Rhiannon Sowerbutts and Peter Zimmerman
Governments have often supported troubled banks whose failure would damage the wider economy. The expectation of such bailouts amounts to free insurance for those who have lent money to these ‘too big to fail’ (TBTF) banks. This amounts to an ‘implicit subsidy’ from the government, with a value that may be as large as £100bn. But where does this money go? We think most of the benefit goes to those who own or work for banks. But verifying this empirically is a challenge requiring further research.
The implicit subsidy: it Hester to go somewhere.
The prospect of government bail-outs reduces the risk that the bank’s creditors incur; effectively the government insures them against credit risk. But, unlike explicit deposit guarantees or insurance, the government receives no compensation and so taxpayers are effectively providing a free subsidy. Estimates of the size of the subsidy in the UK range from £6bn to £100bn (Noss and Sowerbutts (2012)).
Asked by the UK Treasury Committee in 2011 where this subsidy goes, the then chief executive of RBS Stephen Hester agreed that some of this subsidy may “could have fed through to lots of places”, including bankers’ bonuses and the price of customers’ loans. But how can we attempt to quantify the benefit to these different groups? We think there are a number of stakeholders who could benefit from this subsidy – creditors, customers, employees and shareholders – and we examine each in turn.
Our basic idea is to think of the implicit subsidy as a reduction in the cost of an input to the production of financial services – it makes it cheaper for banks to borrow money. Therefore it is a transfer of wealth from (future) taxpayers to bank stakeholders. Standard textbook microeconomics tells us that the distribution of the resulting monetary benefits depends on the scarcity of inputs: in general, the less elastic the supply of a factor in production, the greater the surplus extracted by the supplier of that factor. The competitive environment in the industry also matters because it affects the degree to which changes in marginal costs are reflected in the price paid by the consumer.
Treating the implicit subsidy as a transfer from the government to stakeholders is an over-simplification, because we would expect the existence of the subsidy to distort agents’ incentives and change their behaviour. Some of these distortions are discussed in this post, but they do complicate the issue and make it very difficult to carry out any empirical analysis.
Creditors pass the subsidy on to others…
The subsidy most plainly has a direct effect on creditors, because if a bank cannot fail then their debt will always be repaid. But do creditors capture all of the benefit? With efficient and functioning markets, expectations of a bailout should be fully priced-in and reflected in lower bank funding rates, leaving creditors’ risk-adjusted returns unchanged. If that occurred, then creditors would receive zero net benefit from the subsidy.
But there are several reasons why this may not be the case. Frictions in the market for bank debt – such as restrictive investor mandates and monitoring expenses– mean that that the number of buyers of bank debt is limited. This would mean that pricing does not fully adjust to account for the guarantee, and so some of the subsidy would transfer to creditors.
A number of empirical studies find that TBTF banks do enjoy lower bond spreads. For example, Acharya, Anginer and Warburton (2015) estimate a relationship between bond spreads and indicators of whether a firm is TBTF, controlling for riskiness. They find that the funding subsidy for the largest banks reached 120 bps in 2009-10. This suggests that debt prices do factor in at least some of the government guarantee, so we think that stakeholders other than creditors must benefit from the implicit subsidy.
… and bank customers probably don’t see much of the benefit either.
As Hester suggested, customers of TBTF banks may extract some of the benefits of the subsidy. Lower debt costs for these banks mean that the marginal cost of providing banking services falls, allowing banks to be able to provide the same services at a lower price.
These cost savings are likely to be passed on to customers if the market for consumer credit is competitive. But if the market is not very competitive, then banks may just keep prices the same and retain the surplus. And it is generally recognised that, in the UK, the supply of high street bank credit is very inelastic (Independent Commission on Banking (2011)), so we doubt that customers benefit much from the implicit subsidy in the UK.
The ICB also pointed out that the implicit guarantee introduces a competitive distortion, conferring an advantage on larger banks and raising barriers to entry. In this way, the implicit guarantee could turn a largely competitive market into a non-competitive and non-contestable one.
Another effect of the implicit subsidy is to incentivise banks to take more risks (more on this below). One effect of this may be to allow riskier customers to obtain access to credit which they would not have otherwise. Therefore these borrowers may receive some benefit from the presence of the subsidy, at the expense of safer borrowers and also at some potential risk to wider financial stability.
Employees may be in a position to capture much of the subsidy…
Hester also conceded that employees of TBTF banks may be also to extract some benefit from the subsidy, for example by claiming higher salaries or bonuses. When the expertise for a specific role is scarce, then employees may be price makers and so able to demand higher compensation. However, if there is competition for roles then it may be more difficult for them to extract this rent. Therefore workers with highly specialised roles, or those in more senior positions, are most likely to benefit.
Recent research for the U.S. finance industry (Phillipon and Reshef (2012)) suggests that rents have accounted for 30-50% of the wage differential between financial and rest of the private sector since the 1990s. And, for the U.K., Bell and Van Reenen (2010) find that almost three-quarters of the wage gains for the top income decile of workers from 1998-2008 accrued to finance workers, despite only 12% of the top income decile actually working in finance. This suggests that finance workers are in a position to capture some of the TBTF surplus.
This may be compounded by a problem of perception: if shareholders underestimate the value of the subsidy, then they may assume that their high profits are due to the expertise of bank employees, and may thus be willing to pay more as a result. While numerous articles discuss the distortive effect of the subsidy on executive pay (e.g. Weinberg (2010)) there are currently none which provide an empirical quantitative analysis of the extent to which it actually occurs.
… and shareholders collect any residual value.
Shareholders collect any remaining value of the subsidy that is not captured by other stakeholders. There is certainly empirical evidence that TBTF banks enjoy higher stock prices: O’Hara and Shaw (1990) look at the stock price reaction of those US banks labelled as TBTF by the Comptroller of the Currency in 1984, and find that their share prices rose by 1.3% following the event.
This seems logical: if the subsidy lowers the cost of debt, then the profitability of the bank should increase due to lower future interest payments, without any increase in the underlying risk to shareholders. Moreover, shareholders know that if the bank is rescued by the government in the event of failure, then they will be spared the uncertainties of the bankruptcy process, which they assume to be more costly than the potential dilution of their equity stake in the event of a government bail-out.
This boost to shareholder value may come not only from the direct transfer value of the subsidy, but also because TBTF banks tend to take more risk. Without a guarantee, market discipline means that shareholders’ demand for higher risk-taking is usually countered by an increase in the cost of debt as the firm takes more risk. But, in the presence of an implicit guarantee, this market discipline mechanism is less effective as creditors are insulated from risk-taking. Equity has unlimited upside, while limited liability caps losses at 100%, so shareholders have incentives to take more risk. This means that TBTF shareholders can encourage management to undertake riskier activities or borrow more, without the disciplining effect of higher debt costs. And as risk-taking increases, the probability of a bailout rises, causing the size of the TBTF guarantee to grow further still (Alessandri and Haldane (2009)).
Conclusion
The available evidence implies that most of the implicit subsidy benefits senior bank employees and shareholders. The structure of credit markets suggests that not much of the benefit is passed on to creditors and customers. But further empirical investigation is needed in order to come to any definite conclusions.
Authors: Rhiannon Sowerbutts works in the Bank’s Global Interconnections and Spillovers Division. Peter Zimmerman works in the Bank’s Global Interconnections and Spillovers Division and is currently on study leave for a DPhil at Oxford University.
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.
If you want to get in touch, please email us at bankunderground@bankofengland.co.uk
