The trauma and cost of a public rescue must surely teach the bank management concerned to behave in a more prudent manner, right?
Wrong, according to a recent Bank of International Settlements paper.
Authors Michael Brei and Blaise Gadanecz have asked a simple enough question: what happens to a bank’s lending profile, in terms of risk, following a bailout? To try and answer this, the authors looked at 87 bank holding companies in 14 countries, covering a cool $54 trillion of assets, representing some 52 per cent of the worldwide banking industry. They compared the syndicated loan book, as a proxy for willingness to take on risk, before and after a bailout. In short:
The conclusion is that a bailout made little difference to a bank’s risk appetite (emphasis ours):
We find no evidence that rescued banks reduced the riskiness of their new lending more than non-rescued banks in response to the crisis and the public rescues. Even as lending volumes decreased across the board in 2009, rescued banks continued to write riskier syndicated loans, as reflected by their involvement in the leveraged loan segment and in the spreads charged on the facilities that they originated. We also find, unsurprisingly, that the syndicated lending of banks that later received a bailout was riskier before the crisis than that of non-rescued institutions.
While multiple factors are obviously at play when a bank gets bailed out (such as political pressure to keep corporate customers afloat), the research here suggests that when it comes to risk, bad banks largely carry on regardless.
Brei and Gadanecz also found that the rescued and the non-rescued groups varied in some important ways coming into the crisis. For example, before the crisis, rescued banks actually had a lower average loan-to-asset ratio than non-rescued banks, although rescued banks were typically more dependent on non-deposit funding, making them more vulnerable when liquidity begun to dry up. Following the crisis, rescued banks had a higher ratio of impaired to total loans “either because they were facing more impaired loans or because the rescues were associated with higher recognition of such loans”, note the authors.
Returning to the issue of the syndicated loans, why use them as a risk appetite proxy? Brei and Gadanecz:
With close to $7 trillion of new facilities signed in 2007, syndicated lending has been one of the largest sources of corporate funding. Syndicated loans also form a significant component of banks’ total portfolio of commercial and industrial loans. Importantly, the available information on individual borrowers (like sector or nationality) and loan transaction terms (such as spreads, maturities or guarantees) makes the syndicated loan market a good laboratory for analysing bank risk.
This is how those loans performed over time:
As the charts above show, rescued banks issued more leveraged syndicated loans and the average Libor spreads at the rescued banks’ new loans were much higher than at the non-rescued banks. The average maturity was also higher in the former group, and the loans they issued were downgraded to a greater degree than the loans issued by the non-rescued group. But the post 2007 data on syndicated loans suggests that the bailouts didn’t result in rescued banks slashing their risk appetite:
During the crisis, rescued banks did not reduce the riskiness of their new syndicated lending compared to their non-rescued peers. In fact, our results suggest that the relative riskiness of their lending increased. This is apparent when comparing how the two types of institutions changed their participation in leveraged facilities (relative to their total new signings), as well as the average Libor spread on those signings and the corresponding average maturities.
… as you can see from this table:
As the table shows, the non-rescued banks did reduce their risk. Good (or less-bad) banks behaved better. They made smaller leveraged loans (as a percentage of new signings) and the average Libor spread on new signings fell significantly.
Maybe this a natural and obvious outcome: it was a threat of state intervention (and management sackings) that encouraged banks that had avoided bailouts to stay out of trouble; meanwhile, bailed out banks were already in the mire, so the motivation to slash risk was just not there…
BoE research shows big banks were too big to fail – FT Alphaville