Don’t worry about Target2
Moody’s just slashed Spain’s credit rating three notches — a clear sign that the bank bailout, even though it hasn’t happened yet, is being seen in the markets as decidedly deleterious for Spain’s creditworthiness. Spain’s 10-year bond yield is now 6.75%, up from less than 5% in early March, and approaching the levels at which market access shuts down entirely. Worries over the future of the euro are back — and, like clockwork, whenever those worries appear, people start talking about Target2.
Last week, George Soros warned about the “the Bundesbank’s claims against peripheral countries’ central banks within the Target2 clearing system”; today, in the NYT, Hans-Werner Sinn says that the Bundesbank is owed $874 billion in Target2 money by Europe’s periphery. “Should Greece, Ireland, Italy, Portugal and Spain go bankrupt and repay nothing, while the euro survives, Germany would lose $899 billion,” he writes.
Meanwhile, in a recent report, Jonathan Carmel, of Carmel Asset Management, publishes this chart, with the explanation that “the Bundesbank has replaced the exposure to peripheral debt that the German banks reduced”; he explains that “periphery debt is now the Federal Republic of Germany’s problem”.
These are big and scary numbers, and Sinn in particular is doing his best to scare as many people as possible with their magnitude. Last year, I blogged a Martin Wolf column about Sinn’s theses, and got a lot of pushback as a result. So here’s my attempt at a do-over: my attempt to explain that the chart above really just shows private risks in Germany going down. The line which matters is the blue one, not the green one.
It’s worth backing up here, a little bit. The eurosystem, as it’s known — the system of national central banks, plus the ECB — is highly federalized. The ECB itself sets interest rates, and has a modest balance sheet of its own, but the only banks it deals with are the national central banks. It’s the central banks, like the Bundesbank or the Banco de España, which perform all the liquidity operations, lend money to their commercial banks, and generally keep the euro functioning as a currency.
Every bank in the eurosystem has an account at its respective national central bank — and if you add up all the money in all those accounts, the total is the Target2 balance at the central bank in question. It’s worth mentioning, here, that there’s one thing pretty much everybody agrees on when it comes to Target2: so long as the euro zone stays together, there’s really no problem at all. All the Target2 balances at the various central banks always sum to zero, and the system works efficiently and well.
If a Spanish woman writes a check to her therapist, the money comes out of her account and goes into the therapist’s account. So long as both accounts are at Spanish banks, this is just a transfer from one bank to another, and the Target2 balance at the Banco de España is unchanged. But let’s say our Spanish depositor decides to move €1,000 from Banco Santander to an account at Deutsche Bank. In that case, the balance on her Santander account will go down by €1,000, and the Banco de España will likewise deduct €1,000 from Santander’s account at the central bank. In Germany, €1,000 appears in the Deutsche Bank account, and in the first instance Deutsche Bank will keep that money in its account at the Bundesbank, so the Bundesbank adds €1,000 to Deutsche Bank’s balance.
Essentially, the Banco de España just destroyed €1,000, and the Bundesbank just created €1,000. That’s fine — they’re central banks, and creating and destroying money is what central banks do. But for simple bookkeeping purposes, everything in the eurosystem has to balance. Remember that what we normally think of as assets, banks think of as liabilities. So Deutsche Bank owes €1,000 to our Spaniard — that’s what it means for her to have €1,000 on deposit at Deutsche Bank. In turn, Deutsche Bank has €1,000 on deposit at the Bundesbank, which is to say that the Bundesbank owes €1,000 to Deutsche Bank. And then the chain goes on: the ECB owes €1,000 to the Bundesbank, the Banco de España owes €1,000 to the ECB, and Santander owes €1,000 to the Banco de España, since Santander effectively needed to borrow money from the Banco de España in order to give that money to Deutsche Bank.
This being high finance, obligations to the national central banks here are collateralized, so the Banco de España is holding collateral from Santander which more than covers the €1,000 it’s owed. On the other hand, the Banco de España in turn is not asked to post collateral at the ECB. Central banks don’t do that sort of thing: there’s no need to post collateral when you can just print money whenever you need it.
In any event, it’s easy to see how the Bundesbank’s Target2 balance has been rising of late, as the balances in the periphery have been declining: there’s a flight-to-safety going on, and German banks are (rightly) perceived as being safer than banks in Spain, Greece, and other peripheral countries. Similarly, German banks which lent money to Spanish borrowers — and especially to Spanish banks — are not rolling over those loans. When the loans are repaid, the German banks just keep that money on deposit at the Bundesbank, rather than lending it out again to some country in serious difficulties. Once again, that increases the Target2 balance at the Bundesbank, and whenever that happens, there’s an equal and opposite decrease in the Target2 balance elsewhere.
Now to the naked eye, all of this looks like exactly what it is: money flowing to Germany. It’s people in the PIIGS countries either repaying the money they owe German banks, or moving their money so that it’s on deposit at a German bank. As such, it’s a bit weird that people like Sinn and Soros characterize this money as money which Germany has lent out to the periphery — at heart, the flows are in exactly the opposite direction. But because of the way that bookkeeping works, these flows create internal accounting obligations between the various eurosystem banks, and it’s those internal accounting obligations that Soros and Sinn are seemingly so worried about.
As Karl Whelan says, however, it’s far from clear that those internal accounting obligations are worrisome at all. The eurosystem as a whole is always in balance, and any money which is created in one corner of the euro zone is destroyed in another corner. The only way that these particular chickens could ever come home to roost would be if a country or countries left the euro entirely. And even then, it’s not obvious that the consequences would be particularly bad.
