300 — the beginning of empire consolidation in medieval Europe — to 2000. In my analysis, I use the shares of each country that were under the Prussian and Habsburg empires. My identification strategy assumes that the instruments are plausibly correlated with foreign bank entry but have no other direct or indirect effects on sovereign credit ratings.

Table 6 reports the results of the two-step generalized method of moments (GMM) estimations with robust standard errors. In the first-stage regressions, the instruments are highly significant. The F-statistic allows the rejection of the null hypothesis that the instruments are jointly zero. In the second-stage regressions, the Hansen J-statistic and corresponding Chi² p-value suggest that we cannot reject the null hypothesis of validity of the over-identifying restrictions when more than one instrument is used. Together, these tests provide strong evidence for the validity of the instruments. Foreign bank ownership continues to have a positive and significant impact on credit ratings. While there are always questions about the suitability of instruments, I take these results as evidence that higher levels of foreign bank ownership are associated with the better sovereign credit ratings of host countries.

Table 6.

Instrumental variables estimates of sovereign credit ratings.

GMM	Standard & Poor’s	Moody’s	Fitch
GMM	Model 27	Model 28	Model 29
Second stage
FOREIGN BANKS	0.172*** (0.041)	0.178*** (0.034)	0.137*** (0.025)
GDP (log)	0.656(0.399)	0.954*** (0.337)	1.052*** (0.236)
GDP GROWTH	−0.032(0.129)	−0.033(0.120)	0.044(0.070)
GOVERNMENT BALANCE	0.050(0.115)	−0.034(0.110)	0.047(0.083)
CURRENT ACCOUNT	0.090(0.109)	0.079(0.105)	−0.018(0.056)
EXTERNAL DEBT	−0.004(0.010)	−0.003(0.009)	−0.004(0.007)
BOND DEFAULT	−4.797* (2.673)	−3.044** (1.541)	−4.312*** (1.523)
DEMOCRACY	−0.390** (0.163)	−0.395*** (0.131)	−0.332*** (0.110)
First stage
EMPIRE	13.591*** (4.654)	18.969*** (4.560)	15.935*** (4.355)
POPULATION	−0.000** (0.000)	−0.000** (0.000)	−0.000*** (0.000)
Partial R ²	0.11	0.17	0.18
Hansen J-statistic	1.678	1.940	2.348
Chi² p-value	0.195	0.164	0.126
F-statistic (excluded instruments)	8.36	13.43	15.70
p-value of F-test	0.000	0.000	0.000
Observations	144	140	154

Notes: The control variables in the first-stage regression are omitted to conserve space. Standard errors in parentheses. *p < 0.10; **p < 0.05; ***p < 0.01.

Conclusion

In this article, I have explored whether foreign bank ownership influences the perceptions of sovereign creditworthiness of host countries by CRAs. My statistical results and interviews provide strong support for the hypothesis that foreign banks, as strategic investors, exogenously create or enhance the reputational capital of emerging European governments by making their policies and commitments more credible with CRAs. These results demonstrate that there is a difference between the perceptions of the creditworthiness (default risk) of countries with a substantial level of foreign bank ownership and those without by CRAs and bond investors. Countries with high levels of foreign bank control tend to be assigned more favorable credit ratings by CRAs and find it easier to obtain credit in sovereign debt markets at lower interest rates than do countries where domestic banks with political ties dominate financial intermediation.

This finding holds true in the face of a battery of economic and political controls for alternative explanations and is robust to different measures of creditworthiness and foreign bank influence. I also show that the increase in creditworthiness that we see when countries are financially integrated through multinational banks is not a function of selection, where the same countries also choose to enter the international credit market. Importantly, this study demonstrates that foreign banks have an independent effect that is exogenous to the underlying processes that drive foreign banks into the financial systems of EE countries.

The theory and empirical findings of this article have important implications for the study of government reputation and credible commitments in International Relations. In particular, this study integrates private external actors into the reputation models in International Relations. As such, it provides an alternative reputation-building mechanism. Future research may identify other crucial, but overlooked, external sources of reputation for countries lacking credibility in the eyes of international audiences.

Second, this article can inform governments’ foreign investment policies. Its findings highlight the positive effects of global banking integration and identify foreign bank investors as a source of sovereign borrowers’ reputation in international financial markets. Since the effect of foreign banks hinges on the assumption that they are headquartered in Western industrial democracies, it would also be useful to extend the inquiry of this article by exploring whether that effect holds when these banks come from less-developed countries that have started to play an active role as investors in other less-developed countries.

