Effects of securitization and covered bonds on bank stability

https://doi.org/10.1016/j.ribaf.2020.101196Get rights and content

Abstract

This paper empirically investigates the relationship of securitization and covered bonds with bank stability and highlights that this relationship varies with the level of a bank's involvement in a specific instrument. The study uses the data from 46 securitizing and covered bond issuing listed banks in Europe for 2000–2014. The initial results show that some banks have been heavily involved in securitization activity, while covered bond issuance does not go beyond a certain limit. The results obtained using a quadratic model and a generalized additive model show a U-shaped relationship between securitization and systemic risk of banks. However, some interesting results are obtained for covered bonds. Initial results do not show a significant impact of covered bonds on systemic risk, but further analysis shows the presence of a size effect. The systemic risk of smaller banks increases after the issuance of covered bonds, while larger banks remain unaffected. The study does not support imposing uniform limits on covered bond issuance; rather such limits should be linked to the bank size. However, some regulatory framework is needed to limit banks’ involvement in securitization.

Introduction

Asset-Backed Securities (ABS) — generated through the process of securitization — and Covered Bond (CB) remained widely used funding sources by European Banks. These two instruments are close counterparts but structurally different from each other. Traditionally, ABS issuance was considered having a benign impact on bank risk (Santomero and Trester, 1998, Cantor and Rouyer, 2000), but this view was tarnished after the Global Financial Crisis (GFC). Banks, reliant on ABS for funding, faced liquidity pressure to a point that transformed into a systemic crisis (Blommestein et al., 2011). However, CBs remained relatively resilient during the crisis. The investors’ interest remained intact in this class of assets because of the dual course. Many new banks started issuing CBs to meet their funding needs.

Post-crisis regulations provide favorable treatment to CBs, but a stringent one to securitization,1 because of CBs’ resilience and the collapse of securitization. Many market participants are of the view that such treatment may result in over-reliance on CBs (AFME, 2014, EBA, 2014b, GFMA, 2013), having severe implications for the banks’ stability. Moreover, such a treatment is also undermining the efforts to revive the securitization market. As banks must actively manage the underlying pool of collateral in the case of CBs, therefore, they might not prove to be risk-free. An over-reliance on them may result in risk concentration in the banking system (Anand et al., 2012). The empirical analysis performed in this study is likely to provide a foundation to evaluate such regulations.

The literature studying the impact of securitization on bank risk can be divided into “securitization-stability” and “securitization-fragility” perspectives. Authors supporting the former perspective, argue that securitization helps to shed the risk out of the banking system, thereby improving the diversification (Santomero and Trester, 1998, Cantor and Rouyer, 2000, Jiangli and Pritsker, 2008). However, the proponents of latter perspective are of the view that securitization increases the risk appetite of banks that results in the generation of risky assets, thereby putting the bank stability at stake (Keys et al., 2010, Keys et al., 2012, Kara et al., 2019, Wu and Shen, 2019). While securitization has gained significant attention in the literature, the impact of CBs on banks’ risk remains understudied.

This study makes a two-fold contribution to the literature. First, the study includes CBs in the analysis along with ABS [hereinafter collectively referred to as Securities and Bonds (SB)]. Both instruments share many similarities, making them a good candidate for comparison. Second, the study provides an alternative perspective that I call “securitization-scalability”. I argue that instead of SB issuance per se, their respective volumes determine their implications on bank stability. The intended benefits of these funding sources might not be accessible either beyond or below a certain level. For these reasons, it is often argued that limits should be imposed on the issuance of these instruments, already implemented in some countries like Australia and Belgium in the case of CBs. To the best of my knowledge, this perspective is not presented in any other study.

To evaluate the link between SB issuance and systemic risk of banks, this study performs an extended analysis with data taken from 46 ABS and CB issuing listed banks from Europe between 2000–2014. The analysis focuses on variation in the relationship of SB with bank stability with respect to changes in the scale of their issuance. The study also analyzes the impact of size on the relationship between SB issuance and the systemic risk of banks. The analysis starts with a quadratic model and extends to a partially linear setting that uses the Generalised Additive Model (GAM).

Results show the presence of a U-Shaped relationship between ABS and systemic risk. The issuance of ABS initially helps banks control their systemic risk, but this relationship is reversed when banks continue issuing ABS. This is dubbed by some as ‘securitization beyond limits’, and by this study as “securitization-scalability”. For covered bonds, initial estimations show that CBs do not have a relationship with systemic risk. However, some further analysis, while taking into account the size of banks, provides some interesting results. The small issuance of CBs leads to an increase in systemic risk, but large scale issuance of CBs decreases the systemic risk in the case of small banks. Whereas, bigger banks remain unaffected by the issuance of CBs. I attribute this relationship to bank size and jumbo CBs. These findings do not support the proposal of putting a uniform limit on CB issuance and argue for a framework that could control the limitless issuance of ABS.

