Much is being made at the moment of the idea that banks should have more capital. Predictably, there is huge confusion about what this actually means, and the usual suspects are once again mixing up deposits and capital (deposits are debt) and claiming that QE recapitalises banks (no it doesn't, but it does provide them with liquidity). I don't want to explain the difference again here, but if anyone is still unclear about what "capital" consists of for a bank, read this.
Predictably, banks and other financial institutions are fighting back. Concerns are being expressed about the effect on competition of EU's proposal for money market funds (MMFs) to have capital and liquidity reserves. And banks worried about their return on equity (already shot to pieces) claim that raising more capital would be a) unacceptable to their shareholders b) hugely expensive c) impossible anyway. Meanwhile, Anat Admati and Martin Hellwig, in their book "The Bankers' New Clothes", claim that the banks' arguments are specious: banks in the past have been much more highly capitalised, the Modigliani-Miller model shows that (apart from tax considerations) equity is no more expensive than debt, capital can always be raised if the price is right. This is yet another argument that could run for years and become increasingly political. Personally, I'm not going to take sides. I think they're all missing the point.
You see, there is actually no significant difference between debt and equity. Both are claims on the bank's income (when it is a going concern) or its assets (when it is bust). Nor is debt in any way money that the bank is "looking after": as far as the bank is concerned, debt is funding for things it wants to do (lending, trading...). And debt includes customer deposits.
Customers believe that when they put money in a bank deposit or current (checking) accounts they are putting it in a safe place. But that's not true. Customers are actually lending that money to the bank, which can use that money in whatever way it wishes. And as Andrew Lilico points out, they have no automatic right to return of that money. All they have is a claim. In the event of insolvency, that claim will only be honoured if there are sufficient assets to meet it after settling more senior claims. When a bank fails, the only real difference between a depositor, a bondholder and a shareholder is the seniority of their claims.
Let me explain. A typical bank liability structure looks something like this*:
Here it is reversed, for reasons that will shortly become clear:
I've added some percentages to these to indicate proportions of each type of asset in the capital structure. These may or may not be remotely realistic - that's not the point. The point is to show how losses due to asset writedowns affect corporate liabilities. Let's imagine to start with that the bank has to write down 3% of its assets (to keep it simple I am using the nominal not risk weighted balance sheet):
You can see that, for a bank that meets current UK regulatory requirements for a leverage ratio of 3%, this wipes out the bank's shareholders. They lose their entire investment.
If losses increase, this is who gets hit next:
Junior bondholders (holders of various forms of subordinated debt, including the so-called "Co-Cos" about which there has been much discussion) are bailed in and wiped. Their bonds are converted to equity and they no longer receive interest payments. They become ordinary shareholders, waiting patiently for the bank to be restored to health so their shares can recover their lost value and they can start to receive dividends again. However, sometimes governments interfere with this. When the Dutch bank SNS Reaal was nationalised recently, junior bondholders were wiped along with ordinary shareholders. Shares and subordinated debt were cancelled completely via an expropriation order by the Dutch government. In effect, the Government took over the claims of shareholders and junior bondholders against the rescued bank. There was a fair degree of outrage among investors about this, but it is worth bearing in mind that had the bank been allowed to fail, shareholders and junior bondholders would have lost their entire investment anyway. The Dutch government chose to rescue depositors and protect the financial system. They were under no obligation to make good shareholders and subordinated debt holders.
But if losses rise further, there is a much more complex situation:
Depositors, senior bondholders and wholesale lenders are currently ranked equally ("pari passu") in their claim for settlement. However, debt that is secured directly on assets effectively ranks senior to unsecured debt, because it has a prior claim on certain assets. So if the bank has issued asset-backed securities, or borrowed money via the repo market using its assets as collateral, those creditors effectively rank senior to depositors: the assets backing their debt are not available to settle other claims (they are "encumbered"). If those assets turn out to be worthless, the debt is effectively unsecured and its holders have an equal claim to a share of unencumbered assets.
The UK's Independent Commission on Banking recommended that depositors should rank senior to bondholders ("depositor preference"), and the EU is considering legislation to enforce this from 2018. But currently, once junior bondholders have been bailed in, both unsecured senior bond holders and depositors are fair game. This means ALL depositors - not just large ones. There is no intrinsic difference in claim seniority between large and small depositors. Once depositor preference is established, though, senior bondholders would take losses before depositors.
