Inter-rater differences in sovereign credit ratings How distance influences ratings

(1)

“Trust, but verify” Ronald Reagan

Inter-rater differences in sovereign credit ratings

How distance influences ratings

Master Thesis IB&M by Maarten Barthel Student number: 1921002

Faculty of Economics & Business University of Groningen

Tutor: A. van Hoorn

September 24, 2012

Abstract

(2)

2

Acknowledgement

While writing this thesis a turbulent situation occurred in my personal life, which also had a big influence on my thesis. My parents and many others have helped me through this difficult period and supported me in every way. I am very grateful for the help my dad, Peter Barthel gave me by reading and critically assessing this piece of work. I would also wish to thank my good friends Petra Borkent and Henk Bultena for their feedback. Finally I would like to thank my supervisor Andre van Hoorn for his patience, help and support over the last months. He is been an inspiration and great tutor throughout the entire process.

(3)

3

4

Chapter 4: Data and Method ... 42

4.2 Data ... 43

Moody’s... 44

Standard and Poors... 44

Dagong Global Credit Rating Co ... 45

Data Issues ... 45

4.3 Operationalization of the dependent variable ... 46

4.4 Operationalization of the independent variables... 47

Geographic distance ... 47 Cultural distance... 47 Administrative distance... 48 Economic distance ... 49 Chapter 5 Results ... 50 5.1 Introduction ... 51 5.2 Basic Results ... 51 Geographic distance ... 53 Cultural distance... 53 Language... 54 Administrative distance... 54 Economic distance ... 54 5.3 Empirical model ... 55

Interpretation of the coefficients ... 56

5.4 Robustness Check ... 57

Excluding similar borders... 58

Outliers... 58

Broad categories ... 59

Chapter 6: Discussion and conclusion ... 62

(5)

5

Appendix 1: Country list... 72

Appendix 2: Cases in which Dagong’s sovereign credit ratings are lower than the US rating agencies 2010 ... 73

Appendix 3: Cases in which Dagong’s ratings are the same as those of the US rating agencies 2010... 74

Appendix 4: Overview of the data (1)... 75

Appendix 5: Border US... 79

Appendix 6: Border China ... 80

Appendix 7: Language US ... 81

Appendix 8: Language China... 82

Appendix 9: Empirical model with all variables included (1) ... 83

Appendix 10: Empirical model without border US and border China (1)... 85

Appendix 11: Empirical model without the United Arab Emirates (1) ... 87

Appendix 12: Empirical model with Broad categories (1) ... 89

(6)

(7)

7 Chapter 1 Introduction

Source: www.stockfreeimages.com

(8)

8

1.1 Introduction

Anytime advice is given in a financial matter, there is a potential agency issue. The agent collects information, interprets it and provides a rating. The obtained rating is used by potential buyers or users of the product. This is true for many forms of financial advice like stockbrokerages, private banking, equity research, and also for credit ratings on sovereign bonds. The agent is in business to maximize earnings. Credit rating agencies are subject to agency problems, since rating fees paid by issuers are their principal source of revenue.

In the recent financial crisis large debts of some European countries become a major threat to the world economy. Especially the situation in the PIIGS (Portugal, Ireland, Italy, Greece, Spain) (Gärtner, Griesbach & Florian, 2011), is problematic. High interest rates (see figure 1) and downgraded credit ratings make institutional investors and banks nervous. The fear that some of these countries cannot repay their debt grows every day. The tension between members of the European Monetary Union (EMU) is also growing. Together with the IMF various EMU members are providing support to PIIGS through the European Financial Stability Facility (EFSF) and the recently recognized European Stability Mechanism (ESM) The European Central Bank (ECB) also interferes, by buying sovereign bonds, to keep interest rates below 7%. Despite all this Europe is still in a sovereign debt crisis.

Figure 1

Long term interest rates

(9)

9

To make matters worse, interest rates increase (see figure 1) with the increase of fear of default. Credit rating agencies play a substantial role in these matters. Rating agencies provide sovereign credit ratings which investors use to price sovereign bonds. Sovereign credit ratings are an assessment of the creditworthiness of a government’s ability and willingness to make timely payment of the principal and the interests of its debt (Mellios & Paget-Blanc, 2006; Peters, 2002). The influence of sovereign credit ratings on interest spreads is large (Reinhart, 2002). Like Greece, there are many other cases where downgrades lead to higher interest rates. Regulations are one of the main reasons credit ratings gained so much power and influence. Through regulations such as Basel II (and III), banks are supposed to hold large numbers of triple A assets (highest rating) in their portfolio. Since it is very time consuming, complicated, and expensive to rate so many financial products, banks, financial institutions and others investors rely on specialist to this for them. That is how credit rating agencies became such important players, with extensive influence on macro-economic outcomes (Gärtner et al., 2011).

(10)

10

The role of rating agencies is further jeopardized by their lack of predicting financial crises, and their late response. Although it is not the job of rating agencies to predict financial crises, one would expect credit rating agencies to at least anticipate and foresee something before the crisis unfolds. This was definitely not the case during (before) the crisis in East-Asia in the late 1990’s. Not only did credit rating agencies fail to predict the financial crisis (Radelet & Sachs, 1998), they also exacerbated the crisis by downgrading countries to much after the crisis had started to safe their reputation (Ferri, Lui & Stiglitz 1999). Credit rating agencies made mistakes during the East-Asia crisis in the 1990’s and the banking crisis in 2008 and it looks like they are going in the same direction during the recent debt crisis.

Standard & Poors (S&P), Moody’s and Fitch are three of the major agencies that provide credit ratings (90% market share). Together they are called the “Big Three.” These three agencies are private companies located in the USA (Fitch is partly French owned and has HQs in London and New York) (El-Shagi 2009). Europe and Asia also have rating companies that do similar work, but they are much smaller and their impact is marginal.

The Dagong Global Credit Rating Company (short Dagong) from China rates sovereign bonds since 2010 and has very different outcomes as compared to the “Big Three” American agencies. Dagong presented their first Sovereign Credit Rating Report of 50 Countries in 2010. During the presentation its director challenged the “Big Three”: "Western credit rating agencies are highly ideological, politicized and do not adhere to objective standards. All this is fully supported by evidence which show that the big Western agencies are too close to their customers."

