Yves here. We thought it was important to focus on the operational issues of a Grexit because the commentary on this issue too often muddies multiple issues, which then leads to a poor understanding of what the costs and risks to Greek citizens might be.
The first, and most important, is a default does not mean a Grexit. Countries can and do default on their sovereign obligations, as Russia did in 1998 and Argentina did in 2001, without adopting new currencies. Moreover, default is not a “get out of paying free” card. Even with defaults with mere private lenders, Russia and Argentina restructured their debts, meaning they agreed to a program of payment of a reduced amount. It is also important to recognize that official creditors like the IMF and ECB have far more power relative to sovereign borrowers that private creditors. For instance, the IMF was the only lender to Argentina prior to its default to be repaid in full. That means even when dealing with a borrower of limited means who has defaulted, an official creditor is in a better position to extract what can be gotten than private ones.
Second, when most commentators discuss a Grexit, they focus on two elements: the forcing of a debt restructuring, and a sudden currency devaluation. Argentina again is a precedent of sorts, since when it defaulted, it also abandoned its currency peg with the dollar. Experts widely acknowledge that a rapid currency depreciation is painful even though the longer-term effects are positive. The reason is what economists call a J-curve effect: the trade deficit gets worse before it gets better. A county goes into a devaluation with an established pattern of trade. The immediate impact of a default is that import prices shoot up in the new, depreciated currency value terms while what the country receives on its now cheaper exports goes down. It is only after trade partners have shifted more of their sourcing of imports to buy from the country with the now-bargain-priced goods that the country gets the benefit of the depreciation. The immediate shock of the spike in import prices is an immediate slowdown, shortages, and business failures.
The conventional analyses generally ignore or wave off the element of a Grexit that Nathan Tankus focuses on here: the operational implications of a Grexit. Launching a new currency would be every bit as demanding as launching the Euro was in 1999, and arguably more so, since the Greek government will also have to contend with a large market of Euros circulating as currency and may come to find it necessary, at the outset or later, to offer dual currency payment processing services. We’ve attached a paper at the end that gives an idea of the amount of planning and lead times that were required for this to go smoothly. Now admittedly the Euro launch had an added factor of complexity, which was a large number of existing currencies involved. However, the Eurozone was also able to force citizens off the legacy currencies in 2002, three years after the launch of the Euro. By contrast, with the Euro the currency of Greece’s major trade partners and Greek bank customers already hoarded large amounts of Euros as cash, the Greek government will have a currency it does not control playing a significant role in its economy.
We believe that the impact of the one-two punch of a conventional currency depreciation combined with the disruption of the imposition of a bank holiday, capital controls, and the chaos accompanying an un or underplanned transition to a new currency on an already deeply depressed economy is greatly underestimated by most commentators.
By Nathan Tankus, a writer from New York City. Follow him on Twitter at @NathanTankus
My last post covered how the Euro payments system works. What happens when it doesn’t? In other words, what are the series of operational events that happen when Greece has a non-negotiated default or the ECB ends Emergency Liquidity Assistance?
Before we begin one point needs to be hammered home. Economics, law and politics are all crucial to the story of Greece, but in the moment of default or Grexit they take a back seat to something far more important: organizational capacity. To exploit every opportunity, to get a new monetary system up and running, to devise a way of continuing imports and exports, to decide what to do with the Bank of Greece and generally to decide how to deal with growing mass of consequences hurtling towards them, the government needs large amounts of organizational capacity.
Further, the longer they take getting the “new” system up and running, the more issues will pop up and the more organizational capacity they will need at a future date. (For more on organizational capacity, here’s something Nathaniel Cline and I wrote two years ago.)
This is a large part of why Yves has been so critical of Syriza. Being shut out from the payments system (on any level ranging from having ELA being cut off from the banking system to Iran or de facto Iran-style financial sanctions) and essentially forced to exit will need mobilization of the society on total war levels. The Greek ruling coalition has not prepared the population for the Troika holding firm and letting a non-negotiated default and/or Grexit happen. While we can’t know how much secret planning has happened, we can make inferences from what has already happened to organizations critical to the delivery of essential social services, like the hospitals, due to years of budgetary starvation, the poor state of tax collection and land records, and more immediately, evidence of strained organizational capacity in the negotiations with the Troika. This included poor setting of priorities (for instance, Yanis Varoufakis making himself highly accessible to the media to the point where Greek commentators were critical of time spent on messaging at the expense of negotiating), apparent limited sophistication in dealing with legal or quasi-legal documents (the treatment of the February Eurogroup memo as a success for Greece when it was very unfavorable to them), and possible shortcomings in basic blocking and tackling (creditor complaints about Greece’s delays in supplying requested information needed to evaluate Greek proposals). These may seem like minor things, but remember: What Syriza needs most is time to prepare contingencies for a default and/or a Grexit and build up their organizational capacity.
