Does sovereign debt ratings news spill over
to international stock markets?
q
Miguel A. Ferreira
a,
*
, Paulo M. Gama
b
a
ISCTE Business School-Lisbon, CEMAF, Complexo INDEG/ISCTE,
Av. Prof. Anibal Bettencourt, 1600-189 Lisboa, Portugal
b
University of Coimbra, School of Economics, Av. Dias da Silva 165, 3004-512 Coimbra, Portugal
Received 29 September 2005; accepted 1 December 2006
Available online 25 January 2007
Abstract
The evidence here indicates that sovereign debt rating and credit outlook changes of one country
have an asymmetric and economically significant effect on the stock market returns of other countries
over 1989–2003. There is a negative reaction of 51 basis points (two-day return spread vis-a
´
-vis the
US) to a credit ratings downgrade of one notch in a common information spillover around the world.
Upgrades, however, have no significant impact on return spreads of countries abroad. Closeness (e.g.,
geographic proximity) and emerging market status amplify the effect of a spillover. Downgrade spill-
over effects at the industry level are more pronounced in traded goods and small industries.
2007 Elsevier B.V. All rights reserved.
JEL classification: F30; G14; G15
Keywords: Sovereign ratings; Spillover effects; Stock market
1. Introduction
Does news about sovereign debt rating in one country impact stock markets in other
countries? Our evidence indicates that negative sovereign rating news does spill over.
0378-4266/$ - see front matter 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jbankfin.2006.12.006
q
We thank an anonymous referee, Amar Gande, and seminar participants at the University of Porto for helpful
comments.
*
Corresponding author. Tel.: +351 21 7958607; fax: +351 21 7958605.
E-mail address: miguel.ferreira@iscte.pt (M.A. Ferreira).
Journal of Banking & Finance 31 (2007) 3162–3182
www.elsevier.com/locate/jbf
We focus on the cross-country stock market reaction to Standard & Poor’s (S&P)
announcements of a sovereign credit rating or credit outlook change.
There is published research on this question. Brooks et al. (2004) study the own-country
stock market impact of sovereign debt rating changes. They find that sovereign rating
downgrades have a negative impact on the re-rated country’s stock market (one-day
abnormal returns of 197 basis points), but upgrades have an insign ificant effect. Gande
and Parsley (2005) find asymmetr ic international spillover effects on sovereign debt mar-
kets. Downgrades abroad are associated with a significant increase in sovereign bond
spreads (12 basis points), but upgrades have an insignificant effect. Kaminsky and Schmu-
kler (2002) show that emerging market sovereign rating news is contagious for bond and
stock markets in emerging markets, particularly during periods of turmoil and particularly
for neighboring countries.
The empirical question we address is whether sovereign rating news of one country is
also relevant for other countries’ stock markets. If market players see rating changes as
a country-specific issue with no implications beyond country borders, little information
impact would be expected. At the same time, either rational behavior due to liquidity con-
straints or irrational herding of investors and financial and real sector linkages across
countries can act as transmission vehicles for country shocks (Dornbusch et al., 2000; Kar-
olyi, 2003).
We extend the Gande and Parsley (2005) findings by investigating information spillover
not only across countries but also across markets. That is, we focus on spillovers of credit
rating or outlook of one country (the event country) to stock market return spreads (the
return differential vis-a
´
-vis the US) of all other countries (the non-event countries).
We consider a large set of co untries, including not only emerging markets (18 countries)
but also deve loped markets (11 countries), representing stocks totaling USD 4.9 trillion of
market capitalization in 2002. We explicitly control for recent rating activity worldwide.
We characterize the spillovers in economic terms, i.e., by including controls for capital
flows and level of economic and financial development.
We also present several new results regarding cross-country and cross-market news
spillover at the industry level. The evidence with regard to industry portfolios is of partic-
ular interest, given increasing investor perception that industry factors are becoming more
important than country fact ors in explaining stock returns (see, for example, Cavaglia
et al., 2000).
A sovereign credit rating represents a rating agency assessment of the capacity and the
willingness of a sovereign obligator to meet its debt service payments in a timely fashion.
They are understood by rating agencies as a forward-looking estimate of default probabil-
ity; see Standard & Poor’s ‘‘Sovereign Credit Ratings (2005)’’. In most cases, the sovereign
ceiling doctrine applies; that is, the rating assigned to non-sovereign debt issues (or issuers)
is the same as or lower than the rating assigned to the sovereign of the country of domicile.
Thus, sovereign rating revisions also relate to non-sovereign debt instruments (Radelet
and Sachs, 1998; BIS, ‘‘International Convergence’’, 2004).
1
1
The Basel II Accord provides examples of the sovereign rating ceiling doctrine. Under the standardized
approach to minimum capital requirements for bank claims (option 1), all banks incorporated in a given country
are assigned a risk weight one category less favorable than the risk assigned to claims on the sovereign of that
country. For claims on corporations, no claim on an unrated corporation can be given a risk weight more
favorable than that assigned to its sovereign country of incorporation.
