Macroeconomic News and Their Impact on Sovereign Credit Risk Premia

• Název práce v češtině: Makroekonomické zprávy a jejich vliv na kreditní prémii svrchovaného rizika
• Název v anglickém jazyce: Macroeconomic News and Their Impact on Sovereign Credit Risk Premia
• Klíčová slova: makroekonomické zprávy, suverénní riziko, očekávání investorů, SVE, GARCH, general-to-specific modelování
• Klíčová slova anglicky: macroeconomic news, sovereign risk, investor expectations, CEE, GARCH, general-to-specific modeling
• Akademický rok vypsání: 2012/2013
• Typ práce: diplomová práce
• Jazyk práce: angličtina
• Ústav: Institut ekonomických studií (23-IES)
• Vedoucí / školitel: Mgr. Václav Hausenblas
• Řešitel: skrytý - zadáno vedoucím/školitelem
• Datum přihlášení: 21.06.2013
• Datum zadání: 21.06.2013
• Datum a čas obhajoby: 24.06.2014 00:00
• Datum odevzdání elektronické podoby: 15.05.2014
• Datum proběhlé obhajoby: 24.06.2014
• Oponenti: Mgr. Božena Bobková

Předběžná náplň práce

Tato práce prozkoumá, jak makroekonomické zprávy ovlivní očekávání ohledně platební neschopnosti států. Svrchované kreditní prémie závisí na širokých mezinárodních faktorech, jako sklon k riziku, globální likvidita a nálada na trhu, jakož i globálních a lokálních makroekonomických fundamentálech. Makroekonomická data jsou však dostupná jen jako nízkofrekvenční data, což ztěžuje výzkum jejich dopadu v modelech na datech s vyšší frekvencí. Makroekonomické překvapení, tj. odchylka od trhem očekávané hodnoty, by mělo ovlivnit schopnost státu splatit dluh a tedy by mělo teoreticky úzce spjato se změnou v cenách. Přestože působení domácích makroekonomických překvapení bylo do určité míry studováno pro trh vládních dluhopisů, v literatuře je určitá mezera co se týče tohoto vztahu pro trh suverénních swapů úvěrového selhání (CDS). Jelikož na obě aktiva působí stejné fundamentální riziko a výzkumy odhalily jasné vztahy přenosu informací na obou trzích, předpokládáme, že dopad pro trh CDS bude srovnatelný. Prozkoumáme tedy efekt na obě tyto míry svrchovaného rizika.
Zatímco závěry předchozích studií nebyly jednoznačné co se týče makroekonomických determinantů CDS, a např. Nowak et al. (2011) nenašli žádné signifikantní efekty makroekonomických fundamentálů na CDS spready, předpokládáme, že kvůli náhlému zvýšení citlivosti na fundamentály pozorovaném v krizi budou investoři dávat větší důraz na makroekonomické zprávy než před krizí. Budeme tedy testovat přítomnost strukturální změny ve zkoumaném vztahu, která by se měla objevit během krize, během níž si investoři uvědomili, že dosud silně podceňovaly svrchované kreditní riziko.
Reakce investorů by měla záviset na fázi hospodářského cyklu, což dokládají Beber a Brandt (2010), a může být asymetrická vzhledem ke znaménku zpráv. Kromě efektu na hladinu prémie budeme studovat dopad na volatilitu spreadu a dopad na objem uzavřených CDS kontraktů. Pro makroekonomická překvapení důležitá jejich intenzitou a zemí původu je též možné, že dochází k spillover efektům, tedy budeme zkoumat jejich dopad na ostatní země.

Předběžná náplň práce v anglickém jazyce

This thesis will examine how macroeconomic announcements influence investors’ expectations of sovereign default risk. Sovereign credit risk premia depend on broad international factors such as risk appetite, liquidity and market sentiment as well as on global and local macroeconomic fundamentals. The latter are however generally available only at low frequency which hinders their effect from being well examined in higher frequency models. A macroeconomic surprise, i.e. deviation from what is expected in the market, should impact sovereign’s capability to repay and thus in theory be closely linked to change in prices. Even though the impact of domestic macroeconomic surprises has been researched to some extent for sovereign bond market, there is a gap in the literature with respect to the relationship in sovereign credit default swaps (CDS) market. As both assets are driven by the same underlying risk and researchers have found clear patterns of information transfer between these markets, it is hypothesized the impact of macroeconomic news is going to be similar. We will thus explore effects on both of these sovereign risk measures.
While previous research has not been unequivocal on macroeconomic determinants of CDS, and e.g. Nowak et al. (2011) have found that macroeconomic fundamentals had not been significantly explaining CDS spreads, it is hypothesized that due to sudden rise of sensitivity to fundamentals observed in the bond market, investors will pay more attention to macroeconomic news than before the sovereign debt crisis. We will thus test for a structural break in the relationship which is assumed to emerge during the crisis which had the investors realized they had been vastly mispricing sovereign default risk. Their reaction is assumed to be conditional on the stage of business cycle as documented by Beber and Brandt (2010) and could be asymmetric with respect to positive and negative surprises. In addition to repricing effect, we will examine impact on spread volatility and impact on outstanding CDS volume. For important (with respect to the magnitude of surprise and importance of country of origin) macroeconomic surprises it is also conceivable that there are spillover effects taking place, thus effect in non-event countries will be examined.