The 2008 global financial crisis, which originated in the US subprime mortgage market, may have called into question the foreign bank advantage. Foreign banks were channels for loans in foreign currency, which contributed to credit booms and the accumulation of external debt in some countries of emerging Europe. However, none of these countries had high-leverage, subprime mortgages or other types of toxic assets (Åslund, 2011: 378). While the peak of the crisis appears to coincide with downgrades of the sovereign debt of some countries, this was in the middle of the worst financial crisis since the Great Depression. Still, emerging European countries with a substantial foreign bank ownership fared significantly better than those without. Standard & Poor’s downgraded the credit rating of Slovakia (where 92% of assets was held by foreign banks in 2008) from A+ in 2008 to A in early 2011 but upgraded it back to A+ already at the end of 2011, whereas Fitch left the country’s rating unchanged. Fitch downgraded the sovereign credit rating of Estonia (with the share of assets held by foreign banks amounting to 99%) from A in 2008 to BBB+ in 2009 but upgraded it to A+ in 2011. By contrast, Slovenia, long regarded as the leader in post-communist economic and institutional transformation but reluctant to allow foreign bank ownership (foreign banks held only 26% of banking assets in 2008), experienced a severe banking crisis in 2012. One reason behind this crisis was that the domestic state-owned banks that dominated the banking system lacked discipline and engaged in extensive cross-border borrowing in the run-up to the crisis (Epstein, 2017: 119–122). Consequently, Moody’s sharply downgraded Slovenia’s debt from Aa2 in 2006 to Baa3 in 2015, only one notch away from junk rating, expressing concerns about the country’s banking system. It is reasonable to assume that while citizens and politicians may have lost their trust in the global financial system after the recent crisis, the reputation of multinational banks among their peers and CRAs has remained largely unaffected.

Nonetheless, it should be recognized that the 2008 financial crisis highlighted the risks associated with cross-border banking, and global banks were considered as “transmitters” of the crisis (Cetorelli and Goldberg, 2011). The deteriorated health of some multinational banks and the need to deleverage their balance sheets raised concerns that host countries, particularly in the EE region featuring high levels of foreign bank ownership, would be vulnerable to “cutting and running,” which did not materialize, however (Epstein, 2014, 2017). Multinational banks remained engaged and softened capital outflows from emerging Europe at the peak of the crisis (Berglöf et al., 2009). A vulnerability of host countries to a sudden stop of capital flows during the crisis was particularly reduced when foreign banks conducted their lending through their local affiliates rather than providing direct cross-border loans in foreign currency without having an affiliated presence (Kamil and Rai, 2010). However, foreign banks are not a panacea to the risk of crises, as Tschoegl (2005) notes. Future work may examine more thoroughly whether and how the crisis influenced the ability of foreign banks to transfer good reputations to emerging-market countries.

Finally, this article raises questions pertaining to international financial governance. The recent global financial crisis has shown that financial stability and national policies for banking supervision and resolution do not go well together in a financially integrated economy because banks operate on a global scale, while the power of governments is defined at the national level (Schoenmaker, 2013). To prevent the financial meltdown in the EE region, a new governance framework — the Vienna Initiative — was established in the midst of the crisis at the initiative of Austrian multinational banking groups rather than by a public institution (Epstein, 2014). This novel European private–public coordination platform represents an “alternative coordinated, multi-stakeholder governance framework” (Pistor, 2012: 1). However, in order to build a better international financial architecture, one less prone to shocks and panics, international financial institutions, such as the International Monetary Fund or the Bank for International Settlements, should adopt a “supervisory role for the global systemic banks,” Schoenmaker (2013) pleads.

Footnotes

Appendix

Table 7.

Summary data.

Variable	Mean	Standard deviation	Minimum	Maximum
STANDARD & POOR’s RATING	8.309	2.932	1	14
MOODY’s RATING	7.682	3.931	1	14
FITCH RATING	7.592	3.245	1	14
FOREIGN BANKS	43.228	33.590	0	100
PER CAPITA INCOME (log)	8.741	0.847	6.743	10.212
GDP GROWTH	4.680	6.139	−17.955	34.5
INFLATION (log)	2.030	1.282	−2.995	6.964
GOVERNMENT BALANCE	−2.499	4.146	−17.3	25.462
CURRENT ACCOUNT	−5.008	8.137	−34.254	33.679
EXTERNAL DEBT	113.801	58.276	10.715	333.102
BOND DEFAULT	0.021	0.145	0	1
DEMOCRACY	3.864	6.660	−9	10

