A new tool in the box: dividend restrictions as supervisory policy stimulus

by Ernest Dautović, Leonardo Gambacorta and Alessio Reghezza^[1]

ECB Banking Supervision’s recommendation for banks not to pay out dividends during the coronavirus (COVID-19) pandemic led to increased lending to non-financial corporations. Complying banks’ lending was around 2.2 percentage points stronger than lending by banks not affected by the recommendation. This effect is found to be economically significant but short-lived, and stronger for small and medium-sized enterprises (SMEs) most affected by the pandemic. It did not lead to greater risk-taking by banks. These results suggest that temporarily restricting bank dividend distributions complemented macroprudential, monetary and fiscal policies to support lending during the pandemic crisis.^[2]

The ECB recommendation and its impact

On 27 March 2020 the ECB adopted a recommendation asking euro area banks not to pay out dividends, at least until 1 October 2020. The objective of the recommendation was to boost credit institutions’ capacity to absorb losses and to nudge them to continue funding households and SMEs during the pandemic.^[3]

This recommendation was issued in an environment in which banks tended to distribute a substantial percentage of their earnings as dividends. Figure 1 illustrates the cumulative planned dividend distributions prior to the ECB recommendation and the extent of compliance with the policy. At the same time, in a severe crisis, and from a wider social perspective, dividend restrictions can be justified to induce banks to conserve capital and support lending to the real economy. Empirical evidence suggests that in times of crisis banks tend to expand dividend distributions to signal their strength (Acharya et al. 2012; Abreu and Gulamhussen 2013; Wu 2018).

Banks’ managers were faced with a choice of capital allocation when deciding to follow the ECB recommendation and cut dividends, with different implications for credit supply to borrowers. On the one hand, they might opt for using the surplus capital to increase lending supply, thus supporting the economy and acting countercyclically. On the other hand, they could decide to signal – and increase – their resilience to future shocks by saving capital, and/or to strengthen their loss-absorbing capacity by setting aside provisions. The impact on lending to European corporations was therefore uncertain.

Figure 1: Dividend distribution plans of significant institutions

In Dautović, Gambacorta and Reghezza (2023), we compare banks that fully or partially complied with the ECB recommendation with a group of banks that did not change their plans, and we differentiate bank credit supply stemming from the dividend recommendation from credit supply stemming from other pandemic-related support measures.^[4] Figure 2 shows the co-movement of credit growth and the planned but non-distributed dividends scaled by risk-weighted assets (RWAs). The growth rate of lending increased in the quarter following the pandemic outbreak, with an unconditional average of 18.1%, before declining monotonically in the subsequent quarters. The amount of non-distributed dividends over RWA spiked immediately after the ECB recommendation, remaining persistent until the end of 2020 at 0.24%.

Our results show an overall positive effect of the ECB’s dividend recommendation on euro area banks’ credit supply to non-financial corporations. The effect is greater for SMEs. The results are also stronger for sectors that were more vulnerable to the effects of the pandemic, suggesting that banks channelled the surplus capital to those firms most affected by the pandemic-related lockdowns.

Figure 2: Credit growth and undistributed dividends

We do not find an increase in risk-taking by banks that followed the recommendation. Specifically, lending growth is estimated to be lower when a bank-firm relationship has accumulated impairments, and is null in the case of “zombie” borrowers (i.e. those having above the 95th percentile of impaired loans with a specific bank). Lending growth is lower for banks with high non-performing loan (NPL) ratios. Furthermore, we find that the impact of the ECB’s dividend recommendation is short-lived: the beneficial effect vanishes in the fourth quarter of 2020, and it is mostly concentrated in the second and third quarters of 2020.

Conclusions and policy considerations

Dividend restrictions can be an effective and proven countercyclical supervisory policy tool during times of crisis in the real sector (when non-financial corporations suffer liquidity shortages) – and potentially in financial crisis as well. They can sustain credit growth and have positive real effects, although these effects can be short-lived. The impact of the dividend recommendation tends to be stronger for SMEs and more vulnerable sectors of the economy. Additionally, dividend restrictions can enhance the effectiveness of countercyclical policies in a downturn, showing strong complementarity with government guarantees as a fiscal policy measure.

Implementing a dividend restriction policy can effectively redirect resources from inefficiently high shareholders’ consumption to credit supply. Recent evidence shows that investors’ consumption is excessively influenced by dividend distribution dates (Bräuer et al., 2022). Our results show that undistributed dividends are channelled into loans, which likely have a higher growth multiplier compared to consumption.

It is crucial for dividend restrictions to be temporary in nature to mitigate unintended policy effects, and clear supervisory forward guidance should accompany them. Transparent communication regarding their duration and the rationale behind their implementation can help minimise inefficiencies resulting from an uncertain policy environment. Falling to provide such clarity can undermine financial stability.

Dividend restrictions offer additionally policy improvements. They can simultaneously increase solvency (capital) and enhance loss-absorbing capacity (loan loss provisions). This, in turn, reduces the potential costs borne by debtholders and taxpayers in the event of a bail-in. Moreover, dividend restrictions can address concerns associated with the release and use of macroprudential buffers, as buffer releases can be (mis)used for increased dividend distribution.^[5]

References

Abreu, J. F. and Gulamhussen, M. A. (2013), “Dividend payouts: Evidence from U.S. bank holding companies in the context of the financial crisis”, Journal of Corporate Finance, Vol. 22(1).

Acharya, V. V., Gujral, I., Kulkarni, N. and Shin, H. S. (2012), “Dividends and bank capital in the financial crisis of 2007-2009”, NBER Working Paper, No 16896.

Andreeva, D., Bochmann, P., and Schneider, J. (2023), “Evaluating the Impact of Dividend Restrictions on Euro Area Bank Market Values”, Working Papers Series, No 2787, European Central Bank.

Bräuer, K., Hackethal, A., & Hanspal, T. (2022). Consuming dividends. The Review of Financial Studies, 35(10), 4802-4857.

Dautović, E., Gambacorta, L. and Reghezza, A. (2023), “Supervisory policy stimulus: evidence from the euro area dividend recommendation”, Working Papers Series, No 2796, European Central Bank.

Lee, B.-S. (1995), “The response of stock prices to permanent and temporary shocks to dividends”, Journal of Financial and Quantitative Analysis, Vol. 30(1), pp. 1–22.

Todorov, K. (2020). Quantify the quantitative easing: Impact on bonds and corporate debt issuance. Journal of Financial Economics, 135(2), 340-358.

Wu, Y. (2018), “What’s behind smooth dividends? Evidence from structural estimation”, The Review of Financial Studies, Vol. 31(10), pp. 3979–4016.