Sunday, August 12, 2018

Lira Crash

No, a currency board won't save the Lira, contra Steve Hanke's oped in the Wall Street Journal. Steve:
Turkey should adopt a currency board. A currency board issues notes and coins convertible on demand into a foreign anchor currency at a fixed rate of exchange. It is required to hold anchor-currency reserves equal to 100% of its monetary liabilities,...
Well, that sounds reasonable no? If 100% of the country's currency and bank reserves are backed by US dollars, and the currency is pegged to the dollar, what could go wrong? Don't want Lira? The central bank promises to exchange 1 Lira for 1 dollar and always has enough dollars to make good on the promise. It sounds like an ironclad peg.

Government debt is the problem. Turkey may still have the resources to back its currency 100% with dollar assets. But what about the looming debt? Turkey does not have the resources to back all its government debt with dollar assets! If it did, it would not have borrowed in the first place.

So what happens when the debt comes due? If the government cannot raise enough in taxes to pay it off, or convince investors it can raise future taxes enough to borrow new money to roll it over, it must either default on the debt or print unbacked Lira.

I.e. a currency board run by an insolvent government will fail. The government will eventually grab the foreign reserves.

The Argentinian currency board did fail, and this is basically why.

It's worse in many countries including Turkey for two reasons. One, the government borrows in dollars. It cannot devalue this debt by inflation, so the inflation required to devalue the rest of the debt is higher. From the WSJ editorial, 
A country borrows too much to spur growth in an era of low interest rates and easily available credit. Much of that debt is in U.S. dollars, but the cash flow to finance it is earned by local companies in local currency. By some estimates about half of all Turkish debt is owed in hard currencies.
WSJ is mixing government and private debt here, but they are entwined. When companies borrow in dollars against local currency revenues, they become vulnerable to devaluation. When that happens, the government either bails them out or watches the country collapse. So private debt in dollars becomes government debt, also in dollars.

The currency board can work, if it is part of a package by which the government commits to solve its fiscal problems, either by tax increases (usually, not likely as if there was that much tax revenue around, the government would have already grabbed it), spending cuts, defaults, or some means other than inflation.  But it is the fiscal package, not the currency board, doing the work.

This really is where we differ:
Government finances, state-owned enterprises and trade need not be reformed before a currency board can issue money.
Oh yes they do. Otherwise everyone knows the board will fail. The board can only succeed if it is part of a reform of all the above.

You can see a foundational difference. Steve thinks of inflation as coming from money alone. Control money creation, you control inflation. I think in terms of fiscal theory of the price level. You have to control all government debt to control inflation, sooner or later.

Turkey right now is also a good example to keep in mind for the vast majority of the economic establishment that thought it awful that Greece didn't have its own currency, so that in its fiscal troubles it could do exactly what Turkey is doing now. And, like pre-euro Greece did many times before.
This is nothing new. Inflation has ravaged Turkey for decades. The average annual inflation rates for the 1970s, 1980s, 1990s and 2000s were 22.4%, 49.6%, 76.7%, and 22.3%, respectively.
Those horrendous numbers mask the periodic lira routs. In 1994, 2000-01 and the past few months, the lira has been torn to shreds.
None of this brought great prosperity.


  1. Fiscal theory of the price level only holds because the alternative is experiencing a nasty recession. If turkey backs the lira dollar for dollar and commits to exchanging 1:1 whenever the price deviates, then it cannot devalue. But it's highly unlikely that Turkey with unemployment high and output falling would not try to devalue to save its economy. In the end, you have to pick your poison: either devalue or let your economy crash (temporarily).

    1. That's the opposite of what Sargent reported in "The Ends of Four Big Inflations". All you have to do is back your money, and issue enough to keep people liquid, and you avoid both inflation and recession.

    2. Sargent was making the opposite point. In Sargent's cases, countries that solved their fiscal problem immediately saw inflation decline, with no monetary contrivances needed. Most were able to print more money after the fact, as real money demand increases without inflation. They did not adopt currency boards or other clever schemes. Perhaps common ground is, yes, but the only way to back your money credibly is to solve the fiscal problem.

  2. Just make the Turkish central bank independent of the Turkish government, so the government can't rob the central bank. Then peg the lira at $1=1 lira, and have the central bank purchase $1 of US bonds for every lira issued. Those US bonds will earn some interest, so the bank will have tome earnings with which to pay its operational expenses. That way, the bank won't go broke like the old currency boards did.

