L’objectif de cette thèse doctorale est de contribuer à l’ensemble de la recherche sur les agences de notation. Elle couvre les trois aspects : théorique, conceptuel et empirique. La première partie se focalise sur le Néo-Institutionnalisme. Le Capitalisme d’État propre au marché chinois a été mis en avant, et un bilan comparatif a été dressé entre les marchés chinois et européens. Une analyse dynamique nous permet ensuite d’illustrer l’interaction des agences de notation avec les autorités régulatrices, les émetteurs de titres financiers, et les investisseurs. Dans la seconde partie, nous avons réalisé une vaste revue de la littérature, et proposé d’ajouter les spécificités institutionnelles chinoises aux variables déjà prises en compte dans les études. Au niveau de la méthodologie, certains outils qualitatifs et quantitatifs ont été envisagés. Diverses approches sont choisies en fonction du contenu de chacun des chapitres : le rôle, la stratégie et l’impact des agences de notation en Chine.Premièrement, par une approche comparative, nous avons montré qu'en Chine, le rôle des agences de notation ne se limitait pas à celui qui leur est habituellement assigné. Il semble aussi que les agences de notations se laissent tenter par le « rating shopping », le « split rating » et le « rating inflation », sous la pression des émetteurs. En revanche, les investisseurs sont relativement prudents lors de « multiples ratings ». Deuxièmement, nous avons mis en évidence que les agences de notation internationales ont tendance à noter plus strictement que les agences locales. Nous nous sommes rendu compte que les comportements stratégiques variaient d’une à l’autre, lorsque l’on étudie les déterminants des notations initiales des émetteurs. Troisièmement, en menant une étude d’événements sur le marché chinois des actions, nous avons trouvé des rendements anormaux significatifs sur quelques sous-échantillons des « modifications de notation », après avoir adopté différentes méthodes d’estimation, et mené divers tests paramétriques et non-paramétriques. Enfin, les analyses et résultats que nous avons apportés au cours de chaque chapitre empirique, ont enrichi en retour, nos discussions sur l’environnement institutionnel du marché chinois / In this Ph.D. thesis, we aim to contribute to the entire research on CRAs, with theoretical, conceptual and empirical aspects. We used the New-institutionalism as a theoretical frame of reference to justify the importance to have a good understanding of Chinese institutional characteristics. Chinese State-capitalism and Socialist-market economics, along with the interaction of CRAs with regulators, issuers and investors are devoted to setting the stage. We implemented various methodological approaches that seemed the most appropriate, on the sample that the most suitable, according to each of the objectives that we set, to better comprehend CRAs’ role, strategy and impact, under Chinese market context.Firstly, we found that CRAs were not expected to play “their main role” in assessing the credit default risk for market investment in China, as they are usually assigned to do so in Europe, and other developed countries. After running the frequency test, to register the words most used in the written regulatory and normative documents, we concluded that the expectation of roles of CRAs was also different in Continental China and in Hong Kong. At the expectation of issuers, CRAs also seem to play a role in rating “shopping”, in split rating, and even in rating inflation. Investors are very cautious to the multiple ratings. Secondly, by using logistic models, we found that global CRAs tend to rate more strictly than local CRAs, ceteris paribus. We also realized that the strategical behaviors of Chinese local CRAs are distinctive one from the other, when we studied the determinants “behind” initial issuer ratings. These findings confirmed the preliminary quantitative tests that we executed on the descriptive data. Thirdly, we failed to find informative impact of credit rating changes on the stock market, through a classic event study relying on the market model. Even if we refined the methodology by completing GARCH to OLS estimation model, and non-parametric tests to parametric tests, we only found significant outcomes in some subsamples, and for negative watch-lists.In the end, the analyses we led and the findings we reported from each of these empirical chapters have contributed to enrich, in return, our discussion of Chinese institutional characteristics

This paper contributes to the literature by presenting a detailed summary of existing problems with credit rating agencies and proposing a model of a European Rating Agency that acts as a competitor and benchmark to the established credit rating agencies. European Rating Agency (ERA) as a new entrant will make a difference by increasing rating transparency and reputation cost for rating agencies at the same time. A vital point when developing the idea of an independent rating agency was the minimization of regulation changes. This will help shareholder to better orientate and understand the functionalities of the European Rating Agency which will make its establishment process easier. Reputation cost (capital) are assumed to be the building blocks that support the unique position of current rating agencies. ERA is modelled specifically to target this information and use the fragile concept of trust and reputation to the benefit of all stakeholder.

