Sovereign Debt Crises and Asymmetric Information in Primary Bond Markets

1. 15th World Congress of the International Economic Association Sovereign Debt Crises through the Prism of Primary Bond Market Sebastian Nieto Parra Sciences Po Paris, Chaire Finances Internationales OECD Development Centre  Istanbul, June 2008

2. Motivation • Inefficiency in sovereign bond markets Asymmetries of information between capital markets’ actors • Behaviour and interactions between the three major actors of the Sovereign Bond Market : Governments Investment banks/lead managers Investors • It concerns the advantage of information that investment banks may have over investors.

3. Motivation Structure of the Prices in the Sovereign Bond Market

4. Motivation • Primary and secondary sovereign bond markets provide important information concerning risk perception of capital markets’ actors Risk perception of investors can be measured by the Sovereign bond spreads on the primary and secondary market Risk perception of investment banks can be measured by the remuneration that governments pay to investment banks in order to place bonds (i.e., underwriting fee)

5. Review of the literature Information problems in the emerging sovereign bond market - Flores (2007), and Flandreau and Flores (2007): Historical point of view. Role of underwriters as providers of information for investors. - Nieto-Parra and Santiso (2007): Positive recommendations given by Investment Banks when they are acting as Lead Managers. - Edwards (1997): Information pb between Wall Street analysts and their clients during the Mexican crisis of 1994 - Blustein (2003): Conflict of interest with which Investment banks are faced during the Argentinean crisis of 2001. - Calomiris (2003): Cooperation between research and origination departments. They are not followed by a systematic analysis of the structure of the primary bond market. Information problems for the recent sovereign debt crises.

6. Review of the literature Vast and relevant research literature on the primary corporate bond market 1 Literature related to the determinants of the underwriting fee: West (1967), Sorensen (1979), Higgins and Moore (1980), Kryzanowski et al. (1996), Lee et al. (1996), Altinkihc and Hansen (2000), How and Yeo (2000), Livingston and Miller (2000), Kollo and Sharpe (2002), Livingston and Zhou (2002) and Hua-Fang, (2005) Credit risk and profitability indicators are explanatory variables of the underwriting fee. 2 Relationship between the primary market and the recommendations given by underwriters: - Lin and McNichols (1998), Chen and Ritter (2000), Ljungqvist et al. (2006), Bradley et al. (2003), and Michaely and Womack (1999).

7. Description of the data Period: 1993-2006 Frequency: Annual 29 Countries: EMBI Index (JP Morgan) and countries for whom we have information on underwriting fee Argentina, Brazil, Bulgaria, Chile, China, Colombia, Dominican Republic, Ecuador, Egypt, El Salvador, Hungary, Indonesia, Lebanon, Malaysia, Mexico, Morocco, Pakistan, Panama, Peru, Philippines, Poland, Russia, South Africa, Thailand, Turkey, Ukraine, Uruguay, Venezuela and Vietnam. 1 Structure of the primary bond market (Underwriting fee and Primary sovereign bond spread): Standard issues: (i) ISIN reference number (ii) Coupon rate is not float (iii) Currency denomination (EUR, JPY, USD) (iv) No guarantee for the issue

8. Description of the data 427 issues (67% denominated in USD, 28% in EUR and 5% in JPY) Annual average of the underwriting fee and primary sovereign bond spread of the emerging countries 2 Secondary Sovereign bond spread (EMBI Index, JPMorgan) 3 Information received by investors from investment banks concerning the primary bond market we collect the major investment banks’ publications published by the most important financial actors in emerging countries.

9. Sovereign Debt Crises Standard definition employed mostly on the “early warning models”, a country is defined to be in a debt crisis if: 1. It is classified as being in default by Standard & Poor’s (S&P’s) , OR 2. It receives a large non-concessional IMF loan defined in excess of 100 percent of quota. A variety of crises: 1. S&P’s default: two groups depending on the restructuring case (Pre-emptive and post-default) 2. IMF loans: two groups depending on the vulnerability of public sector (i.e., risk of default of sovereign bonds).

