Tuesday, May 22, 2012

Do Prices Matter? EU Edition

The Euro crisis. One thing sensible people agree on is that the crisis has little or nothing to do with fiscal deficits  (government borrowing), and everything to do with current account deficits (international borrowing, whether public or private.) And one thing sensible people do not agree on, is how much those current account deficits are due to relative costs, or competitiveness.

A thorough dissection of competitiveness in the European context is here; Merijn Knibbe has some good posts critiquing it at the Real World Economics Review. Krugman, on the other hand, defends the competitiveness story, suggesting that the alternative to believing that relative prices drive trade flows, is believing in the "doctrine of immaculate transfer." What he means is, the accounting identity that net capital flows equal net trade flows doesn't in itself provide the mechanism by which trade adjusts to financial flows. A country with an increasing net financial inflow must, in an accounting sense, experience an increasing current account deficit; but you still need a story about why people choose to buy more from, or are able to sell less to, abroad.

So far, one can't disagree; but the problem is, Krugman assumes the story has to be about relative prices. It's not the case, though, that relative prices are the only thing that drive trade flows. At the least, incomes do too. If German wages fall, German goods may become more cost-competitive; but in any case, German workers will buy less of everything, including vacations in Greece. Similarly, if Greek wages rise, Greek goods may be priced out of international markets; but in any case Greek workers will buy more of everything, including manufactured goods from Germany. Estimating the respective impacts of relative prices and incomes on trade flows, or the elasticities approach, is one of the lost treasures of the economics of 1978. Both income and price elasticities solve the immaculate transfer problem, since capital flows from northern to southern Europe were associated with faster growth of both income and prices in the south. But their implications for policy going forward are quite different. If the problem is relative prices, a devaluation will fix it; this is what Krugman believes. If the problem is income elasticities, on the other hand, then balanced trade within Europe will require some mix of structural reforms (easier said than done), permanently faster growth in the north than the south, or -- blasphemy! -- restrictions on trade.

Let's pose two alternatives, understanding that the truth, presumably, is somewhere in between. In the one case, EU current account imbalances are due entirely to countries' over- or undervalued currencies. In the other case, current account imbalances are due entirely to differences in growth rates. One thing we do know: In the short run -- a year or two -- the latter is approximately true. In the short run, the Marshall-Lerner-Robinson condition is almost certainly not satisfied, so a change in prices will have the "wrong" effect on foreign exchange earnings, or at best -- if the country's imports and exports are both priced in foreign currency -- have no effect. In the long run, it's less clear. Do prices or incomes matter more? Hard to say.

So what is the evidence one way or the other? One simple suggestive strand of evidence is the intra- and extra-European trade balances of various countries in the EU. To the extent that trade flows have been driven by price, the deficit countries should have seen larger deficits with other EU countries than with other countries, and the surplus countries similarly should have seen larger surpluses within the union than outside it. Those countries whose currencies would otherwise, presumably, have appreciated relative to other EU members should have shifted their net exports towards Europe; those countries whose currencies would otherwise have depreciated should have shifted their net exports away. Is that what we see?

As is often the case with empirical work, the answer is: Yes and no. From Eurostat, here are trade balances as percent of GDP, within and outside the currency union, for selected countries and selected years.

Intra-EU Trade Balance
1999 2007-2008 2011
Germany  2.0% 4.8% 2.1%
Ireland 19.0% 7.4% 12.6%
Greece -10.0% -9.6% -5.3%
Spain -2.9% -4.0% -0.6%
France -0.3% -3.1% -4.3%
Italy 0.5% 0.5% -0.2%
Netherlands 14.8% 24.5% 27.9%
Austria -3.9% -3.0% -5.0%
Extra-EU Trade Balance
1999 2007-2008 2011
Germany  1.2% 2.8% 4.0%
Ireland 6.1% 7.8% 15.1%
Greece -3.9% -9.1% -4.4%
Spain -2.1% -5.1% -3.8%
France 1.0% -0.1% 0.0%
Italy 0.8% -1.2% -1.4%
Netherlands -11.8% -17.5% -20.5%
Austria 1.4% 2.7% 1.9%

What we see here is sort of consistent with the competitiveness story, and sort of not. Germany did increase its intra-EU net exports about twice as much as its extra-EU net exports over the pre-crisis decade, just as a story centered on relative prices would predict. And on the flipside, the fall in Irish net exports over the pre-crisis decade was entirely with other EU countries, again consistent with the Krugman story. 