Acknowledgements

I am grateful to the Austrian Marshall Plan Foundation for financial support. I would like to express my appreciation to the Center for Transatlantic Relations at the Paul H. Nitze School of Advanced International Studies, Johns Hopkins University and its director Dan Hamilton for institutional support. I would further like to thank Gordon Bodnar, Hein Bogaard, Barry Eichengreen, Rachel Epstein, Leo Feler, Marc Flandreau, Davide Furceri, Ilene Grabel, Pedro Gomes, Daniel Gros, Henry E. Hale, Stefan Huemer, Raymond Hicks, Indridi Indridason, Juliet Johnson, Ronald H. Linden, Thomas Oatley, Robert Wade, John Zysman, seminar participants at the University of Denver, Johns Hopkins University, George Washington University, University of California, Riverside, the National Bank of Slovakia, conference participants at the annual meetings of the International Political Economy Society and the American Political Science Association, and two anonymous reviewers, for helpful comments and suggestions.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author biography

Jana Grittersová is Associate Professor of Political Science and Cooperating Faculty in the Department of Economics at the University of California, Riverside, USA. She received her PhD in Government from Cornell University and a PhD in Economics from the University of Economics in Bratislava, Slovakia. She has previously taught at the University of California, Berkeley and Stanford University, and has worked at the European Commission in Brussels and the National Bank of Slovakia. Her published and ongoing research focuses on the sources of government reputation in international markets, global banking, political perceptions of monetary policy, the impact of interest groups on the credibility of international monetary and financial commitments, the role of non-market coordination in financial-system development, and the political economy of reform. Her book Borrowing Credibility explores how countries with new histories and weak institutions establish monetary credibility in international financial markets.

References

Acemoglu

Johnson

Robinson

(2001) The colonial origins of comparative development: An empirical investigation. American Economic Review 91(5): 1369–1401.

Adams

Mathieson

Schinasi

(1999) International Capital Markets: Developments, Prospects and Key Policy Issues. Washington, DC: International Monetary Fund.

Archer

Biglaiser

DeRouen

Jr (2007) Sovereign bonds and the “democratic advantage”: Does regime type affect credit rating agency ratings in the developing world? International Organization 61(2): 321–365.

Åslund

(2011) Exchange rate policy and the central and east European financial crisis. Eurasian Geography and Economics 52(3): 375–389.

Åslund

Dombrovskis

(2011) How Latvia came through the financial crisis. Washington, DC: Peterson Institute for International Economics.

Barta

(2012) “Rating politics.” EUI Working Paper MWP 08.

Beaulieu

Cox

Saiegh

(2012) Sovereign debt and regime type: Re-considering the democratic advantage. International Organization 66(4): 709–738.

Berglöf

Korniyenko

Plekhanov

, et al. (2009) Understanding the Crisis in Emerging Europe. Working Paper 109. London: European Bank for Reconstruction and Development.

Biglaiser

Staats

(2012) Finding the “democratic advantage” in sovereign bond ratings: The importance of strong courts, property rights protection, and the rule of law. International Organization 66(3): 515–535.

10.

Bonin

(2010) From reputation amidst uncertainty to commitment under stress: More than a decade of foreign-owned banking in transition economies. Comparative Economic Studies 52(4): 465–494.

11.

Bonin

Schnabel

(eds) (2011) The great transformation: From government-owned to foreign-controlled banking sectors, Symposium on Banking in Transition. Economics of Transition 19(3): 397–405.

12.

Brooks

Cunha

Mosley

(2015) Categories, creditworthiness, and contagion: How investors’ shortcuts affect sovereign debt markets. International Studies Quarterly 59(3): 587–601.

13.

Bulow

Rogoff

(1989) Sovereign debt: Is to forgive to forget? American Economic Review 79(1): 43–50.

14.

Butler

Fauver

(2006) Institutional environment and sovereign credit ratings. Financial Management 35(3): 53–79.

15.

Cantor

Packer

(1996) Determinants and impact of sovereign credit ratings. Federal Reserve Bank of New York Economic Policy Review October: 37–54.

16.

Cardenas

Graf

O’Dogherty

(2003) Foreign Banks’ Entry in Emerging Market Economies: A Host Country Perspective. Basel: The Bank for International Settlements. Available at: http://www.bis.org/publ/cgfs22cbpapers.htm

17.