(11)

11

Table 1

Cases in which Dagong’s sovereign credit ratings are higher than the US rating agencies as of December 31, 2010

No. Country Dagong Moody‟s S&P Fitch

1 China AA+(20) A1 (17) A+ (17) AA- (18) 2 Saudi Arab AA (19) Aa3 (18) AA- (18) AA- (18) 3 Russia A (16) Baa1 (14) BBB+ (14) BBB (13) 4 Brazil A- (15) Baa3 (12) BBB+ (14) BBB- (12) 5 India BBB (13) Baa3 (12) BBB- (12) BBB- (12) 6 Indonesia BBB- (12) Ba1 (11) BB+ (11) BB+ (11) 7 Venezuela BB+ (11) B2 (7) BB- (9) B+ (8) 8 Nigeria BB+ (11) - B+ (8) BB (10) 9 Argentina B (7) B3 (6) B- (6) B- (6)

Radu and Cosmin wrote a short paper about the inter-rater differences in credit ratings. Their research is however build entirely on argumentation and lacks empirical support. Nevertheless, their arguments are very useful for my introduction. It is easy to see that China rates itself higher, just as its neighbors India and Russia get better ratings. If I look closer at some of these countries, it is not difficult to see how Radu and Cosmin (2011) find reasons why they are rated higher. Venezuela for instance is a strong critic of the US government and therefore important in China’s ambition to become a global super power. The high rating of Nigeria can be explained by the recent $100 million of Chinese investments and Brazil is a large trade partner of China. There are also 18 cases (see appendix 2) in which ratings of Dagong are lower than those of the “Big Three”. A lot of these countries are western nations like Germany, the Netherlands and France. France and Germany are allies of the USA and NATO members and companies from Germany and France (e.g. car manufacturers) heavily compete with Chinese companies. But German and French companies also compete with American companies, just as China competes with the USA. The argument that Dagong rates NATO members or USA Allies lower does not apply in all cases, for instance Denmark (see appendix 3). Also New Zealand and Australia have similar ratings and they are outspoken allies of the US, as seen in the wars in Iraq and Afghanistan. In the case of Brazil, the connection with China is easy to find through the large Chinese investments in South America. However the US also trades a lot with Brazil and is very interested in the oil fields found of the coast of Rio de Janeiro. Some countries rated better by US agencies have large competitiveness with Chinese products (Germany, France). Another example: Japan, with a debt of 225% of its GDP obtained the same rating as the UK with a 50% debt. Furthermore, 9 countries that were rated higher are non-Western. According to Radu and Cosmin (2011) most of them are countries which China has strategic alliances, or large economic interest with (Radu & Cosmin 2011).

(12)

12

US for its large debt and rise of the debt ceiling. He warned the US for the volatile bond market. At the same time Dagong lowered its rating on US sovereign debt from A-plus to A, with a negative outlook. Dagong has drawn attention to its rating with a number of controversial calls on government credit worthiness. Their differences are large, but no empirical research1 has been done to compare those with the three mayor credit rating agencies S&P, Fitch and Moody’s.

During the recent Euro crisis many politicians have suggested that rating agencies are politically and economically motivated and that the American and Chinese rating agencies lower ratings on other countries, in order to improve their competitive position. The Eurogroups chairman, Juncker said in March that he was “not happy” with the rating agencies role in the European2 debt crisis (Bloomberg, 2011). Also the Wall Street Journal (Dean & Back, 2011) questions whether the rating agencies are playing the right role.

American politicians have accused Chinese rating agencies and vice versa. The Chinese Dagong is not taken very seriously by investors outside China and has therefore not received less criticism. These political motivated rating events might be true, but is there not any evidence. Downgraded countries will always try to find reasons other than “this downgrade is justified”. It could just be very true that a downgrade is correct without politicians admitting it. For a European politician it is easier to blame rating agencies, than to take the blame on themselves. Politicians admitting to failure would be a first.

Radu and Cosmin (2011) describe various reasons that sound logical, but they lack empirical evidence and their arguments only hold for individual examples. Their arguments can be well understood trough geopolitical reasoning. Geopolitics is the practice of analyzing the use of political power over a territory, and the relationship between political power and geographic space. It is a method of foreign policy analysis to explain and predict international political behavior (Evans & Newnham, 1998). In this particular case, geopolitics is used as an argument to explain differences in ratings. The geopolitical argument suggests that American or Chinese rating agencies would up/downgrade a country’s rating for political or economic reasons. They would do this to protect the country’s foreign policy, interests, ambitions and investments.

I am not convinced by these arguments and think that it is too easy to claim that rating agencies from both the US or China give countries a better or worse rating on purpose. Sinclair (2010) and El-Shagi (2009) explain that credit rating agencies are merely referees and that we are witnessing an example of “kill the messenger”. Although credit rating agencies have made mistakes in the past, they have never been scandalized or suffered a significant loss in franchise value as a result of misconduct (Smith & Walter, 2001).

1_{Except a comparison by Dagong itself, see “Sovereign Credit Rating Report of 50 Countries in} 2010” http://www.dagongcredit.com/ and a paper by Radu & Cosmin, 2011

(13)

13

Ratings are created by people and based on the information people have about a country. A lack of knowledge creates uncertainty, something investors fear most. Uncertain investors demand higher interest rates. I think that rating agencies process insufficient knowledge in a similar way and therefor give a lower rating. I believe that a good flow of information is needed for a proper rating. If this flow of information is jeopardized or biased in any way, this will influence the rating agents and lower the rating. This thesis focusses on these inter-rater differences and aims to find an explanation for this. I will test the influence of distance on ratings by looking at the differences between US ratings and Chinese ratings.

1.2 Research question

Rating agencies create ratings by using macro-economic fundamentals (qualitative) and their opinion or interpretation of a country’s economic situation. The economic fundamentals, that rating agencies use are similar for both Chinese and USA agencies. The structural characteristic of these fundamentals are the same. The interpretation of fundamentals however can differ between agencies, especially if the home-country is different. Radu and Cosmin (2001), politicians like Juncker and many others call these differences politically biased. I believe that these differences should be studied properly before such statements can be made. In thesis I ask and answer the question:

Which factors explain inter-rater differences in sovereign credit ratings given by American and Chinese rating agencies??

This research question is very broad but I explain all my steps in more detail along the way. My main goal is to find empirically supported factors that explain the differences between ratings given by American and Chinese credit rating agencies.

I follow up on the work of Cantor & Packer (1996) who were among the first to study the determinants of credit ratings. To understand differences one first needs to understand what defines sovereign ratings, and how their impact affects economies world-wide. In the second chapter of this thesis I focus on the basics of sovereign credit ratings. This chapter overviews the history of credit ratings and discusses the relevant academic literature on this topic. Step-by-step I explain the definition of sovereign credit ratings and describe the history of rating agencies and how they work. To understand how ratings are created I continue by writing about determinants of ratings. A theoretical model by Cantor & Packer (1994) predicts 90% of credit ratings by Moody’s and S&P. In the next section of the literature review I describe in details how ratings impact countries and regions. In the final section I describe the most recent example: the debt crisis in Europe.