Keeping the Troika at the table without having to agree to third rails like pension gutting means showing up to the meetings on time, providing detailed proposals on all the smaller matters and dancing until you get to the big issues. In other words, use the same kind of tactics banks and other corporate interests use to gut regulatory rule-making in the United States. Hopefully, when you’re finally forced to choose whether or not to agree to the major reforms, you’ve pushed the process to the point where the demands aren’t so bad and if they still are, you’ve preserved enough of a veneer of civility that you can move the discussion to impasse alternatives that reduce the costs of negotiation failure to both sides. (Both pundits and financial analysts have mentioned idea like a “managed default” or a “managed Grexit.”) In other words, playing the game has the upside at a minimum of extending the timetable and also increases the odds of cooperation in mutual damage control. If that fails, more aggressive options are still open.
In addition, you’ve bought yourself time to plan for an exit and can reasonably claim — and perhaps leak portions of detailed and very technical plans showing this — that you can exit relatively smoothly as the final bargaining chip. Instead the attitude of Syriza leadership seems to be nicely summed up by this quote from Yanis Varoufakis: “[W]e’re dealing with a system of cronyism and corruption. That’s what we have to tackle. But, instead, we’re debating pharmacy opening times.”
To drive this point home, let’s go back to what Varoufakis dismissively refers to: pharmacy opening times. This is part of the negotiation of structural reforms, something that the Greek government agreed to when it signed the Eurogroup memorandum in February. And recall Varoufakis at that time mentioned how he and the so-called institutions agreed on 70% of those reforms.
Despite Varoufakis’ complaint, pharmacy hours are an important issue wrapped up in the Troika’s desire to “reform” the pharmaceutical retail industry. Greece sets prices of over-the-counter drugs while making it illegal for everyone but licensed retail pharmacies. They also regulate how these retail pharmacies function in all sorts of ways, including setting opening hours. Easing and eliminating these rules is about shrinking this industry and ultimately about eliminating jobs. This week pharmicists went on strike because Syriza conceded on supermarkets selling medications.
That means this set of issues is not beneath Varoufakis at all. In fact, negotiating over issues such as this is his job. Now, he does have a point that he should not be going into the weeds on every deal point, but someone on his team should be handling matters like this. If this is winding up on his desk because there’s no one else to handle it, that demonstrates how desperately overstretched his ministry is already, a poor sign for its ability to take on far more complex and urgent tasks. In other words, you can argue that criticizing Syriza for this is unfair since they were forced into negotiations without having the organizational capacity to conduct them, but then how would they have the organizational capacity to conduct a Grexit?
Let’s return to the most urgent tasks that a Greek government would face in the initial hours of a Grexit: the payments system. The misunderstanding I laid out last post surrounding the relationship between the Greek government and the Bank of Greece has convinced people that taking over the Bank of Greece is a trivial matter. I said then that taking over the Bank of Greece is a trap.
To restate from my first post, the reason I’m skeptical of taking over the Bank of Greece is that it makes the Greek government responsible for the depreciation of asset values posted as collateral for ELA and paying Euro denominated interest when otherwise they are not financially responsible for the Bank of Greece. (The ECB is.) Further, this assumes that the ECB and European powers would be happy for the Greeks to take on this obligation. Even worse, this could be seen as an expropriation that may qualify (especially depending on how rabid other member states get) as a “clear risk of a serious breach by a Member State of the values referred to in Article 2 [human rights, democracy and the rule of law].” This would mean suspension from the European Union, which would have very serious consequences for them ranging from agricultural supports all the way to access to Target2. (Net spending by the EU in Greece, that is spending – taxes, was about 2.5% of GDP in the latest available figures.)
Now someone may argue that “Greece is taking over the Bank of Greece because they’re defaulting and exiting the Eurozone, all of this is illegal and none of their Euro debts matter at this point”. I disagree with this view.