M.A. Ferreira, P.M. Gama / Journal of Banking & Finance 31 (2007) 3162–3182 3163
The stock market should be expected to react to a sovereign rating downgrade because
a downgrade can affect a country’s ability to borrow in international markets, and thus
contribute to a credit crunch, which negatively impacts the stock market. Other mecha-
nisms as well reveal a link between sovereign rating and stock markets. Sovereign rating
can provide information on the future economic health of the rated country that is not
otherwise available to stock market participants, and governments can take policy actions
that directly affect companies’ future prospects (e.g., raising corporate taxes to compensate
for increased debt service following a downgrade). Moreover, because many institutional
investors can hold only investment-grade instruments, rating downgrades and (upgrades)
may have a negative (positive) impact on security prices.
Our work is related to the literature on stock and bond market correlation. Campbell
and Ammer (1993) show that stocks and bonds react in the same direction to news about
fundamentals, with the exception of news about inflation. Other evidence on low (or even
negative) stock-bond correlatio n is inconclusive as to the expected direction of the reaction
to a rati ng revision (Connolly et al., 2005). At the corporate bond level, there is theoretical
support for both a positive and a negative reaction to corporate bond rating downgrades;
see, for instance, Zaima and McCarthy (1988).
Our major findings can be summarized as follows. First, we find rating changes in one
country incorporate valuable information for the aggregate stock market returns of other
countries. The spillover effect is asymmetric, both in direction of the reaction and in terms
of economic impact. A one-notch rating downgrade abroad is associated with a statisti-
cally signi ficant negative return spread of 51 basis points on average across non-event
countries. No significant impact is found for rating upgrades.
Second, controlling for time-invariant characteristics that proxy for underlying similar-
ities between countries affects the asymmetric nature of spillovers. We control for the cul-
tural, regional, and institutional environment as well as level of economic and financial
development. We find that geographic distance is inversely related to the spillover impac t.
This is consistent with the hypothesis that rating news has a more pronounced effect in
countries nearby where the information asymmetry is moderated. We also find that the
downgrade impact is more pronounced in emerging stock markets. This is consistent with
the hypothesis of a more pronounced common information spillover across emerging
markets.
Finally, we show that rating downgrades also hav e a significant effect on local industry
portfolio return spreads. Sovereign rating downgrades abroad are associated with a highly
statistically significant negative two-day return spread of industry portfolios vis-a
´
-vis their
counterpart industry in the US. The negative effect of downgrades is pervasive across
industries, but it is more pronounced in traded-goods and small industries.
2. Research design
We discuss our data, rating events, and test procedures separately.
2.1. Data
We examine the cross-country spillover effects of sovereign rating revisions using the
S&P hist ory of sovereign rating for the cou ntries analyzed by Gande and Parsley (2005)
that are included in the TF Datastream Global Equity Indices database. The data cover
3164 M.A. Ferreira, P.M. Gama / Journal of Banking & Finance 31 (2007) 3162–3182
the period from July 3, 1989, through December 31, 2003. The starting date represents the
first complete month S&P debt rating and credit outlook information are ava ilable.
We prefer the S&P foreign currency long-term rating history over other agencies’ rating
history because of data availability. Moreover, S&P tends to be more active in making rat-
ing revisions, and tends to lead other agencies in re-rating (Kaminsky and Schmukler,
2002; Brooks et al., 2004; Gande and Parsley, 2005 ). Foreign currency rating announce-
ments by S&P also seem to convey a greater own-country stock market impact and seem
not to be fully anticipated by the market (Reisen and von Maltzan, 1999; Brooks et al.,
2004).
2
A preliminary analysis of our data shows that the re-rated country two-day [0, 1] return
spread relative to the US is, on average, 146 basis points on S&P rating downgrades
announcement days. This result is consistent with the own-market wealth effects of down-
grades documented by Brooks et al. (2004).
The countries in our dataset must meet two criteria. They have publicly traded US dol-
lar-denominated sovereign debt, and country-level portfolio total return index data are
available in the TF Datastream database. The 29 countries meeting these criteria are:
Argentina, Austria, Belgium, Brazil, Canada, Chile, China, Colombia, Denmark, Finland,
Greece, Hungary, Indonesia, Ireland, Israel, Italy, Korea, Malaysia, Mexico, New Zea-
land, Philippines, Poland, South Africa, Spa in, Sweden, Thailand, Turkey, the UK, and
Venezuela. Thus, we build a geographically balanced sample that includes both emerging
and developed countries. The stock market indexes consider ed here represent about 80%
of each country’s stock market capitalization and are constructed using similar methods
across countries.