But for the other countries, it's not so simple. The increase of the Euro-era Greek deficit, for instance, was entirely the result of increased imports from non-Euro countries. Euro-area trade, and non-Euro exports, were approximately constant in the ten years from 1999. This is more consistent with a story of rapid Greek income growth, than uncompetitively high Greek prices. Similarly, the movement toward current account deficit of Spain was mostly, and of Italy entirely, a matter of trade with non-EU countries. This is not consistent with the relative-price story, which predicts that intra-EU trade imbalances should have grown relative to extra-EU imbalances. Note also that today, Germany's net exports to the rest of the EU area are no higher than when the Euro was created, while Greece and Spain have substantially improved their intra-EU balances; but all three countries have moved further toward imbalance with extra-EU countries. This, again, is not consistent with a story in which trade imbalances are driven primarily by the relative price distortions created by the single currency.

Conclusion: Krugman is right that how much relative prices have contributed to intra-European current account imbalances, is a question on which reasonable people can disagree. But as a doctrinaire Keynesian, I remain an elasticity pessimist. It seems to me that we should at least seriously consider a story in which European current account imbalances are due to relatively rapid income growth in the periphery, and slow income growth in Germany, as opposed to changes in competitiveness. A story, in other words, in which a Greek exit from the Euro and devaluation will not do much good.

UPDATE: While I was writing this, Merijn Knibbe had more or less the same thought.


  1. Excellent, as always, and I don't really have anything to add other than I wish you would post more often.

    I am curious what you think about the Varoufakis piece though. The "animal spirits" argument seems like a bit of a cop-out to me. Have you done much reading about Argentina's "Jefes y Jefas de Hogar Desocupados" program?


    Many, many professional economists I read seem, IMHO, overly focused on increasing net-exports as a way out of recessions even though the current account is usually a relatively small percentage of GDP for most countries. Those making the "increase net-exports" argument seem to be conceding that the problem is a lack of effective demand so why not just tackle the lack of effective demand directly and domestically through a public employment or transfer payment system like Argentina did? The Jefes y Jefas de Hogar Desocupados program seems to be an important, but often ignored part of Argentina's recovery. Probably just as important as the rebound in net-exports and certainly more so for the lower classes in Argentina. Thoughts?

  2. I wonder if there isn't a little more to the price story in regard to the extra-Euro trade imbalances with Greece, Italy and Spain. The Euro after all does not only influence prices and trade within Europe but it also effects the price of Greek goods in relation to countries outside of Europe as well. Greece with the Euro presumably has a stronger currency in relationship to non-euro currencies than Greece with the Drachma, and any potential currency effect on the trade balance would be distorted by the overarching Euro region in a way that might not occur if Greece had its own currency.

    Overall I think your skeptcism over the price story is warranted, but I also note that most of the discussion of devaluation tends to focus on Greece vs Germany, and not on what would happen between Greece and the rest of the world.

  3. Ragweed,

    Yes, the Euro presumably results in an overvalued peripheral real exchange rate relative to the rest of the world, and an undervalued german real exchange rate relative to the rest of the world. But as a matter of logic they have to be even more over/undervalued relative to each other.

    Also, it's not shown in the charts I give, but the fact that the worsening of the Greek current account under the Euro was 100% a matter of a growing import share, with exports/GDP unchanged, tends to support the idea that the growth of the Greek c.a. deficit was due to rapid income growth rather than an overvalued exchange rate.

  4. pk prolly buys your conclusion

    notice his post basically disparaging forex adjustment solutions on grounds of short run impact

    but long run adjustment requires forex changes
    notably zone versus non zone price levels

    inner zone that translates as adjusted rates of national inflation

    to do this at all round full tilt expansion back to high employment
    requires a savage german wage boom

  5. "german real exchange rate relative to the rest of the world."
    i doubt that
    i think here we are talking about premium products not subject to commodity prices
    ie one international price

  6. Hey paine!

    I don't know about that. My sense is Krugman sees a real devaluation as necessary and sufficient to resolve the current account imbalance, and that exit from the Euro will straightforwardly produce the required devaluation. I think he's still orthodox enough -- he certainly used to be! -- to think that relative prices are what really matters for trade. In the linked post, he writes

    We know that huge current account imbalances opened up when capital rushed to the European periphery after the euro was created, and reversing those imbalances must involve a large real devaluation.

    We "know," it "must": not much wiggle room there.

    Note that Greece has been running continuous current account deficits at least since 1980. I think it will take more than devaluation to change that.

