Thursday, 4 April 2013

Cyprus and the financing of banks

The final deal agreed to restructure the Bank of Cyprus involves the bail-in of senior bond holders and large depositors (over 100,000 Euros). What this means is that in return for the seizure of some of their money, bond holders and depositors will be provided with shares in the resolved banks. They will become part-owners of the bank.

From the point of view of small depositors, this looks rather good. It ensures that their deposits are protected not only now, but in the future. I pointed out in a previous post that small deposits are only protected to the extent that their sovereign can afford them (or to the limit of the amount that can be raised from other banks): but if large deposits and senior bonds can be bailed-in, deposit insurance should always  be affordable - shouldn't it? That is, of course, assuming there are any. I shall return to that shortly.

The Bank of England's Financial Policy Committee has recommended recently that banks should have more capital. Contrary to popular opinion, capital is not "cash" - it is shareholders' funds. It is the difference between the bank's debt and its assets. And borrowing - of any kind, including from the central bank - is NOT "capital", except under certain circumstances that I shall explain. When banks have insufficient capital, they usually have three ways of creating more:

- they can issue more shares. In the first instance the share offer would normally go to existing shareholders - what we call a "rights issue".

Bank shareholders are now somewhat wary of rights issues. RBS famously did a rights issue in April 2008 to shore up its damaged balance sheet after the ABN AMRO acquisition. It failed less than 6 months later, and shareholders are now proceeding with legal action against RBS. Rights issues by damaged and undercapitalised banks would not be popular. And rights issues by partly-nationalised banks are a demand for more taxpayer money.

- they can retain earnings. This means that instead of dishing out profits to shareholders in the form of dividends or employees in the form of bonuses, they keep them. But that of course assumes that they are making profits at all. Not all banks are. And banks that are under pressure to recapitalise by retaining earnings don't necessarily give customers a good deal. They need to maintain a wide spread between interest earned on lending and interest paid on deposits, so borrowers are charged a lot and savers are paid little. We have seen widening credit spreads ever since the financial crisis.

- they can sell off or wind down portfolios of risky assets which need more capital. Or they can shrink their balance sheets, reducing both assets and debt, by selling off bits of themselves - for example, RBS's sale of the Direct Line insurance company.

This requires some explanation. Bank capital requirements as set by regulators currently use what we call "risk weighted assets", which are a way of assessing the relative risk of different types of asset. Without going into details, the effect of risk weighting is to require banks to have more shareholders' funds if their balance sheets are riskier. But the calculation of risk weightings is something of a black art, so there are numerous opportunities for banks to game the weightings and pass assets off as less risky than they actually are. Additionally, assets become riskier under certain circumstances: for example, prime mortgages become sub-prime as house values fall and interest rates rise. A large fall in house prices resulting in a lot of home owners having negative equity in their properties can render mortgage lenders insolvent due to the increased capital requirements for what are then in effect unsecured loans.

So clearly, if banks are short of capital, it is in their interests to reduce riskier lending. Unfortunately this tends to be the lending that governments like them to do - loans to small businesses, loans to first-time house buyers. Hence Mervyn King's directive that banks should increase capital "in ways that don't harm lending". I am struggling to see how this is possible. In the short-term, lending must either be more expensive (because of the need to retain earnings) or scarcer - or both. In the long-term, of course, increased capital enables banks to lend more to higher risks without putting depositors' money at risk. So a drive to recapitalise banks involves short-term pain for long-term gain.

Except in Cyprus. And possibly in the rest of the Eurozone too. Because the deal that has now been struck turns senior bonds and deposits over 100,000 Euros into contingent capital. Contingent capital is debt (bonds and deposits) that can be converted into equity (shareholders' funds). The UK's ICB has recommended that banks should, in addition to equity, hold a substantial amount of contingent convertible debt to provide additional protection to depositors in the event of a bank failure. There has been much discussion about bonuses being paid in the form of "CoCos" (contingent convertible bonds), and in Europe there have been some ideas about changing the legal status of senior bonds so that they can be bailed-in (converted to equity) if necessary. But no-one has ever suggested that deposits could be bailed in - until now. Suddenly the game has changed.

If this deal is used as a model for bank resolution in other countries as well, it means that Eurozone banks have suddenly acquired the means of recapitalisation in distress. But it carries much larger implications than that.

Firstly, it means that the only real deposits in banks are insured ones. All other placings are contingent shareholdings. That includes the liquid assets of businesses - used to pay suppliers and employees - and large amounts of funds in transit, for example during house purchase. People and businesses will have to think carefully about how they move large amounts of funds around in future, if they can be seized without warning and converted to illiquid equity.