We also use data on several country-specific control variables (Table A.1 in the Appen-
dix details the variable definitions and data sources). Classification of countries as emerg-
ing or developed is based on Morgan Stanley Capital International, S&P, and ISI
Emerging Markets. A country is classified as emerging if it is listed as emerging by at least
one of these sources.
3
We consider bilateral dummy variables for sharing a common language, adjacency (or
common land border), legal tradition, and membership in a formal trade bloc, either the
North American Free Trade Agreement (NAFTA), the Mercado Comun del Sur (Merco-
sur), the European Union (EU), or the Association of South East Asian Nations
(ASEAN). We also explicitly control for physical distance between countries, computed
as the great circle distance between capital cities. These variables are intended to control
for historical factors that may influence spillover effects because they proxy for similarities
between countries that could heighten common spillover effects; see Gande and Parsley
(2005). Geographic factors akin to our control variables are standard controls in the liter-
ature explaining cross-country economic flows and also relate to linkages across stock
markets (Rose, 2000; Portes and Rey, 2005).
We explicitly control for crisis periods by including dummy variables for the European
Exchange Rate Mechanism crisis of 1992–1993, the Tequila crisis of 1994, the Asian Flu of
1997, and the recent crises in Russia, Brazil, Turkey, and Argentina. These crisis periods
2
Sovereign rating history comes from the S&P website: http//www.standardandpoors.com.
3
Greece is the only country in the sample that was upgraded from emerging to developed either by S&P or
MSCI; we classify it as an emerging market. Countries classified as developed are Austria, Belgium, Canada,
Denmark, Finland, Ireland, Italy, New Zealand, Spain, Sweden, and the UK.
M.A. Ferreira, P.M. Gama / Journal of Banking & Finance 31 (2007) 3162–3182 3165
include a total of 49 rating events. Finally, we use the Bekaert and Harvey (2000) and
Bekaert et al. (2003) ‘‘official liberalization’’ dates to control for emerging market segmen-
tation from the world market due to regulatory constraints on international capital
flows.
2.2. Ratings events
We define a rating event as a change in either the explicit credit rating or the credit out-
look assigned to a specific sovereign foreign currency debt. This is consistent with recent
work on the spillover effects of sovereign rating revisions that accounts for effective rating
announcements as well as information on imminent rating actions in a comprehensive
credit rating (CCR) measure. The changes to CCR define our rating events.
Table A.2 in the Appendix presents the details on the numerical coding of the CCR mea-
sure. First, we map letter explicit rating to numerical codes by a linear transformation to a
scale from 0 (the lowest rating, SD/D) to 20 (the highest rating, AAA). Next, we add the
credit outlook information (on a scale between 1 for a negative credit outlook and +1
for a positive credit outlook) to the rating numerical code. Any non-zero change in the com-
prehensive credit rating measure defines the events of interest: ‘‘upgrade’’, a positive change
resulting from an upward move in the (letter) credit rating of the sovereign or from a favor-
able revision in the credit outlook; and, ‘‘downgrade’’, a negative change resulting from a
downward move in the (letter) rating or from an unfavorable revision in the credit outlook.
Table 1 describes the soverei gn rating events sample. There are 106 upgrades and 109
downgrades between July 1989 and December 2003. The vast majority of events are
announced individually (for one country on a given day), although multiple-event days
occur for 14.1% of the upgrades, and 3.7% of the downgrades (see Panel A). The time clus-
tering of events can also be evaluated by looking at the average time elapsed between
events and the time periods in which they occur. Panel B of Table 1 shows that just over
50% of the events (54 upgrades and 59 downgrades) occur within a window of two weeks
(ten trading days). Panel C shows that just over 45% of the events (54 upgrades and 44
downgrades) are announcements made after 1998.
The strong temporal association of events suggests the use of a short event-window in
evaluating the impact of rating revisions and to explicitly control for worldwide recent rat-
ing activity. The use of a long event-window can bias results because stock returns in the
(longer) event wind ow can incorporate rating changes in countries beyond the country
being evaluated. Moreover, if markets see rating revisions in the context of recent rating
activity, today’s reaction will be a function of prior rating revisions.
In fact, Kaminsky and Reinhart (2000) show that domestic markets become sharply
more susceptible to crises elsewhere if a core group of countries (not one single country)
are already affected. If the same type of behavior characterizes home–country reaction to
sovereign rating changes abroad, this implies that events in other countries can cumulate.
Panel D of Table 1 divides the rating events by emerging or developed country. Not
surprisingly, the vast majority of events, about 85%, occur in emerging markets. Investi-
gating whether rating news also affects developed stock markets has been overlooked in
the literature.
Finally, Panel E of Table 1 divides rating events according to the change in the CCR
measure. The vast majority of events are one-notch changes, although half-notch events
3166 M.A. Ferreira, P.M. Gama / Journal of Banking & Finance 31 (2007) 3162–3182