(14)

14

discussion making of people -and with that influences ratings- is the key element in my thesis. The distance between the sample countries on one side, and the USA or China on the other side forms the basis. The distance between countries is well understood through the four dimensions of Ghemawat (2001). On these dimensions I will base what might explain inter-rater differences in sovereign credit ratings.

The fourth chapter deals with the operationalization of the variables. I use the four dimensions distance by Ghemawat (2001) but I have to transform them in order to use them in my empirical model. The empirical model is then tested in chapter five. First I test the variables separately and then in the complete model. In the third section I develop several robustness checks to test the validity and strength of the model.

(15)

15 Chapter 2: Literature review

“When I was young I thought that money was the most important thing in life; now that I am old I know that it is.“

(16)

16

2.1 Introduction

This chapter overviews and summarizes the recent literature on the topic of sovereign credit ratings. I start at the beginning, going back more than 100 years and describe the history behind the rating industry. I will then explain what defines sovereign bonds are and how credit risks are assessed and rated by the various rating agencies. Credit ratings are not without several controversies. To understand ratings better I explain what economic factors and fundamentals figures determine sovereign credit assessment theoretically. This is primarily based on the work of Cantor and Packer (1996) and elaborated later on by many others. The financial crisis in East-Asia in the late 1990s is used to illustrate some of the problems with credit rating agencies.

Then I continue with the self-fulfilling prophecy of ratings. Ratings influence macro-economic fundamentals that are later used to create the credit assessments. Especially interest rates are highly influenced by credit events. The final section of this chapter is used to describe the key elements of the recent debt crisis.

2.2 Sovereign credit risk Sovereign bonds

Countries borrow money by handing out sovereign bonds, a debt security issued by a national government within a given country (Eiteman, Stonehill & Moffet, 2010). By doing this they create public debt. Most sovereign bonds have a maturity of 10 years, but other maturities like 15 and 20 years are possible. The nominal value presents the amount that is borrowed. On top of the nominal value comes the interest. Interest is associated with the risk. When the risk of default or restructuring of debt increases the risk premium increases (Reinhart, 2002). Bonds can be traded on most large stock exchanges in the world and for long time they were seen as low risk or risk-free investments. The recent debt crises in Europe however changed this view.

(17)

17

assessments by rating agencies to judge whether to buy or sell certain bonds (Juko 2011).

Sovereign risk and default

Of the different types of sovereign risk, one of the most important is the risk of default or the restructuring of debt (most cases both happen simultaneously) (Gapen, Gray, Cheng & Yingbin, 2005). This is also the type of sovereign risk discussed in the literature surrounding the topic. Gapen et al. (2005) explain sovereign default in comparison to corporate bankruptcy. Sovereign defaults do not trigger a well-defined bankruptcy process (like with corporate defaults) that applies equally across countries and investors. The sovereign debt market is characterized by the absence of a bankruptcy code. While corporations can go bankrupt, disappear or restart, sovereigns have no choice, a country cannot stop existing. In the event of default the lender cannot gain access to the debtors’ assets (Mellios & Paget-Blanc, 2006). Sovereign default triggers a complicated restructuring process whereby pre-default liabilities are exchanged for post-default liabilities (Gapen et al., 2005). In this restructuring process, holders of sovereign liabilities do not receive similar legal claims to ownership of sovereign assets in the event of a default (e.g. bondholders cannot assume control over a country, possess public sector entities or liquidate assets). Instead, the holders of sovereign debt have a claim on restructured debt of lower value in the event of default. The Russian default on bonds in 1998 illustrates this problem (see Mellios & Paget-Blanc, 2006).

The creditworthiness of a sovereign borrower is depending on the willingness to repay the debt, not just its ability. This raises the question why sovereigns would repay their debt. The lender cannot gain access to resources and there are no legal options to enforce payments. First there is the willingness of countries to maintain a good reputation to preserve future access to credit markets (Mellios & Paget-Blanc, 2006; Grossman & Van Huyck, 1988). Second, there is the possibility of economic, legal and political sanctions (Bulow and Rogoff, 1989). Finally, a default could have a negative impact on imports, exports, foreign direct investments and other forms of international trade (Rose, 2002; Mellios & Paget-Blanc, 2006). A recent example in repaying debts is the ongoing battle between the Dutch and British governments on one side, and the Icelandic government on the other. The Dutch and British government repaid bank account holders of Icesave the money they lost during the Icelandic banking crisis in 2008. Ever since, the repayment of this debt to The Netherlands and Britain has been a point of debate (Felton & Reinhart, 2008). The Dutch and British use international pressure to withhold Icelandic membership to the EU to enforce payment.

(18)

18

similar information, they all want to know the risk of their investment. This creates a demand for evaluations of bonds. While large banks have the resources to maintain their own assessments, smaller investors feel forced to use ratings from agencies even if they do not believe in their qualities. If enough investors trust and use (buy or sell bonds) a rating, the market will respond anyway, so in a way every investor is dependent on credit rating agencies. This generates a market for enterprises which provide quality, trusted ratings (El-shagi, 2009). Credit rating agencies like Moody’s have stepped in this gap. Given the large economies of scale and the needed reputation the credit rating industry is highly concentrated around three rating agencies in the US. There is also a handful smaller, specific, agencies but new entries are hardly observed (Ferri, Lui & Stiglitz (1999).

2.3 Credit ratings History

Rating agencies began existence in the last decades of the 19th century (Cantor & Packer, 1994). Louis Tappan founded the first mercantile rating agency in 1841, Robert Dunn acquired this firm later and published their first guide book on ratings in 1859. John Bradstreet had published a similar ratings book two years earlier. Although ratings started in the 19th century, it was not until 1909, when John Moody started rating railroad bonds on a regular basis that the industry really kicked off. By the mid-1920s, nearly 100 percent of the US bond market was rated by Moody’s (Sinclair, 2010). In 1916 Poors Publishing Company, in 1922 Standards Statistic Company and in 1924 Fitch Publishing Company respectively published their first ratings. The first two merged in 1941 to become Standard & Poors. Duffs and Phelps started publishing some ratings in 1932but started providing bond ratings on a wide spread of companies in 1982.

