Snow fences for Europe

The debt mountain could use some snow fences, by the_junes

A recent post by Karl Smith  got me thinking about sovereign bonds: is it possible to use the disciplining power of the market in a currency union at all? Or in any country for that matter?

One argument against using the market is that you need to be willing to let countries default — with all that entails. According to a report by the German newspaper Süddeutsche Zeitung, Italy and France are now trying to remove private sector involvement (PSI) from the European Stability Mechanism (ESM) that is bound to replace the EFSF in 2013 because the PSI for Greece allegedly has caused contagion (I doubt that, but that is not the point today).

Another related argument against default (or PSI) is that sovereign borrowing works, until it doesn’t. In other words, the market does not discipline the sovereign through interest rates, but once it does, the self-fulfilling nature of the crisis leads to default. This mechanism, as far as I can see, works for countries with a central bank as lender of last resort as well. It will just hit at a later point in terms of debt load.

Karl proposes “administrative discipline”. Well, good luck with that. Not only is the record of European fiscal oversight dismal, it is also potentially a good way to destroy the friendship between countries in Europe, the original purpose of the European project. It might work, and maybe we need this leap of faith in Europe in the current situation, but without substantial treaty changes (with referenda and all that), it is almost impossible.

But I wonder whether there is, at least in theory, another market-oriented way. For instance, we could structure a country’s liabilities as those of banks, as the IMF suggested in 2004. Let me propose a 4-tier structure: equity, cocos, bonds and deposits.

  • “Deposits” would make up the lower, say, 40% of sovereign debt (relative to GDP). They would be senior to all other claims and the risk-free asset of that country.
  • “Bonds” make up the next, say, 20% of sovereign debt. They are slightly riskier than the deposits, but still a safe investment for pension funds and others given that we are just at the Maastricht criterion of 60% of GDP in debt at this point.
  • “Cocos” are contingent convertible bonds for the next 20% of debt, that convert to “equity” once a certain trigger event happens. Barry Eichengreen suggested cocos, but for slightly different reasons.
  • “Equity” is the most junior claim for debt above 80% of GDP. „Equity“ claims are effectively bonds as well. Its coupon payment could be tied to GDP growth of the country to give it a equity-like flavor, but still, there is a principal repayment and other bond-like features.

Regarding the “cocos”, the trigger would be that “equity” needs to take a haircut. Then “cocos” are converted 1:1 into “equity”, after the haircut on the original “equity” is determined. This could be very helpful because it leaves an empty 20% of the seniority structure that can be filled with new coco bonds that arguably will have a lower yield than the marginal equity. Countries in an acute crisis will probably not have a budget surplus and could therefore refinance their deficits through issuing new coco bonds, before having to issue new “equity”.

As far as I can see, there could still be a self-fulfilling “equity” crisis. But it would be limited to the equity part; the more senior claims, when they need to be rolled over, are much less affected and the “deposits” probably not at all (unless markets price in the probability of a Euro exit).

The mental picture I have in my head is that of snow fences against avalanches: yes, there may be an avalanche in one sector of the debt mountain, but it will be blocked by the first array of snow fences, or if it is larger, by the second. You can safely ski under the third.

This setup makes sovereign default a credible threat: it would give regulatory priority to “deposits”, forbids European banks to hold “equity” and thereby isolate the banking sector to a large extent from the sovereign. You also get market signals regarding the sovereign debt, or at least for “equity” and “cocos”.

A few final remarks:

  • Costs: it might not increase the overall costs, as the senior debt would be much cheaper. With the usual debt setup, there is always the danger of dilution such that more debt reduces the value of existing debt. This danger has a strict limit in this model.
  • Incentives: the incentive to borrow short-term is lower, as short-term debt will not be that much cheaper with explicit seniority.
  • Transition: The transition to this new world could be done right away: Spain or Italy would ask the IMF and the Paris Club for help in negotiating the transformation of existing sovereign bonds into these four different categories. Portugal and Greece would do the same. There would be some default on the equity in some countries, but the senior parts are protected, and new cocos could be issued.
    Other countries would make the slow transition, treating all existing bonds as “deposits” and issuing new “bonds”, “cocos” and “equity” over time.
  • Effects: the markets know in this model that some parts of sovereign debt are safe, and others are not. At the moment, the shift from having an ample supply of „safe“ assets to a situation where nothing seems safe, is a threat to the financial system, something that would be avoided with the snow fence setup. The forced transition of countries like Italy and Spain also takes away the self-fulfilling nature of the debt, and reduces it to the “equity” part, and possibly the “cocos”.

Please, do poke holes into this proposal, as I am really interested in whether it could work. Also, please post links to related research or proposals that I might not be aware of.

Karl Smith’s Italian assumption

Karl Smith has a post on Tyler Cowen’s recent remarks about Germany and the Euro. I like Karl’s way of framing the issue: Italy has a primary surplus, and therefore just needs to find new lenders for old ones. There is no real transfer of resources, but a necessary coordination of savers in order for this to work. And the ECB could fulfil that role.

Except, it is not quite true unless you make the important assumption that Italy will have a primary surplus in the future. Who knows that? The cruelty of a currency union is rather obvious, and Italy will struggle, much as Germany did. What is more, the ECB intervention would work against the effort to reform the country and regain competitiveness.