    1. "Then peg the lira at $1=1 lira, and have the central bank purchase $1 of US bonds for every lira issued."

      So you are saying that Turkey should adopt the US dollar but call it a lira. And how would Turkey pay for the dollar denominated assets that the central bank would hold to back the new currency?

  3. I cannot believe that somebody, in 2018, argued for a currency board on a large economy with free capital movement. It is absolutely insane. I agree with the fiscal theory of the price level (and developed the argument for the lay person in my forthcoming book) but we don't even need it to rule out a currency board. We don't even need good models of speculative attacks. Just basic macroeconomic common sense. No currency boards! Hanke is 100% wrong.

  4. Fun fact: Turkey’s public debt to GDP ratio is significantly lower now than in the past (40.1% in 2010, 28.3% in 2017). Since they actually increased their current account deficit during this period, who in Turkey was racking up all the debt? As nearly as I can tell from googling all the usual suspects, a huge chunk of the foreign currency debt is owed by non-financial corporates and to a lesser extent non-government financials. (If I’ve misread the numbers, administer dope-slap now.) As John points out in the post, if the government is going to guarantee the private debt, none of this matters. A currency board can’t solve the problem. But if the government can credibly commit to allowing a bunch of these firms to fail (and note, that might include some banks), it could just work. Turkey is a mess but it’s not nearly as big of a mess as, say, Zimbabwe or Venezuela. Unlike those places, Turkish fiscal policy is only a little bit imprudent and so they can easily make a credible commitment to a slightly more cautious fiscal policy.

    The much more interesting question—which is the same question we should ask about Greece, Argentina and every other such place—is what’s happening in the financial sector to let these kinds of problems get so big? I’m pretty sure the answer is that it’s mostly the result of bad regulation (notice, I didn’t say not enough regulation) that creates moral hazard and all sorts of unfortunate incentives.

  5. "vast majority of the economic establishment that thought it awful that Greece didn't have its own currency,"

    If I understand correctly the new government in Italy wants to abandon the Euro for precisely the same reasons people thought Greece should abandon the Euro.

  6. Ditto. I love Steve Hanke from my Mexico days, but world has changed dramatically. Local deficits funded in dollars. His op-ed illustrates your point on fiscal vs monetary factors behind inflation. The US is the case in point. The Fed talks about inflation, but the Treasury deficits are ultimately the engine. Thus as QE unwinds, we reduce monetary base and bank deposits as burden of public debt finance if shifted back to private sector from Treasury alter ego -- namely the Fed. May be time to resurrect "crowding out" as a mainstream economic theory.

  7. Well-meaning US first-rank university economics professors are an independent source of risk for over-indebted countries. A very recent instance is Greece. Stiglitz, Sachs, Krugman and James Galbraith, while sitting in their offices and secure in their tenure, kept giving false hopes to the Greeks that their corrupt and clientelist state could keep going for ever longer, by some, supposedly clever, combination of blackmail and monetary and exchange tricks. They burdened the Greeks with the huge cost of delaying adoption of the necessary fiscal and other reforms, but did not bother offering an apology, except a half-baked one by Krugman.
    The deleterious effects of large capital inflows are extremely long-lived. Once abundant capital becomes available, competition forces domestic firms to invest in marginally profitable projects. The ones who resist, get acquired. They shift investment from the production of tradeables, that are expoosed to international competition, to the production of non-tradeables, hence sheltered from the cold winds of international markets. Thus, even in the absence of fiscal profligacy, market forces push the recipient of capital inflows into crisis.
    A good account of the process is the one by the redoubtable Ioannou brothers :

    Dimitris Ioannou argues forcefully, if light-heartedly, that the real default is not financial, but intellectual.

    George J. Georganas

  8. "Well-meaning US first-rank university economics professors are an independent source of risk for over-indebted countries." This! Perfectly encapsulates the Hanke op-ed.

  9. John,

    Thought this would be of interest to you:

    "In addition, he played a central role in drafting and bringing about the inclusion of the so-called Hanke Amendment in the 1993 Foreign Operations Appropriations Bill. This measure, sponsored by Senators Phil Graham, Bob Dole, Connie Mack, Jesse Helms, and Steve Symms allowed U.S. contributions to the International Monetary Fund to be used for the purpose of establishing currency boards."

