Recapitalising the banks through enhanced credit support: quasi-fiscal shenanigans in Frankfurt

Last week the Eurosystem performed a €442bn injection of one-year liquidity into the Euro Area banking system.  They did this at the official policy rate – the Main refinancing operations (fixed rate) – of 1.00 percent, against the usual collateral accepted for Longer Term Financing Operations, effectively anything euro-denominated, not based on derivatives and rated at least BBB-.  It was a fixed-rate tender, that is, the ECB was willing to meet any demand at the 1 percent interest rate, as long as eligible collateral was offered; 1121 banks participated in the operation.

You will not be surprised to hear that this was the largest one-day ECB/Eurosystem operation ever.  Even more remarkable than its scale are the terms on which the one-year funds were made available.  There can be no doubt that this operation represents both a subsidy and a gift from the Eurosystem to the banks that participated in the operation.  I hope to clarify the distinction between a subsidy and a gift in what follows.

First, it is clear that a 1.00  percent interest rate for collateralised borrowing at a one-year maturity is well below the cost at which the participating banks could have funded themselves at a one-year maturity.  Twelve-month Euribor averaged 1.64% in May.  This is, of course, an unsecured interbank rate.  Secured rates would be lower.  On the other hand, Euribor is a ‘cheap talk rate’, based on what up to 43 panel banks believe a typical prime bank can borrow at from another prime bank.  Citing the Euribor website, “A representative panel of banks provide daily quotes of the rate, rounded to two decimal places, that each panel bank believes one prime bank is quoting to another prime bank for interbank term deposits within the euro zone.” A prime bank is a bank that is not yet insolvent.

Although the Euribor procedure is less likely to lead to an understatement of the true cost of unsecured borrowing than Libor, which, according to the Libor website “is calculated each day by asking a panel of major banks what it would cost them to borrow funds for various periods of time and in various currencies, and then creating an average of the individual bank’s figures.”, neither measure is a substitute for the only meaningful measure-  one based on borrowing rates charged and paid on actual loans or deposits rather than on a panel bank’s guess at the rates at which that bank or some other bank might be able to borrow.  Why any contracts are based on ‘cheap talk’ rates like Libor and Euribor is a mystery to me, but that is a subject for another post.

Other measures of private source borrowing costs for banks include deposits.  Interest rates on new household deposits up to 1 year maturity were 2.24%, new deposits from non-financial corporations with agreed maturity up to 1 year averaged 1.39% in March 2009.  Household deposits up to the deposit insurance limit are ‘safe rates’, guaranteed by the government.  The rate on German government bonds at a one-year maturity was around 0.80% at the time of the Eurosystem’s mega-operation.

You may think that this implies that the cost to the banks of borrowing from the Eurosystem for a year – 1.00% – does not imply a subsidy, as the banks’ borrowing from the Eurosystem is secured against collateral. You would be right if the collateral consisted of German government bonds. My guess (I don’t have hard information) is that this was not the case, and that instead the borrowing banks stuffed the Eurosystem with the worst quality collateral they could put their hands on, subject to the constraint that a rating agency had rated it at least BBB-.  Given the well-established practice of Eurozone banks that are eligible counterparties of the Eurosystem in repos and at the discount window, to carefully structure collateral packages that just meet the letter of the ECB’s collateral eligibility requirements, I am happy that I am not responsible for vetting and verifying the credit risk present in the portfolio of that increasingly speculative, highly leveraged entity known as the Eurosystem.

What can the banks do with the €442 that they have borrowed at 1.00 % for up to a year from the Eurosystem?  They could invest it in secured loans to households, e.g. mortgages. New floating rate and up to one year fixed mortgages in the Eurozone paid 3.66% interest on average in March 2009.  Or they could put it into government debt.  At least as regards German one-year central government securities, there is no ‘money machine’  allowing one-year funds borrowed at 1.00% to be invested in a safe asset yielding more than 1.00 percent.

But if you are willing to bet that repo rates will not rise above above1.5% over the next year, and not above 2.5% over the next two years, you have something close to a money machine.  The Euro Area yield curves, based on AAA-rate Euro Area central government bonds showed spot rates at one year maturity of 0.93 percent, at two years maturity of 1.53 percent and at three years maturity of about 2.5% (eyeballing Chart 26 on page S45 in the ECB’s June 2009 Monthly Bulletin for this last figure).  The instantaneous forward rates at one-year maturity were 1.43% in May 2009 and 2.77% at two years maturity.

When is there a subsidy?