10. Sovereign Debt Crises Sovereign Risk Countries Note: * denotes countries that experienced also a currency crisis during the 12 months prior and following the sovereign debt crisis. See next section for the definition of currency crises.

11. Hypothesis Efficiency of the sovereign bond market Market inefficiencies can arise when information is often asymmetrically held by market participants. In order to test market inefficiency in the emerging sovereign bond market, the null hypotheses used are the following: Hypothesis 1: Prior to sovereign debt crises, investors are not perfectly informed on the quality of the sovereign bonds issued by risk countries. By contrast, investment banks observed this risk before the onset of crises. Hypothesis 2: This asymmetric information is above all present in sovereign risk countries exposed with high public finances difficulties.

12. Hypothesis Efficiency of the sovereign bond market These hypotheses are validated when : H1: Prior to sovereign bond crises investment banks demand a high underwriting fee for “bad” countries with respect to the sovereign bond spread priced by investors for these countries. H2: By differentiating among sovereign debt crises, we note this effect is above all existent on countries that present sovereign risk difficulties.

13. Stylized Facts Fees and Sovereign Bond Spreads during Crises (Annual Basis)

14. Stylized Facts Fees and Sovereign Bond Spreads during different types of Crises Sovereign Risk Countries No Sovereign Risk Countries

15. Stylized Facts Underwriting and Primary Sovereign Bond Spreads 1993-2006(Annual basis)

16. Econometric analysis where is the underwriting spread received by investment banks from countryi in period t, is the sovereign bond spread (i.e., primary or secondary bond spreads) and it is taken in basis points (bp) is a dummy variable that takes the value of 1 for countries placed prior to the onset of a sovereign debt crisis (between T-3 and T-1) and 0 otherwise. is defined as the product of SBS and crisis is a time dummy variable. OLS and Fixed Time Effect estimation

17. Econometric analysis 1. An increase of 100 bp of the bond spread only implies an increase of 0,03 per cent of the underwriting fee. 2. H1: On average prior to crisis, countries paid 0,52 per cent of extra fee. This variable is statistically significant at 1 per cent. 3. H2: When we take into account ONLY sovereign risk countries, the fixed cost that these countries have to pay to investment banks is high (0,94 per cent of the proceeds).

18. Econometric analysis

19. Fee and financial markets’ actors • Availability of this information: Bloomberg and Dealogic However it is not accessible at the day of the issue: According to a member team of Dealogic at the end of 2007 “for about 80 % of large deals (more than US$200m equivalent) we should have the fee within 1 day”. Impact on secondary market prices… • Do investors concern themselves with underwriting fees? Questionnaire to investors in Wall Street (Alliance Bernstein, Alliance Capital, Fidelity, GE Asset Management, GMO, Goldman, Invesco and Western Asset)

20. Fee and financial markets’ actors Seven investors of the eight interviewed argue that underwriting fee is of no concern in investment decisions. Moreover they do not perceive underwriting fee as a good indicator of credit risk. 3. Investment banks’ publications on emerging sovereign bond markets. 12 investment banks covering the period 1997-2007ABN AMRO, Barclays Capital, Bear Stearns, Citigroup, Credit Suisse, Deutsche Bank, Dresdner Kleinwort Wasserstein, Goldman Sachs, JP Morgan, Lehman Brothers, Merrill Lynch and Morgan Stanley. Underwriting fee is not a piece of information given by investment banks to institutional investors.

21. Fee and financial markets’ actors 4. Why, therefore, do investors not pay attention to the evolution of underwriting fees? This is puzzling in that useful, publicly available information is not tracked by investors to help improve allocation of their emerging market fixed income assets.

22. Conclusions • investment banks price sovereign default risk well before crises and even before investors. This result suggests that investment banks hold an information advantage over investors • Investment banks’ behaviour differs depending on the type of sovereign debt crisis. Before crises investment banks charged a higher underwriting fee to countries presenting public bond vulnerabilities with respect to other sovereign crises. • There is a puzzle in that it appears that investors are not using potentially useful (and public) information in order to allocate efficiently their portfolios.