واکنش بازار ارز به خبر اوراق قرضه

We analyse the reaction of the foreign exchange spot market to sovereign credit signals by Fitch, Moody’s and S&P during 1994–2010. We find that positive and negative credit news affects both the own-country exchange rate and other countries’ exchange rates. We provide evidence on unequal responses to the three agencies’ signals. Fitch signals induce the most timely market responses, and the market also reacts strongly to S&P negative outlook signals. Credit outlook and watch actions and multiple notch rating changes have more impact than one-notch rating changes. Considerable differences in the market reactions to sovereign credit events are highlighted in emerging versus developed economies, and in various geographical regions.

Currency crises in emerging markets highlight the importance of exchange rate behaviour for economic stability and attract interest in exchange rate dynamics. Such events include the depreciation of Latin American currencies in the 1980s, the collapse of the Mexican peso in 1994, the crashes of the Turkish Lira in 1994 and 2001, the 1999 currency crisis in Brazil, and the collapse of the currency board in Argentina in 2002 (Loría et al., 2010). During the 1997 Asian crisis, unstable exchange rates were partly responsible for the collapse of the trade and other sectors of the economies in Indonesia, Korea and Thailand. The recent global financial crisis was accompanied by exchange rate volatility, including the US$-euro. Credit rating agencies (CRAs) play a central role in international financial markets through disclosing credit information, not only via rating changes but also via outlook and watch actions. While rating changes communicate permanent changes in issuer credit quality, credit outlook and watch are supplemental instruments to signal potential rating adjustments.1 Prior studies show that outlook and watch signals are at least as important as rating changes in their market impact (e.g. Hand et al., 1992, Kaminsky and Schmukler, 2002 and Sy, 2004). Each CRA has a clear reputational goal of providing timely and high quality signals to financial markets. Nevertheless, CRAs were criticised for failing to properly assess the risks in the securities backed by sub-prime mortgages prior to the recent financial crisis. In response to the perceived role of CRAs in the crisis, the International Organisation of Securities Commissions (IOSCO) revised the Code of Conduct Fundamentals for CRAs in 2008, and a formal regulation on CRAs was approved in 2009 by the European Parliament. These actions seek to increase competition in the rating industry, and imply ongoing scrutiny of the relative performance of CRAs. This paper analyses the relative information content of sovereign credit actions of the three major players in the rating industry: Moody’s Investors Service, Standard and Poor’s (S&P) and Fitch Ratings. Using an extensive data sample for 1994–2010, we examine how the foreign exchange spot market reacts to credit events for 124 developed and emerging economies, and also investigate spillover effects (i.e. the impact on other countries’ exchange rates). Sovereign ratings are particularly important for emerging economies because risk is greater and information flows can be of lower quality than for developed economies. Investors pay close attention to sovereign ratings when investing capital in emerging countries. Credit risk changes are more frequent in emerging economies and large changes can occur quickly and unpredictably. Many market participants believe that there is an added value in multiple ratings (Baker and Mansi, 2002). We examine relative CRA reputations, by investigating whether sovereign credit signals released by a particular CRA have a stronger influence than those of other CRAs. We also analyse whether any information lead of one CRA relates to specific type of signals, and whether it is evident for developed and/or emerging economies. Alsakka and ap Gwilym (2010) examine the relative timeliness of sovereign rating actions by five CRAs (Moody’s, S&P, Fitch, and two Japanese agencies JCR and R&I), and show that Moody’s is the first mover in upgrading sovereigns, while S&P rating changes are the most independent of other CRAs. Alsakka and ap Gwilym (2012) find that Moody’s tends to be the first mover for positive outlook and watch signals, while S&P (Fitch) shows the least (most) links with other CRAs’ outlook/watch actions. We focus the analysis on two perspectives: (i) the relative information content of the sovereign actions by the major CRAs for the foreign exchange market, highlighting any informational lead of a given CRA in a specific economy (developed or emerging), particular geographical region or a specific type of credit signal (positive versus negative events, and rating changes versus outlook actions versus watch announcements); and (ii) the own-country exchange rate responses and the regional spillover effects. We contribute to the existing literature on the market impact of credit ratings in five respects. First, while prior studies on the information content of sovereign ratings have considered equity and bond markets and currency crises, the literature offers little evidence on foreign exchange market reactions.2 Second, we provide evidence on both national and regional spillover effects of sovereign credit signals. Third, whereas prior research on CRAs’ actions has mainly centred on rating changes, we investigate the relative impact of rating changes, outlook signals and watch events. Fourth, we extend the methodology previously applied in the literature on the information content of ratings by employing a logit-type transformation of the numerical rating scale to account for non-linearity (see Sy, 2004). Finally, there has been a little prior empirical analysis of the relative information content of the credit signals of different CRAs. The main findings are summarised as follows. First, positive and negative sovereign credit signals affect the own-country exchange rate and other countries’ exchange rates. Second, there are unequal responses to the three CRAs’ signals. In the case of the own-country reactions, we find that Fitch signals induce the most timely reactions, particularly in developed countries, and S&P offers informative negative outlook signals, especially in emerging economies. Moody’s demonstrates an information lead for upgrades in developed economies and for downgrades in emerging economies, while the reverse is true for S&P. The significant impacts of Moody’s positive outlook and watch events (S&P negative watch signals) are only seen in emerging (developed) economies. In addition, the market generally reacts more strongly in the cases of negative news, larger rating adjustments and outlook and watch events. For regional spillover effects, we find informational leads in favour of: (a) Fitch actions in Latin America, (b) positive actions by Moody’s and negative signals by both Moody’s and S&P in East and South Asia; and (c) negative events by S&P and Fitch in the Middle East and North Africa and Sub-Saharan Africa. Credit signals from the three CRAs significantly contribute to the contagion in Europe and Central Asia, with the exception of Fitch positive events. The rest of the paper is organised as follows. Section 2 discusses the importance of sovereign ratings and the previous literature on their impact on financial markets. Section 3 describes the data set, while Section 4 presents the methodology used to examine the effects of sovereign credit signals. Section 5 discusses the results, and Section 6 concludes the paper.

