Hans-Werner Sinn, the CEO of CESifo, claims that the ECB is conducting a secret bailout strategy that is utterly dangerous for the stability of the eurozone and is doing harm to the German economy. Martin Wolf and Felix Salmon have accepted his thesis. However, it is plainly wrong.
Claiming that Hans-Werner Sinn, Martin Wolf and Felix Salmon are all wrong at the same time on the same issue is quite a bold statement.
Hans-Werner Sinn is one of the smartest and most renowned German economists. Financial Times columnist Martin Wolf is, according to Wikipedia, nothing less than “one of the world’s most influential writers on economics.” And Reuters journalist Felix Salmon runs one of the best blogs on economics and finance.
I think, however, that they have all got it wrong with regard to risks allegedly hidden deep inside the balance sheets of European central banks.
Hans-Werner Sinn has stirred this debate with some outlandish claims about a “stealth bailout” that allegedly is being conducted by the ECB and the Bundesbank. Sinn is smelling disaster:
“If markets sense the end of the line, the Eurozone may face a crisis like the one Britain faced in 1992.”
Martin Wolf bought Sinn’s argument and concludes that central banks in the Eurozone:
“have been financing their governments. Let us call a spade a spade: this is central bank finance of the state.”
Felix Salmon thinks that Sinn’s figures show that the insolvency of an individual country’s central bank“is a real possibility, now”.
Showing that all three got it wrong requires a close look at arcane details of the European interbank payment system and the money creation in the Eurozone.
Bear with me, please.
The Target2 payment system
At that heart of the debate is the interbank payment system for cross-border money movements in the EU. This mechanism is called “Target2”. The amounts of money moved through Target2 are almost inconceivable. On an average day, 2299 bn Euros change hands via the system. Within four business days, the amounts settled equal the annual GDP of the Euro area. 94 percent of the transactions are taking place between private banks. (See: “Target 2 Annual Report”, 2010, p.7)
Since the start of the financial crisis, the Bundesbank and several other central banks have accumulated huge claims against the ECB. The German claims rose from zero in 2007 to 177.7 billion Euros at the end of 2009 and 325.5 billion Euros at the end of 2010. The Bundesbank earns interest on these positions, the rate equals the interest for the main refinancing operations.
At the same time,Greece,Ireland,PortugalandSpainhave accumulated large Target2 liabilities. According to the Ifo Institute, at the end of 2010 they equalled 340 bn Euros. The biggest debtor was the Irish national bank (146 bn Euros), followed by the Greek national bank (87 bn Euros). Sinn makes a fuss about the fact that those numbers look similar to the current account deficits of those countries and states that the Target2 balances
“reflect Ireland’s past current account deficits with other Eurozone nations that have not been financed by inflows of private or public capital, but rather by the Irish Central Bank’s money creation.”
He notes that ”Spain took less and Ireland more than their respective current-account deficits” but does not bother to try explain why this might have been the case.
How do these claims and liabilities arise?
Let’s take the following example: Assume an Irish farmer wants to buy a German tractor. (This example was first used by my colleague Stefan Ruhkamp in the Frankfurter Allgemeine Zeitung and was later picked up by Hans-Werner Sinn.) Let’s assume that the farmer is a customer of Anglo Irish Bank and the tractor manufacturer is a customer of Germany’s Commerzbank.
The farmer transfers the money for the tractor electronically from his account at Anglo Irish to Commerzbank. However, the money does not change hands directly between both banks. It takes a detour via the Target2 payment system. Technically, Anglo Irish transfers the money to the Central Bank of Ireland. These guys hand it over to the ECB which channels the money through to the German central bank. The Bundesbank eventually transfers the payment to Commerzbank.
This transaction affects the balance sheets of all parties involved. The Bundesbank has recently described this in its Monthly Bulletin (March 2011, p. 34):
“If, for example, foreign funds are transferred to a bank that participates in Target2 via the Bundesbank [in my example: Commerzbank], this results in a liability of the Bundesbank to this bank (such as in the form of a credit to this amount on the banks current account).
In return, the transaction generates a Bundesbank claim for the same sum against the sending central bank. This central bank then in turn debits the account of the originating commercial bank.”
