Robert Hills, John Hooley, Yevgeniya Korniyenko and Tomasz Wieladek.
When funding conditions became much more difficult in the recent financial crisis, how did UK banks react? Did they adjust their domestic and external lending to different degrees? Did foreign-owned banks behave differently from UK-owned banks, and did it make a difference whether they were a branch or a subsidiary? Did the other features of their balance sheet make a material difference to their lending behaviour? Our research suggests that the answer to all of these questions is “yes”.
This is an important set of questions because it helps us to establish the extent to which international activities of banks act as a cross-border transmission mechanism for liquidity shocks – and indeed whether they amplify those shocks. In a recent Bank of England Staff Working Paper (also in the IMF Economic Review), we looked to answer these questions, using a proprietary dataset on individual UK-resident banks’ activities.
Given the UK’s importance as an international financial centre, the results of an exercise of this sort limited to UK-resident banks would be of interest in and of themselves. But our work forms part of a much broader project. The International Banking Research Network (IBRN), which was established in 2012, brings together central bank researchers from around the world. Central banks are in a unique position to understand developments in the banking sector, since they often collect individual bank-level data that are typically not available to others – even to other central banks. But if each central bank research team investigates a common research question and uses similar methods, they are able not only to make cross-country comparisons across different jurisdictions, but also to start to put together a map of how shocks propagate globally. The first IBRN project (of which our paper was a part) is described in this VoxEU post by Claudia Buch, James Chapman and Linda Goldberg, and the full set of papers was published in a special issue of the IMF Economic Review.
Our research team looked at how UK-resident banks with different characteristics reacted to a rise in funding costs during the crisis. The basic idea is that the effect of the liquidity shock on bank assets can be identified by comparing institutions with different balance sheet characteristics (a ‘difference-in-difference’ panel regression), which requires conditioning on both the ex ante characteristics of the individual bank, and demand conditions common to all banks. Specifically, we run a set of regressions, over the period 2006-12, in which we look at the effect on various dependent variables (the flow of total external and domestic lending to banks and non-banks, the flow of private credit, commitments and liquid assets) of balance sheet characteristics such as commitments, capital, total assets, share of domestic deposit funding, and holdings of liquid assets. The liquidity variable is the difference between the 3-month unsecured interbank lending rate (LIBOR) and the OIS, an interest rate swap derived from the overnight rate, both in sterling.
We found that UK-resident banks that grew their balance sheets quicker relative to their peers pre-crisis responded by decreasing their foreign lending by more than other banks. Specifically, the results suggest that, for a 1pp increase in the LIBOR-OIS spread, a bank whose balance sheet is twice as large relative to its own historical average and its peer group would contract total lending as a share of its balance sheet by around 37% more than the average bank, primarily by cutting its external lending by 47% more. But, interestingly, they also increased their domestic lending – by about 11% more than the average bank. How might we interpret this? One possibility is that, in advance of the crisis, banks that grew their balance sheet quickly relative to their historical average and their peers might have been more aggressive risk-takers. But it is difficult to know this with certainty.
We saw broadly the same pattern for both UK-owned and foreign-owned banks. UK-owned banks with higher pre-shock asset growth cut their external lending by more than average (17%, though the coefficient is not significant), offset by expanding their domestic lending by 42% more. That might suggest a form of home bias – a widely documented feature of bank behaviour in many countries since the onset of the crisis.
We also found that banks with relatively low reliance on wholesale funding both cut back external lending and raised domestic lending by more. This is consistent with the idea that banks with a funding model more reliant on stable retail deposits used the disruption in wholesale funding as an opportunity to gain market share in the UK (as hinted at in this paper).
There was some evidence that the commitment ratio was also an important element of the liquidity shock transmission: banks with a high commitment ratio tended to cut back on commitments, while raising domestic lending. This was particularly pronounced for foreign banks and lending to non-banks, which suggests that this might reflect their desire to protect existing domestic lending relationships at the expense of new commitments. Interestingly, our results suggest that banks affected by official interventions did not discriminate in that way, perhaps as a means of passing on the support to real economy borrowers.
So the bottom line from the UK data is that certain balance sheet characteristics (rapid pre-crisis asset growth, reliance on wholesale funding) seem to have played a particularly significant role in determining the extent to which banks retrenched their cross-border lending during the crisis, contributing to the international transmission of the liquidity shock. And that tentatively suggests some areas for policymakers to focus their attentions when seeking to limit the transmission of future shocks.
Robert Hills works in the Bank’s International Surveillance Division. John Hooley and Yevgeniya Korniyenko work at the IMF and Tomasz Wieladek works at Barclays Capital. They all worked at the Bank of England when the paper was written.
If you want to get in touch, please email us at firstname.lastname@example.org. You are also welcome to leave a comment below. Comments are moderated and will not appear until they have been approved.
Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.