There is wide interest in empirical research on exchange rates due to their important role in global economic and financial performance. There is prior literature regarding the effects of sovereign rating changes on the bond and equity markets, but there is very little prior evidence on the impact of sovereign credit events on the dynamics of foreign exchange markets. In addition, very few studies have properly considered the relative information content of sovereign credit actions across the three major CRAs (for any financial market). This paper identifies whether any CRA demonstrates an informational lead on the basis of reactions in the own-country exchange rate and other countries’ exchange rates to sovereign credit actions. We study three types of credit signals across different economies and geographical regions. We use an extensive sample of daily sovereign ratings and credit outlook and watch assigned for 124 countries by Fitch, Moody’s and S&P during 1994–2010. We also apply a logit-type transformation of numerical rating scores to account for potential non-linearity in the rating scale. There is evidence that sovereign credit signals affect the own-country exchange rate, and that there are unequal reactions to the signals of the three CRAs. We provide some evidence on an immediate information lead in favour of Fitch relative to both Moody’s and S&P, which contrasts with evidence from stock and bond markets. We also find a unique market reaction to S&P in the case of negative outlook signals. We identify stronger market reactions to negative versus positive credit signals, and to large rating adjustments and outlook/watch announcements versus one-notch rating changes. The rating changes are, to some extent, anticipated when they are preceded by outlook or watch signals, and therefore, considering the outlook and watch status is a very important element in assessing the complete credit opinion. Importantly, and in contrast to prior studies on the information content of sovereign credit actions for bond and stock markets, we find significant responses to rating upgrades by the three CRAs and to positive watch signals by Moody’s and Fitch. The impact of credit signals varies considerably (and across CRAs) in emerging versus developed economies, as follows. (i) The effects of Moody’s downgrades and positive outlook and watch signals are only significant in emerging economies, while upgrades and negative watch signals by Moody’s are only significant in developed economies; (ii) The effects of upgrades and negative outlook actions by S&P are only significant in emerging economies, while downgrades and negative watch signals by S&P are associated with significant currency depreciation in developed economies only; and (iii) The immediate exchange rate responses for upgrades and positive watch signals by Fitch appear only in developed economies. The results show significant effects of a sovereign credit signal of one country on the exchange rates of other countries (i.e. spillover effects), and we identify inter-agency differences across regions in this case. We provide evidence of stronger responses to Fitch relative to both Moody’s and S&P in ‘LA’, as only Fitch sovereign credit events spillover to non-event currencies in ‘LA’. There is a contrast for ‘E S Asia’, where Moody’s positive events demonstrate a lead, yet negative events by both Moody’s and S&P contribute to contagion in this region. Negative signals by S&P and Fitch (not Moody’s) spillover to non-event currencies in the ‘ME-Af’ region. Finally, CRA signals significantly spillover to non-event currencies in ‘EU-CA’, with the sole exception of Fitch positive events. It is important to reflect on the inter-agency and cross-country differences which are summarised above. Several reasons can drive differential market reactions. CRAs take credit actions at different times and this will often reflect their different interpretations of the same economic, financial or political factors for a given country. Good recent examples of this are the sovereign ratings of Portugal and Ireland, which were both rated at speculative grade by Moody’s during 2011 while retaining investment grade ratings from S&P and Fitch. This also reflects the common occurrence of split ratings (i.e. CRA differences of opinion), especially for emerging markets and in crisis periods. Another factor is that a given CRA may demonstrate a consistent ‘first-mover’ advantage in certain markets or regions (e.g. Alsakka and ap Gwilym, 2010 and Alsakka and ap Gwilym, 2012). Similarly, Hill and Faff (2010) find that the ‘leading’ agency varies across economies and time periods. Reputational effects are a further aspect, whereby issuers are more likely to hire a CRA if it has a strong reputation in a specific market or region (Duff and Einig, 2009 and Hill and Faff, 2010). The strongest spillover effects are observed for the two regions which have more highly developed and integrated capital markets (‘EU-CA’ and ‘E S Asia’). For the ‘EU-CA’ region, strong spillovers within short time windows arise from Moody’s positive actions and S&P negative actions, which are both consistent with indications of ‘first-mover’ status in Alsakka and ap Gwilym, 2010 and Alsakka and ap Gwilym, 2012. The results offer an important contribution in evaluating and understanding the link between sovereign credit signals and the dynamics of exchange rates. Our evidence on the relative national and international information content of sovereign credit events by the three CRAs will interest many market participants. The findings are also particularly relevant to the credit rating industry given the expectation of increased competition in light of the recent and forthcoming regulatory changes in Europe and North America.