Saturday, January 24, 2015

Grexit: An Escape to More of the Same

The upcoming Greek election has renewed interest in the idea of Grexit. This option is often presented to the Greek public as desirable given that it would restore an independent monetary policy to the nation.

Beware, this is dangerous advice. The euro isn't a glove that you can take on and off, it's a Chinese finger trap; once in, it's tricky to get out. Even if Greece were to formally leave the euro, odds are that it would remain unofficially euroized, leaving it just as bereft of an independent monetary policy as before. The real trade off in a Grexit-or-not scenario is between formal membership in the euro with some say in monetary policy, no matter how small, or informal membership without any say whatsoever.

The optimists, say someone like Hans-Werner Sinn, advise the Greeks to leave the euro and adopt a new currency. The value of this new drachma would immediately collapse. As long as prices in Greece are somewhat sticky, Greek goods & services will become incredibly competitive on world markets, spawning an export/tourism-led recovery. By staying on the euro, however, Greece forfeits the exchange rate route to recovery. Instead, Greece's competitiveness can only be restored via a painful internal devaluation as wages and prices adjust downwards.

While the optimists tell a good story, they blithely assume a smooth switch from the euro to the drachma. Let's run through the many difficult steps involved in de-euroization on the way to an independent monetary policy. All euro bank deposits held at Greek banks must be forcibly converted into drachma deposits, and speedily enough that a bank run is preempted as Greeks desperately try to evade the corral by moving euros to Germany. At the same time, the Bank of Greece, the nation's central bank, needs to issue new drachma bank notes, the public being induced to use these drachmas as a medium of exchange.

Now even if Greece somehow pulls these two stunts off (I'm not convinced that it can), it still hasn't guaranteed itself an independent monetary policy. To do so, the drachma ₯ must also be adopted as the unit of account by the Greek public. Not only must financial markets like the Athens Stock Exchange begin to publish stock prices in drachmas, but supermarkets must be cajoled into expressing drachma sticker prices, employees and employers need to set labour contracts in terms of drachmas, and car dealership & real estate prices need to undergo drachma-fication.

Consider what happens if drachmas begin to ciruclate as a medium of exchange but the euro remains the Greek economy's preferred accounting unit. No matter how low the drachma exchange rate goes, there can be no drachma-induced improvement in competitiveness. After all, if olive oil producers accept payment in drachmas but continue to price their goods in euros, then a lower drachma will have no effect on Greek olive oil prices, the competitiveness of Greek oil vis-à-vis , say, Turkish oil, remaining unchanged. If a Greek computer programmer continues to price their services in euros, the number of drachmas required to hire him or her will have skyrocketed, but the programmer's euro price will have remained on par with a Finnish programmer's wage.

As long as a significant portion of Greek prices are expressed in euros, Greece's monetary policy will continue to be decided in Frankfurt, not Athens. Should the ECB decide to tighten by lowering interest rates, then Greek prices will endure a painful internal deflation, despite the fact that Greece itself has formally exited the Euro and floated a new drachma.

We know that a unit of account switch (not to mention successful introduction of drachma banknotes) will be hard for Greece to pull off by looking at dollarized countries in Latin America. To cope with high inflation in the 1960s and 70s, the Latin American public informally adopted the U.S. dollar as an alternative store of value, medium of exchange, and unit of account. Even after these nations' central banks had succeeded in stabilizing their own currencies, however, dollarization proved oddly persistent. This is referred to as hysteresis in the economics literature. Economists studying dollarization suggest that network externalities are the main reason for hysteresis. When a large number of people have adopted a certain standard there are significant costs involved in switching over to a competing standard. The presence of strong memories of past inflation may also explain dollar persistence.

In trying to de-euroize, Greece would find itself in the exact same shoes as Latin American countries trying to de-dollarize. Greeks have been using the euro for 15 years now to price goods; how likely are they to rapidly switch to drachmas, especially in light of the terrible performance of the drachma relative to other currencies through most of its history? Those few Latin American countries that have successfully overcome hysteresis required years, not weeks. If Greece leaves the euro now, it could take decades for it to gain its own monetary policy.

