A common measure of profitability is return on equity, or RoE. In the run-up to the recent financial crisis many banks increased their RoE by boosting net income and operating with lower levels of equity, driven by investors demanding higher returns and a loosening of regulatory standards. This caused the RoE of many banks to reach levels in excess of 15%. Banks took more risks in their pursuit of a higher RoE, and many ran into trouble once the crisis hit.
To prevent this situation from reoccurring, regulatory requirements have been tightened since 2008. For example, the Basel III accord requires banks to increase both quantity and quality of their capital. Higher equity allows banks to absorb larger losses, but also dampens their RoE.
Negotiations are currently under way in the Basel Committee about how risk weights must be set. These weights are used to determine risk-weighted assets as a basis for capital requirements. As risk weights go up, banks are required to maintain more capital, causing their RoE to go down. This debate is also important to banks that use internal models to set their risk weights. The Basel Committee is considering to subject the use of internal models to restrictions, in order to improve consistency and comparability between the models banks use.
It is against this background that the question arises of what could be a feasible RoE level. Many European banks still use double-digit RoE targets, and often refer to tightened regulations as obstacles to achieving those targets. However, the changed economic environment also plays a role.
In a new study, DNB makes an assessment of a feasible RoE level in the longer term, based on data from the three Dutch large banks – ABN AMRO Bank, ING Bank and Rabobank. The study presents a number of scenarios that combine the potentially stricter capital requirements with other factors that may affect profits. The scenarios are not predictions, but aim to determine a plausible range for RoEs under different assumptions, which reflect the current uncertainty surrounding the further tightening of regulatory requirements. It must also be taken into account that the scenarios provide an average picture of banks, whereas RoEs of individual banks will vary in practice.
In mid-2016, the RoE of the major Dutch banks averaged 7.3%. In a balanced scenario, it could go down slightly to some 7%, provided that banks are able to achieve most of their planned cost savings and pass on part of their regulatory costs to customers. In a scenario in which prudential requirements are tightened to a limited extent only and banks can easily cope with them, causing RoEs to end around two percentage points above current levels. In an alternative scenario, however, they may go down about two percentage points, due to significantly more onerous prudential requirements and a lack of compensatory measures.
In sum, pre-crisis RoEs are unlikely to return. Even under favourable economic circumstances, banks will need to make a substantial effort to maintain their profitability at the current level. To do so, they must make sizeable cost savings and pass on part of their additional costs to higher margins without losing much of their market shares.
The structural fall in RoEs does not necessarily make banks less attractive investments. After all, stricter rules and changed business models have made banks safer, so the risk premium above the risk-free interest rate which shareholders demand may be lower than before the crisis. Furthermore, the risk-free interest rate has dropped in recent years and may well remain low for a long time to come.