No description available.

Il rating creditizio è un elemento molto importante per le imprese in quanto ha impatto, ad esempio, sul costo del capitale (Bhojraj and Sengupta, 2003; Campbell and Taksler, 2003), sul prezzo delle azioni e dei titoli obbligazionari (Dichev and Piotroski, 2001; Hand et al., 1992). Anche le imprese che ottengono il rating per la prima volta, tipicamente soggetti di dimensioni più piccole e con una storia più recente, hanno molto interesse per il rating. Nel principale contributo della mia tesi (“Is Indebtedness always negative for Credit Ratings? Empirical evidence on Newly Rated Firms”), misuro empiricamente, per il periodo dal 1985 al 2013, se i soggetti che hanno ottenuto un rating per la prima volta ottengono una valutazione differente rispetto agli altri operatori. Dalle mie analisi emerge che i soggetti che hanno ottenuto un rating per la prima volta ottengono valutazioni lievemente più negative rispetto agli altri operatori ma, sorprendentemente, emerge che coloro che hanno un maggiore indebitamento hanno delle valutazioni migliori. Negli altri due articoli della mia tesi sviluppo ulteriori analisi sui rating. Nel primo articolo (“Credit Rating Agencies: a Review of Recent Academic Studies and Key Practical Implications”), presento una sistematizzazione della letteratura accademica sui rating e sulle agenzie di rating. Nel secondo articolo (“The Dynamics of Credit Rating Standards”), esamino se le agenzie di rating hanno modificato i propri standards nel corso del tempo. / Firms care deeply about their credit ratings, since ratings influence, for instance, firm’s cost of capital (Bhojraj and Sengupta, 2003; Campbell and Taksler, 2003), bond and stock market prices (Dichev and Piotroski, 2001; Hand et al., 1992). This is also true for newly rated firms, typically smaller, in a younger stage of their life cycle and with a shorter track record compared with other issuers to show to the external financial stakeholders. In the main paper of my thesis (“Is Indebtedness always negative for Credit Ratings? Empirical evidence on Newly Rated Firms”), I test the impact of being newly rated firms on credit ratings over the period from 1985 to 2013. I report a negative but pretty low effect on rating outcome for the entire sample of newly rated firms but, surprisingly, I find a strong positive relation between highly levered firms and credit rating. I develop additional research on credit rating in the other two papers of my thesis. In the first paper (“Credit Rating Agencies: a Review of Recent Academic Studies and Key Practical Implications”), I present a systematization of the latest academic contributions on credit ratings and credit rating agencies. In the second paper (“The Dynamics of Credit Rating Standards”), I examine long-term issuer credit ratings and I focus on the time variable to study how credit rating agencies have modified their standards over years.

Credit risk represents one of the most significant risks which a bank must face, and therefore, its intention is effectively manage and measure this risk. However, management and measurement methods are supervised and influenced by national regulators. Banking regulatory supervision plays a significant role among others in determining minimum capital requirements that serve as buffer against losses stemming from credit risk. This thesis provides theoretical foundation of regulatory approaches - standardized and internal rating based (IRB) approach - used for quantification of regulatory capital to credit risk as well as empirical application of such approaches on created portfolio of corporate loans. As a part of IRB method I suggested a model composed of financial ratios estimating probability of default using logistic regression. I founded out that rather the use of combination of financial ratios from different groups of ratios with slight dominance of profitability ratios forms final model. Therefore, superiority of solvency ratios in modelling cannot be proved on my portfolio. After estimating and determining necessary parameters I quantified the minimum regulatory capital requirements to credit risk under standardized and IRB approaches prescribed by Basel III. In the end, the results are...

