Policy initiatives in several European countries have targeted bank profits to secure public funds. These measures include introducing windfall taxes on excess profits and, in some instances, encouraging higher, taxable dividend pay outs. At face value, the case for taxing banks appears intuitive: people are facing hardship and banks are making record profits. This column argues, however, that the recent boom in bank profits is set to fade in the next two years. Despite generally adequate capital levels, most banks should therefore lock in profits into buffers to absorb future shocks and ensure a steady provision of credit even in adverse scenarios.
Senior Economist International Monetary Fund
Banks earn interest on longer-dated loans and securities and pay interest on shorter-dated deposits. Their interest rate margins typically increase when policy rates rise or the yield curve steepens, as returns on floating rate assets rise faster than payments on relatively stickier deposits. This positive effect on margins is particularly strong when policy rates rise from low levels or, as in recent years, switch back from being negative to positive, as shown by Adalid et al. (2023) and Beyer et al. (2024) among others. Since deposit rates generally do not turn negative (at least not in the case of retail deposits), banks’ margins compress when policy rates become negative. On the other hand, higher interest rates generally affect borrowers’ ability to repay their debt and raise loan loss provisions, reducing bank profits over time.
In our recent research (Chen et al. 2024), we compiled balance sheets and income statements from over 2,500 European banks. Evidence based on this new broad dataset shows that in 2023 the return on assets (ROA) for European banks increased sharply, on average doubling the level prevailing during 2015-19 (see Figures 1 and 2).
Figure 1 European banks’ return on assets
Figure 2 European banks’ net interest margins
This is in stark contrast with the lacklustre performance of the pre-pandemic years. After the Global Financial Crisis (GFC), European banks’ ROA fell sharply and, despite some improvement over the following years, it never recovered anywhere near its pre-GFC levels. The drop was even larger and more persistent when measured in terms of return on equity (ROE), as tighter regulatory standards compelled banks to build larger capital buffers, further compressing ROE (see Figures 3 and 4).
Figure 3 European banks’ return on assets by size
Figure 4 European banks’ return on equity by size
Two important factors have contributed to the recent boom in bank profits. First, in the last tightening cycle, deposit rates have been slower to respond to rising policy rates than loan rates (and even slower than in previous cycles). Abundance of bank liquidity significantly reduced competition for deposits in the aftermath of the pandemic. Therefore, banks, which had been reluctant to pass on negative interest rates to retail depositors, were able to raise deposit rates only gradually when policy rates turned positive again, achieving wider net interest margins. Second, resilient labour market conditions and healthy corporate balance sheets have held down non-performing loans so far, allowing lower provisioning expenses.
Passthrough to deposit rates could soon accelerate, as noted by Messer and Niepmann (2023) among others, and record bank profits are not here to stay. As bank reserves are drained and deposit holders switch banks or reallocate funds from sight to term deposits in search of better remuneration, bank funding costs are increasing with a lag, and profits are already being compressed. Moreover, non-performing loans may still increase as the lagged effect of earlier shocks and tighter financial conditions weaken asset quality only over time, as shown by Ari et al. (2021). Pressure on borrowers’ ability to service their debt will gradually subside, as central banks lower policy rates. But, at that same time, bank net interest margins will likely decline, as deposit rates, which lagged loan rates in the tightening cycle, are set to fall by a smaller extent.
To investigate these channels, we model net interest margins as a function of a short-term sovereign yield and a term premium reflecting the maturity mismatch between banks’ assets and liabilities. As expected, net interest margins are found to be a concave function of short-term yields. In other words, margins are more sensitive to changes in short-term yields when interest rates are close to the lower bound (and when banks financed themselves primarily through retail deposits which are sluggish compared to other forms of funding). Figure 5 shows the path of the median model-implied net interest margin during 2024-26, conditional on the assumed path for sovereign yields and term premia. The grey area indicates the projection period. A significant share of the increase in net interest margins recorded in 2023 is projected to fade, with only a small residual amounting to no more than 200 basis points extending over 2024-26 on relatively high short-term sovereign yield (compared to pre-pandemic years), and a gradually steepening yield curve. In short, almost 90% of the increase in net interest margins recorded by European banks in 2023 is projected to fade by 2026.
Figure 5 European banks’ actual/projected interest margin
In the meantime, most structural factors that have eroded the profitability of European banks in the past two decades have remained largely unaddressed. The ECB (2017, 2019) stresses that overcapacity and low competitive dynamics are important structural hurdles to bank profitability. If overcapacity in certain markets, low operational cost efficiency, and limited adoption of digital technologies are not tackled appropriately, they will continue being a drag on profits and capital on top of the effect of narrowing net interest margins.
Against the backdrop of temporarily high bank profits, 12 European countries have introduced new taxes on banks since 2023. In a companion paper (Maneely and Ratnovski 2024), we offer a detailed review of all the new bank taxes. The design of these measures reveals significant heterogeneity in terms of their tax base, rate, duration, and burden. With only a few exceptions, the annual expected tax revenue varies between 0.25% and 1% of bank risk-weighted assets. But, even in the case of very low, de-facto symbolic levies, ad hoc taxes introduced in response to a temporary surge in profits may hamper the predictability of the business environment and dampen investor confidence, raising the cost of equity financing and potentially inhibiting banks’ ability to finance the recovery from the current cyclical downturn. Moreover, such heterogeneity of policy responses may contribute to an uneven playing field among banks from different countries potentially further undermining efforts towards a banking union.
Another concern is that taxes on bank profits or one-off, taxable dividend payouts may draw on financial resources that banks could have otherwise allocated to their capital base. While taxing bank profits could be a convenient way to raise fiscal revenue, the possibility of rising credit risk suggests that these profits could be used more appropriately to build or preserve bank capital buffers. The need to build buffers clearly differs across countries or individual banks. It may be less urgent where banks have already channelled profits towards voluntary buffers, but for other banks there may be a more compelling case.
In this context, if taxes are to be raised, arrangements that include options to allocate profits to bank capital are therefore preferrable as they ensure that such capital remains locked in and unavailable for the near term. We have long advocated for higher countercyclical capital buffers (CCyB), including through the possibility of setting a positive rate when cyclical systemic risks are not elevated, in other words a positive neutral rate. As argued in our research (Maneely and Ratnovski 2024) and by Miettinen and Nier (2024), high bank profits now offer an opportunity to increase CCyB. Especially for countries that still have zero or low CCyB rates (see Figure 6), allocating temporarily high profits to releasable capital would considerably increase banks’ resilience to future shocks.
Figure 6 CCyB rates in participating Banking Union countries (%)
Structural considerations also remain key when it comes to bank profits. Since the GFC, fundamentals have not improved enough to drive European bank profits to a level consistent with the compensation demanded by market participants to invest in and hold banks’ equity over the medium term, as noted by Di Vito et al. (2023). Not long ago – especially in the period between the GFC and the COVID-19 pandemic – discussions of low profitability for European banks were prevalent in the literature (e.g. ECB 2015, 2018, 2019, as well as Detragiache et al. 2018). Those earlier concerns appear to have rapidly given way to more sanguine views among some policymakers. However, most of the structural factors that have eroded the profitability of European banks in the past two decades have yet to be addressed and will continue weighing adversely on banks’ capital.
There is significant scope to strengthen bank business models, including through increased diversification in target markets and funding sources, improve cost efficiency, and promote broader adoption of digital technology. Further progress towards a banking union within the EU could help bring about significant efficiency gains.
Source: cepr.org