    1. the basic problem is of course
      the MNCs want these slowly closing distortions
      if they can beat up a national economy they will
      especially if they can hold it by the throat forex wise
      and bash at it or us its fiddled forex to bash another national market by shooting in imports

      the more i look at the global market game i see the hedge row policy and the mow down the hedge rows as part of a beat em coming and going strategy
      MNCs can play

      only a very alert nation state can hope to gain
      a big slice of the gains from trade in this game

      yes china can play it

      south africa brazil ?

      we'll see

  7. i think here we are talking about premium products not subject to commodity prices

    But the question is how true this is of trade in general. I think this description applies to most traded goods.

    I don't think you can deny -- and I'm sure PK wouldn't -- that *if* real exchange rates are the main drivers of European current account imbalances, intra-European imbalances should be larger than extra-European imbalances. No?

    1. " how true this is of trade in general"

      profound point
      perhaps too profound

      the models cut exchanged products
      into tradeable /non trabeable
      then assume a world price for tradeables

      and thats not even accurate about raw commodities
      at the point of final sale !!

      the introduction of novel products
      and the follow on "commodification process "
      can be arrested at many points
      it is not automatic ..obviously

      cisterns of rent can be built
      and edges of oceans
      and distant borders are places where this happens often

      german producer goods have brand rents
      and lots long ago lost any reward to innovation aspect
      they are simply monopoly products

      the stiglitz-dixit solution
      a continuum of unique
      but grossly substitute-able "products"
      simply conceals one assumed can opener
      in the process of discarding another
      beautifully classic and endemically sown
      can opener assumption

      too much to say ...
      and i'm lazier then a bachelor bull walras

      let this suffice

      a system of national markets
      allow multiple prices for the same commodity

      and innovation creates an interval
      for surplus profit or economic rent

      these two can tango
      the reward to innovation can quickly become
      a rent sump
      a market rigging

      all to grimly common
      as the busy trans nat outfits
      from national market to national market

      using borders like billiards players use banks

    2. the three layers here need a distinction

      inside euro zone
      outside euro zone EZ but inside common market EU
      and of course outside common market

      the numbers are what they are

      but the euro could do some serious targeted devals with non zone EC members

      even if only to reduce trans bordr corporate arbitrage profits
      if not to close trade gaps

      ---too often this distinction is lost if the only goal is balanced trade ---

      the intra zone north/south arbitrage is certainly
      all pk seems hep to
      and i agree this need not be the biggest gap maker
      for club med players

      i hunt for "unearned profits"
      not trade imbalances per se

      over time you can be de industrialized thru arbitrage
      even while trade remains mostly in balance

  8. Of course you and I and PK all agree on the desirability of higher German wages -- what a commenter on Crooked Timber once called the "Drink finer wines, drive nicer cars, and party harder!" plan for Europe. But PK thinks this would improve things mainly by raising the relative price of German goods, while I think it would help mainly by raising German income.

    1. i agree with you
      the real international markets are about import absorption mostly
      at the macro level
      and arbitrage profits at the "firm" level

    2. the facts are clear i think the romney robinson school of thought ruled in the 50 to 71 era

      and we only can talk seriously about unleashing forex instruments apres the gold window drop
      by potus-issimo dick nixon

      and yet with the ability for oligop trans nat firms
      with some degree of product differentiation
      to price discriminate
      from one national market to another
      and the ability to resist the substitution effect by not repricing according to forex moves

      ---of course firm level hand made pricing algorithms
      have much more ideosyncratic complexity
      then my words here encompass ---

      the point to me of aligning forex
      with purchasing ower parity
      has all to do with disabling the trans nats ability to hop around making windfall profits that are not driving technical progress
      but only playing off trade area and production area market anomalies

      --often of their own creation or at least sustenance ---

      to simply turn your back on forex markets blinds you to where and to whom
      gains from trade go

      the focus is reversed
      we need to use class to count gains from trade
      and nations to formulate trade policy

      firms formulate trade policy
      and national stats
      are used to keep score

    3. wow what a delightful sentence fragment in that last comment

      "and yet with the ability for oligop trans nat firms
      with some degree of product differentiation
      to price discriminate
      from one national market to another
      and the ability to resist the substitution effect by not repricing according to forex moves..."
      allows firms to sustain imbalances in relative prices to exploit for arbitrage profits

      the china trade the last decade
      was a great example of a trans oceanic profit slurry
      for trans nats-multi nationals
      thanx to grossly non purchasing power parity forex

      now this isn't the only game on market earth
      there's tech patent and corporate privacy transfer games tax and reg games market access games state manipulation games ...the king of all wage games
      etc etc
      but all of these games require
      the corporate ability to prevent fully flexible
      and fully sustained forex rates
      in particular forex rates that are adjusted as pk wants
      by nation states
      in the interest of domestic
      non oligop non comprador mostly non capitalist class elements

      ---hence the call for
      a economic class based accounting ---