Should the eurozone be mutualised?

This comment was published in a live debate hosted by The Economist

The debate on Eurobonds is at risk of missing the forest for the trees. Discussions about their various colours and shapes, their incentive properties, and the conditions for implementing them without implicit transfers from responsible to irresponsible governments are all of significant importance, as discussed in a recent paper by Bruegel and IMF scholars. But the principal issue is why we may need them in the first place. Paul De Grauwe makes a strong case for Eurobonds based on first principles. Let me add an argument that has often been overlooked in recent debates.

Investors worldwide need to know what is the safe euro asset. For America they know what it is—the Treasury bond—and they are stockpiling it in spite of their doubts about the soundness of American public finances. For the euro zone they thought they knew what it was—all euro-zone government bonds—until they paid a price for this wrong belief. Now they tend to consider that it is the German bund—and perhaps, to some extent, a few others, depending on conditions and risk appetite.

It is important to note that for an asset to be considered safe, underlying quality is neither necessary nor sufficient. The benchmark asset has to be of good enough quality, but it does not need to be of superior quality—and superior quality does not guarantee an asset serves as benchmark. So the rent that accrues to its issuer should not be confused with the benefits of sound policies. It comes on top of them. The premium the US Treasury benefits from is not a reward for fiscal responsibility. It is the privilege that accrues to the issuer of the principal safe asset in world finance. Similarly, if the German bund becomes the reference asset for the euro zone a rent will accrue to Germany irrespective of the relative quality of its budgetary policy.

The question then arises: can we have a monetary union where one country is (or a few of them are) providing the safe asset of reference? If things go this way, Germany will benefit from a significant rent. Current estimates of the premium the US Treasury benefits from are around 50 to 100 basis points. Take the low end of this estimate and assume Germany from now on benefits from a 50 basis points premium on its debt securities. This would translate into something like an annual €10 billion gain. Even half of that sum would be meaningful.

There is nothing wrong with benefiting from a rent, except that it rarely comes without strings attached. As students of the international monetary system emphasise (see, for example, the work of Pierre-Olivier Gourinchas of Berkeley and Hélène Rey of LBS), an exorbitant privilege implies exorbitant duties. The dollar's exorbitant privilege is to be the world currency. The exorbitant duty is that America has extraterritorial monetary responsibilities. At irregular intervals it must step in and save the system, very much like an insurer. In 2008 and recently again the US Federal Reserve entered into swap agreements with a series of partner central banks to give them the means to provide dollar liquidity to their banks. Should these claims have failed to be honoured, American taxpayers would have had to cover the loss.

The relevance for the euro zone is that if Germany's partners realise that Germany gets a rent, they will request that it takes on special responsibilities for the stability of the euro. In part, this is the case already: the now-famous Target 2 balances highlight the special role of Germany as a hub of the euro system. But only in part, because potential losses on the Target 2 balances are to be shared among all euro-zone countries. In the same way losses on loans to Greece or other programme countries would be shared among all non-programme countries, including Italy and Spain.

Could Germany collect the rent and refuse to behave as the insurer of the system? I very much doubt it could lastingly. This would be resented by partners and would create significant acrimony. Already, the Italian prime minister, Mario Monti, has spoken of the "creditocracy" of countries that see themselves as creditors of the others. His tone is indicative of the risk there would be in the persistence of a structural asymmetry between first-tier states benefiting from privileged access to the bond markets and second-tier states.

Can Germany accept being the insurer of the system? This is also unlikely. It would imply a willingness to consider transfers to partners in situations of crisis, or at least the risk thereof. What has happened since the start of the euro crisis shows that the German political system is not comfortable with this idea. Furthermore, it would imply more or less direct German oversight of partner countries, which is hardly palatable.

This is what makes Eurobonds attractive. For Germany, to agree on them would basically amount to sharing the rent with partners, on the condition that they accept a new budgetary regime based on ex ante rather than ex post control. This would be functionally equivalent to what Germany did 20 years ago when it agreed to share its currency with partners, on the condition that these partners would accept German monetary principles. For all the troubles we have gone through recently, this contract has by and large been respected: prices in the euro zone have remained remarkably stable.

So in the end Eurobonds could be the gift Germany makes to its partners in exchange for them locking in budgetary discipline (through, for example, a system like the Blue Bonds-Red Bonds scheme scheme of Jacques Delpla and Jakob von Weizsäcker or the gradual scheme proposed by the Padoa-Schioppa Group).

Obviously for the trade to be acceptable it should not cost Germany more than the rent it currently benefits from. Rather than be eliminated, the rent should be shared, ideally at little cost to Germany. This is where we are brought back to the minutiae of mechanism design.