DUDEK Jeremy

Systemic risk measures generally aim to identify systemically important financial institutions (SIFIs) that would allow regulators to allocate macro-prudential capital requirements in order to reduce risk stemming from such institutions. Among widely-cited are the measures of tail dependence in financial institutions’ equity returns, such as ΔCoVaR of Adrian and Brunnermeier (2011) and Marginal Expected Shortfall (MES) of Acharya et al. (2010). This paper compares nonlinear and linear approaches to modeling return dependence in the estimation of the ΔCoVaR and MES. Our results show that while the refined and complicated estimation techniques are able to produce more accurate value of institution’s systemic risk contribution they do not greatly improve in terms of identifying SIFIs compared to simpler linear estimation method. Modeling dependence linearly sufficient to identify and rank SIFIs.

ETFs and index funds have grown at very rapid rates in recent years. Originally launched totrack some large liquid indices in developed markets, they now also concern less liquid assetclasses such as emerging market bonds. Illiquidity certainly affects the quality of the replication,and in particular, liquidity might increase the tracking error of any index fund, i.e., thedifference between the fund and the benchmark returns. The tracking error is then the firstcharacteristic that investors consider when they select index funds. In this paper, we beginfrom the CDS-bond basis to simulate the tracking error (TE) of a hypothetical well-diversifiedfund investing in the emerging market bond universe. We compute the CDS-bond basis andthe tracking error for 9 emerging market sovereign entities: Brazil, Chile, Hungary, Mexico,Poland, Russia, South Africa, Thailand and Turkey. All of these countries are included inthe MSCI Emerging Market Debt in Local Currency index. Our sample period ranges fromJanuary 1, 2007 to March 26, 2012. Using a Regime Switching for Dynamic Correlations(RSDC) model, we show that the country-by-country tracking error is reduced by the diversificationat the fund level. Moreover, we show that this diversification effect is less effectiveduring crisis periods. This loss of diversification benefits is the main risk of index funds when they are designed to create a liquid exposure to illiquid asset classes.

Fund managers face liquidity problems but they have to distinguish the market liquidity risk implied by their assets and the funding liquidity risk. This latter is due to both the liquidity mismatch between assets and liabilities and the redemption risk due to the possible outflows from clients. The main contribution of this paper is the analysis of contagion looking at common market liquidity problems to detect funding liquidity problems. Using the CDS Bond Spread basis as a liquidity indicator and a state space model with time-varying volatility specification, we show that during the 2007-2008 financial crisis, there exist pure contagion effects both in terms of price and liquidity on the emerging sovereign debt market. This result has strong implication since the main risk for an asset manager is to get stuck with an unwanted position due to a dry-up of market liquidity.

No summary available.

This thesis is articulated around three financial risks: liquidity, contagion and systemic risk. The latter have been the focus of attention since the 2007-08 crisis and will continue to be relevant in light of the events that the financial markets are experiencing. The first chapter of this thesis presents a funding liquidity factor obtained by interpreting a contagion phenomenon in terms of market liquidity risk. In the second chapter, we propose a meta-measurement of this market liquidity. The latter takes into account all the dimensions present in the definition of liquidity by looking at the dynamics of several liquidity measures simultaneously. The objective of the third chapter is to present a model of market returns that allows for the inclusion of funding liquidity in the estimation of the DCoVaR. Thus, this work proposes a new measure of systemic risk with countercyclical behavior. Finally, we focus on the non-linearity assumption of the dependence structure between market returns and financial institutions' returns. At the heart of systemic risk measurement, this assumption appears to be restrictive since it has little impact on the identification of the riskiest firms but can considerably complicate the estimation of these measures.

The last crisis sheds light on the importance to consider liquidity risk in the financial industry. Indeed, liquidity had a predominant role in propagating the turmoil. In contrast, controlling for liquidity is a difficult task. The definition of liquidity links different dimensions that are impossible to fully capture together. As a consequence, there exist a lot of liquidity measures and we find in the literature some solutions to take into account more than one dimension of liquidity but also liquidity measures considering a long lasting liquidity problem. In this paper, we focus on drastic illiquidity events, i.e liquidity problems reported by several liquidity measures simultaneously. We propose a Meta-Measure of liquidity called MLiq and defined as the probability to be in a state of high liquidity risk. We use a multivariate model allowing to measure correlations between liquidity measures jointly with a state-space model that endogenously defines the illiquid periods.