It is important to have in mind that only a certain type of money can be transferred via Target2 – the so called central bank money. This is money which has been directly created by central banks. Economists also call it “monetary base”. Only two kinds of money constitute central bank money: physical cash (banknotes and coins) and deposits of commercial banks at the central bank. Neither the money on my private current account nor the money which is being created if I use my overdraft facility is central bank money. This money has been created by the private banks, not by the central bank. It turns into central bank money when I withdraw cash at a cashpoint.
Target2 imbalances arise if a central bank receives more payments than it sends. Here is the more technical explanation by the Bundesbank:
“The resulting claims and liabilities generated at the national central banks by the multiple transactions over the course of a day normally do not fully balance out. (…) [T]he outstanding claims and liabilities of all the national central banks (…) are transferred to the ECB at the end of the business day, where they are netted out.
The resulting Target2 (net) balances hence arise from the cross-border distribution of central bank money within the Eurosystem’s decentralised structure.”
The Target2 (im)balances
What is driving those large claims and liabilities? According to the Bundesbank (p. 35), the main reason is that private banks in Germany are hoarding central bank money because they are very risk averse at the moment:
“While funds tended to continue to flow into German banks from abroad due to non-bank payments and their own operations, after the onset of the crisis they were less willing, and in some cases unable, to lend these funds to foreign institutions on the interbank market. Instead, they gradually curtailed their refinancing operations with the Bundesbank.”
In early 2007 German banks borrowed 250 bn Euros from the Bundesbank. By the end of 2010 this had fallen to 103 bn.
In contrast, prior to the crisis private banks in Germany lent significant sums abroad. According to the Bundesbank (p. 35), in those days:
“credit institutions’ (short-term) net external position in particular acted as a kind of “offsetting item” in the balance of payments. (…) [M]ost years saw outflows of funds (net capital exports) in banks’ short-term credit business. Hence temporary Target2 positions were quickly reduced by private capital flows.”
The Bundesbank does not mention another important source for the Target2 imbalances: There’s a silent bank run on deposits going on in the Eurozone periphery. People withdraw their money from bank accounts in the crisis countries and transfer it to banks in countries they regard as more stable. In 2010 this problem was particularly severe inIreland. According to the “Irish Independent”, depositors withdrew 110 bn Euros from Irish banks last year. The paper reported in February:
“In November the group of 15 banks lost €26.7 bn and by December the monthly rate of deposit loss soared to €40.3 bn.
Depositors from outside Ireland are withdrawing their cash at the fastest rate, pulling out more than €35 bn in December and €91 bn in the full-year. Non-Irish deposits include cash from multinational companies doing business here and cash held by investment funds and pension funds.
The depositors have little motivation to keep their funds inIreland, and many have been spooked into withdrawing money as the credit ratings of Irish banks deteriorated.”
The same thing is happening inGreece. According to the statistics of the Greek central bank, the deposits of households and non-financial corporations fell by 16.1 bn Euros or 17% between January 2010 and March 2011.
If the money is transferred to banks in other countries of the Euro area, these payments drive up the Target2 imbalances as well. One indication that this is really happening is the fact that tiny Luxembourg, which is an important financial hub in Europe, has the second biggest Target2 claims (68 bn Euros). (From this perspective, Sinn’s claim that the Target2 balances “made it possible for the GIPS to continue living beyond their means” is deeply questionable.)
2) The interpretation of the facts
What does all this mean economically? Are those Target2 imbalances dangerous?
“No,” says the German Bundesbank.
“Yes,” say Hans-Werner Sinn, Martin Wolf and Felix Salmon.
“there is no immediate change in the level of risk to the Bundesbank due to the rise in its Target2 settlement accounts. This risk is not directly related to the Target2 positions and arises from the risks associated with the Eurosystem’s liquidity supply.”
Losses, however, are possible, the Bundesbank concedes:
“An actual loss will be incurred only if and when a Eurosystem counterparty defaults and the collateral it posted does not realise the full value of the collateralised reﬁnancing operations despite the risk control measures applied by the Eurosystem.”
