On and off target

Hans-Werner Sinn 14 June 2011



My exposure of the Target2 balances issue (see here, here and here for German-language articles, and this website for an English-language one) has triggered a heated debate that has seen economics pundits take positions on either side of the argument. In essence, what I say is that the massive imbalances that have accumulated in the Target2 settlement system amount to a hidden bailout of the stricken economies in Europe’s periphery by the European Central Bank (ECB). My critics say that this is not so (Whelan 2011, Buiter et al 2011). Much of the criticism appears to stem from misunderstanding what admittedly is a very complex issue.

Let’s take the critiques one by one.

  • The tolerance shown by the ECB towards the Target balances was wrong. It was not fundamentally wrong, as I have stressed more than once. It was in fact the right thing to do when the crisis began and the parliaments had no time to react. But it nonetheless was a de facto bailout, a fiscal measure that was facilitated by lowering the standards for the collateral for refinancing operations and providing Emergency Liquidity Assistance credits. The policy is now in its fourth year. There has by now been ample time to hand the issue over to the parliaments of Europe.
  • The Bundesbank is liable in the same volume as its lending to the Eurosystem. This is incorrect and I never said that. While nearly all the Target debt of the GIPS countries (Greece, Ireland, Portugal, and Spain) is countered by a claim of the Bundesbank, the extra liability affecting the Bundesbank as a result of the Target balances, should the GIPS countries default, is limited to the German capital share in their Target debt. My first calculation of this amount, published on April 3 in the German daily Süddeutsche Zeitung (Sinn 2011a), came to a value at risk for the Bundesbank of €114 billion if all GIPS countries were to collapse and the bank collateral, often government bonds, lost its value. This sum was and remains correct, as it was calculated exactly as explained. Incidentally, in naming this figure, I say nothing at all regarding the probability of it falling due.
  • Central bank credit creation is crowded out in Germany because the ECB fixes the monetary base. This argument is neither right nor have I ever made it. Crowding out occurs not because the supply but the demand for money is limited. Given the payment habits, economic activity and the ECB interest rate, only a given amount of central bank money is needed. Any excess liquidity brings no benefit and only involves interest costs. The German commercial banks can, at present, borrow as much money from the Bundesbank as they wish, but they do not want to. For this reason, the Bundesbank’s refinancing operations, i.e. the credit it gave to commercial banks, shrank in Germany basically to the same extent as money was flowing in through the Target2 system. The credit that the Bundesbank would have been creating in Germany was displaced by the jointly guaranteed credit it de facto gave via the central bank system to the GIPS countries.

The facts of the matter will perhaps become clearer if we bear in mind what the Target deficits of the GIPS countries, a hefty €340 billion by the end of last year, actually are. Essentially, they are the portion of the money created by the corresponding national central banks that was used for the net acquisition of goods and assets from other Eurozone countries. I gave this definition in my article of May 4 in the Frankfurter Allgemeine Zeitung, and it took me quite a bit of thinking to distil it. The €340 billion are in fact a loan to the GIPS from the euro community, which, like any other loan, made it possible for them to purchase more goods and assets abroad that would otherwise have been the case. In terms of liability and the transfer of resources actually involved, it differs little from short-term Eurobonds whose revenue is given as credit to the GIPS and entail a joint liability of all the Eurozone countries. The only difference with real Eurobonds is the fact that the central banks of the GIPS countries can dispose of such credit at their leisure as long as they offer collateral and that the Bundesbank cannot refuse to accept the purchase of the implicit Eurobonds.

The possibility of taking on Target loans encourages a self-servicing attitude. This is why the US central bank system excludes it, as explained in the articles mentioned above. And I find this relevant for Europe because, from an economic perspective, the 12 districts in the US do not differ much from the 17 euro countries. A district that wishes to import more goods than it exports must receive private credit from another district or hand over marketable assets, and a district whose citizens wish to acquire net assets from other districts must export more goods than it imports. It is not allowed to fulfill its wishes by cranking up the money-printing press as in the Eurozone.


Buiter, Willem, Ebrahim Rahbari, and Jürgen Michels (2011). “TARGETing the wrong villain: Target2 and intra-Eurosystem imbalances in credit flows”, Global Economics View, CitiBank, 9 June.

Sinn, Hans-Werner (2011a), “Kredite, Kredite, Kredite”, Süddeutsche Zeitung, 3 April.

Sinn, Hans-Werner (2011b), “The ECB’s stealth bailout”, VoxEU.org, 1 June.

Sinn, Hans-Werner and Timo Wollmershäuser (2011), "Target Loans, Current Account Balances and Capital Flows: The ECB’s Rescue Facility" CESifo Working Paper Nr. 3500, 24 June.

Whelan, Karl (2011). “Is there a hidden Eurozone bailout?”, VoxEU.org, 9 June.