The remainder of this paper is organized as follows. Section 2 provides a review of relevant literature. Section 3 discusses the theoretical underpinning of the model being tested here. Section 4 provides the empirical model to test the suggested relationship and explains the characteristics of the data including the sample details and descriptive statistics. Section 5 provides the empirical results of the model tested herein. This section provides results of the quadratic model and generalized additive specification of a partially linear model. Section 6 discusses the results and concludes.

Section snippets

Securitization and bank stability

Researchers have argued that securitizing banks follow an aggressive business strategy (Chen et al., 2017, Bakoush et al., 2019). They start issuing loans to subprime borrowers in the absence of prime borrowers, resulting in lax screening and lower lending standards (Keys et al., 2010, Keys et al., 2012, Kara et al., 2019). This situation threatens the stability of the banking system by creating imbalances in the credit markets, thereby increasing the fragility of the financial system (Altunbas

Theoretical underpinnings

Banks use SB to mitigate their funding and liquidity risks. They might help banks meet these objectives, resulting in a reduction in the systemic risk at an initial stage. The problem originates when banks increase their reliance on one instrument. In the case of ABS, it may result in three main effects. First, the bank's reliance on traditional liquid assets is decreased and securitization becomes a major source of liquidity generation. This creates concentration risk. Second, a liquidity glut

Systemic risk measures

A good risk measure should account for the contribution of an FI to the aggregate risk, along with the stand-alone risk. Banks have been widely using Value at Risk (VaR) to measure their risk before the GFC. However, this measure does not account for the tail risk and the contribution of a financial institution to systemic risk. Many of the previous studies use the firm's stock Beta, by following Capital Asset Pricing Model (CAPM), as a measure of systemic risk (Battaglia et al., 2014, Nijskens

Results and discussion

This section presents the empirical results of various tests including the findings of the robustness checks.

Conclusion

The empirical analysis performed in this study provides important insights about the securitization and CB markets. This study examines the impact of these instruments on systemic risk and bank stability. Contrary to other studies, this study investigates the possibility of a non-linear relationship of these instruments with the banks’ risk and stability. The empirical analysis suggests the presence of a U-Shaped relationship between systemic risk and securitization. The implications of the ABS

Author statement

The paper was written as part of European Doctorate in Law and Economics, a joint doctorate programme of University of Bologna, University of Hamburg and Erasmus University of Rotterdam. The funding was provided by Education, Audiovisual and Culture Executive Agency of European Union. The paper is not written jointly with any one else and not submitted for publication in any other journal.

Acknowledgement

Author is highly indebted to express his deep gratitude to Sergio Pastorello (University of Bologna), Alessio M. Paccess (University of Amsterdam), and Jonathan Klick (Penn Law School) for their useful guidance and comments.

References (45)

  • L. Laeven et al.

    Bank size, capital, and systemic risk: some international evidence

    J. Bank. Finance

    (2016)
  • R. Nijskens et al.

    Credit risk transfer activities and systemic risk: how banks became less risky individually but posed greater risks to the financial system at the same time

    J. Bank. Finance

    (2011)
  • M. Prokopczuk et al.

    Credit risk in covered bonds

    J. Empir. Finance

    (2013)
  • A.M. Santomero et al.

    Financial innovation and bank risk taking

    J. Econ. Behav. Organ.

    (1998)
  • R. Trapp et al.

    Derivatives usage, securitization, and the crash sensitivity of bank stocks

    J. Bank. Finance

    (2016)
  • M.W. Wu et al.

    Effects of shadow banking on bank risks from the view of capital adequacy

    Int. Rev. Econ. Finance

    (2019)
  • V.V. Acharya et al.

    Capital shortfall: a new approach to ranking and regulating systemic risks

    Am. Econ. Rev.

    (2012)
  • V.V. Acharya et al.

    Measuring systemic risk

    AFA 2011 Denver Meetings, Stern School of Business

    (2010)
  • T. Adrian et al.

    The Shadow Banking System: Implications for Financial Regulation

    (2009)
  • AFME

    High-Quality Securitisation for Europe: The Market at a Crossroads

    (2014)
  • K. Anand et al.

    Covered Bonds, Core Markets, and Financial Stability

    (2012)
  • D. Angelos

    Covered bonds – ripe for expansion?

    Briefing to European Parliament, European Parliamentary Research Service

    (2015)
  • Cited by (0)

    The paper was written as part of European Doctorate in Law and Economics, a joint doctorate programme of University of Bologna, University of Hamburg and Erasmus University of Rotterdam.

    View full text