Issuers’ credit ratings change by one or more notches when credit rating agencies provide new ratings. Unique to the literature, I study the influences affecting multi notch versus single notch rating upgrades and downgrades. For Standard & Poors data, I show that rating changes with multiple notches provide more information to the market than single notch rating changes. Consistent with prior literature on the business cycle, I show that investors value good news rating changes (upgrades) more in bad times (recession) and that investors value bad news rating changes (downgrades) more in good times (expansion). I model and test probit models using variables capturing the characteristics of the previous issuer’s credit rating, liquidity, solvency, profitability, and growth opportunity to determine the classification of single notch versus multi notch rating changes. The determinants of multi notch versus single notch rating changes for upgrades and downgrades differ. Business cycle influences are evident. Firms that have multi notch rating upgrades and downgrades have significantly different probit variables vis-à-vis firms that have single notch rating upgrades and downgrades. The important characteristics for determining multiple notch upgrades are a firm’s prior rating change, prior rating, cash flow, total assets and market value. The important characteristics for determining multiple notch downgrades are a firm’s prior rating change, prior rating, current ratio, interest coverage, total debt, operating margin, market to book ratio, capital expenditure, total assets, market value, and market beta. The variables that differ for multi notch upgrades in recessions are cash flow, net income, operating margin, market to book ratio, total assets, and retained earnings. The variables that differ for multi notch downgrades in expansions are a firm’s prior rating change, current ratio, interest coverage ratio, debt ratio, total debt, capital expenditure and market beta. The power of the explanatory tests improves when the stage of the business cycle is considered. Results are robust to consideration of rating changes across rating categories, changes from probit to logit, alternative specifications of accounting variables, lags and leads of recessions and expansions timing, Fama and French industry adjustments, and winsorization levels of variables.

Die Arbeit ANLEIHERATING UND BONITÄTSRISIKO untersucht die Zusammenhänge zwischen dem Rating von Anleihen, dem Risiko dieser Anleihen sowie ihren Risikoprämien (Spreads). Dazu wird in einem ersten Schritt - auf analytischer Ebene - untersucht, wie Ratings vergeben werden und welchen Einschränkungen sie unterliegen. Die wichtigsten Einschränkungen für den Kapitalmarkt hinsichtlich der Ableitung von quantitativen Risikogrößen (Ausfallwahrscheinlichkeit und ggf. Ausfallschwere) sind: Ordinalität und Relativität der verwendeten Skalen; die Zeitverzögerungen bei Ratingänderungen; die Intransparenz, Subjektivität und Urteilsunabhängigkeit des Ratingverfahrens bzw. der Ratingagenturen; die Erstellung von auftragslosen Ratings; die mangelnde Vergleichbarkeit von Ratings (zeitlich, zwischen Emissionen und zwischen Agenturen); die fehlende Äquidistanz von Ratings; das Risiko fehlerhafter Ratings. In einem zweiten Schritt wird empirisch untersucht, in welcher Höhe vom Markt Spreads (Überrenditen gegenüber den als risikolos geltenden Bundesanleihen) - je nach Rating - für bestimmte Anleihen gefordert werden. Datenbasis sind Kupon-Anleihen (ohne Sonderrechte) des DM-Euromarktes mit täglichen Kursen im Zeitraum Januar 1990 bis Dezember 1995. Die Untersuchungen bestätigen die Relativität: Ratings definieren nur im langfristigen Durchschnitt die Renditeabstände zwischen den verschiedenen Klassen. Kurzfristige Veränderungen der Spreads hingegen hängen von Veränderungen des allgemeinen Zinsniveaus und dem Verlauf der Zinsstrukturkurve ab. Diese beiden Größen weisen dabei einen negativen Einfluß auf, d.h. Zinserhöhungen führen zu Verringerungen der Spreads. Grundsätzlich gilt dabei aber, dass diese Effekte umso stärker ausfallen, je geringer die Bonität der Anleihen ist. Zusätzlich hängen die Spreadänderungen auch (positiv) von Veränderungen der Spreads der jeweils anderen Klassen ab. Als mögliche Erklärungsansätze bieten sich für diese Beobachtung auch an, dass der Markt Schwankungen der erwarteten Ausfallwahrscheinlichkeiten antizipiert und/oder Veränderungen der allgemeinen Risikoeinstellung vorliegen.

