Bank capital regulation: How the implementation of CRD IV affects Danish financial institutions’ capital structures

Abstract

This thesis finds that three capital requirements are especially important for banks and that breaches of these have different but all severe consequences. These requirements are the MDA requirement, the Pillar 2 requirement and the minimum capital requirement. These requirements make the bank’s capital structure decision especially challenging compared to corporate firms. The thesis uses a model by Harding, Liang and Ross (2009) that estimates the optimal capital structure of banks by taking into account deposit insurance, capital requirements, bankruptcy costs and tax-advantaged debt. The model assumes that banks face bankruptcy if they fail to comply with their capital requirement but that some of this cost is balanced by the government’s deposit insurance. When applied to the cases of Danske Bank, Sydbank and Jyske Bank, however, it dramatically overshoots the optimal capital ratio compared with the observed capital ratios and the banks’ own proclaimed target capital ratios. The shortcomings of the model are the riskless interest rate’s dramatic influence on results as well as the fact that it fails to take into account risk-weighted assets. The extended model includes a second capital requirement, which is modeled as the combined capital requirement. This model overshoots the optimal capital structure marginally more than the base model. A new model is developed specifically to take into account the new regulatory framework in Denmark, where systemically important banks are not allowed to be liquidated but where shareholders and owners of banks’ AT1 capital face the risk of loss even though the bank is not liquidated. The predicted capital ratios of this model for Danske Bank, Sydbank and Jyske Bank are more realistic but still much higher than the banks’ own capital ratio targets. The shortcoming of the model stems from the current extremely low riskless rate’s large influence on the result. The riskless interest rate is negatively correlated with the optimal capital ratio and the asset volatility and the combined capital requirement are positively correlated with the optimal capital ratio. The combined capital requirement only has a marginal impact, which implies that the phase-in of higher capital requirements until 2019 will have little impact on Danish banks’ optimal capital ratio. The main factor in the optimal capital structure decision is the risk of expropriation and the risk that tax benefits are lost in the process. The results further imply that bank managers, investors and regulators should be careful not to forget to focus on the capital ratio in terms of total assets. It is tempting just to look at the capital ratio in terms of risk-weighted assets, as this is the ratio that is monitored in relation to capital requirements. However, the average risk weight can differ substantially between banks without an apparent difference in the riskiness of the banks’ assets. Therefore, a bank with a very low average risk weight but an asset volatility that is comparable to banks with higher average risk weights, should target a higher capital buffer in terms of risk-weighted assets. The potential introduction of a leverage requirement can help on this issue. The implementation of CRD IV is predicted to increase Danish financial institutions’ optimal capital structures marginally. It is also found that SIFIs’ optimal capital structures will not be significantly different from non-SIFIs.