Secondly, it means that the only real creditor of banks is the government. Deposit insurance is paid for by banks but is a government directive, and as we saw in the financial crisis, when the funding for deposit insurance is insufficient, governments step in to top it up with taxpayers' funds. In Cyprus, the top-up has been provided not by taxpayers but by bailing-in large depositors. A preference order has indeed been established for the resolution of banks, but it's not what might be expected: the order is shareholders, contingent capital holders (junior AND SENIOR debt holders and large depositors), government. Insured depositors are effectively creditors of government, not banks. But insured depositors are still at risk if there is insufficient contingent capital to bail out government AND government is so highly indebted that it cannot borrow more to make small depositors whole. They are last in the preference order, but that doesn't mean they are safe from a government default.

Could this happen? Well, yes, I think it could. Bailing-in senior debt and large deposits this time came out of the blue, giving little time for them to escape - although figures show that a large amount of money was withdrawn from Cyprus banks prior to the deal. But now the precedent has been set, and Dijsselboem made his incautious remarks about this possibly being a model for future bailouts (he was silenced of course, but the damage had been done), who is going to place large deposits in Eurozone banks, except at much higher interest rates? And who is going to buy senior bonds, except at a deep discount? By bailing-in large deposits and senior bonds in this manner - contrary to existing terms & conditions, and with very little warning - the Eurozone leadership may have ensured that banks end up more thinly capitalised, and with far higher funding costs. This places small depositors at greater risk - especially in countries where sovereigns are already highly indebted and suffering from reduced tax take due to falling GDP.

It also carries serious consequences for lending. In the Eurozone periphery, interest rates to households and businesses there are already much higher than they are in the core, and lending is scarcer. This can only get worse as banks are forced to raise interest rates to attract large deposits and buyers for senior bonds. And these countries are already in recession: investment is falling and GDP collapsing. If the credit crunch intensifies due to rising interest rates to borrowers, it will make an already bad situation much worse.

Even in Cyprus, the bailout does not look like a good deal: small depositors may have been rescued, but they will pay with their jobs and their livelihoods. But the economic collapse there was going to happen anyway.  What is much more worrying is the effect on the Eurozone periphery. If large depositors and senior bondholders take fright because of this deal - which seems highly likely - the economic effects could be disastrous.

Related links

Sham guarantee - Coppola Comment
A failure of compassion - Coppola Comment
Dijsselboem, do remember that careless talk costs lives.... - FT Alphaville
and the rest of the excellent (and now very extensive) FT Alphaville "A Cypriot Precedent" series
Still crunching - The Economist


  1. It's important to note that this only applies to deposits in banks in countries that don't control the central bank issuing the currency of the deposit, such as Euro deposits in Eurozone countries. Sterling deposits in UK banks are much safer from default (if not from inflation and devaluation), because the UK government can print as much extra currency as is needed to bail them out. Euro and dollar deposits in UK banks are obviously riskier.

    1. Good point. Though the Fed's dollar swap lines do reduce the risk for dollar deposits, and as UK government debt is acceptable collateral at the ECB, Euros can be obtained if necessary.

  2. Haven't the effects of the pre-Cyprus Eurozone policies already been disastrous for the periphery ? The new template for bank resolution will turn whole EZ into a catastrophe.

  3. Frances,

    You describe the effect of exposing bondholders and depositors to risk as “disastrous”. I think that is going too far.

    Obviously exposing those bank creditors to more risk will induce some of them to flee to ultra safe forms of saving (e.g. National Savings and Investments in the U.K.). And that will reduce bank lending, which is deflationary, which would raise unemployment. But governments or central banks can perfectly well make up for that deflationary effect with stimulus, which would put employment levels back to where they would otherwise have been.

    In fact if stimulus took the form of fiscal policy plus QE, that would amount simply to government creating new money and spending it into the economy. The net result would be that everyone would have more money and thus WOULD NOT NEED to borrow so much.

    In the process, THE SHAPE of the economy would be changed somewhat: there’d be less lending based economic activity, but that doesn’t bother me. As for those poor bondholders, it’s a scandal that they should ever have been PROTECTED from risk. In any normal business that goes wrong, shareholders are wiped out first, followed by bondholders.

    1. I am, of course, talking about the Eurozone not the UK. I've used UK bank examples when talking about approaches to recapitalisation because a large part of my readership is British. But UK-based depositors are not involved in the Cyprus bail-in. Even those in Laiki bank, which have now been rescued. And as a commenter above notes, the UK has its own central bank which can of course rescue depositors if necessary. Deposits in countries that have control of their own currency are far less risky.

      I refer you to the total inaction of the ECB regarding monetary policy, the fact that Cyprus has no central bank of its own, and the fact that there is no EU "government" capable of taking any such stimulative action. There will be no supportive fiscal policy and no QE. Therefore the economic effects of deposit flight from risky banks in the periphery are indeed likely to be disastrous.

      You've also failed to notice that this post is about depositors, not bondholders. I might agree with you that large depositors should not expect to be bailed out. But the price they will demand for the removal of the implicit guarantee will be higher interest rates. Those will inevitably be passed on to borrowers.

  4. If the primary aim was to simply rescue the payments and clearing system from collapse(and to hell with the economy) could this policy be described in any way successful or will it be totally self defeating? Was there a better way to achieve that aim?