Cetorelli

Goldberg

(2011) Global banks and international shock transmission: Evidence from the crisis. IMF Economic Review 59(1): 41–76.

18.

Claessens

Van Horen

(2008) Foreign bank presence in developing countries 1995–2006: Data and trends. Social Science Research Network. Available at: http://ssrn.com/abstract=1107295

19.

Claessens

Van Horen

(2012) Foreign Banks: Trends, Impact and Financial Stability. Working Paper 12/20. Washington, DC: International Monetary Fund.

20.

Corsetti

Pesenti

Roubini

(2001) Fundamental determinants of the Asian crisis: The role of financial fragility and external imbalances. In: Takahashi

Krueger

(eds) Regional and Global Capital Flows: Macroeconomic Causes and Consequences. Chicago, IL: The University of Chicago Press, pp. 11–41.

21.

Cull

Martinez Peria

(2010) Foreign Bank Participation in Developing Countries: What Do We Know about the Drivers and Consequences of This Phenomenon? Policy Research Working Paper 5398 (August). Washington, DC: World Bank.

22.

De Grauwe

(2012) Mispricing of sovereign risk and macroeconomic stability in the Eurozone. Journal of Common Market Studies 50(6): 866–880.

23.

De Haas

Naaborg

(2006) Foreign banks in transition countries: To whom do they lend and how are they financed? Financial Markets, Institutions & Instruments 15(4): 159–199.

24.

De Haas

Van Lelyveld

(2006) Foreign banks and credit stability in Central and Eastern Europe. A panel data analysis. Journal of Banking & Finance 30(7): 1927–1952.

25.

Detragiache

Tressel

Gupta

(2008) Foreign banks in poor countries: Theories and evidence. Journal of Finance 63(5): 2123–2160.

26.

Dinç

(2005) Politicians and banks: Political influences on government-owned banks in emerging markets. Journal of Financial Economics 77(2): 453–479.

27.

Dreher

Voigt

(2011) Does membership in international organizations increase governments’ credibility? Testing the effects of delegating powers. Journal of Comparative Economics 39: 326–348.

28.

Epstein

(2014) Assets or liabilities? The politics of bank ownership. Review of International Political Economy 21(4): 765–789.

29.

Epstein

(2017) Banking on Markets: The Transformation of Bank–State Ties in Europe and Beyond. Oxford: Oxford University Press.

30.

Financial Times (2008) Latvia takes over Parex after a run on bank. November 9.

31.

Flandreau

Flores

(2009) Bonds and brands: Foundations of sovereign debt markets, 1820–1830. Journal of Economic History 69(3): 646–684.

32.

Flandreau

Gaillard

Packer

(2011) To err is human: US rating agencies and the interwar foreign government debt crisis. European Review of Economic History 15(3): 495–538.

33.

Focarelli

Pozzolo

(2001) The patterns of cross-border bank mergers and shareholdings in OECD countries. Journal of Banking & Finance 25(12): 2305–2337.

34.

Giannetti

Ongena

(2009) Financial integration and firm performance: Evidence from foreign bank entry in emerging markets. Review of Finance 13: 181–223.

35.

Goldberg

(2009) Understanding banking sector globalization. Staff Papers 56(1), International Monetary Fund.

36.

Gray

(2009) International organization as a seal of approval: European Union accession and investor risk. American Journal of Political Science 53(4): 931–949.

37.

Gray

(2013) The Company States Keep: International Economic Organizations and Investor Perceptions. New York: Cambridge University Press.

38.

Grittersová

(2014) Transfer of reputation: Multinational banks and perceived creditworthiness of transition countries. Review of International Political Economy 21(4): 878–912.

39.

Grittersová

(2017) Borrowing Credibility: Global Banks and Monetary Regimes. Ann Arbor, MI: University of Michigan Press.

40.

Gros

(2003) Who needs foreign banks? Working Paper 998, Centrum for European Studies.

41.

Grosjean

(2011) The weight of history on European cultural integration: A gravity approach. American Economic Review 101(3): 504–508.

42.

Hau

Langfield

Marques-Ibanez

(2013) Bank ratings: What determines their quality? Economic Policy 28(74): 289–333.

43.

Hauner

Jonas

Kumar

(2007) Policy credibility and sovereign credit: The case of the new EU member states. Working Paper 07/01, International Monetary Fund.

44.

Hellman

Murdock

(1998) Financial sector development policy: The importance of reputational capital and governance. In: Sabot

Skékely

(eds) Development Strategy and Management of the Market Economy (vol. 2). Oxford: Oxford University Press.