    Seems like economists and political leaders from the right (Graham, Dole, Mack, etc.) are all in favor of currency boards.

  10. What did you're favorite dog have to say about currency management? "Oh Yea: A sovereign (Treasury combined with the Federal Reserve Bank), like the US, that:
    a. issues,
    b. borrows in, and
    c. floats
    its own currency, can NEVER run out of cash."

    Don't borrow in a currency not your own. If you're private sector does so, and currency collapses, let the companies go under, but inject sufficient local currency to maintain full employment and have companies recover under new ownership.

    Don't know how Erdogan is going to handle this, but the Pup's guess is we will have moved on to another story within a week. Total nothingburger.

    1. Once again Greece, been there, done that, in 1973-1994. This injection of money "to maintain full employment" ends up propping countless non-viable businesses and wholly bankrupt banks (they are banks only in name). Nobody wants to hold the currency, hence one needs capital controls to stem its continuous devaluation. Capital controls generate further distortions and corruption. Any truly productive economic activity goes underground. Parasites flourish in business, especially any well-connected ones, who will hire the employees designated by the one who provides the loans for "full employment". Real interest rates turn negative, thus a disincentive for firms to earn profits and accumulate capital. Savers protect any savings by buying foreign currency at the black market. The list goes on. But, yes, most probably this particular mess is Turkey's problem and not the rest of the world's.
      George J. Georganas

    2. Greece borrowed in foreign currency during this period.

    3. Not really in the period 1973-1994 and certainly not in the scale in which it was able to borrow, once it was in the eurozone. It was not for want of trying. Simply, nobody would lend it any serious money at reasonable rates of interest.
      George J. Georganas

  11. "should" to what end? Governments do not benefit 1:1 with the well-being of their subjects. And, the individuals working for governments don't benefit 1:1 with the well-being of the government.

    People in the Turkish Government will continue to do what is good for them, Turkey be damned. They know what's good for their subjects. That's not the problem.

    There is a longstanding correlation between corruption and inflation. The puzzle here is a political one, not an economic one.

  12. Erdogan has been trashed for claiming that the lower the interest rate, the lower will be the inflation rate. Correct me if I'm wrong, but isn't that the exact same economic line taken by the Grumpy Economist himself?

    1. I don't think so. See p. 91 of "michelson morley fisher and occam" here
      which replaces the paper you cite. (And thanks, that one is old and I need to take it down.)

      Briefly, fiscal policy underlies everything. If the government is broke, you're gong to get inflation. For lower rates to lower inflation, the rate decline has to be long lasting, people believe it's long lasting, you have to wait for a possible transitory movement in the other direction, and fiscal policy has to be running surpluses. Turkey seems to be replaying some of the forces of the 1990s asian crises -- private companies borrow in foreign currency, and the government explicitly or implicitly backs them.

    2. Professor:

      RE: "... If the government is broke, you're gong to get inflation. ... ..."
      • What the hell does this mean? How can a govt that issues, borrows in, and floats its own currency ever be broke? If it wanted more cash and inflation was inching up, what would prevent it from raising taxes?

      RE: "... For lower rates to lower inflation, the rate decline has to be long lasting, people believe it's long lasting ... and fiscal policy has to be running surpluses. ..."
      • Lower interest rates reduce the amount of net financial assets injected into the economy, reducing the amount of potential spending. Its not a huge effect on spending but its definitely in the down direction. (These may also be offset by the upward pressure on spending by leveraged companies who may now have more money to spend and may want to invest more.)
      • Whether fiscal policies are in deficit or surplus is irrelevant. Its the direction and affect. Whether the affect is sufficient to offset other forces is a different story.

      RE: "...Turkey seems to be replaying some of the forces of the 1990s asian crises -- private companies borrow in foreign currency, and the government explicitly or implicitly backs them. ... ..."
      • The foreign private sector borrowing has already been executed. Did that have anything to do with Erdogan? - interesting question.

  13. John,

    "For lower rates to lower inflation...and fiscal policy has to be running surpluses."

    Fiscal policy must reduce the outstanding debt - surpluses / deficits are ancillary.


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.