First, what would force them out of the Euro in practice is the cutting off of ELA assistance. Legally however, a member state can’t be forced out of the Euro which is what the ECB action would amount to. The Greek position should be that they have been forced by the illegal actions of the European Union to build a new payments system and thus they aren’t liable. Further, they should state they would welcome the reintroduction of liquidity support so that a drachma in their payments system would equal to a Euro and effectively be a Euro. A way to back up this argument would be to declare that these new liabilities are legal tender only as long as ELA to Greece has stopped but that they are acceptable in payment of taxes and loans no matter what. In other words, legally treat them like Electronic Tax Anticipation Notes (ETAN).
Despite all the issues with taking over the BoG, the Greek government might need to do it simply because of the time it takes to program a new payments system. Even if they take over the IT system from the Bank of Greece, a lot of software will need reworking to separate from the Euro IT wise. In correspondence, Richard Smith (who worked on the IT transition to the Euro) made this quite clear:
Richard Smith: Sounds pretty horrible. The similarish reverse operation that set up all this bliss for Greece a decade and more ago, namely redenomination from Francs, DEM etc to EUR, at fixed rates, took several elapsed months of analysis, design coding, testing, planning and rehearsal, just on one well-maintained system. It worked fine but we were really glad we rehearsed it and had time. Nothing important that we were interfacing to fell over, either (a related risk).
Goodness knows what bringing the Drachma back at no notice and unplanned would actually be like for IT systems. I really don’t like the sound of it much, especially if somehow there would have to be two currencies in parallel and a floating rate
Nathan Tankus: Is there any way for them to secretly build this system on their own and roll it out to the banks over a two to three week bank holiday? What if they impose capital controls and for payment settlement purposes jerry rig it so that drachma equals 1.5 Euros or some such number?
Richard Smith: It sounds like a bit of a stretch. Not sure where you’d hide all the programmers and analysts. Even asking the questions that enabled you to work out which systems needed modding would attract attention and I’d have thought there would be hundreds or thousands of systems.
Simply going all-drachma [using the existing IT system of BoG and commercial banks] is comparatively unterrifying (more or less a rollback of what they did in 2001) but would still take a lot of highly visible lining up. Long bank holiday for sure, plus lots of Drachmas to distribute quietly to all sort of places in advance, with nobody ever, ever talking about it; cash economy for a bit. One of those times when you’d rather everything was on paper, still.
That entire discussion assumes that people will be there to run the IT system. Problems greatly compound if the ECB transfers all the staff out after cutting off ELA or even after the Greek Government comes in. To recap, let’s walk through what each option might look like.
Taking over the Bank of Greece
1. Seize the headquarters of the Bank of Greece.
2. Explain to the staff that the Bank is no longer following the directives of the ECB and salaries will be paid in physical Tax Anticipation Notes and that they can deposit them in the bank once a payment system has returned which requires their diligent effort.
3. Declare a bank holiday.
4. Assign relatively knowledgeable Government officials to meet with subdivisions of the Bank to lead discussions about what needs to be changed in the payment system to convert from Euros to ETAN in their sections.
5. Start dispatching Government officials to private banks after receiving information about the Bank of Greece side. Inspect their IT system and start preparing them for IT changes to introduce ETAN. In the meantime, bring in auditors to examine their books and monitor for transactions.
6. Compare the estimated timelines to what needs to be in place for trade to resume and production to continue. Possibly recruit IT people from private banks sympathetic to the cause to bolster the amount of people working around the clock.
7. Contact the ECB and let them know what is happening and that the Government hasn’t decided to repudiate Target balances. Request confirmation that paying interest on Target balances will be sufficient to maintain access to Target2.
Ignoring the Bank of Greece and Setting up a New Central Bank
1. Declare a bank holiday.
2. Find a building to place a number of IT systems that have been preprogrammed for this eventuality. Start hiring staff immediately.
3. Immediately send officials to outside the Bank of Greece, announce over the speakers what is going on and request any officials loyal to Greece or who at least care for the suffering of Greece to leave the building and join them to rebuild the Greek Payments system. Do this indefinitely until most staff have left Greece or joined up.
4. Send officials out to Recruit retired bank IT people and those at private banks.
5. Send officials to private banks to demand that they disconnect from the BoG payments system and start working out the logistics for switching to the new system. Inspect their IT system and start preparing them for IT changes needed to introduce ETANs. In the meantime bring in auditors to examine their books and monitor for transactions.
