Working Paper Series 30 Mar. 2020 - The effect of possible EU diversification requirements on the risk of banks’ sovereign bond portfolios
Non-technical summary
In the current Capital Requirements Regulation, special treatment is reserved for exposures of European banks to government bonds denominated in domestic currency. Banks are not required to fund with capital their investments in sovereign bonds denominated in euro, which are considered de facto riskless. Furthermore, sovereign exposures are not subject to any concentration limits and can represent a large part of banks’ capital. As a result, there are strong regulatory incentives for banks to hold disproportionate amounts of domestic sovereign debt for capital and liquidity reasons.
Recently, policymakers proposed the introduction of capital rules and diversification requirements for euro area government bond holdings [ESRB, 2015, Juncker et al., 2015, Veron, 2017]. The rationale behind this regulation goes beyond improving banks’ risk management and resilience to sovereign risk. Regulators want primarily to weaken the so-called doom loop between sovereigns and banks that has emerged especially since the financial crisis in 2007-08. This strong nexus, represented by high balance sheet exposures of banks toward sovereigns, allows countries with weak finances to heavily affect their banking system, with banking sectors in distress more likely to receive government financial support. Additional to a reallocation of banks’ sovereign exposures, other possible ways to break the doom loop would be to improve countries’ fiscal soundness and/or enhance banks’ equity requirements. Increasing diversification in banks’ sovereign portfolios is considered a necessary step for the introduction of the joint European Deposit Insurance Scheme (EDIS), which requires the national deposit protection schemes to be combined [Juncker et al., 2015]. However, if banks are largely exposed to their own sovereign debt, a joint deposit guarantee scheme might result in a sharing of fiscal risk [Weidmann, 2016].
To inform the policy discussion, this paper analyses the effects of possible responses of banks to a new diversification requirement. We use a sample of 106 European banks included in the EBA stress test dataset over the period June 2013 to December 2015. These banks cover approximately 70% of banking assets in each of their countries and across the EU. Sovereign exposures represent a large part of their total assets and are much larger than banks’ capital.
In this paper, we point out how the standard definition of diversification, quantified by looking at the distribution of asset holdings, might be too limited. Other dimensions, such as risk exposure and factor exposure, need to be explicitly taken into account to better describe the status quo of banks’ portfolio allocations as well as to model the potential consequences of a rebalancing, forced by the new regulation, which is intended to reduce home bias and favour more overlapping portfolios.
Given that the reduction of risk is a major reason for a diversification requirement, our results suggest caution before its adoption. We examine the question with a set of risk tools that have not been used in a banking regulation context, but are especially relevant to understanding the sources of risk for the current sovereign debt portfolios of European banks and the impact of the likely responses to limits placed upon single sovereign exposures. First, we identify the common risk factors of European sovereign risk through an independent component analysis and introduce different diversification measures used to evaluate portfolios in the financial services industry. Using simple rebalancing rules, we find that the likely portfolios that result from such higher diversification requirements will generally increase the risk of most banks in the euro area. As a first step, we focus on portfolio variances of countries’ banking systems and of individual banks. Then, because tail risks play a key role during a crisis, we also estimate the impact of portfolio rebalancing on value-at-risk by using risk aggregation techniques developed by Bernard and Vanduffel [2015].
We focus on the assumption that banks would choose a sovereign bond portfolio that most closely matches the risk/return profile of their current portfolio, always preferring a less risky one. This is a strong assumption and our results depend upon it, but it is not unrealistic on either theoretical or empirical grounds. Another possible scenario that we analyze is that of banks rebalancing toward safer bonds, i.e. “flight-to-quality”. We find that, under this assumption, banks would achieve lower return portfolios with similar or lower risk profiles to the current ones.
Copyright: European Central Bank
Directorate General Communications
Sonnemannstrasse 20, 60314 Frankfurt am Main, Germany
Tel.: +49 69 1344 7455, E-mail: media@ecb.europa.eu
Website: www.ecb.europa.eu
page source https://www.ecb.europa.eu/