Topics:  EU policies Europe's nations and regions

Tags:  Eurozone crisis


In a series of papers and statements, Professor Sinn has drawn our attention to the fact that starting in 2007 the TARGET2 system has been used to channel long-term capital flows to the so-called GIPS countries (Greece, Ireland, Portugal, Spain). His findings and arguments have caused a heated and partly emotional discussion. Some of the heat is due to a limited understanding of the economic significance of the TARGET2 balances.
In my view, Professor Sinn has a very important point that should not be too difficult to understand, once we interpret what has been going on in the past 4 years of the “TARGET2 history” against the backdrop of standard theory of balance of payments adjustment within fixed-rate systems. After all, that is the lens through which we must look at the Eurozone, as long as a full economic union with a unified fiscal policy is not established. The relevant question concerns the type of adjustment mechanism that the Eurosystem relies on in case individual countries run into balance of payments crises. For that is what the GIPS countries were facing by 2007 when huge negative lending positions of their domestic public and private sectors could no longer be financed by private capital imports. We have repeatedly seen fixed-rate systems break down because of flawed adjustment mechanisms, the most prominent case in point being the Bretton Woods system.
He doesn’t frame it that way, but Sinn’s key point is that the TARGET2 system, although not at all designed for this purpose, has in recent years been turned into a mechanism akin to the adjustment mechanism present in the Bretton Woods system, but with the important qualification that the GIPS countries are de facto playing a role similar to that of the reserve currency country, i.e. the US in the Bretton Woods system. Once we realize this, we also realize that Sinn was right in raising the issue.
This analogy may seem a bit far-fetched at first sight, but it becomes quite clear once we acknowledge that a positive TARGET2 balance of a country’s central bank indicates the amount of central bank money circulating in that country that was originally created in some other Eurozone country. Sinn has consistently pointed out, particularly in his working paper with Timo Wollmershäuser[1], that in the present context this is the crucial economic significance of TARGET2 balances. The Bundesbank’s own description of the TARGET2 system in the March 2011 Monthly Bulletin similarly portrays it in this way.
But exactly the same description applies to what happened in the final years of the Bretton Woods system every time the US was financing its excess of absorption over income by means of printing dollars. As other countries were obliged to guard their dollar exchange rates, these additional dollars would end up as assets in those countries’ central bank balance sheets, i.e., as those countries’ portion of the monetary base that had originated in the US. Unless they were to sterilize their build-up of foreign exchange reserves, those countries would be facing the specter of inflation. This was the ultimate reason why the Bretton Woods system finally broke down in the early 1970s.
Now, in any true monetary union we would not expect each and every country’s monetary base to have originated domestically. We would perhaps expect the fraction of the monetary base of the union that has been created in any one country to broadly reflect the size of that country, say in terms of its GDP, relative to the size of the union. Also, we would not worry about nonzero TARGET2 balances, unless there is a systematic trend or build-up of huge TARGET2 balances and we have reason to believe that this reflects a balance of payments crisis. In the aforementioned working paper, Sinn clearly shows that in this sense, indeed, we had no reason to worry about the TARGET2 balances up to 2007.
But that was the year when everything changed. The figures meticulously drawn up and presented by Sinn and Wollmershäuser are compelling.  In a matter of 3 years, a group of countries commanding a mere 18 percent of a monetary union’s GDP has become the source of 66 percent of that union’s monetary base.
If we agree that China, through its massive build-up of dollar reserves, has recently been financing US expenditure, then we should also be able to agree that the TARGET2 balances that have accumulated since 2007 amount to the Eurozone as a whole, via the ECB and the Eurosystem, have financed GIPS expenditure, whatever the position of their capital account during those years (see below). Moreover, if we agree that China faces the risk of a huge loss in the real value of its central bank’s dollar holdings in case of dollar devaluation, then we should also be able to agree on a similar risk for Eurozone countries, in line with their ECB capital shares, from partial write-offs of the assets behind the TARGET2 balances in case of default. And finally, just as its huge dollar reserves constitute an incentive for China to avoid a dollar devaluation, the TARGET2 balances constitute an incentive for the ECB to avoid scenarios that lead to write offs on the GIPS assets present in the non-GIPS monetary base.
The economic significance of TARGET2 balances outlined above does not imply that they can be seen as literally mirroring current account deficits, let alone as being causal for such imbalances. Nothing of this sort can be read out of anything that I have seen or heard from Hans-Werner Sinn. For one thing, this would ignore capital movements which, unlike in the Bretton Woods system, are an important element of the Eurozone. But Sinn and Wollmershäuser provide a prominent and detailed treatment of capital movements, including in particular capital flight, in their paper. Moreover, I see no contradiction at all between Sinn’s point and the empirical result that over the entire history of the TARGET2 system the pattern of balances does not correspond to the pattern of current account deficits and that negative positions evolved in line with capital flight, such as for instance in Ireland. Indeed, all of this is explicitly discussed by Sinn and Wollmershäuser. 
Reading causality into whatever co-movement we observe between TARGET2 balances on the one hand and either the current account or capital movements on the other seems very strange; even thinking about it seems futile. But I know of no statement that Hans-Werner Sinn has made to this effect. His point is that the TARGET2 balances are an ill-conceived mechanism of adjustment to balance of payments crises in GIPS countries, not their cause.
Professor Sinn was absolutely right to raise the question of whether the Eurosystem has a well-functioning mechanism to deal with balance of payments crises. Addressing this issue is long overdue. His analysis suggests with impeccable logic that the answer is no. Allowing the TARGET2 system to support an adjustment mechanism that lets deficit countries, whether the GIPS in the present situation or countries closer to the core in the future, enjoy benefits equivalent to that of having a reserve currency hardly stands the test of a well-functioning adjustment mechanism. Moreover, if the crisis arises in the historic context of exploding public deficits, it also blurs the distinction between monetary policy and fiscal policy and the respective decision-making procedures and institutions that we have so far rightly cherished.

[1] Sinn, Hans-Werner / Wollmershäuser, Timo: Target Loans, Current Account Balances and Capital Flows: The ECB’s Rescue Facility, June 2011 <http://www.cesifo-group.de/portal/page/portal/ifoHome/b-publ/b3publwp/_w...

Professor of Economics and Public Finance, University of Munich; President, CESifo