When the financial market became a more globalized market the rating agencies started looking and rating across borders. Before that, rating agencies were little known outside the USA. Until the mid-1990s most European and Asian companies relied on their market reputations to secure financing (Sinclair, 2010). Since the late 1980’s the demand for sovereign credit rating expanded (Cantor & Packer, 1996). More governments with greater default risk were engaged in the international bond markets and investors needed trustworthy ratings. The uprising of emerging markets like Brazil, Russia, India and China (BRIC countries), also increased the need for ratings. Many investors invest their money in emerging markets and want the same rating agencies that they know from the USA to rate these markets as well.

(19)

19

investors put on sovereign credit ratings is high. Rating agencies nowadays provide a wide range of ratings for developed, emerging and developing countries.

Agencies

Nowadays three major rating agencies hold more than 90% of market share (El-shagi, 2009). The three largest credit rating agencies are Moody’s Investors Service, Fitch Ratings and Standard and Poors (S&P). Rating agencies provide a wide range of ratings on corporate and sovereign bonds and various other financial products, and give recent updates on their creditworthiness. This paper focusses on sovereign credit bonds; however the occasional jump to corporate credit ratings is inevitable. Rating agencies assign a grade to a borrower according to its degree of relative creditworthiness (Mellios & Paget-Blanc, 2006; Peters, 2002). The primary purpose of obtaining a rating is to enhance access to private capital markets and lower debt issuance and interest costs. Early theoretical work indicated that rating agencies play a role as information gatherers. They can reduce an issuing firm’s capital costs by certifying its value in the financial market. By doing this they solve or reduce the information asymmetry between investors and issuers (Ramakrishnan & Thakor, 1984; Millon & Thakor; 1985 and UN, 2008). For sovereign borrowers, there is evidence of a clear correlation between bond spreads and rating.

Regulations

Regulations force banks and institutional investors to hold certain capital requirements and back them up with high rated assets and bonds. Although the regulations differ per country the main ideas are similar and therefore easily explained. The basis for regulations on banks is the Basel II and III agreements. Under Basel II agreements, banks are supposed to hold large numbers of triple-A assets (highest rating) in their portfolio. Crotty (2009) describes how banks only need small amounts of capital to support triple-A securities. Furthermore, in the USA the major financial institutions have to hold assets of high level ratings and even a high number of triple-A ratings from one of the three major credit rating agencies. As soon as a rating drops to junk level, many investors have to sell them for regulatory reasons. Basel III continues on the basics of the Basel II agreement but tightens and strengthens bank capital requirements and introduces new regulatory requirements on bank liquidity and bank leverage (UN 2008). The idea of the agreements is that capital requirements of banks should be closely linked to asset risk (El-shagi, 2009). The Basel III agreement is being implemented in stages starting in 2011 and will be fully implemented in 2019. It is estimated that these new agreements have a negative impact of 0.05% to 0.15% on the world GDP growth (Jones, 2010).

Sovereign credit ratings

(20)

20

however buy bonds in currencies other than dollars and euro’s keeping the need for local currency ratings also high. However foreign currency ratings, which form the basis of this thesis, are still the prevalent and influential ratings in the international bonds markets (Cantor & Packer, 1996 and Mellios & Paget-Blanc, 2006). Usually the ratings on foreign currency denominated bonds are slightly lower than those of local currency bonds (Mellios & Paget-Blanc, 2006).

(21)

21

Table 2

Rating numbers per agency

Source: Mellios and Paget-Blanc (2006)

The conflict of interest: charging bond issuers

Initially the agencies provided public ratings free of charge to the issuers and charged the user or reader (Cantor & Packer, 1994). They financed their operations through the sale of publications. These publications however were easily copied and spread and they did not generate enough to sustain the intensive coverage. Rating agencies started to find ways to charge issuers of bonds for a rating. When Penn Central defaulted in the 1970’s, it bond investors suffered an $82 million loss. Trust in bonds went down as the US went into a recession. Issuers needed solid ratings to sell their bonds on the market. Issuers with lower ratings pay higher interest rates (embodying larger risk premium) than higher-rated issuers (Ferri et al., 1999). Besides affecting the cost at which issuers can borrow, ratings determine the extent of potential investors. Furthermore, regulations and statutes can forbid institutional investors to invest in bonds below certain level. The growing demand for ratings boomed the rating business and the agencies found ways to charge issuers.

Charging issuers of bonds for a rating has been a platform for debate ever since. It could give rating agencies the incentive to assign higher ratings to satisfy issuers and clients. There is a large conflict of interest in this business model of rating agencies (Crockett, Harris, Mishkin and White 2003). “Conflicts of interest occur Number Broad S&P Moody's Fitch Dagong Interpretation

21 8 AAA Aaa AAA AAA Highest quality

20 7 AA+ Aa1 AA+ AA+

19 7 AA Aa2 AA AA High Quality

18 7 AA- Aa3 AA- AA-

17 6 A+ A1 A+ A+

16 6 A A2 A A Strong payment capacity

15 6 A- A3 A- A-

14 5 BBB+ Baa1 BBB+ BBB+

13 5 BBB Baa2 BBB BBB Adequate payment

capacity

12 5 BBB- Baa3 BBB- BBB-

11 4 BB+ Ba1 BB+ BB+

10 4 BB Ba2 BB BB Likely to fulfill payment,

ongoing uncertain

9 4 BB- Ba3 BB- BB-

8 3 B+ B1 B+ B+

7 3 B B2 B B High risk obligation

6 3 B- B3 B- B-

5 2 CCC+ Caa1 CCC+ CCC+

4 2 CCC Caa2 CCC CCC Current vulnerability to

default

3 2 CCC- Caa3 CCC- CCC-

2 1 CC Ca CC CC

1 1 C C C C In bankruptcy or default

(22)

22

when a financial service provider, or an agent within such a provider, has multiple interests that create incentives to act in such a way as to misuse information” (Crockett et al., 2003, page 2). Rating agencies gave high ratings to packages with subprime-mortgage which should never have gotten these ratings. The Wall Street Banks like Lehman Brothers, Morgan Stanley and Merrill Lynch had so much power that they could force S&P, Moody’s and Fitch into giving their products high ratings (Gärtner et al., 2011, Lewis & Einhorn, 2009, Crotty, 2009). This led eventually to the collapse of the financial market in september 2008 and the end of investment bank Lehman Brothers (Grymbaum, 2008).

Rating agencies could also be put under political pressure to rate the US sovereign and corporations higher to protect its economy and trade. Although no empirical evidence has been found so far, the European Committee and researchers like Ferri have suggested that a German or European rating agency should be formed to counterbalance the weight of the US rating agencies. Ferri made this claim in 2007, before the crisis in the EMU-zone started (Ferri & Lacitignola, 2007).