Ratingsysteme sind ein zentraler Bestandteil der Kreditrisikomodellierung. Neben der Bonitätsbeurteilung auf der Ebene der Kreditnehmer und der Risikoquantifizierung auf der Ebene der Ratingklassen spielt dabei die Bildung der Ratingklassen eine wesentliche Rolle. Die Literatur zur Ratingklassenbildung setzt auf modellfreie, in gewisser Weise willkürliche Optimierungsverfahren. Ein Ziel der vorliegenden Arbeit ist es, stattdessen ein parametrisches statistisches Modell zur Bildung der Ratingklassen einzuführen. Ein geeignetes Modell ist im Bereich der Bruchpunktschätzung zu finden. Dieses Modell und die in der mathematischen Literatur vorgeschlagenen Parameter- und Intervallschätzer werden in der vorliegenden Arbeit dargestellt und gründlich diskutiert. Dabei wird Wert auf eine anwendungsnahe und anschauliche Formulierung der mathematisch-statistischen Sachverhalte gelegt. Anschließend wird die Methodik der Bruchpunktschätzung auf einen konkreten Datensatz angewendet und mit verschiedenen anderen Kriterien zur Ratingklassenbildung verglichen. Hier erweist sich die Bruchpunktschätzung als vorteilhaft. Aufbauend auf der empirischen Untersuchung wird abschließend weiterer Forschungsbedarf abgeleitet. Dazu werden insbesondere Konzepte für den Mehrklassenfall und für abhängige Daten entworfen. / Rating systems are a key component of credit risk modeling. In addition to scoring at borrowers’ level and risk quantification at the level of rating classes, the formation of the rating classes plays a fundamental role. The literature on rating classification uses in a way arbitrary optimization methods. Therefore, one aim of this contribution is to introduce a parametric statistical model to form the rating classes. A suitable model can be found in the area of split-point estimation. This model and the proposed parameter and interval estimators are presented and thoroughly discussed. Here, emphasis is placed on an application-oriented and intuitive formulation of the mathematical and statistical issues. Subsequently, the methodology of split-point estimation is applied to a specific data set and compared with several other criteria for rating classification. Here, split-point estimation proves to be advantageous. Finally, further research questions are derived on the basis of the empirical study. In particular, concepts for the case of more than two classes and for dependent data are sketched.

I syfte att undersöka psykometriska egenskaper hos Affektiv självskattningsskala (AS-18) fyllde 88 patienter med diagnos bipolär typ I (N=46) eller typ II (N=42) i självskattningsskalorna AS-18 och MADRS-S vid två tillfällen med en dags mellanrum. Principalkomponentsanalys för AS-18 genomfördes med extrahering av två komponenter. Items laddade i de delskalor de tillhörde. Intern konsistens mättes med Cronbachs alfa och överensstämmelse med Cohens kappa. Test-retest-reliabiliteten beräknades. Grupperna bipolär typ I och typ II skiljde sig inte åt, vid rutinuppföljningsbesök, vad gäller skattning av mani eller depression. Studien gällande AS-18 visade att faktorstrukturen från tidigare studier kunde bekräftas, att test-retest-reliabiliteten var hög och att AS-18 är användbar även för patienter med bipolär typ II-diagnos. Skalan hade sammanfattningsvis goda psykometriska egenskaper. / <p>Psykoterapeutexamensarbete (PTU)</p>

Universiẗat, Diss.--Osnabrück, 2007.