45.

Johnson

(2016) Priests of Prosperity: How Central Bankers Transformed the Postcommunist World. Ithaca: Cornell University Press.

46.

Kamil

Kulwant

(2013) Global Deleveraging and Foreign Banks’ Lending to Latin American Countries. Economía 13(2): 1–29.

47.

King

Sinclair

(2003) Private actors and public policy: A requiem for the new Basel capital accord. International Political Science Review/Revue internationale de science politique 24(3): 345–362.

48.

Kulish

(2009) Goal of unified Europe falters amid businesses’ scramble to survive down-turn. New York Times, 23 August, p. A6.

49.

Levine

(2001) International financial liberalization and economic growth. Review of International Economics 9(4): 688–702.

50.

Mishkin

(2006) The Next Great Globalization: How Disadvantaged Nations Can Harness Their Financial Systems to Get Rich. Princeton, NJ: Princeton University Press.

51.

Mora

(2006) Sovereign credit ratings: Guilty beyond reasonable doubt? Journal of Banking and Finance 30(97): 2041–2062.

52.

Moreno

Villar

(2005) The increased role of foreign bank entry in emerging markets. In: Bank for International Settlements (ed) Globalisation and Monetary Policy in Emerging Markets, vol. 23. Basel: BIC, pp. 9–16.

53.

Mosley

(2003) Global Capital and National Governments. Cambridge: Cambridge University Press.

54.

Panizza

Sturzenegger

Zettelmeyer

(2009) The economics and law of sovereign debt and default. Journal of Economic Literature 47(3): 651–698.

55.

Pistor

(2012) Governing interdependent financial systems: Lessons from the Vienna Initiative. Journal of Globalization and Development 2(2): 1–25.

56.

Raiffeisen Research (2012) CEE Banking Sector, Report, June. Vienna: Raiffeisen Bank International.

57.

Rose

(2004) Do we really know that the WTO increases trade? American Economic Review 94(1): 98–114.

58.

Saiegh

(2005) Do countries have a “democratic advantage”? Political institutions, multilateral agencies, and sovereign borrowing. Comparative Political Studies 38(4): 366–387.

59.

Schoenmaker

(2013) Governance of International Banking: The Financial Trilemma. Oxford: Oxford University Press.

60.

Schultz

Weingast

(2003) The democratic advantage: Institutional foundations of financial power in international competition. International Organization 57(1): 3–42.

61.

Shen

C-H

Huang

Y-L

Hasan

(2012) Asymmetric benchmarking in bank credit rating. Discussion Paper 13, Bank of Finland.

62.

Sinclair

(2005) The New Masters of Capital: American Bond Rating Agencies and the Politics of Creditworthiness. Ithaca, NY: Cornell University Press.

63.

Standard and Poor’s (2006) Sovereign defaults at 26-year low, to show little change in 2007. Standard and Poor’s Global Credit Portal, September 8.

64.

Standard and Poor’s (2011) Default, transition, and recovery: Sovereign defaults and rating transition data, 2010 update. Standard & Poor’s Global Credit Portal, February 23.

65.

Stasavage

(2007) Cities, constitutions, and sovereign borrowing in Europe, 1274–1785. International Organization 61(3): 489–525.

66.

ANR

(2002) Emerging market bond spreads and sovereign credit ratings: Reconciling market views with economic fundamentals. Emerging Markets Review 3(4): 380–408.

67.

Sylla

(2002) An historical primer on the business of credit ratings. In: Levich

Reinhart

Majnoni

(eds) Ratings, Rating Agencies, and the Global Financial System. Boston, MA: Kluwer, pp. 19–40.

68.

Tomz

(2007) Reputation and International Cooperation: Sovereign Debt across Three Centuries. Princeton, NJ: Princeton University Press.

69.

Tschoegl

(2003) Financial crises & the presence of foreign banks. Wharton Financial Institutions Center Working Paper Series No: 03-35.

70.

Tschoegl

(2005) Financial crises and the presence of foreign banks. In: Honohan

Laeven

(eds) Systemic Financial Distress: Containment and Resolution. Cambridge: Cambridge University Press.

71.

Vogel

Winkler

(2010) Foreign banks and financial stability in emerging markets: Evidence from the global financial crisis. Frankfurt School of Finance & Management Working Papers No. 149.

72.

Wooldridge

(2002) Econometric Analysis of Cross Section and Panel Data. Cambridge, MA: MIT Press.