Understanding this sentence is rather difficult. I think it refers to the failure of a commercial bank like Anglo Irish or Commerzbank, not a sovereign government. Remember that private banks and not governments are using Target2, hence “Eurosystem counterparties” cannot refer to governments. For getting central bank money, private banks have to hand over collateral like government bonds. If a private bank goes bust and their collateral turns out to be partly worthless, the central bank would suffer a loss.
Of course, the way the Bundesbank describes possible losses is a bit cheeky. If things go wrong, the most likely scenario is that things will work the other way round. Let’s assume thatGreeceeventually defaults on its debt. This might push private banks which own a lot of Greek government bonds into the abyss. If those banks have borrowed from the central bank and handed over Greek bonds as a collateral, this might make some Target2 claims worthless.
However, the central banks in the Euro area would share those losses, as the Bundesbank stresses:
“Any actual loss would always be borne by the Eurosystem as a whole, regardless of which national bank records it. The cost of such a loss would be shared among the national banks in line with the capital key.”
Does the central bank finance public debt via Target2?
Hans-Werner Sinn claims that the Target 2 balances
“come close to short-term Eurobonds”
Martin Wolf also subscribes to this view:
“because national central banks have lent against discounted public debt, they have been financing their governments. Let us call a spade a spade: this is central bank finance of the state.”
Felix Salmon sees the Target2 numbers as a
“finger in the eye of Article 123 the Lisbon Treaty, which bans monetary financing.”
Sorry, but I don’t understand what these guys are talking about.
Let us remind ourselves: The Target2 claims are neither loans to private banks nor loans to governments. They are positions on the central bank’s balance sheet which are related to the payments between private entities within the monetary union. Target2 claims existed well before the financial crisis but – at least according to the Bundesbank – were “quickly reduced by private capital flows.”
I’m even more puzzled by Martin Wolf’s argument that the central banks have been financing their government. He comes to this conclusion because central banks have lent against public debt. As far as I understand monetary policy, this is true for any open market operation of a modern central bank. Wikipedia offers a good explanation:
“The primary tool of monetary policy is open market operations. This entails managing the quantity of money in circulation through the buying and selling of various financial assets, such as treasury bills, government bonds, or foreign currencies. Purchases of these assets result in currency entering market circulation (while sales of these assets remove money from circulation).”
This happened in the Eurozone before the crisis, is happening right now and will happen in the future. However, it’s called money creation, not financing of government debt. The ECB does not buy bonds straight from the government, and the limited purchase of Greek bonds on the secondary market by the ECB that is happening does not affect the Target2 balances.
Do Target2 imbalances distort capital flows in the monetary union?
According to Hans-Werner Sinn, the Target2 imbalances are highly dangerous. Interestingly, he does not stress the risk of losses. He claims that those imbalances distort capital movements in the Euro area. In a nutshell, Sinn argues that the credit supply for the German economy is crowded out.
At this stage, I think Sinn’s way of reasoning is getting utterly bizarre.
Sinn takes the Irish farmer who buys a German tractor as an example. Sinn assumes that the farmer asks his bank for a loan:
“Unable to borrow from other European banks, or only at high premiums, his bank turns to the Irish Central Bank, which “prints” and lends out fresh Euros for the purpose. This raises the Irish Central Bank’s assets and liabilities. The farmer then transfers these Euros through the central bank system to the German producer. This means that the Irish central bank’s money base (its liabilities) shrink back to normal but the Bundesbank’s money base (liabilities), in the first instance, increase by this same amount.”
Until this point, his way of reasoning is completely in line with the description by the Bundesbank. However, Sinn introduces a mind boggling twist. He writes:
“[T]he new money coming into the German economy as a result of the payment for the tractor is likely to crowd out normal German money creation by way of the Bundesbank’s lending to German banks.”
This is a really weird point and completely at odds with everything I’ve learned about money creation.
Let us remind ourselves: The German banks currently shy away from risky loans and do not want to get more central bank money from the ECB. If the bankers changed their mind, they could easily get more central bank money in the next tender operation. In fact, getting more central bank money is much easier than before the crisis.