This isn’t anywhere near a comprehensive checklist of things they would need to start doing, nor is every strategy necessarily the right one for the particular conditions they emerge from (for example paying interest on Target2 balances or converting euro deposits to ETAN deposits). In particular ETAN have a lot of legal benefits for Greece and make a possible rejoining of the Eurozone simpler, but might have even more difficulty gaining acceptance on the foreign exchange market then drachma would. I’ll turn to the foreign exchange market issues of introducing a new currency in part 3.
Moreover, there is the issue of the large volume of Euro banknotes that have already been withdrawn and which will likely be hoarded and only used for emergency purchases like seeing the doctor or, in the worst case, food. This will strengthen the already relatively strong black market economy and undermine the imposition of drachma or ETAN tax liabilities, a necessary precondition for their widespread adoption. There is also the issue of contracts where in one case they need to all say drachma when they said Euro and in the other they need footnotes saying when the Euro payments system is broken ETAN are legal tender. That is an enormous job to say the least and it needs to be semi-functional by the time they reopen the banking system. What I think this exercise does illustrate, however, is how many discrete activities would need to take place to get the new system up and running, and how time consuming it would be. Further, I think one starts to get the sense of how much additional organizational effort watching the bank staff imposes.
Now one could try arguing that “All we need is actually to get enough drachma in circulation to have things start functioning.” See the Addendum at the bottom of the post on why that’s not trivial either, even assuming you’ve got the new currency ready to go. And remember, having the economy go into an even deeper freeze due to a protracted bank holiday and the imposition of capital controls (which will also hurt tourism during the peak season) will not merely reduce incomes all over the country, increasing distress, but will lead to business failures, so a meaningful portion of the damage will persist even after the money system starts functioning again.
Nearly a century ago the Bolsheviks faced almost an identical problem. In December of 1917 Lenin made this statement to the Central Executive Committee to justify nationalization of the banking system:
Among the employees of the banks were men who held the interests of the people close to heart, and so they said: ‘The bankers are deceiving you, make haste to stop their criminal activity which is aimed against you.” And so we made haste. We know that this is a complicated measure. None of us, not even those who have a training in economics, will undertake to prepare it. We shall employ experts, people who are skilled in such work, but only after we have the keys in our own hands. Then we can even employ as consultants ex-millionaires…
We wished to follow the path of conciliation with the bankers, we supplied them with credit for the purpose of financing business, but they resorted to a sabotage of unprecedented dimensions, and so experience compelled us to adopt different measures in order to insure control… If we do not affirm the decree right now the bankers will take all the necessary steps to shatter our economy [Banks, Credit, and Money in Soviet Russia, by Arthur Z. Arnold, with a foreword by H. Parker Willis. New York: Morningside Heights, Columbia. University Press, page 60]
In this context Arthur Arnold comments, “Apparently the bankers had no difficulty in evading the control set up by the communists, whose ranks included very few who knew anything about the banking, business, accounting or auditing. Especially weak was the control in the provincial towns where the banks had branches.” In other words, they didn’t have the operational capacity to prudentially regulate the banking system and preserve the integrity of the payments system so a nationalization was the only feasible option. The question of nationalizing the banking system is one I’ll return to.
Discussion of capital controls returns us to the argument Lenin makes above. The IT issues with restarting a banking system in ETAN (or Drachmas for that matter) and making sure this new system doesn’t undermine their attempts to survive the crisis might be so complex that the Greeks might have to nationalize the banking system. It may be that no other strategy could successfully transition to a non-Euro economy in the time frame required. Thus, at least temporarily, deposits would be accounts directly held with the central bank and the central bank would make direct decisions about industrial projects to finance. This doesn’t end the organizational capacity needed, it just minimizes it. While they’re rebuilding the payments system they need to make loans in physical cash or large denomination scrip (to pay suppliers), start figuring out which firms owe what to which banks, make it clear that the payments system freeze doesn’t mean they are in arrears or that the debts end. Direct financing of production and investment by Government many precedents but the one that strikes me the most (and one I’ve wanted to write about another time) is the Defense Plant Corporation, a subsidiary of the Reconstruction Finance Corporation set up in World War II that financed a lot of war production and the creation of new necessary industries (like artificial rubber).
Greece has this series of policy choices before it not because it is now run by “radical leftwingers”, but because during a crisis the most valuable resource a country has is its ability organize and reorganize society. Neither committed capitalists, committed socialists or committed fascists can afford to burn organizational effort holding onto bugaboos about private banks, the abolition of money, or any other pet policy.
