In the aftermath of the financial crisis, in several countries new levies on the financial sector have been proposed and in some cases implemented. We focus in particular on the recent introduction of a securities transaction tax (STT) in Italy. A peculiarity of the Italian STT is that it only concerns stocks of corporations with a market capitalization above € 500 million. We exploit this feature via a differences-in-differences approach – comparing taxed and non-taxed stocks before and after the introduction of the tax – and via a regression discontinuity design – comparing the performance of stocks just above the threshold with those just below. Focusing on the regulated market, we find that the new tax reduced liquidity, but it left transaction volumes and returns substantially unaffected. There is also evidence – although not conclusive – that the tax increased volatility.

Banks play a crucial role in providing liquidity to borrowers, particularly during crises (Kashyap et al., 2002 [33]). The existence of multiple lending relationships between banks and borrowers has been seen as an element that reduces the risk of liquidity shortage for debtors (Detragiache et al., 2000). In this paper, we aim to show how the interaction of these two aspects with solvency and liquidity requirements might have implications for the stability of the banking system, which might still need to be fully analyzed. We show that if other sources of liquidity are unavailable or too costly for banks, multiple lending might be a key element in a systemic liquidity shortage and a large drop in lending to the economy. These findings are particularly relevant for understanding how macroeconomic shocks, such as the relatively recent outbreak of COVID-19, could impact the real economy, as well as for assessing the implications of alternative banking resolution mechanisms.

We study the impact of higher capital buffers on bank lending and risk-taking behaviour, at different time horizons following the initial policy decision. Employing a regression discontinuity design and confidential centralised supervisory data for euro area banks from 2014 to 2017, our research uniquely explores the effects of the EU policy on other systemically important institutions (O-SIIs) through a quasi-randomised experiment, exploiting the induced policy change and discontinuity of the O-SII identification process. Our findings show that the introduction of the O-SII buffers resulted in a short-term reduction in credit supply to households and financial sector, followed by a medium-term shift towards less risky borrowers, particularly in the household sector. We find a temporary cut in loan growth post-capital hikes, succeeded by a rebound in the medium-term. Our results substantiate the hypothesis that higher capital buffers can positively discipline banks by reducing risk-taking in the medium-term. At the same time, evidence suggests a limited adverse impact on the real economy, characterised by a temporary reduction in credit supply restricted to instances of macroprudential policy tightening.