As an alternative illustration of the power of network externalities, consider the multi-year plans made by Slovakia (pdf, fig 2) prior to switching over to the euro, or the Czech Republic's timeline when it makes the changeover. Each step must be broadly communicated and telegraphed long ahead of time so as to ensure that all members of a nation are properly coordinated, thus ensuring the network effects engendered by the incumbent currency can be overcome. These euro changeover plans weren't adopted a few days before the switch, but often as much as a decade before.

In sum, I fail to understand how Greece can ever expect to enjoy the effects of a drachma-induced recovery if the odds of drachma-fication or so low, especially given the sudden nature of a Grexit. At least if it stays part of the euro, Greece has a say in how the ECB functions thanks to the Bank of Greece's position in the ECB Governing Council. And at least Greece's inflation rate and unemployment rate will be entered into the record as official data worth considering by ECB monetary policy makers. For just as the Federal Reserve doesn't consider Panamanian data when it sets monetary policy (Panama being a fully dollarized nation), neither would the ECB care about Greek data if Greece were to leave the euro, though still be euroized.

Basil Halperin responds.


  1. If you think about how Greece could successfully introduce a new currency, it's pretty clear that there's only one practical way: by credibly pegging it to the euro. Eventually, after a few years perhaps, they could consider floating it and regain monetary policy control. But that won't help with the current crisis.

    So I agree completely and I'm surprised by the number of professional economists who suggest that Greece can benefit (short term) from leaving the euro. It's bad advice.

  2. All the issues mentioned here have little merit.
    1. "All euro bank deposits held at Greek banks must be forcibly converted into drachma deposits, and speedily enough that a bank run is preempted".
    Forcible conversion is probalby unnecessary. The big money has already fled from Greek banks. Much of that which remains is likely to be voluntarily converted to drachmas in order to pay taxes and otherwise do business in Greece. This will be especially the case if drachma prices are lower than Euro prices.

    2. "The Bank of Greece... needs to issue new drachma bank notes".
    Yes, but this not a serious problem. With competent planning, drachmas could be printed in advance, and then be quickly put into circulation.

    3. "euro remains the economy's preferred accounting unit, even as Greek drachmas begin to circulate as a medium of exchange"..."hysteresis".
    This discussion is a bit confused.
    It seems to be recognised that drachmas would immediately become a medium of exchange. This because all government expenditures, transfer payments and tax payments would be in drachmas.
    The use of Euros as a medium of exchange for domestic transactions would quickly decline to almost zero because of Gresham's Law.

    Regarding the unit of account, maybe prices and wages would be expressed in both Euros and drachma initially. Or in US$. Or any other unit. It really doesn't matter what is the unit of account.

    What matters is the level of Greek prices and wages, expressed in whatever unit of account is preferred. In contrast to the current arrangements, with Grexit "Greek goods & services will become "competitive on world markets, spawning an export/tourism-led recovery". I would add to this, that domestically produced import substitutes would also become more competitive within Greece.

    1. "The use of Euros as a medium of exchange for domestic transactions would quickly decline to almost zero because of Gresham's Law."

      Do you know what Gresham's law means?

      And of course the use of euros would decline to almost zero, just like the use of U.S. dollars rapidly declined to zero in Latin America.

    2. My error - Gresham's Law is irrelevant.

    3. Sorry to butt in here a day late and a drachma short, but why is Gresham's law irrelevant? If you have two circulating currencies, then Gresham's law (if I understand it correctly) states that the bad money (here, drachmas) will drive out the good (Euros), in this case likely due to hoarding. What am I missing?

    4. Gresham's law doesn't apply when two currencies circulate at floating rates to each other. Only when the government or some other body requires that merchants accept them at a fixed rate to each other does it begin to have an effect. If the stipulated fixed rate overvalues one of the currencies, then the undervalued currency will disappear into hoards.