Hypotheses:
1. Macroeconomic surprises are associated with repricing effect of sovereign risk that takes place in a short window around the announcement.
2. On announcement days there is higher anticipated volatility and conditional volatility (on surprise effect).
3. There is an asymmetry in reaction to good and bad news (documented e.g. by Nowak et al. (2011)), whose presence is conditional on the level global market stress indicators.
4. CEE sovereign risk premia react to regional/global macroeconomic surprises (German/Eurozone, US).
5. There has been a structural break in the relationship during the sovereign debt crisis – which made the relationship stronger, as investors received the “wake-up call”.

Methodology:

This thesis will employ event study approach as well as regression techniques (probably GARCH model for panel data) on data from Central and Eastern Europe countries. The dependent variables will be two common measures for sovereign risk using daily frequency: change in CDS premium (alternatively - adjusted for the aggregate movement in CDS spreads) and change in government benchmark bond spread. Macroeconomic surprises are defined as deviations from analysts forecast scaled by standard deviation of all surprises for a given series.
Firstly, similar to Ismailescu & Kazemi (2010) and Havlíček (2013), the relationship shall be examined using the standard event study approach, i.e. select certain event window and measure the abnormal performance of CDS, i.e. whether in most cases the adjusted spread change reacts in the anticipated way, using chi-square test for equality of proportions or various t-tests; possibly splitting the sample in quantiles.
Secondly, similar to Büttner and Hayo (2012) and Andritzky et al. (2007), the relationship will be examined using a regression framework taking into account data properties. It is expected that CDS spread changes will show volatility clustering, hence a GARCH-type model will be used. GARCH method enables the examination of surprises’ effect on volatility through the conditional variance equation. The control variables will include broad factors that account for global and/or regional risk premia. These include risk appetite proxy, risk free rate as well as movements of other key financial markets (money market, stock market, foreign exchange) in a given country.
GARCH model could be specified as a panel data model with a simplifying assumption on asset correlation, which could produce unbiased results as Büttner and Hayo (2010), who studied time varying correlations within CEE-3 using a DCC-MGARCH, found the coefficients for time varying correlation matrix not to be significantly different from zero for government bond market.Finally, the possibility of a structural break in the relationship will be analyzed either by splitting the samples, including a dummy variable into the regression or by a model with endogenous unknown break.


Literature:
1. Büttner, D., & Hayo, B. (2012). EMU-related news and financial markets in the Czech Republic, Hungary and Poland. Applied Economics, 44(31), 4037-4053.
2. Andritzky, J. R., Bannister, G. J., & Tamirisa, N. T. (2007). The impact of macroeconomic announcements on emerging market bonds. Emerging Markets Review, 8(1), 20-37.
3. Özatay, F., Özmen, E., & Şahinbeyoğlu, G. (2009). Emerging market sovereign spreads, global financial conditions and US macroeconomic news. Economic Modelling, 26(2), 526-531.
4. Fender, I., Hayo, B., & Neuenkirch, M. (2012). Daily pricing of emerging market sovereign CDS before and during the global financial crisis. Journal of Banking & Finance, 36(10), 2786-2794.
5. Nowak, S., Andritzky, J., Jobst, A., & Tamirisa, N. (2011). Macroeconomic fundamentals, price discovery, and volatility dynamics in emerging bond markets. Journal of Banking & Finance, 35(10), 2584-2597.
6. Ismailescu, I., & Kazemi, H. (2010). The reaction of emerging market credit default swap spreads to sovereign credit rating changes. Journal of Banking & Finance, 34(12), 2861-2873.
7. Havlíček, T. (2013). Credit rating agencies and their impact on the bond markets of EU countries. Master thesis at IES FSV UK. Supervisor: doc. Roman Horváth, Ph.D.
8. Büttner, D., & Hayo, B. (2010). News and correlations of CEEC-3 financial markets. Economic Modelling, 27(5), 915-922.
9. Beber, A., & Brandt, M. W. (2010). When It Cannot Get Better or Worse: The Asymmetric Impact of Good and Bad News on Bond Returns in Expansions and Recessions. Review of Finance, 14(1), 119-155.