Elsevier

Finance Research Letters

Volume 27, December 2018, Pages 113-117
Finance Research Letters

Systematic risk and banks leverage: The role of asset quality

https://doi.org/10.1016/j.frl.2018.02.015Get rights and content

Highlights

  • Panel of the largest 97 commercial banks listed in 11 EU countries for the period 2005–2016.

  • The standard version of leverage ratio is informationally limited for bank's investors.

  • A three-step bank leverage adjustment is proposed for sterilizing the effect of provisioning and incorporating the effect of credit risk on equity beta coefficient.

  • Banks’ asset quality affects the relationship between leverage and systematic risk.

Abstract

We analyse how bank asset quality interacts within the relationship between leverage and systematic risk. We elaborate three leverage adjustments for sterilizing the effect of provisioning and incorporating the effect of non-performing loans and total credit risk exposure. We test the model on a sample of 97 European banks from 2005 and 2016. Controlling for size, findings show the relevance of a combined effect of leverage and asset quality as a systematic risk component. NPLs are found to be one significant variable of market risk. Results demonstrate that simple leverage is pointless for verifying the financial riskiness of banks.

Introduction

The global financial crisis left a huge stock of bad loans in European banks balance sheets (EBA, 2016). As widely acknowledged, non-performing loans (hereafter NPLs) shrink profitability forcing banks to extensive write-offs and provisions, reduce managerial flexibility limiting lending supply and require incremental capital buffers either to face additional asset risk and cover loans losses.

The increase of capital requirements imposed by regulatory authorities under the Basel framework (or additionally required for specific situations) contribute to support banks in NPLs’ losses absorption and, more generally, foster the resilience of the banking system to potential negative shocks. Despite in the view of bankers raising new equity capital is costly, empirical evidence suggests an inverse relation between equity capital and systematic risk across all the main banking markets (Baker and Wurgler, 2015, Kashyap et al., 2010, Miles et al., 2013, Toader, 2015). With specific regards to EU banks, Haq and Heaney (2012) document a non-linear relation: an increase in equity of less capitalized banks is associated with lower systematic risk while, for highly capitalized banks, further increase of equity determine higher systematic risk. Accordingly with Calem and Rob (1999), this evidence can be explained with the strict connection and sensitivity between equity and banks’ asset risk for which the more the risk a banks is exposed to, the greater the equity capital needs. Thus the “disciplinary effect” driven by a capital increase tends to be larger for banks holding lower equity in relation to risks, for the effect of aligning the capitalization close or above to the market average (or comply to the minimum requirement). On the opposite, for adequately-capitalized banks, a capital increase is perceived by the market as a signal of strategy change toward more risky investments so that investors might undertake the risk of being exposed to higher asset risk (Haq and Heaney, 2012).

Hence it is clear that the risk associated to banks leverage is not only a matter of financial indebtedness, but of asset risk as well. For this reasons, in this paper we improve the well-known Hamada equation by taking into consideration the interactions between leverage and asset quality. To do so, we rely on a sample of 97 EU banks over the period 2005–2016 applying a panel fixed effect regression model in which the independent variables are the interactions between leverage, banks’ credit portfolio quality and the size of risk exposure. In our regression model, we control also for bank size and time fixed effects to capture time-varying heterogeneity in the average riskiness of banks assets that varies from year to year.

Findings show the relevance of NPLs as a systematic risk component and strong explanatory power of our credit-risk-adjusted-leverage on banks equity beta.

We contribute to the extant literature providing evidence of the combined higher capital requirements-NPLs effect over bank's cost of equity in the Euro-area, which may have additional implications in the perspective of loan pricing and, more generally, from a banks value creation standpoint.

Our model suggests that leverage ratio used as in its standard version is informationally limited for investors since banks, compared to other industrial companies, are subject to capital requirements in relation to risk exposures. Therefore, leverage should incorporate asset risk adjustments for providing information that is efficient for financial markets.

The reminder of the paper is organized as follow. Section 2 develops the model. The research design and data appear in Section 3. Section 4 discusses empirical results and Section 5 concludes the paper.

Section snippets

The model

The extant literature (Baker and Wurgler, 2015, Kashyap et al., 2010, Miles et al., 2013, Toader, 2015), defined Beta coefficient in the light of Hamada (1972) equation:βE=VEβAwhere βE is the common stocks beta, V is the enterprise value, E is the market capitalization and βA is the asset beta. Generally, to investigate the effect of capital requirements on systematic risk, market values are usually replaced by accounting data:βE=Leverage*βAwhere leverage is the ratio of Total Assets to Equity

Research design and data

The first aim of our empirical analysis is to test whether the three adjustments presented in Section 2 improve the explanatory power of the standard leverage on systematic risk. The determinants of equity beta are tested through the following linear regression:Betajt=α1+γ1Leveragejt+γ2Sizejt+t+ɛjtwhere: for bank j = 1 to N and time t = 1 to T Beta is the raw equity beta calculated using weekly returns from previous two years; Leverage are the four configurations of leverage defined in Eqs. (2)

Results

In Table 3 we show the estimates of Eq. (6) with the four configurations of leverage, obtained through the fixed effect estimator. According to the significance of NPLs on beta (Das and Sy, 2012), we tested our model controlling for such additional variable, also for showing the incremental effect of the proposed leverage adjustments. In column 1 and 2, leverage is the simple leverage ratio (Leverage 1), while in column 3 and 4 leverage is the leverage ratio corrected for loan loss reserves (

Conclusions and implications

Does asset quality have an impact on the relationship between leverage and systematic risk? In this paper, we address this question proposing a three-step re-elaboration of bank-specific leverage which incorporates the idiosyncratic credit risk of banks. In addition, the study tests the role that banks size exerts in explaining systematic risk.

Based on a dataset of 97 European banks from 2005–2016, our analysis provides empirical evidence of the effect of asset quality in explaining banks’

Acknowledgments

We thank Ettore Croci, Maurizio Polato, Riccardo Ferretti and participants at the 2017 Wolpertinger Conference for helpful observations.

References (20)

There are more references available in the full text version of this article.

Cited by (17)

  • Interconnectedness between stock and credit markets: The role of European G-SIBs in a multilayer perspective

    2024, Journal of International Financial Markets, Institutions and Money
  • Determinants of connectedness in financial institutions: Evidence from Taiwan

    2023, Emerging Markets Review
    Citation Excerpt :

    In addition, supervisors also use these indicators to assess a bank's soundness. Van Oordt and Zhou (2019) find that banks with lower capital ratios are associated with a significantly higher level of systemic risk, and Beltrame et al. (2018) find a combined effect of leverage and asset quality on systemic risk. Furthermore, Brunnermeier et al. (2020) find that non-interest income is positively correlated with the systemic risk for U.S. banks.

  • A two-stage general approach to aggregate multiple bank risks

    2021, Finance Research Letters
    Citation Excerpt :

    Banks’ business activities will inevitably produce a variety of risks, such as credit, market and operational risks (Basel Committee on Banking Supervision, BCBS for short, 2006; Imbierowicz and Rauch, 2014; Beltrame et al., 2018).

View all citing articles on Scopus
View full text