SOVEREIGN RISK RATINGS IN LATIN AMERICA: MEASURES OF ECONOMIC CONDITIONS OR POLICY PARADIGM COMPLIANCE?

• Author: Pinheiro, Diogo L.

INTRODUCTION

In our daily lives, measures of risk have become increasingly commonplace. We have our personal credit scores, which are supposed to signal to lenders the likelihood that we will repay our debts; we are aware of how certain actions we take and how certain parts of our history affect our perceived risk to insurers, with the accompanying changes in premiums we pay for certain protections; and so on. A similar process has also affected nation-states and national governments. Investors, banks, and all other sorts of economic actors have tried to devise measures that allow them to evaluate and compare just how risky it is to invest in any particular nation. In an era of increased flows of capital globally, these measures are supposed to serve as a sort of measuring stick for different possibilities of investment, and they can have a significant impact on a nation's prosperity.

The business of measuring risks associated with credit markets is big business--one now dominated by credit rating agencies (CRAs). (1) They include such well-known entities as Moody's and Standard and Poor's. With thousands of ratings issued each year, it is not surprising that one of the main CRAs, Moody's, had net income in excess of $1.7 billion in 2020. (2) Indeed, different sources have put the estimated operational profit margins for the main CRAs at somewhere between 40 and 50 percent. (3) These are sizable profit margins that are supposedly payment to these CRAs to use their reputation to assuage (or warn) investors of potential risks in the credit market. The power of these agencies is such that they have even been written into regulations and laws holding the distinguished position of nationally recognized statistical research organizations (NRSROs). (4)

Despite their financial success, these CRAs have come under intense attacks from all sources: the popular press, academics, and even the International Monetary Fund. (5) According to the bulk of the criticism, these CRAs effectively have a license to "print money" because of the role governments have established for them. (6) This creates an oligopolistic market where agencies are given the power to decide what are acceptable risks, backed up by significant regulations that are not affected by hits to their reputations that are caused by inaccurate or bad ratings. Thus, not surprisingly, there is extensive literature discussing the extent to which CRAs' success in the corporate bond market is driven by producing accurate information or by shoddy regulation. (7) However, little research addresses the impact that these agencies have had on development economics and the policies of developing nations. (8)

As late as 1981, the biggest agencies only rated a handful of countries--with most located in the developed world and all of them earning the highest ratings. It was only in the last couple of decades that these agencies have started to rate developing nations--and thus becoming part of the larger trend of measuring "country risk." (9) Country-based risk ratings are a weakly defined set of measures that intend to encapsulate the risks involved in a growingly interconnected world economy. (10) Banks, institutional investors, and financial news services like the Economist Intelligence Unit all have their own measures of country risk, either published or for internal use. The most basic definition of what they involve is that they include "all additional risks induced by doing business abroad, as opposed to domestic transactions." (11) A country's credit risk is called sovereign risk. It refers to the risk of default by the central government of a given nation. The sovereign risk rating, with a few exceptions, becomes the "ceiling" for all other entities located within that country--ranging from corporations to local governments. (12) As such, beyond the impact on a country's access to credit markets, these ratings also affect all local corporations and local governments. As Thomas Friedman once famously said: "There are two superpowers in the world today in my opinion. There is the United States and there's Moody's Bond Rating Service. The United States can destroy you by dropping bombs, and Moody's can destroy you by downgrading your bonds. And believe me, it's not clear sometimes who's more powerful." (13)

Given their visibility and widespread usage by corporations and governments alike in explaining their decisions, it comes as a surprise that there is no unified theory of sovereign risk that sustains the practical efforts of so many agencies. The key notions used by these agencies come from standard financial theories. "Risk," "rewards," and the language of the models employed is familiar to those who have worked and studied financial markets more generally, but "sovereign risk" is measured in an ad hoc manner, with significant, often undisclosed differences between agencies. The ratings are given in letter grades, and the agencies themselves do not provide any guidelines as to what these letter grades mean in concrete probability terms. (14) This is the case even as these ratings have been included in the latest attempts to create a set of international banking regulations through the Basel II accords of the Bank for International Settlements. (15)

This paper is concerned with the relationship between sovereign risk ratings and policy in the Global South. In particular, I am concerned with the determinants of country risk rating and their evaluation of policies adopted, and how their actual performance fits with existing economic and sociological theories. Policy and political factors are at the heart of sovereign risk ratings. In fact, all three agencies major agencies (Moody's, Standard and Poors, Fitch) use that as a selling point for their ratings. (16) One of their claims is that their experts and their inclusion of "qualitative" information in their rating process allow them to be forward-looking and identify potentially risky situations that the nuts and bolts of data analysis cannot. And yet the impact that policy has on these ratings remains under-analyzed. (17) If certain policies are rewarded with better ratings while not being useful predictors, then they serve as an important policy diffusion mechanism, rewarding conformity with international paradigms.

This is not to suggest that scholars have not tried to understand just what determines sovereign ratings. But to the extent that they have, they have treated them as a mostly economic variable, with little if any political implications. There is a vast literature that attempts precisely to pinpoint what factors affect them, but it rarely, if ever, focuses on policy and political issues. (18) To the extent that possible sources of bias have been considered, these have tended to focus on potential bias against poor nations, a bias for downgrading countries more aggressively than upgrading them, or biases that result from ignoring the quality of institutions. (19) The few exceptions to this can be found in the work of Glen Biglaiser and colleagues. (20) In the studies that measure policy impact, the key finding is that the adoption of neoliberal reforms regarding international trade has a significant and positive impact on sovereign ratings, but no other type of neoliberal reform has a significant positive impact on ratings. (21)

My focus here is on developing nations. More precisely, I will focus on Latin America, as a number of studies on the spread of neoliberalism have done in the past. (22) I begin with a discussion of the historical background and theoretical underpinnings of sovereign ratings. Next, I explain why I focus on Latin America, and why that is important and significant. This is followed by a description of the measures used in this paper and the methods I use in estimating the determinants of sovereign risk ratings and the determinants of a nation's likelihood of default. The paper closes with a discussion of my findings.

A BRIEF HISTORY OF CREDIT RATING AGENCIES

Credit ratings predate the appearance of credit rating agencies. In the mid-1800s, credit reporting agencies started to develop scores that encapsulated the bulk of the information contained in the reports. These agencies wrote extensive, qualitative reports on companies describing their credit history and trustworthiness. Barry Cohen describes how the R. G. Dun and the Bradstreet's Commercial agency first developed credit reports for companies seeking loans in the mid-1800s. (23) More than simply solving problems of lack of information in a growing economy, the assignment of specific, predetermined values to the creditworthiness of a company was a way to legitimize and confer objectivity to their analysis, even when the information they had available was not enough to allow them to make such precise judgments, but that was still a relatively minor part of a credit report and a credit reporting process. The CRAs that began to appear in the early 1900s, however, devoted themselves solely to estimating those letter grades.

The first organization devoted only to risk assessment--the initial CRA--appeared in 1909, when John Moody published Analysis of Railroad Investments. (24) The idea behind it was simple. Moody regularly published a statistical manual that contained most of the important data on railroad companies, as well as a rating designed to measure the likelihood that a certain company would default. (25) Other CRAs would soon emerge. Shortly after these companies started rating US municipalities, they also began rating sovereign debt. (26) By 1924, all four major rating agencies (Standard Statistics and Poor's had not yet been merged) issued sovereign risk ratings. However, a wave of defaults and the Great Depression made this perhaps one of the best examples of "institutions forgetting" something that was once institutionalized. (27) By the 1930s almost half of the nations rated had defaulted on their debt...