Over the past week, it’s become clear that the real leverage the European authorities have over Greece is via the banking system. What does Greek need continued loans for? Not to pay for public expenditures, thanks to the primary surplus. Not to pay for imports — Greece has a (small) trade surplus. Not to service current debt, if it defaults. What does need to be financed, is the flow of deposits out of Greek banks to the rest of Europe.
So what happens if that financing is cut off, as the ECB is threatening? The usual answer is collapse of the Greek banking system, followed immediately by a forced exit of Greece. But the other night I was talking to some friends about the situation, and we found ourselves wondering: What concretely are the mechanics of this? What is the exact chain of events from an end to ECB financing to Greek exit from the euro? I don’t know the answer to this, but the more I think about it, the less confident I am in the conventional wisdom.
What concretely does it mean that the ECB is providing liquidity support to Greek banks? As far as I can tell, it is this. When a holder of a deposit in a Greek bank wants to make a payment elsewhere, either to purchase a good or asset outside Greece or to move the deposit elsewhere, the Greek bank must transfer an equal quantity of settlement assets to the bank receiving the deposits. These settlement assets are normally acquired on the fly, by issuing a new liability in the interbank market, but if other banks are unwilling to accept the liabilities of Geek banks, they can be borrowed directly from the ECB, against suitable collateral. This is the lending that the ECB is threatening to cut off.
What if the Greek banks can’t acquire settlement assets? Then other banks will not accept the deposits, and it will be impossible to use deposits in Greek banks to make payments. Depositors will find their accounts frozen and, in the normal course of events, the banks would be shut down by regulators.
But Greece still does have a central bank. My understanding is that much of the day to day business of central banking in Europe is carried out by the national central banks. In principle, even if Greek banks can’t acquire settlement assets by borrowing from the ECB, they can still borrow from the Greek central bank. This doesn’t help with payments to the rest of Europe, since reserve balances at the Greek central bank won’t be accepted elsewhere. But I don’t see why the Greek central bank can’t keep the payments system working within Greece itself. If the Greek central bank is willing to provide liquidity on the same terms as the ECB, what’s going to force the Greek banks to shut down? It’s not as though there’s any Europe-wide bank regulator that can do it.
In a sense, this is a kind of soft exit, since there will now be a Greek euro that is not freely convertible into a non-Greek euro. But I don’t see why it has to be catastrophic or irreversible. Transactions within Greece can continue as before. And for routine trade it might not make much difference either, since the majority of Greek imports come from outside the EU. Where it would make a difference is precisely that it would prevent Greek depositors from moving their funds out of the country.  In effect, by cutting Greece off from the European interbank payment system, the ECB will be imposing capital controls on Greece’s behalf. You could even say that, if the threat of cutting off liquidity support can trigger a run on Greek banks, actually doing so will ensure that there isn’t one.
Now maybe I’m wrong about this. Maybe there is a good reason why the Greek central bank can’t maintain the payment system within Greece. But I also think there’s a larger point here. I’m thinking about the end of the gold standard in the 1930s, when breaking the link with gold was considered an unthinkable catastrophe. And yet the objective basis of the money system in gold turned out to be irrelevant. I think, in the same way, the current crisis may be revealing the reflexive, self-referential nature of money. On a certain level, the threat against Greece comes down to: “You must make your money payments, or we will deprive you of the means to make your money payments.”
The rule of the money system requires that real productive activity be organized around the need for money. This in turn requires that money not be too freely available, but also that it not be too scarce. Think of Aunt Agatha in Daniel Davies’ parable. Suppose her real goal is to run her nephew’s life — to boss him around, have him at her beck and call, to know that he won’t make any choices without asking if she approves. In that case she always has to be threatening to cut him off, but she can’t ever really do it. If he knows he’s getting money from her he won’t care what she thinks — but if he knows he isn’t, he won’t care either. He has to be perpetually unsure. And in keeping with Davies’ story, the only thing Jim actually needs the money for, is to continue servicing his debt to Aunt Agatha. The only real power she has is a superstitious horror at the idea of unpaid debts.
In this way I’ve tentatively convinced myself that all Syriza needs to do is hold firm. The only way they can lose is if they lose their nerve. Conversely, the worst outcome for the ECB and its allies would be if they force Greece into default — and everyone watches as the vengeful money-gods fail to appear.
UPDATE: It turns out that Daniel Davies is making a similar argument:
Capital controls are arguably what Greece needs right now – they have
balanced the primary budget, and they need to stop capital flight.
From the ECB’s point of view, I’d agree that the move is political, but
it also means that they are no longer financing capital flight.
a sensible negotiated solution here – with a lower primary surplus than
the program (in which context I think Varoufakis’ suggestion of 1.5% is
not nearly ambitious enough), a return to the structural programs (the
Port of Piraeus really does need to be taken out of the political
sphere), and an agreement to kick the headline debt amount into the far
future (in service of which aim I don’t think all the funny financial
engineering is helping).
The fall-back is a kind of soft exit,
with capital controls. But the massive, massive advantage of capital
controls over drachmaisation is that they preserve foreign exchange.
Greece imports fuel and food. With capital controls, it can be sure of
financing vital imports.
The fact that Davies is thinking the same way makes me a lot more confident about the argument in this post.
 Greek banks would also presumably be limited in their ability provide physical cash to depositors, but I don’t think this is important.