Financial markets are today so interconnected that they are fragile to contagion. Massive investment funds with very short horizons in -and out- flows can generate contagion effects between markets. Since 2010, investors are willing to get a liquid exposure to the EMsovereign debt. As a consequence, some asset management firms started to propose products to track the performance of this asset class. However in that case, the fund manager faces a mismatch of liquidity between assets and liabilities and needs some tools to manage the liquidity of his investments. The main contribution of this paper is the analysis of contagion looking at common market liquidity problems to detect funding liquidity problems. Using the CDS Bond Spread basis as a liquidity indicator and a state space model with time-varying volatility specification, we show that during the 2007-2008 financial crisis, there exist pure contagion effects both in terms of price and liquidity on the emergings overeign debt market.This result has strong implication since the main risk for an asset manager is to get stuck with an unwanted position due to a dry-up of market liquidity.

Emerging economies have passed an important stress test during the period 2008-09 and are now the key drivers for global growth of the world economy. Financial markets are today so interconnected that they are fragile to contagion. The issue of financial contagion was historically concerning Emerging Markets (EM). These latter attract foreign investors and massive investments funds in -and out- flows on very short horizons can be a source of contagion effects between markets. The analysis of the sovereign debt markets and particularly related CDS markets is of interest since it is at the very center of a new phenomenon: banks are not anymore the main source of systemic risk but sovereign economies are. As foreign investors represent the most of the volume traded, capital flows in these markets should also impact FX market. Their analysis is thus also central to this study. Indeed, the main risk for an asset manager is to get stuck with unwanted sovereign debt due to a dry up of market liquidity. The main contribution of this paper is the analysis of contagion looking at common markets liquidity problems to detect funding liquidity problems. We use the Credit Default Swap bond spread basis and the deviations from the Covered Interest Parity as liquidity measures respectively for sovereign debt and FX markets. Moreover, we distinguish interdependence and pure contagion using a state-space model with a time-varying volatility specification and we apply it to both returns and liquidity indicators.

The aim of this thesis is to improve the management of financial risks through the employment of econometric methods. We focus on liquidity (market and funding), contagion and systemic risk, which have attracted a particularly large interest in the last years of financial turmoil. Firstly, we construct a funding liquidity factor based on the contagion effects that market liquidity risks encounter. This procedure can be useful to provide a better management of the liquidity mismatch among the assets and liabilities of a fund. Secondly, we propose a meta-measure of liquidity which incorporates multiple liquidity measures through the use of a conditional correlation model. As a result, we are able to detect drastic liquidity problems by using a single measure. Thirdly, we propose a new modeling framework for financial returns by adding an extra component related to funding liquidity to the standard DCoVaR model. In this way we obtain a countercyclical measure of systemic risk. Finally, we study to which extent a change in the estimation method affects the identification of systemically relevant Financial Institutions. In particular, the most popular measures aim at capturing the nonlinearity of the dependence structure between financial firms and market returns. We show, however, that similar results can be obtained by simply assuming a linear dependence, which can also largely simplify the estimation.

Systemic risk measures generally aim to identify systemically important financial institutions (SIFIs) that would allow regulators to allocate macro-prudential capital requirements in order to reduce risk stemming from such institutions. Among widely-cited are the measures of tail dependence in financial institutions’ equity returns, such as ΔCoVaR of Adrian and Brunnermeier (2011) and Marginal Expected Shortfall (MES) of Acharya et al. (2010). This paper compares nonlinear and linear approaches to modeling return dependence in the estimation of the ΔCoVaR and MES. Our results show that while the refined and complicated estimation techniques are able to produce more accurate value of institution’s systemic risk contribution they do not greatly improve in terms of identifying SIFIs compared to simpler linear estimation method. Modeling dependence linearly sufficient to identify and rank SIFIs.