Karl knows all this, of course. It is therefore odd that he seems to deny the risk for the German taxpayer (that ultimately stands behind the ECB) in a lender-of-last-resort operation. And if markets know this risk, and they know that Germany’s public finances are less good than they are commonly portrayed, Karl’s solution may not work in the medium term.

There would have to be some form of collateral, or IMF pressure or the like to supplement the ECB buying Italian debt as I don’t share Karl’s belief in European institutions. But do read Karl’s analysis, it is a useful way to approach the issue.

Real factors in Germany

Tyler Cowen has a short remark on German growth:

German economic growth is surging; again, real factors matter much more than fiscal policy or fiscal austerity.

Karl Smith disagrees:

This seems consistent with a story where both fiscal and monetary policy are in the drivers seat and real factors are relatively unimportant. Indeed, I am not quite sure what the real effect story would be?

Let me offer a graph on German nominal GDP (NGDP) and nominal wage growth (Source: Destatis):

NGDP in Germany increased above its „adjustment period“ trend between 2006 and 2008, fell sharply thereafter just to return to its long run trend in 2010|Q3.
The y-o-y nominal wage growth indicates that Germany was in fact still on this relatively low trend of ≈2% nominal GDP growth (even though this is likely to change in the near future). In other words, Germany is one of the few countries that does not have aggregate demand (AD) problems. But why is this so?

One reason is that German is highly dependent on exports and therefore demand from emerging economies like China. Maybe this is what Tyler refers to as „real factors“?! That monetary or fiscal policy did not have to unduly force AD back on trend, but that it was naturally pushed back there? But is that really true for China that imports US monetary policy via an undervalued exchange rate? Sure, German AD was not driven by bubbly sectors like real estate before the crisis, but as David Beckworth has convincingly shown, the decline of these sectors wasn’t a big problem in the US before overall AD started falling sharply.

Another reason is that the future path of monetary policy in the Eurozone will be relatively loose for Germany, given that the periphery has to undergo massive adjustments in wages and prices. ECB monetary policy was therefore relatively accommodative for Germany (if measured by NGDP returning to trend quickly) and is probably expansionary by now. I am not sure whether Tyler agrees, but this seems to me one of the most important reasons for Germany’s performance.

Yet another reason is that Germany had a labor market policy in place called Kurzarbeit, a massive automatic stabilizer (aka fiscal policy): Workers would temporarily work reduced hours, keep their job and get a subsidy from the government to make up for the loss in income. This has kept unemployment low and skilled labor in the appropriate firms so they can now pick up where they left off. In a way, unemployment was temporarily spread over many workers instead of laying off a few. The costs were high, but probably worth it: the employers lost €8bn, the employees €3bn and the government subsidy was €8.6bn in total.

Sorry, Tyler, I am not convinced that real factors are mainly at work here.

Competitiveness, Dani Rodrik Edition

Ever since I discussed Germany’s „competitiveness“, I have been wondering what a proper definition of that term might be. Paul Krugman and others have criticised the standard notion – measures of relative costs and prices – as misguided. Countries don’t compete in the sense that firms do. The current account surplus of a country is probably the worst possible measure, as I have explained for the German case. I therefore suggested a definition of competitiveness as „wages being at market-clearing levels“, but surely this definition doesn’t capture the whole concept.

Karl Smith offers a critique of Krugman et al:

[A]ll of these critiques hinge on a false premise. That it is just as good for us to have capital in the US as to have capital in China. That it is just as good to have the smartest minds and the best Entrepreneurs in the US as it is to have them in South Korea.

This is wrong. …

Put quite simply, it is much better to be in a rich country than in a poor country and resources are in fact scarce. That other countries attract capital and talent will necessarily mean that the US does not.

Some people become confused on this because they forget that countries matter, that joint equity exists and in a very real sense you are in it with your fellow Americans.

Even though the quantity of „talent“ and „capital“ is not fixed, I agree that domestic economic growth and the ability to generate and sustain it should be part of a comprehensive definition of competitiveness.

This made me think of Dani Rodik’s notion of Growth Diagnostics. In a nutshell, Dani’s argument is that there are always multiple constraints on growth in an economy. Removing all of them is impossible, and because of second-best interactions removing some of them might even be harmful, or at least not beneficial. Moreover, removing constraints is always politically costly.

Combining these insights provides a rule for economic growth policy: a country should remove only the most binding constraint(s). Identifying these most binding constraints is far from easy, and the framework that Dani offers provides only a roadmap on how „to confront those difficulties in a systematic way“. Essentially, you go through this process from top to bottom:

If a country’s „competitiveness“ encompasses full employment and economic growth, we could use Dani’s framework to define it. So my suggested definition would then be:

Competitiveness is the institutional ability of a country to constantly overcome binding constraints on growth.

Unemployment is most likely not a characteristic of such a country, so wages will be at market clearing levels (even though extreme cases with unemployment might exist). Paul’s criticism is also included as „competing against other countries“ is part of this definition only insofar as attracting investment and talents is a binding constraint on growth. This also neatly nests Karl’s arguments.

Is such a definition enough to prevent people from using „competitiveness“ in the wrong context? I doubt it. But if economists can agree on a definition, it will be easier to explain why competitiveness of countries is such a difficult and somewhat dangerous concept.


Get every new post delivered to your Inbox.

Join 486 other followers