Cantor & Packer (1994) see the reputation of credit rating agencies as a moderator for this kind of behavior. If credit rating agencies lost the confidence of investors in their ratings, issuers would no longer believe that ratings were to lower their costs. Every time a rating is assigned the agency puts its reputation on the line. Cantor & Packer find no evidence or mayor scandals to support the claim that rating agencies would rate higher to satisfy issuers. The paper by Cantor and Packer however was written in 1994, long before the financial crisis and scandals in 2007/2008. A more recent paper by Sinclair (2010) follows up on this topic and does not agree with the statement by many that the conflict of interest in ratings is a problem. He acknowledges the conflict of interest exists but no more than in universities in which students pay intuitions that support the salaries of the professors who grade their examinations. Rating agencies, like universities, manage this dilemma, through codes of conduct (Sinclair, 2010). Sinclair sees rating agencies as referees during a soccer match. They do not make the rules; they follow them and can therefore not be blamed as much as is done by politicians and economists.

Solicited and unsolicited ratings

(23)

23

agencies is another reason that unsolicited ratings are lower. It is easier to upgrade a rating, than to downgrade one.

S&P claims that since 2001 they have not done any unsolicited ratings outside the USA, Moody’s on their part says they have not done any unsolicited ratings in Europe. They both also specify in their ratings whether they were solicited or not and that they give issuers the opportunity to participate in any stage of the process.

2.4 Determinants of credit ratings Determinants

Ratings are based on quantitative and qualitative variables that the rating agencies consider reliable, (Cantor & Packer, 1996; Peters, 2002). The quantitative variables contain economic and financial figures that can easily be found. The quantitative measures are available through the various websites of rating agencies and can be recalculated. The qualitative measures remain mostly secret. It is the opinion that rating agencies have about the country’s ability and willingness to pay interest and principal. Their opinion is not based on quantitative measures and cannot be recalculated, which leaves room for debate.

(24)

24

Afonso (2003) adds that external debt has an important influence on developing countries. Afonso (2011) repeats his study, using a slightly different (improved) methodology and with data from 1995 till 2005. His results confirm the determinants for sovereign credit ratings in table 3. Afonso also makes a distinction between long-term and short-long-term impact on ratings. Government effectiveness, external debt, foreign reserves and sovereign default are determinants for the cross-country dimension and therefore only have a long-term impact. The per capita income, real GDP growth, government debt and government deficit are determinants on the short-term.

Table 3

Determinants of sovereign credit rating

Source: Cantor & Packer 1996 and Mellios & Paget-Blanc 2006.

Afonso (2011) further emphasizes the importance of the fiscal variables. The main conclusion however states that the determinants by Cantor & Packer (1996) are the main determinants of sovereign credit ratings. In total 40% of the ratings were estimated correctly, and 75% lie within one notch from the actual rating. In this thesis I will focus on the six most important determinants as measured by Cantor & Packer (1996) and displayed in table 3.

Determinant Explanation

Per capita income

An increase of the per capita income implies a larger potential tax base and a greater ability for a country to repay debt. GDP growth An increasing rate of economic growth tends to decrease the

relative debt burden. Moreover, it may help in avoiding insolvency problems.

Inflation A low inflation rate reveals sustainable monetary and exchange rate policies. It can also be seen as a proxy of the quality of economic management.

External debt A large debt makes it harder for a country to repay lenders.

Level of

economic development

Developed countries are integrated within the world-economy and are less inclined to default on their foreign debt in order to avoid sanctions from the lenders.

(25)

25

Table 4

Factors determining credit ratings, principal components analysis

Source: Mellios & Paget-Blanc (1996)

Mellios & Paget-Blanc (2006) do a similar study but use data on more recent years and add the ratings given by Fitch. Their principal components analysis (PCA) (table 4) on 50 variables separates thirteen factors that explain sovereign credit ratings. The explanatory factors by Mellios & Paget-Blanc (1996) are highly correlated with the determinants by Cantor & Packer. Their derived model predicts 74% of the ratings within one notch, 95% is explained within two notches.

Rating differences: differences in the USA

Hill et al. (2010) study the differences in credit rating by Moody’s, Fitch and S&P. With data on 129 countries from 1990 until 2006 they show that the sovereign ratings often disagree but remain within one or two notches from each other. In an earlier study by Ammer and Packer (2000) is shown that although the rating agencies differ in methodology, the ratings show great similarities for both corporate and sovereign credit ratings. Hill et al. (2010) also study the methodology of rating agencies and find that changes in GDP growth, the direction of the rating event, and probabilities derived from the rating levels equation, are all significant factors for the three rating agencies. For S&P, the change in Institutional Investor rating and market risk premium are significant, for Moody’s the change in GDP per capita is significant, and for Fitch the changes in external debt are also significant.

All theoretical models of the determinants of sovereign credit ratings have an error somewhat between 10% (within 2 notches difference) and 25% (within 1 notch) in predicting the ratings of the three rating agencies. This is partly due to the

Factor Percentage of variance explained Cumulative percentage of variance explained Development level 21.05 21.05 Public indebtedness 11.60 32.65 Quality of governance/political stability 9.70 42.35 Economic growth 7.42 49.77 Money supply 5.95 55.72 External liquidity 4.96 60.68 External indebtedness and

openness

4.33 65.01 Inflationary pressure 3.52 68.53 Net investment inflows 3.12 71.65 Size of the economy 3.04 74.69

Competitiveness 2.67 77.36

Debt servicing 2.34 79.70

(26)

26

undefined qualitative criteria by credit rating agencies. The minor differences are also due to the impact the ratings could have on these variable. The impact of credit ratings is further discussed in the next paragraph.

2.5 The impact of ratings Impact

Sovereign yield spreads seem to increase when ratings decline (Cantor & Packer, 1996; Reisen & Maltzan, 1999, Gärtner et al., 2011). Some economists believe that rating agencies promote herding due to their role as information multipliers and as such, they might accelerate crises (Ferri et al., 1999). Since the price of sovereign bonds is defined by the underlying value and interest (risk premium) the effects of herding behavior would be seen in the interest yields. The assumed risk grows larger as the ratings are downgraded, and interest rates increase.