      If the Greeks were to introduce a new drachma, it would likely float against the euro, so Gresham's law doesn't apply.

  3. I'm not sure I fully understand why a Grexit wouldn't be substantive. Wage contracts, pensions, rental agreements, loan repayments, Greek treasury bond servicing, etc are all in Euros now. Let's imagine that Greece decrees that all of those contracts swap over to being drachma denominated as of Monday. Then on Monday the Drachma floats, surely those contracts require some sort of legal recourse if the creditors want to enforce Euro denominated payments. As far as I can see it all boils down to who controls those laws.

    As far as I can see, whether a shop displays current prices in Euros or Drachmas isn't the crucial point, it's existing long running payment contracts that really make the difference. Its the burden of those contracts that is hobbling Greece.

    The switch over to the euro entailed just such a swap in the denomination of pre-existing contracts didn't it? So it does have a legal precedent.

    Perhaps an example is the swap from the Reichsmark to the Deutsche Mark in Germany in 1948

    1. "I'm not sure I fully understand why a Grexit wouldn't be substantive."

      The point of this post was very specific, about the ability of Grexit to lead to an independent monetary policy. It's not a comment on debt write downs. Having an independent monetary policy comes down to the control of an economy's price level. Whether a shop displays current price in euros or drachma is a therefore a crucial point. If only a portion of an economy's population of prices are expressed in drachmas, then that country's central bank only partially controls the price level.

    2. Thanks, sorry for not grasping that.

      But perhaps then the point is that it isn't independent monetary policy that is the crucial issue here.

      Perhaps the crucial thing Greece needs is for existing contracts to be in Greece's own free floating currency and for Greece to have its own central bank standing by as a buyer of last resort so as to ensure that Greek treasury bonds don't have default risk.

      Many small countries (such as Singapore) just have rough exchange rate targets as being what largely governs monetary policy (rather than inflation targets). And yet they have functioning fiat currency systems whilst Greece frankly doesn't.

    3. If it was easy for existing contracts, especially rent and wages, to be re-denominated from a stronger to a weaker unit, wouldn't the Latin American governments have been able to escape dollarization decades ago? The reason that a literature has sprung up on dollarization, de-dollarization, and hysterisis is because shifts like these don't occur between Sunday night and Monday morning.

    4. I'm just trying to get my head around this. I really appreciate your blog and the effort you take in explaining all of this.

      My understanding was that the Latin American governments did not have the political power to cause contracts to change. The military in those countries was beholden to the creditor interests. Perhaps even the CIA would have taken part in regime change had such measures been taken. But it really wasn't down to what customers and industry in those countries were needing in terms of a workable currency arrangement. Employers could have hired workers readily enough with local currency wage contracts. Suppliers would still have wanted to sell goods even using local currency.

      A switch from a currency to wriggle out of insurmountable debts was done in Germany in 1948 after all wasn't it? I think it is all about the political power of the creditors. In Germany in 1948, that power was fairly non-existent. In Latin America it was huge. Perhaps in Greece now it is not so strong?

      I'm struck by how currently lots of Polish household apparently have mortgages denominated in CHF. I'm fairly certain that such mortgages would not be allowed to be marketed to retail customers in the UK. I think countries such as the UK (and I guess Canada too) take legal measures to ensure the national currency is used. I was interested to see though that this gold clause was upheld by a US court

    5. The 1948 example had the following features: 1) The existing currency was being canceled; 2) the existing currency was a failed one; 3) the new currency was to be supported by the strongest nation in the world. If Greece were to leave the euro, none of the same factors exit: 1) The euro is not being canceled; 2) the euro is not a weak currency; and 3) a new drachma would not receive US support. So 1948 Germany is a bad analogy to Grexit.

