Tuesday, March 26, 2013

Don't shackle Target2

Like Guntram Wolff over at the Bruegel blog, I hope that the much-rumoured capital controls on Cypriot deposits don't get enacted. So far the Euro authorities seem to have done everything right, albeit in a slow and circuitous manner. Insolvent banks are being closed, uninsured depositors, unsecured creditors, and shareholders are being bailed in, and solvent banks are slated to reopen.

Wolff's main concern is that capital controls threaten the very meaning of a monetary union:
With capital restrictions, the value of a euro in Cyprus is no longer worth the same as a euro held by any other bank in the eurozone. A euro in Nicosia cannot be used to buy goods in Frankfurt without limits. Effectively, it means that a Cypriot euro is not a euro any more.
Enact capital controls and we'd see the emergence of an entirely new currency trading pair CYP€:onshore€, with Cypriot euros trading at a discount. The discount would emerge since the ability of CYP€ to buy things outside of the island of Cyprus is limited. It would be a less liquid euro than "mainland" euro, and therefore would get penalized with a liquidity discount.

A Eurosystem in which euros are heterogeneous would technically be workable. For an analogy, look at China. The Chinese yuan has several different prices. Mainland yuan (CNY) typically trades at a discount to yuan in Hong Kong(CNH) and yuan in Taiwan (CNT). I've cribbed a chart below that shows the spread. Chinese capital restrictions prevent arbitrage forces from reducing the gap. Foreigners would prefer to buy cheaper CNY than more expensive CNH and CNT, but they can't because restrictions on capital inflows into China prevent them from doing so. Chinese companies would like to borrow in Hong Kong rather than China since they'd be borrowing high-value CNH and repatriating it, thereby lowering their cost of funding. But capital outflows are also limited.*

Source: HSBC Global Research

Just like capital controls prevent the closing of the CNY-CNH spread, the introduction of European capital controls would lead to the emergence of the CYP€-€ spread. What are the dangers of Europe adopting a Chinese model of multiple prices for the same currency?

As Wolff points out, the Eurosystem already has a tool to deal with flight from banks: the ECB's incredibly powerful Target2 clearing system. When Cypriot banks reopen and depositors start to transfer deposits to Germany, Target2 will accommodate these flows by stepping between the two banking systems, simultaneously acting as a creditor to Cyprus and a debtor to Germany. Like any lender of last resort, Target2 will be vigorous in lending, providing whatever assistance is required to Cypriot banks facing liquidity shortfalls.**

The great thing about Target2 is that its mere presence has the ability to prevent a bank run from ever being kick-started. If Cypriot depositors know at the outset that the incredibly powerful Target2 will accommodate their fears, why should they be fearful? Target2 is like Chuck Norris, as Nick Rowe and Lars Christensen would say. Its mere presence is enough to create powerful self-fulfilling counter-effects.

Cyprus wants to soften potential deposit flight with capital controls rather than leaving Target2 to do all the work. One wonders if these controls would help at all. Controls are porous, and investors will find cunning ways to get around them.

Worse is the precedent this sets. If capital controls are used as a substitute for Target2, the Euro risks losing a major stabilizing force come the next crisis. Say that it is 2016 and doubts spring up concerning Finland's banking system. If Finnish depositors know that Target2 will accommodate all deposit outflows from solvent Finnish banks, then they realize that they have nothing to fear, and the panic will subside on its own accord. But if Finnish depositors think that Cyprus-style capital controls will be put in place to prevent deposit outflows, the panic will only be exacerbated as people try to withdraw money from Finland before capital controls cause FIN€ to trade at a discount. Anyone who gets through the gate before it closes can't lose, so everyone tries to get through the gate. This effect is perverse, since the very rumour of capital controls leads to their actual adoption. With capital controls, a European bank panic is self-fulfilling—with Target2, that same panic is self-correcting.

Leave Target2 free to be the regulator of European liquidity flows—don't use capital controls. Haven't we already learnt this lesson? It was Draghi's speech about Euro convertibility from last summer that helped reduce yield spreads and stop the intra-European bank run. Gavyn Davies read that speech as a reaffirmation of unlimited Target2 power, and so did I.*** Never shackle Target2.

* The Chinese are moving to less capital controls. Spreads are already declining, and at some point the CNY-CNH/CNT differential will be no more. 

** The provision of these LOLR services is subject to Cypriot banks providing collateral. But the winding up of insolvent Cypriot banks and the haircutting of depositors *should* have insured that the remaining quantity of Cypriot banking liabilities have been pruned so that they equal the quantity of remaining collateral.