Certainly a Greek exit would be small enough not to worry about at all. Greece has a negative Target2 balance of about €100 billion. What that means is that Greek banks owe the Bank of Greece €100 billion, which is fully collateralized; and that in turn the Bank of Greece owes the ECB €100 billion on an unsecured basis. If Greece were to chaotically devalue and default, then it’s entirely reasonable to assume that the Bank of Greece would default on those obligations to the ECB, and would keep the Greek banks’ collateral for itself, to help prop up as much as possible the nascent drachma.
If that happened, the rump eurosystem — the remaining 16 central banks, plus the ECB — would take an accounting write-down of €100 billion. They have €86 billion in capital, and another €400 billion in capital they can create any time they want, just by revaluing their gold reserves. So coming up with €100 billion wouldn’t be hard — especially since the whole concept of an insolvent central bank is a little bit silly. If the capital of a central bank stopped being positive and started being negative, then in practice nothing at all would change. Central banks can never go bust, because they can print money.
But what if the entire eurosystem fell apart, and every country reverted to its own national currency? In that case, it’s still hard to see how there would be much of a hit to Germany. Germany’s banks, like Deutsche Bank, would see their Target2 balances redenominated from euros into Deutschmarks. And the Bundesbank would have a theoretical claim on the ECB, but at this point the ECB would barely exist. But that’s fine, it could simply declare that all those euros were now Deutschmarks, since the Bundesbank can create as many Deutschmarks as it wants.
The hidden assumption underlying Sinn’s doom-mongering is essentially that if the euro fell apart, German taxpayers would have to write a trillion-euro check to the Bundesbank, to make up for all the money that the Bundesbank would never be able to collect from the ECB. But that just isn’t realistic. Here’s Whelan:
The new Deutschemark would, like the euro, be a fiat currency and there would be no need for all D-marks to be fully backed by hard assets held by the Bundesbank.
If German officials were concerned about the need for the Bundesbank’s balance sheet to show assets greater than liabilities, then they could agree for the Bundesbank to write itself a cheque equal to the value of the TARGET2 credit and to top it up each year with interest. There would be no need to also top up its liabilities, so the Bundesbank’s technical solvency will have been restored without raising any taxes on German citizens.
I suspect some may suggest that a failure to fiscally recapitalize the Bundesbank would produce a currency that people will have no faith in and/or that this will result in inflation. However, this approach would do nothing to change the amount of money circulating in a post-EMU Germany. And a cheque tossed in an empty vault can’t trigger hyperinflation. More likely, because the value of a fiat currency depends largely on the faith of citizens that the quantity of the currency will be kept in limited supply, is that the new Deutschemark will appreciate significantly, with the result being deflation rather than inflation.
To put it another way: yes, the Bundesbank would essentially be printing a trillion euros’ worth of Deutschmarks, which isn’t a very Bundesbanky thing to do, and is in theory inflationary. But if you’re creating a new currency, then you need to print that currency. And so long as German banks kept those Deutschmarks on deposit at the Bundesbank, and remained shy about lending them out to borrowers in other countries, the money supply in Germany wouldn’t actually increase at all.
Should Greece, Ireland, Italy, Portugal and Spain go bankrupt and repay nothing, while the euro survives, Germany would lose $899 billion. Should the euro fail, Germany would lose over $1.35 trillion, more than 40 percent of its G.D.P. Has the United States ever incurred a similar risk for helping other countries?
Insofar as Sinn is talking about Target2 balances here — and those balances account for the majority of these numbers — I just don’t think he’s right. For one thing, as Whelan points out on his blog, the US did actually incur rather more than 40% of GDP in costs associated with helping other countries. In 1941, the US national debt was less than 40% of GDP; in 1946, it was more than 120% of GDP. Oh, and 400,000 American men died fighting, too. Much of that can reasonably be considered self-defense, but a lot of it was much-needed aid for the rest of the free world.
But more to the point, German depositors wouldn’t lose any money, German banks wouldn’t lose any money, and the German government wouldn’t lose any money. The only entity which might be considered to have lost money would be the Bundesbank — but really the Bundesbank would just have converted all of the euros in Germany to Deutschmarks. If the euro ceased to exist, obligations within the eurosystem would cease to exist as well — indeed, the whole eurosystem would cease to exist, since its entire raison d’etre is to support the euro. All those internal accounting conventions would disappear in a puff of smoke, and every national central bank would be on its own, running its own currency and looking after its own banks.
It’s maybe comforting to think that today’s euros are somehow real and that tomorrow’s hypothetical Deutschmarks are not real, and that therefore if tomorrow those euros ceased to exist and were replaced by Deutschmarks, that there would be a loss of hundreds of billions of euros. But that’s not how fiat currencies work. Tomorrow’s Deutschmark is no more or less real than today’s euro, and in fact the most likely problem with the Deutschmark is not that it would be weakened when the Bundesbank printed lots of it, but rather that it would be such a popular currency that it would soar in value, making German exports uncompetitive.
There’s no doubt that there would be absolutely massive costs associated with a breakup of the euro. But let’s not exaggerate matters by including Target2 balances in those costs. They’re little more than accounting conventions, really: they’re a way of making sure that the eurosystem always sums to zero. If you’re the kind of person who thinks that the Target2 balances are real liabilities, then you’re also the kind of person who thinks that a bank run in Spain, where deposits flee for Germany, is bad for Germany and good for Spain — since it only serves to exacerbate those Target2 imbalances.
By Sinn’s logic, it would be good for Spain or Italy to leave the euro, since they would default on their Target2 obligations and thereby find themselves incredibly rich — they would have borrowed hundreds of billions of euros from the eurosystem, and then would have no need to ever repay that loan. If you believe that, then feel free to take Sinn seriously. But it seems clear to me that if those euros cease to exist, then all that matters are the bilateral relationship which the national central banks have with all the banks in their country. And those bilateral relationships, built on fully-collateralized loans, wouldn’t be affected by Target2 accounting conventions at all.