Reisen & Maltzan (1999) studied 103 sovereign rating events and the effect on the yield spreads 30 days after and 30 days before the rating event. They also examine the moment a country is put on the watch list. The impact of ratings is measured by yield spread on sovereign bonds. When only one rating agency changes the rating or outlook the impact is insignificant. This changes when the rating event is confirmed by one of the others. Rating announcements by two or three agencies can produce significant impact on yield spreads in the expected direction (Reisen & Maltzan, 1999). A discussion paper by the UN (UN 2008, page 9) summarizes: “Reisen and Maltzan have found that ratings are correlated with sovereign bond yield spreads.” During and after the Mexican crisis in 1994-1995 (part of their sample), Reisen & Maltzan (1999) find a two-way causality between ratings and interest yield spreads. Mishkin (1996) describes high interest yield spreads as one of the main factors that started the Mexican crisis. International capital markets react to changes in the ratings and this is reflected in sovereign bond yield spread. Reisen & Maltzan (1999) also show a highly significant impact during announcement, for instance when sovereigns are put on a watch list. Furthermore the study finds that there is a significant effect during downgrades, but not during upgrades. Stated diffently, interest spreads go up after downgrades, but do not go down after upgrades.

Economic and financial integration between markets and the growing globalization has made countries vulnerable to contagion. Most crises are not limited to one country, but spread among the region or the world. Kaminsky, Reinhart and Vegh (2003) find that financial linkages are the main reason of contagion, trade linkages are of less influence.

(27)

27

that these spillover effects are highly correlated with capital and trade flows. In line with Kaminsky et al. (2003) earlier work, Gande and Parsley further (2005) suggest that the spillover effect is stronger through capital and financial linkages than through trade linkages.

Pro-cyclical behavior

Rating agencies were caught by surprise when the East Asian crisis started in the late 1990’s. Once the crisis had started, their reaction was strong and all countries (except Malaysia) were downgraded from investment to non-investments grades (Ferri et al., 1999). Some countries (Indonesia and South Korea) were downgraded more than six notches in a very short period, which is highly unusual. Ferri et al (1999) focus on the pro cyclical behavior of rating agencies. They show, by using the determinants of Cantor & Packer (1996), that ratings of four East Asian high-growth countries before the crisis were slightly higher than expected. During and after the crisis they see that the ratings decline much more than the determinants and economic fundamentals would justify. Ferri et al. (1999) then continue by stating that it is apparent that the rating agencies attach higher weights to their qualitative judgment than to economic fundamentals, especially during crises. One year after the crisis the ratings by agencies and modeled ratings converge towards each other. According to Ferri et al (1999) this is expectable because the rating agencies have so much impact on a country and its potential investors that the economic fundamentals deteriorate.

Mulder and Perrelli (2001) suggest that Ferri et al (1999) used the wrong sample to support their claim of pro-cyclical exaggeration of the crisis. The estimations by Ferri et al (1999) ended shortly after the beginning of the crisis. One could argue for example that the crisis was still deepening and that the downgrades correctly anticipated this deepening. Therefore, Mulder and Perrelli (2001) redo the study with a longer time range of data. Their findings however confirm Ferri et al (1999). Mulder and Perrelli (2001) find significant overshooting of the credit ratings during the crisis in East Asia. Their findings show the self-fulfilling prophecy of ratings and show how rating agencies try to protect their reputation at all costs. To protect this reputation, rating agencies have to be pro-cyclical (Reinhart, 2001; Cantor & Packer, 1996). Investors expect them to lower ratings during crises, even if the economic fundamentals do not indicate this. Rating agencies made the protection of their reputation protection more important than giving quality ratings. Reinhart (2001) shows that credit rating agencies systematically fail to predict crisis. Hence, if ratings were not pro-cyclical they would be able to predict crises by lowering ratings before the crisis, not after. Reinhart also names reputation as an important reason for this pro-cyclical behavior.

(28)

pro-28

cyclical role. This is discussed earlier in this section with the discussion of the paper by Reisen and Maltzan (1999).

Mishkin (1996) uses the Mexican crisis as an example for his theory on financial crises and names the increase of uncertainty, the exchange rate and the increasing interest rates as main factors of financial crises. The increase in interest rates can be linked to the impact of sovereign credit ratings. Ferri et al (1999) use data from the period that the Mexican crisis took place and prove the claim of pro-cyclical effects. I assume that the Mexican crisis also holds some evidence for the pro-cyclical behavior of sovereign credit ratings Although not explicitly stated by the authors, I believe that the paper by Mishkin (1996), as well as the paper of Reissen and Maltzan (1999), both support the views of Ferri et al. (1999), Reinhart (2001) and Cantor & Packer (1996) that rating agencies are pro-cyclical.

Many authors believe that the rating agencies have worsened the crisis; others even go further by stating that rating agencies caused the East Asian crisis (See Hillebrand, 2001). However, not all academic literature supports this pro-cyclical claim. El-shagi (2009) claims that the empirical evidence claiming pro-cyclical effect of ratings is not convincing enough. Although ratings have an impact on capital flows and the economy of countries, it is not proven to be more substantial than other factors or minor disturbances in the market. El-shagi (2009) explains that there is correlation between sovereign credit ratings and interest spreads, but that this correlation does not mean causality. According to him, the argument that rating agencies caused or worsened the crisis is unsupported (El-Shagi, 2009) because agencies started downgrading the countries after the start of the crisis. El-shagi (2009) further claims that the arguments of pro-cyclical effects are unconvincing. In all countries, the last rating downgrade occurred at the peak of the crisis or even later. If ratings would have been the cause of acceleration, some further capital flight would be visible after the last downgrade. Any reaction around the end of the crisis would mean that the agencies could not predict the end of the crisis. The evidence from the Granger causality, test used by Reisen and Maltzan (1999), is not sufficient to prove causality between ratings and interest. It is likely that sovereign credit ratings had short-term impacts of financial markets. Many of the papers have focused on a time horizon of several days, surrounding the rating event which could indicate the expected effect, but short-term effects are not evidence for acceleration of the crisis.

(29)

29

2.6 Crisis of the PIIGS

Many countries in Europe are suffering under the current debt crisis which threatens the continuation of the Euro (Gärtner et all, 2010). The crisis that started in Greece quickly spread to Portugal, Cyprus, Spain, and Italy. Ireland is also suffering under the crisis, but their problems are slightly different. The large public debts of these European countries are the core of the current crisis (Stein, 2011).. The crisis in Europe has its roots in the financial and banking crisis in 2008. Many financial institutions and banks had to be bailed out or be supported with public funds in 2008 and 2009 in both the United States and Europe.

