ECB QE. To mutualise or not?
Had a great lunch with an economist friend who alerted me to something that is obvious, but I had not thought of, and who then explained it wonderfully. (And then returned to find Twitter full of it too.) That there is a decision to be made about how ECB QE gets done that didn't arise in the case of fiscal unions like the UK or the US.
Recall the Bank of England's QE was done on the balance sheet of the Asset Purchase Facility, a special purpose vehicle - ironically enough given the role of these beauties in the storm of the financial crisis - from whose profits and losses the main balance sheet of the Bank of England was insulated. In the case of the UK, the main thought experiment was that gilts would be bought at relatively low interest rates (high prices), pushing up the price still further [so perhaps a small capital gain if sold back to the private sector quickly], but that the portfolio would be sold off when interest rates had returned close to steady state levels, ie when prices had fallen. This matters for the Eurozone too. But the real issue is the possibility that the ECB buys the bonds, and they then become worth much less, either because people later worry more about default, or one of the peripheral governments actually does default, or exits, financially pretty much the same thing.
There is no single fiscal authority in the Eurozone to play the same role as the Treasury in the UK. And, be sure that the collection of states there does not want to play this role anyway.
On the face of it there is a choice about whether the ECB chooses to put its own balance sheet at risk in the event that purchases of member state sovereign debt imply losses. Or whether to delegate the task of purchase to member central banks. In which case, it will be the member state governments that have to make good losses, and not the ECB, whose losses fall, in proportion to the shareholding, on member state governments. That might sound like a distinction without a difference. But not so. The purchases most likely to imply losses are those of troubled 'periphery' sovereigns, Ireland, Portugal, Greece, Spain, Italy. The delegated model of QE ensures that those losses fall where they are incurred. The centralised, mutual model, redistributes losses somewhat away from the 'periphery' [in quotes because, come on, Spain and Italy hardly peripheral] towards the North.
A motive for operating the delegated way is of course to make it more likely that the Northern ECB voting representatives will accept it. A bad reason why they would be more likely to accept it is that it implies, under certain myopic views of the world, a better outcome for their taxpayers. Bad because they are not meant to be casting their vote on national grounds at all. An acceptable reason for them to like it better is that it is an operation that is less 'fiscal' sounding than the mutualised model. And would be therefore less subject to legal challenge (for example from the German constitutional court). [Though, as I wrote before, there is an element of intellectual redundancy about trying to discern the monetary or fiscalness of a policy].
Note that the fiscal element comes without losses crystallising. Just as the fiscal subsidy to banks that are too big to fail in the private sector comes about without them actually becoming insolvent, in the form of low, subsidised borrowing rates.
If you see QE as something precisely that's meant to shift sovereign default risk off governments that can bear it and onto those that can, then the mutualised model is of course better. But really, the decentralised version is the one that resembles the operation that the BoE or the Fed did. How so, given that, legally, at least, the Fed and BoE operations were entirely centralised? Well, relative to the counter-factual of no QE, QE there had no redistributive implications within the state. Centralised QE in the Eurozone would. Personally, I think that since pretty large redistributions are exactly what's needed there, this moderate risk shuffling fiscal redistribution is desirable. But it is what it is.
If the bad scenarios unfolded for peripheral sovereigns, there would no doubt be a terrible further convulsion in the financial system, and insuperable pressure for further action by the fiscal authorities themselves. But I suppose that the hypothetical Teutonic master of the ECB universe might at least comfort himself with the thought that they had not authorised a small subsidy up front, and that there was at least a small chance that transfers would never happen, or be insufficient, and that an economically disastrous but morally satisfying outcome would be assured.