It should not have to be said, but I feel compelled to explain why I’m spending so much time on the payments system. An orderly and efficient payment system is crucial to the continuance of economic transactions. It allows businesses to buy more inputs to production, enables retailers and shops to meet customer demands for goods and services, helps finance that demand, permits the speedy transfer of supplies around a geographic area, etc. When the payments system is disrupted, even for a short period of time, all of these vital activities cease or require special agreements at all levels to keep going at even a minimal, uneven level. Two years ago in Cyprus, an emergency bank holiday was declared and capital controls installed. The bank holiday only lasted for twelve days yet supply chains started drying up instantly. An ex-Cyprus central bank governor told the Guardian:
Supplies of food are being exhausted and there are cases of raw materials like iron and timber being held up in customs because importers don’t have the cash to pay for them … No one expected, myself included, that the EU, ECB md IMF, would behave like this. Cyprus has been treated very badly … Where is the solidarity principle that is supposed to underline Europe?
Even 6 months later after the banks had reopened and capital controls were loosened, businesses were still having trouble getting basic supplies. This happened because their working capital was largely frozen and/or written down in the bail-in, or they had made payments but their suppliers were frozen because their own working capital had been largely frozen and/or written down. It is important to emphasize that their nearly two-week bank holiday was only to resolve insolvent institutions, not to build or significantly modify the payments system. Cyprus experienced a storm. It is not an exaggeration to say that freezing the Greek payments system would be like a financial hurricane and this one certainly looks to be a category five.
On getting currency in circulation, hoisted from older comments by Naked Capitalism regular Clive who has worked with ATMs:
Way way a-back when dinosaurs roamed the Earth and I was a junior in my TBTF I worked in the ATM Operations unit. It was a big, big unit. The logistical effort is considerable in maintaining an ATM estate — and that is under BAU conditions. Cash is bulky. The security requirements are considerable. The ATM hoppers are designed for specifically sized notes (the specification is exacting — if the notes are not precisely the same height, width and, crucially, paper grade, the ATM will jam up quickly; this applies to note counters as well). You can’t just max out high denomination notes either — a functioning cash currency needs sufficient quantities of various notes and coinage in the right proportions. Physically stocking, balancing and making an ATM ready for service once depleted takes a finite amount of time and that time isn’t trivial. And most ATMs are “dual control” — you need two operators for fraud prevention. Like I say, for a single TBTF to maintain a network of ATMs takes several hundred head office staff and a lot more in-situ staff. Then you need armoured cars, guards, access to premises. That’s many hundreds more needing to be planned, rostered and actually turning up each day. It is a major undertaking — and this is under normal, day-to-day operational conditions. Factor in secrecy (which would be inevitable, certainly in the run-up until the last few days anyway) and it is even more difficult.
And then you have self-service tills in supermarkets, vending machines, ticket machines and so on. Again, the specification of the physical currency is critical. For stores, marking prices is a labour-intensive job.
If Varoufakis were here having a cup of tea with me, I’d tell him that, from first-hand experience, don’t fcuk with the currency unless you absolutely have no options left whatsoever. From what he’s pronounced — and as Yves has repeatedly said, this has been his consistent message — I think he gets that already….
Like I already said, the physical cash handling is just one aspect. And to take your example, okay, unloading the cash from the truck takes 10 minutes. But how does the cash get onto the truck ? Where are the regional distribution centres ? How much cash goes to which ATM ? In what denomination ? Who is the hand off to once the cash gets off the truck ? And when the new currency is loaded, the ATM has to be taken offline until the old currency to new currency switch over date — so the ATM network is progressively disabled as old currency is swapped out to the new currency.
Yes, there is infrastructure, processes and people in place to handle this task now (how else would the ATM network keep going). But it is sized for normal operating parameters! A typical ATM will run for between 3 to 10 days between replenishment (some really busy sites like stations or shopping malls after a weekend will run “dry” in a day, but these are the exception). So everything is geared up to service the ATM’s cash needs in a gradual, progressive fashion. I can tell you without any doubt at all there is not enough pre-existing capacity in the cash holding centres, the fleet of transport vehicles, trained people or the IT systems to cope with a currency swap out in a weekend or even a whole week.
And that is before you get to physical constraints like having to redesign the ATM hoppers. Creating a new currency with the same specification as the old one is a non-starter. It is called counterfeiting.
Put it this way — this has all happened before. The Euro countries swapped over from their previous national currencies to the new Euro. The planning and implementation took the best part of a decade to do in an orderly way. You’d be talking weeks or months — best case — for Greece to do the same.
The last thing Greece needs is more chaos.