The ECB has expanded its reﬁnancing operations and expanded the collateral framework. The main refinancing operations are conducted “as fixed rate tender procedures with full allotment.” This basically means that the banks can get as much central bank money as they want, provided they hand over enough collateral and pay the interest rate. In March 2011 the ECB reiterated that this will be done “as long as necessary.”
Here, Sinn’s way of reasoning becomes quite fuzzy:
„The crowding out will not necessarily occur, but it is the normal case to be expected as, given Germany’s GDP and given Germany’s payment habits, the commercial banks only need a certain amount of Euros for circulation in Germany.”
The next sentence is even weirder:
“Moreover, strict crowding out is inevitable if the ECB controls the overall stock of central bank money in the Eurozone by way of sterilising interventions or auctioning off limited tenders.”
This statement comes close to saying:
“Today is a sunny day. But if it were raining, it would not be nearly as beautiful outside.”
The ECB does not control the overall stock of central bank money in the Eurozone. The Bundesbank tried this in the 70s and fortunately abandoned this strategy a long time ago.
Nowadays, the ECB’s task is
“to influence the level of short-term interest rates to ensure that price stability is maintained over the medium term.”
The central banker also look at monetary aggregates like M3 and assess “their implications for future inflation and economic growth.” However, they do not have a specific target for either M1 or M3. And Hans-Werner Sinn should know that.
Additionally, as mentioned above, the ECB is not auctioning off limited tenders.
Sinn carries on regardless:
“ultimately, the credit to the Irish farmer comes from the Bundesbank at the expense of a similar credit provided to the German economy.”
This is utter nonsense.
A slight variation of the thought experiment involving the Irish farmer and the German tractor illustrates this.
Imagine the farmer decides not to trust the banks any more. Instead of making an electronic payment, he puts the cash into suitcase, takes a plane toGermanyand pays the manufacturer of the tractor in cash. The manufacturer pays the Euro notes into his bank account and his bank ends up with new central bank money.
Does this operation limit the ability of the private bank to lend more money to the German economy?
Does this have influence the next tender operation of the ECB?
And, as an additional question to Wolf and Salmon: Does this by any means involve any financing of public debt by any central bank?
I absolutely think not.
Does this situation resemble 1992?
Sinn claims that the current situation is as dangerous as the situation in 1992:
“That was when the British pound collapsed because the Bank of England had fewer deutschmarks and francs to sell than George Soros was buying.”
I don’t see the parallel at all. Nowadays, we are talking about one currency, not two or three. I cannot imagine how anybody could technically bet against the ECB or individual central banks of the Eurozone with regard to the Target2 balances in a way Soros did in 1992 against the Bank of England. In fact, even if this were possible I doubt it would be very smart thing to do. The fundamental difference to 1992 is that in those days, the Bank of England could not print deutschmarks or francs. The ECB, however, can print as many Euros as it wants.
Does it economically matter in which country Euros were created?
Sinn, however, is convinced that there has to be a limit and the whole procedure cannot go on for ever:
“If every year a further €100 billion is granted to the GIPS as Target loans, the stock of credit given by non-GIPS central banks to their commercial banks via refinancing operations will shrink by the same amount.”
So what? I personally do not care if the Euro notes in my wallet are printed by the Bundesbank or the national bank ofGreece.
A Euro is a Euro is a Euro, or isn’t it?
What am I missing?
In a German version of his thoughts, Sinn himself concedes that after the tractor purchase neither the monetary base in Ireland nor in Germany has changed.
Does it economically matter if the central bank money which is used in Germany has been created in Germany? Is central bank money “made in Greece” of inferior quality?
I don’t think so.
Hans-Werner Sinn had a scoop when he observed that there are significant imbalances in the Target2 system. Those imbalances impressively show that the banking crisis inEuropeis far from over and that private savers, as well as banks, deeply distrust banks in the periphery.
However, Sinn got carried away by his observation. The Target2 imbalances are the consequence of the financial crisis, not the result of a deliberate bailout policy of the ECB.
I think there is convincing evidence that his conclusions are plainly wrong and should not be taken seriously.
If you happen to like this blog, why don’t you like me on Facebook as well?