The most senior of all claims are the official sector - central bank funding and other public sector loans. Only if a bank was so deeply in trouble that all deposits were wiped would the official sector take losses. I suspect that government would step in long before that point was reached.
The Brown-Vitter proposed legislation in the US would force banks to increase their Tier1 capital ratio (unweighted) to 15%. Anat Admati wants it to be 25%. That would mean that bank asset writedowns would have to be far greater before depositors were at risk. But it wouldn't eliminate the risk completely. And it would place other people - or potentially the same people, in a different way - at risk.
Capital is not simply "money that absorbs losses". It is people's savings. The principal shareholders in banks are pension funds - which invest the money that people save for their retirements. Losses for these people are just as serious as losses for depositors: although the consequences may not be felt immediately, money lost through investment failure may mean a materially lower standard of living in retirement, as Equitable Life pensioners could tell you. If we force banks - and potentially corporations too, since highly-geared corporations are a risk to their stakeholders - to finance themselves much more with equity than debt, losses will still fall on ordinary people. It's the same money, just in a different form. We should not forget this.
My walk through capital structure above I hope showed that when a company is failing, it does not matter whether you call your investment equity or debt - what matters is the seniority of your claim. Losses are the same whether it has 15% equity or 3% equity. Increasing the proportion of equity does not make it less likely to fail. All it does is increase the likelihood that creditors will get their money back.
The distinction between debt and equity made by, among others, Anat Admati is misleading. In insolvency, debt and equity are fungible. It does not matter whether the debt is called "subordinated", or secured on assets: if losses due to asset writedowns are sufficiently large and widespread, all debt is effectively converted to equity. Creditors have no more intrinsic right to return of their money than shareholders. They get paid first, but they don't necessarily get paid completely. Nor should they be. Lending money to anyone is risky. Banks are no exception. As long as it is clearly understood that creditors can, and should, lose money in insolvency, there is no reason for vast increases in equity in bank capital structures, because debt is effectively equity anyway.
I know that everyone is now going to shout - "BUT WHAT ABOUT DEPOSIT INSURANCE"? The reason for protecting small depositors with deposit insurance is a social one. It has nothing to do with preventing bank runs, really - the best way of preventing a bank run is to ensure people know that banks can't run out of money (liquidity), which means a central bank doing a good job as Lender of Last Resort. Bank depositors are assumed to be naive people who don't understand finance, so need protection from risky banks. I don't think this is acceptable, really. These people are only too happy to put their money in banks when they could put it elsewhere. It needs to be made clear to them that their money is no safer in a bank than it would be in an investment fund - and then let them make an INFORMED choice about what to do with their money. There are advantages to bank deposits even if they aren't fully safe. There is in my view a case for protecting transaction accounts from losses, because we have become so dependent on banks for payments that allowing current accounts to take losses would cause real hardship to many people. But these are political considerations, and I am only expressing a personal view. They have nothing to do with the nature of deposits. All bank deposits are investments. "The value of investments can fall as well as rise, and the return of the investment is not guaranteed." Why isn't this statement on every bank deposit account agreement?
I am certainly not suggesting that banks increasing the proportion of equity in their capital structure is a bad thing. On the contrary, there is plenty of evidence that heavy reliance on debt finance can be destabilising not just for banks but for corporations, too, because of the cost of debt service and the risk that creditors will foreclose. Personally I would eliminate the preferential tax treatment of debt, which encourages debt financing at the expense of equity. But we should not buy into the idea that increasing the proportion of equity in the capital structure makes banks "safer". It doesn't. The only thing that really makes banks safer is limiting the risks they can take and ensuring they are well managed. When, please, are we going to regulate lending properly?
Liquidity matters - Coppola Comment
Cyprus and the financing of banks - Coppola Comment
Tarullo's speech on capital and regulation - FT Alphaville
Brussels to clamp down on shadow banking - Financial Times (paywall)
Equity capital requirements - The Economist
Rant at me about property rights - Andrew Lilico
The Bankers' New Clothes - Anat Admati & Martin Hellwig
Independent Commission on Banking Final Recommendations (summary)- KPMG
EU Assembly seeks depositor preference in bail-in law - Bloomberg
Brown-Vitter bill analysis & commentary - Davis Polk
State of the Netherlands nationalises SNS Reaal - NL Government
Theory & practice of corporate capital structure - Deutsche Bank
* Yes, you are right, this does look very much like a CDO tranche structure. That's because it is. A CDO is a financial company and its tranche structure is a corporate capital structure.