*** See this comment as well as my first Never Shackle Target2 post.


  1. I blogged on the idea of Bitcoin-assisted arbitrage (not as eloquently as you):
    In the first post I identify potential for profitability that results from the controls and in the second I reach essentially the same conclusion as you with regard to the self-reinforcing effect of such controls.

    1. Interesting scheme there with bitcoin and capital controls. Thanks for sharing.

  2. Chuck requires an asymmetric payoff to work. In other words, the payoff from heeding his threat is much greater than that of ignoring it. This does not apply in the case of bank deposits.

    Take a Cyrpus deposit holder. The potential cost of ignoring Chuck's threat of unlimited ELA funding is a wire transfer and some time filling out applications for a German bank deposit. The potential cost of heeding the threat is 40%+ should the Cyprus bank experience a future haircut, devaluation, or capital control.

    Separately, Target2 poses enormous moral hazard for Germany, and Cyprus is a great example. The country emerges from its "bail out" with a defunct offshore banking system, an overvalued currency, and a crippling austerity. The Cypriot PM, realizing this, has every incentive to default/exit the Euro AFTER the ECB finances a bank run with further ELA assistance, and the EU contributes its 10b Euro loan. Its risky to make an unsecured loan to someone that has every incentive to declare bankruptcy as soon as they cash the check.

    1. This comment has been removed by the author.

    2. Diego, if the Chuck Norris effect isn't the right analogy, then call it the Moses effect. As long as an infinitely powerful being ensures that there is always a passage from A to B, then large groups of people will never flee from A to B based on expectations of that passage closing. Capital controls interfere with the Moses effect... the possibility that they might be implemented would create the very flight that Moses is designed to prevent.

      Europe really needs the Moses effect on its side, at least till it's more well-integrated.

  3. JP - one comment: if Cyprus has reached the limit on their allowed ELA (its hard to know for sure but they are likely close) then even if there were no capital controls Cypriots would have a very hard time sending out deposits. Cypriot banks would either fail because they couldn't access reserves or would have to pay high interbank rates from Euro banks outside Cyprus to facilitate their customers' requests (perhaps leading them to fail quickly). The banks could also sell their highly illiquid (most assets are bank loans) assets to fund withdrawals but this would also create a bad dynamic. Thus providing more ELA is also likely to be necessary to prevent more bank closures. The ECB has said they do not object to ELA continuing in Cyprus but I doubt they will provide more.

    So even if there were no capital controls Cypriots might have a hard time sending their deposits elsewhere via the Target2 system.

    1. Hi Will.

      If they're that close to the limit, I guess the ECB will just have to widen the range of collateral they accept and apply appropriate discounts. This famous line from Bank of England Director Harman on the crisis of 1825 comes to mind:

      "We lent it [pounds] by every possible means and in modes we had never adopted before; we took in stock on security, we purchased Exchequer bills, we made advances on Exchequer bills, we not only discounted outright, but we made advances on the deposit of bills of exchange to an immense amount in short, by every possible means consistent with the safety of the Bank, and we were not on some occasions over-nice. Seeing the dreadful state in which the public were, we rendered every assistance in our power."

      Here's another quote from a new blog that I recently discovered:

      "If you can get accurate prices and take an appropriate haircut...then you can Repo anything"

      The ECB should start accepting with haircuts stocks of olive oil, Cypriot farmland, you name it. I'm exaggerating obviously, but you get where I'm coming from. There's always good collateral out there somewhere, and if there isn't, haircut it till it is.

  4. Had a similar question. Why does the Cypriot government fear bank runs? It must not be a liquidity issue. They must be thinking of devaluing or the rest of the Cypriot banking system must be more insolvent than publicly acknowledged. Perhaps the system is stuffed with underwater vacation real estate loans that will only worsen with the coming depression, and in that case the only solution is devalue/exit to save the tourism industry.

  5. You don't understand the real issue for depositors & creditors. If your account is located in Netherland, Germany, Austria in a too big to fail bank implicitely backstopped by the local government you'll have a better treatment. This holds true even if you are a senior creditor. So it's very rational to reshuffle your portfolio and move your money in a country where you could hope to be bailed out. Obviously this could sound like a self fulfilling prophecy, but it's a fact some states could support their financial system some states couln't: credit market is already balkanised in the EZ.
    While you may assume Mrs Merkel won't let Deutsche Bank to fail *whatever it takes*, the same isn't true for peripheral banks: there you know your money will be at the mercy of what Eurokrats.