The debt crisis in the EMU started in Greece. The Greek government used wrong statistics to gain access to the EMU with the help of Morgan Stanley in the late 1990’s. Morgan Stanley created special bonds, swaps, and other complicated financial products which made Greece’s financial situation look very good in the short term, but created a large debt for the future. Ten years later, Greece got into trouble because it could no longer repay its debt and interest. This large public debt triggered the downgrading of Greece’s credit rating and a large widening of the interest yield spread especially when compared with German bonds in late 2009 to April 2010 (Stein, 2011). To refrain from these kinds of problems the Stability and Growth Pact (SGP) was signed by all member states of the EMU as part of the Maastricht Treaty. The government budget deficits of many European countries however are in excess of the Stability and Growth Pact (SGP). Despite many warnings, multiple violations of the Pact occurred. Many of these violations were made by Germany and France, the two largest members (Issing, 2011). Greece has one of the largest debts, which is way beyond the 60% that is allowed.

Large public debt and government mismanagement was also the cause of the crisis in Portugal that soon erupted after the Greek problems (Stein, 2011). Basically the governments of Greece and Portugal caused the crisis because they created the debt. In figure 1 it is clearly shown that the debts of Greece and Portugal kept rising until the beginning of the crisis. In Spain and Ireland the causes lay in the private sector (banking and housing sector). The bubble/crash in these markets caused the economies to fail. In Ireland the housing and mortgage market boomed, in 2008 the people and business in Ireland had borrowed the equivalent of 60 % of GDP for property. The banking sector in total exceeded the GDP of Ireland in that year. When the financial crisis occurred in the housing market the banking sector collapsed. The net worth of banks vanished and they became insolvent. The government had no other choice than to bail-out or take large equity stakes in most banks. Although Irelands public debt was low prior to the crisis, the bail-out package created a large deficit and was basically the cause of the Irish debt crisis.

Figure 2

(30)

30

Source: Stein (2011), EconStat and IMF World Economic Outlook

The crisis in Spain also finds its roots in the private sector. Housing prices boomed in from 2003 – 2007. The growth in debt was facilitated by the availability of cheap credit through the international financial market. When the prices of houses dropped, the banks were unable to repay their loans to the international investors. The Spanish government had to save many banks and financial institutions. The effect of this bail-out package is seen in the sudden rise of debt in 2008 and 2009. Other countries in the EU also suffered from the economic crisis and had to bailout their banks. They also had to help Greece with cheap and restructured loans.

The problems in Greece, Italy, and Spain have resulted in a deep crisis in the EU which even puts the future of the Euro at risk. “The euro, which together with the single market is the foundation for growth and prosperity– also in Germany – is in danger. If we don’t deal with this danger, then the consequences in Europe are incalculable, and the consequences beyond Europe are incalculable too . . . because if the euro fails, then Europe fails. But if we do deal with this danger, the euro and Europe will be stronger than before.” These were the words of Germany’s Federal Chancellor, Angela Merkel, in a speech to the German Bundestag on 19th May 2010 on the measures to stabilize the euro (Issing, 2011). Until this day the problems in the EU and the Euro-zone have not been resolved. There is even been spoken about a Greek exit from the Euro.

Rating agencies have a significant impact on markets and play a big role in the recent downgrades of the PIIGS and the interest rates on their bonds. The rating agencies play a key role in the recent crisis. Therefor it is important to know whether these rating agencies can be trusted and are not biased by home country effects. Earlier research on the banking crisis in 2007 (Crotty, 2009, Gärtner et al., 2011) is not giving very positive signs about the rating agencies’ trustworthiness.

(31)

(32)

32 Chapter 3: Research Design

“Never spend your money before you have earned it. “

(33)

33

3.1 Introduction

Over the past years several researchers and politicians have blamed rating agencies for causing or worsening crises. One argument they give is political motivated, claiming that agencies use their power to help their home-country. In this paragraph I discuss the theory and arguments behind those ideas which are closely related to chauvinism and protectionism. I however, do not think those arguments explain these differences. They also lack empirical evidence. I believe that cultural differences, language and distance have a big influence during the rating process. I explain the theory of bounded-rationality in relation to these factors. To determine the distance between countries, four dimensions of (geographic, cultural, administrative and economic distance) CAGE, by Ghemawat (2001) are introduced. I believe that these distances explain how inter-rater differences originate.

3.2 Theoretical Explanations

Chauvinism and “new” protectionism

Governments have many incentives to protect their own economies. Sentiment, chauvinism and upcoming elections are all reasons for governments to enforce laws, rules and regulations that help their home-country. This does not only affect the home-country, it also affects other countries and even the world economy. It is called protectionism (Enderwick, 2011).

(34)

34

Ideas behind protectionism are first of all to protect the country’s own economy and trade. A second argument is that trade and foreign investments increase dependency on other nations and large foreign firms. Several countries do not want their airlines, financial institutions, defense contractors and media industries in foreign investors’ hands. The third argument is that countries want to prevent a race to the bottom where cheap products from abroad destroy local price levels (Enderwick, 2011). Since the start of the financial crisis in 2008 there has been a debate about the “new protectionism” (Enderwick, 2011). The bail-out and help for banks by governments is a good example of this “new protectionism” or “financial protectionism”. Financial protectionism is defined as a nationalistic change in banks’ lending behavior, as the result of public intervention. It leads domestic banks either to lend less or at higher interest rates to foreigners (or both) (Eichengreen & O’Rourke, 2010 and Enderwick, 2011). Since governments own (parts) of banks since the crisis they can stimulate this behavior.

Government support can be seen as protectionism, just like saving large corporations (e.g. France helping Peugeot). Financial protectionism goes further than just support: it reduces foreign bank lending and investments (Rose & Wieladek, 2011). Although it is a sparse and new field of research and the empirical evidence is still weak, the arguments are sound and make sense theoretically. Rose and Wieladek, (2011) are among the first to find evidence that the behavior of foreign banks in the UK seems to be consistent with financial protectionism. They do not however find evidence to support that British banks do the same with foreign lending or that nationalized British banks changed their lending behavior in any substantive way. Leaders of the G20 in 2009 in Pittsburg3_{have said: “We will not retreat into financial}

protectionism” which, according to me, practically means: we know it is there, but we deny it! Also the former Prime Minister of England, Gordon Brown and several senior European Bankers have talked about the issues of financial protectionism. In an interview with the Wallstreet Journal4, Brown says he sees banks retreating in the

3_{http://www.g20.org/Documents/pittsburgh_summit_leaders_statement_250909.pdf} 4_{February 1, 2009. Interview Transcript: Gordon Brown. Wallstreet Journal.} http://online.wsj.com/article/SB123354149126338139.html

Box 1: Chauvinism

The term chauvinism dates back to the times of Napoleon. The eponym chauvinism comes from the legendary, possible apocryphal French soldier and patriot Nicholas Chauvin. He is supposed to have served in the First Army of the French Republic in La Grance Armée of Napoleon. According to the legend he was born in Rochefort around 1780 and enlisted at the age of 18. He was a brave and honorable soldier. The story tells that he was wounded 17 times. For his loyalty and dedication Napoleon supposedly presented him the Sabre of Honor.

