What caused liquidity disruptions during the financial crisis?

The Global Crisis that started with the Lehman Brothers failure in September 2008 and intensified, especially in the Eurozone, with the sovereign debt crisis after April 2010 was largely centred on dry-ups in wholesale funding liquidity, in stark contrast to historical systemic crises where the runs were mainly by retail depositors. Importantly, there has been a geographical fragmentation of liquidity in global markets, notably around the sovereign debt crisis, partially unwinding the financial globalisation trend of the last two decades. The main responses to combat these tensions have been central banks’ non-standard monetary policy actions.

However, empirical evidence on wholesale turmoil and the effects of monetary policy remain scarce, mainly due to the relative lack of comprehensive micro-datasets since wholesale transactions are mostly over-the-counter. In this column, we present micro evidence on the impact of financial crises and monetary policy on the supply of wholesale funding liquidity using a detailed loan-level (lender-borrower-level) dataset of interbank lending based on the Eurozone TARGET 2 data.1

Maturity shift: First tightening in term lending, then overnight lending

For the Eurozone, we document that in the first two weeks after Lehman’s failure the total turnover in overnight markets actually increased slightly (Figure 1a), i.e. no sign of an imminent liquidity dry-up (Abbassi et al. 2014). However, term-lending volumes decline immediately after Lehman’s collapse by up to 60% as compared to the pre-Lehman level. About two weeks after Lehman, when term-lending volumes discontinue dropping after the massive long-term liquidity freeze, the total overnight lending activity starts decreasing, at a similar degree as observed for term-lending turnover. With the intensification of the sovereign debt crisis, there is a further shift from term-lending to shorter maturities and a stronger tightening for term-credit conditions, which is stronger for borrowers headquartered in countries with sovereign debt problems (Figure 1b).

Figure 1.

Cross-border liquidity dries up during crises, heterogeneous effects for core and periphery banks

Figure 2 shows that specifically cross-border lending constrains tighten during crisis times and impair the intermediation role of interbank markets in the Eurozone. In particular, we document a reduction of access to cross-border loans, and conditional on granting a cross-border loan, a reduction of loan volumes and an increase in interest rates. Interestingly, we find a strong segmentation in domestic versus cross-border lending on all margins (i.e. funding access, volumes, and prices) not only during the sovereign debt crisis, but also in the period after the Lehman’s failure in 2008 when sovereign debt problems were not pronounced in the Eurozone.

Figure 2.

During the sovereign debt crisis, however, there is important heterogeneity depending on the countries where the borrower is headquartered. Supply restrictions to cross-border overnight funding in crisis times are only binding for borrowers from Troika-rescued periphery countries (i.e. Portugal, Ireland, and Greece, which were rescued by an IMF-EU-ECB consortium), with a reduction of up to 99% at the worst moment of the crisis (Figure 2a). Moreover, for the granted cross-border loans, spreads paid by banks headquartered in these countries for the relatively few granted loans increase by 31 basis points relative to core country banks. For the large periphery banks headquartered in Italy and Spain, the crisis-induced effect on access to overnight credit is not different from core country banks, but spreads increase by 12 basis points as compared to spreads paid by core country banks. Our results therefore suggest that beyond counterparty risk and sovereign default risk or liquidity hoarding, other factors such as bank-to-bank frictions are important drivers of interbank distress during financial crises that contribute to the segmentation of the Eurozone interbank market.

Price dispersion, credit constraints, and monetary policy

The interbank market – unlike other credit markets – allows to exploit the price dispersion from different lenders on identical credit contracts, i.e. overnight uncollateralised loans during the same day (or morning) for the same borrower. We find that the average price dispersion before Lehman’s failure is virtually zero as different lenders charge similar prices from the same borrower during the same day. On the worst day of the Lehman crisis, 10% of all banks face an average price dispersion of 80 basis points during the day; for some borrower banks, the price dispersion is as large as 100 basis points. Figure 3 depicts the mean and the respective percentiles of the price dispersion distribution during the Lehman (a) and sovereign (b) period, where the results are similar but quantitatively at a smaller scale.

Figure 3.

Moreover, our findings suggest that the price dispersion relates to borrower characteristics. Specifically, riskier borrowers and borrowers from countries with strong sovereign debt problems are subject to high price dispersion during crisis periods. At the bank-to-bank level, cross-border loans for the same borrower in the same day have higher interest rates during the crisis, even after controlling for different loan volumes and other conditions. Further, borrowers pay higher rates from lenders with a previous strong lending relationship, but are more likely to obtain funding and higher loan volumes. None of these effects prevails in the pre-crisis periods.

To assess the effectiveness of the main non-standard monetary policy measures during the financial crisis and the sovereign debt crisis (before the 2015 QE), we analyse the impact of the Eurozone’s switch to the fixed-rate full allotment (FRFA) and the 3-year long-term refinancing operations (LTRO). We find that the announcement of the Eurozone’s fixed-rate allotment policy in October 2008 and the announcement of the first 3-year long-term refinancing in December 2011 significantly reduced the price dispersion and diminished it almost to the pre-crisis level (Figure 3a and 3b). Moreover, the quantitative effects are significantly stronger for banks headquartered in countries with severe sovereign debt problems (Greece, Italy, Ireland, Portugal, and Spain).

Conclusion

In sum, our analysis of the Eurozone interbank money market during the Lehman and the European sovereign debt crisis shows strong tightening in funding access, volumes, maturity, and prices (specifically for cross-border loans), as well as the substantial different interbank prices for the same borrower in the same morning during the crisis periods. These results suggest that bank-to-bank frictions impaired the functioning of this market severely, in particular for cross-border transactions. Interestingly, this was not only the case for the sovereign debt crisis but also after the Lehman failure. This might suggest that even the financial integration of the money market, which is typically considered to be the most integrated financial market in the Eurozone, is not very resilient. Finally, non-standard monetary policy measures by the Eurozone help to mitigate these liquidity disruptions, with stronger effects in countries under distress.

This article is published in collaboration with VoxEu. Publication does not imply endorsement of views by the World Economic Forum.

To keep up with the Agenda subscribe to our weekly newsletter.

Author: Puriya Abbassi is a Financial Stability Expert, Deutsche Bundesbank. Falk Bräuning is a PhD Candidate, VU University Amsterdam and the Tinbergen Institute. Falko Fecht is a Professor for Financial Economics and DZ Bank Endowed Chair of Financial Economics, Frankfurt School of Finance and Management. José-Luis Peydró is an ICREA Professor of Economics at UPF, Barcelona GSE; Research Professor and Research Associate, CREI; and CEPR Research Fellow.

Image: A man looks at a stock quotation board outside a brokerage in Tokyo May 11, 2012. REUTERS/Toru Hanai