    1. I don't think this had anything to do with rescuing the payments and clearing system. It was all about breaking a country that was perceived by the German electorate as living off the proceeds of crime. Given that, there was never going to be a solution that didn't involve trashing the Cypriot economy. Hence my comment to that effect in the post.

    2. Thank you Frances - just trying to understand what it is all about.

    3. The UK, or Russian, or Chinese, or US electorates could all have saved tiny Cyprus and yet didn't either, why single out the Germans? The vendetta idea is silly, I don't think the German electorate cares about what Cyprus (or anyone else) does as long as they don't have to pay for failures there too often and too obviously.

    4. Why would the UK, or Russia, or China, or US bail out a Eurozone member whose banks have been torpedoed by the partial default of another Eurozone member?

    5. What about compassion? You want Germans to be more compassionate than you want your own group to be.

      Being part of not of the the eurozone as currently setup is not that relevant. The current eurozone setup gave member states the right to run their banking sector as they wished, very badly if that was what took their fancy, which was balanced by no central bail out funds. Seems fair enough, if less robust than a banking union with balanced regulation and a safety net, which we're getting to, but can't really be done in retrospect, indeed the Cyprus case is part of the building work in establishing banking union and the rules of play. In the meantime the Cypriot government got 10B, and will probably get more in debt restructuring or direct aid in the future. That it will be drip fed rather than paid as a lump sum is good, as people usually mismanage windfall lottery wins.

      There's a question of scale as well, the Germans don't have infinite funds either, and if they apply compassion fairly, they'll quickly bankrupt themselves if you scale pro-rata a generous Cyprus package to the other peripherals (who are in many ways more deserving).

    6. I totally disagree with you that being part of the eurozone is not relevant. It is the heart of the matter. Cyprus's problems, like those of the other periphery states, stem primarily from the poor construction of the Euro - particularly, in this case, the lack of a common banking system. And the proximate cause of their banking collapse was the EU's hamfisted handling of the Greek bailout, particularly the partial default in 2012. I don't really see why "my group" should pay to bail out a country which has got into difficulties because of its membership of a badly-constructed currency union which we have chosen not to join.

      The Germans have benefited considerably from their membership of the Euro. The rhetoric we have heard from German politicians in relation to the Cyprus mess is no doubt electorally popular (in an election year) but it is unpleasant and inaccurate, just as the earlier rhetoric about Greece was. Demonising other nations to justify a political stance that pleases your own supporters is jingoistic and demonstrates to my mind how weak the commitment of German politicians to the ideals of the European Union really are. In the end they are only interested in their own political fortunes.

      FWIW I know that Germany can't possibly bail everyone out and there is no alternative but to establish the principle of depositor contribution. It is the attitude of German politicians that I am criticising. I stand by my comment above about their intention to break Cyprus in order to gain electoral popularity in their own country.

  5. Surely canny holders of large deposits will downsize their cash holdings and buy into real, non-confiscatable assets? Presumable this will cause the prices of real assets to increase?

  6. "who is going to place large deposits in Eurozone banks"

    where else could they place their large deposits?

    - place euros in UK/Swiss/etc banks: that doesn't help as these sovereigns couldn't guarantee the deposits if the whole eurozone large deposit business moved to them, so would have to adopt the eurozone bail-in principles sooner or later. If not, UK/Swiss/etc taxpayers pay for eurozone banking, which would be very good for the eurozone.

    - convert deposits to government bonds: very good too, it makes the dependency much clearer, takes banks out of the equation (so they can't privatise the upside of the government guarantee anymore), and sinks interest rates governments borrow at.

    - invest it: good, private stimulus, growth, etc.

    - convert euros to G3 currencies for wealth store purposes: that's the best of both worlds, a stimulus effect via euro xrate collapse, and foreign taxpayers paying for eurozone banking.

    All very virtuous, isn't it?

    1. You talk about "the eurozone" as if it is a homogenous group. But it isn't. Any stimulatory effect from capital flight will benefit core countries at the expense of the periphery - as it has done throughout the Euro crisis. That would apply whatever form the capital flight takes. Stimulus in the core will raise inflationary pressure in eurozone metrics, encouraging the ECB to tighten monetary policy - which is the LAST thing the periphery needs. It's potentially disastrous.

    2. I agree there's clearly some (inconvenient) frictions but perhaps not as much as those who are analysing the thing in "country" terms think there is. It's actually pretty hard for the North alone to move the whole eurozone inflation needle to get the ECB to do bad things, because movable goods prices can't diverge materially, so only Northern immovable services can inflate and are to a large extent compensated by the deflation of Southern immovable services -- which contributes to put things back into balance so is also virtuous.

      That said of course it would be nice if the ECB/Eurogroup could find some way for monetary policy to flow more directionally to the South, so as to shorten and smooth out the pain, but in a dynamic system it will tend to get there anyway sooner or later I think.