(35)

35

home-country systems. This could not only be dangerous for the British banking sector and economy, but even disastrous for emerging markets like the Central and Eastern European countries. There is no World Trade Organization (WTO) for financial disputes, which makes dealing with financial protectionism far more complicated.

Home-country bias in investment portfolios

Evidence and theoretical thoughts on financial protectionism can also be found in the theories about home-country bias in investments portfolios. Home country bias is the investors' natural tendency to be attracted to investments in domestic markets. Investors tend to focus more on their home markets, and the companies that do business within these markets because they are familiar with them (Hasan & Simaan, 2000). Home country bias is a well-documented phenomenon and discussed very briefly in this thesis5. Research has shown that international portfolio diversification results in better returns for US investors (Driessen & Laeven, 2007). Hence: it is better to invest in more countries than in the home-country only. This positive effect has been dropping in the US each year for the past two decades as US companies become more internationalized, but is still positive compared to domestic investment only. The main reason is that spread of risk is associated with more diversification, and spreading risk gives higher returns. The positive effects are largest for developing and high risk countries. U.S. investors ignore these figures and have invested 90% of their portfolio within in the US in 2007 (Driessen & Laeven, 2007), and they even tend to invest less in companies with an international or global scope. In smaller portions, the home country bias is present between neighboring countries, or countries within a free-trade zone like the European Union or the NAFTA. Investors that go abroad invest their money in countries that are close by and well known.

Home country bias also exists in product evaluation. Consumers are often positively biased in their evaluation of domestic products (Verlegh, 2007). This tendency is higher in areas or moments in times with high unemployment. Consumers see foreign competition as a threat to local manufacturers. Hence: when an economy slows down, people and businesses tend to be more chauvinistic. If this is the case for investors and consumers, then there is a good reason to expect that the corporations that rate financial products (like sovereign bonds) do a similar thing. The question is if this is the reason that credit ratings differ. In the next section I go deeper into the process of rating.

Bounded rationality

Cantor & Packer (1994) described the 7 most important determinants for sovereign ratings, per capita income, GDP growth, inflation, external debt, level of economic development, and default history. These determinants are based on economic fundamentals which are the same whether you are Chinese or American. While the economic fundamentals are similar, the interpretation can vary. Credit

(36)

36

ratings are opinions by people that work at the various rating agencies. Panels of researchers, experts and analysts of the rating agencies base their rating partly on economic fundamentals (quantitative), and partly on common sense, views on politics and judgment of the country sovereign credit ratings (qualitative). Therefore a Chinese agency can have a different perspective on a country than an American.

Many factors like geographic distance or cultural differences can influence how information about a country is perceived. The culture of China is very different from the American culture. Let me explain this with an example. If both a Chinese agency and an American agency provide a credit rating for Canada, it is much easier for the American agency to find and process the information. Not only is the culture of Canada close to the American culture, they also speak the same language. Furthermore they are neighbours. China does not have a border with Canada, nor do they speak the same language, or have a similar culture. For a Chinese agency it is therefore much more complicated to find and process the information. Rating agencies are profit-based companies and have limited resources (e.g. time, money etc.) to find and process information. Imperfect information leads to higher uncertainty and high uncertainty leads to higher costs, or in this case lower ratings. A similar effect also occurs when investors demand higher interest on their investments because the uncertainty is high.

The whole idea of increasing costs is in close relation with to the transaction cost theory. A transaction cost is a cost, incurred during an economic transaction. The costs can be made within the company (internal) or outside (external). Both are subjected to distance since a larger distance leads to a loss of information. The farther away a country is, the more time and costs are involved in gathering reliable information.

Maintaining their reputation gives rating agencies the incentive to keep ratings low. “Dependency on reputation has become the principal market-driven safeguard against exploitation of any potential conflicts of interest in the ratings’ business” (Cantor and Packer, 1996).

(37)

37

3.4 Distance between countries

Measuring the distance between countries and comparing them with each other is not as easy as one would expect. The easiest way to measure the distance between two countries to take the kilometers between both capital cities; this is called the absolute geographical distance. But geographical distance can be measured in many other ways, trough airline fares or the cost of shipping goods. Other dimensions of distance are cultural distance, administrative distance, and economic distance (Ghemewat 2001). Figure 3 shows the four dimensions of distance and their expected influence on credit ratings. Each of the dimensions is further discussed in the remaining of this section.

Figure 3

Inter-rater differences

Geographic distance

The first variable for measuring the distance between two places is the distance in kilometers (Conley & Ligon, 2002), the absolute geographic distance. Generally speaking, the further a country is away, the harder (and more costly) it is to do business with. Prices for shipping increase and also knowledge sharing becomes more complicated. It also takes more time to travel and differences in time-zones makes communication tougher. It is easier and cheaper to do business with countries that are physically closer. If two countries share a border this effect is even stronger because the flow of knowledge is very easy.

Cultural distance

Culture is the way people interact, communicate, work and live within a group. It distinguishes life in one group from other groups. Cultures can differ between groups, regions, countries and companies. Everything people do is influenced by the national (or regional) culture (Hofstede 1984). Differences in religion, social values and language can create distance between two countries. The

(38)

38

information flow between countries is highly influenced by culture. If two countries have similar cultural beliefs and values it is much easier to interpret the information flow, but when the cultures are very different there is a loss of information. Language also plays an important role here. If the language is similar, or if the same language is spoken by a large minority, the cultural gap is much smaller. This leads to a much better flow of information. It is also seen in other relations such as trade. Countries that speak the same language trade three times more (Ghemewat 2001). If the languages are not similar people get lost in translation which leads to an imperfect information flow.

Administrative distance

Countries that have similar institutions or government policies have a relatively small administrative distance. All sorts of agreements between countries decrease the administrative distance. Furthermore the political relation is important, also from a historic perspective. Colonial ties for instance have a large impact on trade. Tariffs, trade quota, protectionism, and restriction on foreign direct investments (FDI) are examples of attempts to increase distance. Countries have a tendency to protect their labor market, national champions (big firms), agriculture, and food supply. They also try to protect national security by keeping vital services (e.g. telecommunication, infrastructure, airlines) within national hands. Defense contractors or weapon manufactures are also subject to much government interference. They all increase the distance between countries.

Inter-rater differences in sovereign credit ratings How distance influences ratings