  4. In Latin America, I think it was especially true that the local currencies were basket cases because the government debt was denominated in USD. That was the key step that ruined everything else as far as I can see. It meant that the governments had to exchange tax receipts that were in local currency for USD in order to service the government debt. If they had had local currency denominated government debt (as a developed world government would have), then the local currencies would not have been sliding. Then local people would not have wanted to have financial assets denominated USD and so the local currency would have been sought after and as a result stable.

    1. The UK had a very brief dabble with USD denominated government debt and it was an utter fiasco.
      It makes it all the more appalling that much of the world has used such a system for decades. Basically the Greek euro denominated government debt creates just the same problem doesn't it?

  5. "Economists studying dollarization suggest that network externalities are the main reason for hysteresis. When a large number of people have adopted a certain standard there are significant costs involved in switching over to a competing standard. The presence of strong memories of past inflation may also explain dollar persistence."

    I think it is more fundamental than that, it is a reflection of the strength of the respective economies themselves. Especially if these countries are major traders, "hard currency", that is a currency with 'currency' has to be earned. Import dependent countries need to buy the dollars with something, which means a competitive value added export sector.

    The problems in Greece do not really relate to government debt, that is really a symptom, but the structure of its trade. These problems were hidden in the pre-Euro days through devaluation, but this did not do anything for industrial progress.

    Greece's problem existed pre-Euro and during the Euro. Its problems go back a long way.

  6. As you point out, the euro has only been in circulation for 15 years. The idea that this short period of time has created an overwhelming social pressure or inertia in favor of the euro as a unit of exchange seems odd. Most living Greeks clearly remember a time when they used drachmas.

    If olive oil producers post a euro price for their olive oil for foreign customers and a drachma price for drachma-using domestic customers, and the price ratio doesn't reflect the euro-drachma market exchange rate, then market forces will compel the producers to re-price. And if most Greeks are being paid in drachmas, then obviously olive oil producers will be compelled to post drachma prices if they want to sell most of their olive oil.

    Why not nationalize the banks, including the Bank of Greece. Then "tax" all euro-denominated accounts at 100% and transfer all of those euros to government accounts. At the same time issue an equal number of drachmas and credit each of those accounts with drachmas through a direct transfer on a 1-1 basis. Everybody wakes up with drachmas in their accounts. Use the taxed euros to pay off euro-denominated government debts, and hold the rest as foreign reserves.

    Pass legislation that requires every labor contract and existing private domestic debt relationship to be re-interpreted in drachma terms 1-1. Companies will be compelled to accept payment for their products in drachmas in order to meet their own payment obligations to employees.

    Pass additional legislation by which the Greek government stands ready to assume all euro denominated debts owed abroad (credit card, whatever), and those private debts are then replaced with drachma debts to the Greek government. Basically, you submit your credit card bills and other bills from foreign firms to a government office. They pay with the taxed euros - then they charge the debtor a drachma debt on a government account. (What to to with those debts then becomes a matter of policy.)

    There are a lot of state run services in Greece. If Greece refuses to accept euro in direct payment for those services, or for payment of taxes, while offering euro-for-drachma exchange at par, they can reduce the domestic circulation of euro notes rapidly.

  7. I have some additional comments to how much effort it was/is to switch TO the Euro. During a switch, commercial banks are obligated to exchange the old currency for Euros AT A FIXED EXCHANGE RATE for several months (I recall half a year, but I'm too lazy to check). The central bank is obligated to do the exchange for a longer time (I recall that in some cases forever but I might be wrong). And the exchange rate is set a long time in advance (I recall about half a year).

    Several weeks in advance, there was a huge information campaign for people to familiarise themselves with the new coins and notes (there still are, albeit smaller, information campaigns as the plastic 5€ and 10€ notes were introduced recently). People could buy a sample set of coins/notes to try it out directly.

    Furthermore, in some countries (e.g. Slovakia), there was no meaningful political or public opposition to the switch to the Euro. There was no debate. Everyone basically tacitly acknowledged it. This made the switch easier.

    Compared to that, the logistics and PR in Greece appear much more complex. Greek population does not appear to want to stop using the Euros either. I guess we'll have to see what happens.