It is possible for the ECB/Eurosystem to provide Euro Area banks with funds at a rate well below the rate at which the banks could have funded themselves elsewhere, without this implying a subsidy from the ECB/Eurosystem to the banks.  There is a subsidy only if the rate charged by the ECB to the banks is less than the ECB’s risk-adjusted opportunity cost of funds.  Let i(b) be the banks’ risk-adjusted cost of borrowing, i(l) the banks’ risk-adjusted lending rate, i(ecb)the Eurosystem’s lending rate to the banks and ithe Eurosystem’s risk-adjusted opportunity cost of funds.  Then the subsidy provided by the ECB per € lent out is i(ecb)-i .  The joy of the borrowing banks is measured by i(b) – i or by i(l) – i.  But i(b)-i- both risk-adjusted rates, is not a subsidy from the Eurosystem to the banks.  It is financial manna-from-heaven, reflecting the superior risk-sharing capacities of central bank and the sovereigns behind it (one hopes).

Assume that, in May, the ECB’s opportunity cost of funds (other than through base money issuance) at a one-year maturity equals the average Eurozone AAA-rate sovereign borrowing rate – 0.93 percent.  If the banks’ collateral is safe, there is no subsidy but a slight tax, 0.93 – 1.00= -0.07  or seven cents per € borrowed.

If instead the collateral offered by the banks is without value, their secured borrowing is equivalent to unsecured borrowing.  The one-year Euribor rate, 1.64%, provides a lower bound on the true unsecured borrowing rate of the banks.  I believe it is safe to assume that most of the collateral offered to the ECB in this operation was rubbish.  The supply of one-year funds was open-ended (demand-determined) and it is plausible to assume that the banks did not demand more than €442bn because they ran out of collateral and exhausted their capacity to transform pig’s ear securities into silk purse collateralisable assets.

The risk-adjusted rate of return to the Eurosystem on its lending to the banks can hardly be more than 0.70%, given the poor quality of the collateral offered and the dreadful state of the balance sheets of many Euro Area banks.  In that case there is a subsidy from the ECB to the banks of just over 0.25 percent, say € 1 bn.  While this is a small number, on the gargantuan scale on which bank losses and bailouts are measured these days, it is clearly inappropriate for the central bank to engage in quasi-fiscal operations of this nature.  Subsidies should be voted by the appropriate parliaments, not distributed by unelected technocrats.

The total increase in profits to the Euro Area banks from this operation is a multiple of the subsidy, and can be measured by the difference between the safe lending rate of the banks and the rate charged by the ECB.  Depending on which use of funds you consider, this could amount to 1.5% of the €442 bn (if the money is invested in 3-year government instruments and nothing too nasty happens to short rates over the next 3 years) or 2.5% (minus a discount for default risk) one-year housing loans, that is, around €6bn or €10bn. While most of that is not a subsidy, it is a gift from the Eurosystem to the banks.  If the ECB wants to play Santa Claus, I know of more deserving recipients of their largesse than the banks.

Interest subsidies and gifts are a slow, inefficient and inequitable way to recapitalise the banks.  Japan pursued this strategy and created zombie banks that brought the country a lost decade.  The right way to recapitalise banks is to have a mandatory conversion of unsecured bank debt into equity.  If there is insufficient unsecured debt, the tax payer can come in as recapitalisor of last resort, but only in exchange for equity or some other claim on any future upside.

When the ECB’s enhanced credit support is mainly a slow and inefficient mechanism for recapitalising the banks – the ECB recently estimated short-term capital needs in the banking system of the Euro Area at about €280bn –  without giving the taxpayers and other citizens of the Eurozone a claim on the banks in exchange, it turns the ECB into an agent of the banks (or more precisely of those in control of the banks and of the banks’ unsecured creditors) rather than of the 340 mn citizens of the Euro Area.  The ECB should avoid such capture by narrow sectional interests and opt to act in the public interest instead.

Maverecon: Willem Buiter

Willem Buiter's blog ran until December 2009. This blog is no longer active but it remains open as an archive.

Professor of European Political Economy, London School of Economics and Political Science; former chief economist of the EBRD, former external member of the MPC; adviser to international organisations, governments, central banks and private financial institutions.

Willem Buiter's website

Maverecon: a guide

Comment: To comment, please register with, which you can do for free here. Please also read our comments policy here.
Contact: You can write to Willem by using the email addresses shown on his website.
Time: UK time is shown on posts.
Follow: Links to the blog's Twitter and RSS feeds are at the top of the page. You can also read Maverecon on your mobile device, by going to