Many good economists, like Simon Johnson and Paul Krugman for example, have said something about permanent productivity differentials and optimal currency areas I simply do not understand. (Lots of other people say the same thing, but when good economists say it and I don't understand it I get worried.) Maybe they are making some implicit assumption that I'm just missing. Or maybe I am misunderstanding what they are saying. Or maybe, just maybe, what they are saying doesn't make any sense.
{Update: Hmmm, on re-reading both those economists a third time, maybe they are saying something different from what the others are saying. But it's still not clear to me.)
This is what I think they are saying (I could be wrong):
If two countries have a permanent difference in productivity levels, (or a permanent difference in productivity growth rates, or do not converge in productivity over time), then that is one reason why those two countries do not belong in the same Optimal Currency Area. Unless there are fiscal transfers from the high productivity country to the low productivity country.
That's what doesn't make sense to me.
I understand why asymmetric shocks, including asymmetric shocks to productivity, are one reason why two countries do not belong in the same OCA. Because real exchange rates may need to adjust quickly in response to asymmetric shocks, and nominal wages and prices might be slow to adjust, and nominal exchange rates can adjust more quickly.
And I think I understand why fiscal transfers from the country experiencing temporarily high productivity to the country experiencing temporarily low productivity might help alleviate the problems a fixed exchange rate creates in the presence of those asymmetric shocks. Because fiscal transfers might mean the real exchange rate doesn't need to adjust as much in response to asymmetric shocks.
But I don't understand why the exchange rate regime matters for permanent productivity differentials. And so I don't understand why fiscal transfers would make the exchange rate regime matter less. You can't make something matter less if it doesn't matter at all.
Assume that Canadians are permanently only half as productive as Americans. Always have been and always will be. It's something in the water.
Does that mean it would be better to peg the exchange rate at 50 cents US rather than at par, just so that Canadians could feel good about having the same dollar incomes as Americans, even though we are getting paid in different dollars that are worth only half as much? Would it make any difference if our dollars were the same, but we only got paid half as many? I can't see why it should make any difference.
And sure it would be nice if the rich Americans gave us poor Canadians some fixed fraction of their income every year. But it would be equally nice whether we convert their dollars into the same number of Canadian dollars that were worth the same amount, or into twice as many Canadian dollars that were only worth half the amount.
Maybe, just maybe, there is indeed some sort of permanent money illusion, so that Canadians would insist on being paid the same as Americans only if we call our currency by the same name, even though we are only half as productive. So we suffer permanently higher unemployment that could be eliminated if we switched to calling our dollar after an aquatic bird so people stop making the comparison with American incomes. But if that's the case, maybe we should also pay lower productivity workers in cents, rather than in dollars, so they are satisfied getting the same million cents salary as those who get a million dollars.
I don't think this is what good economists like Simon Johnson and Paul Krugman would be assuming. There must be something else. Some other hidden (to me) assumption they are making. What is it?
Maybe they are assuming that permanent productivity differentials cause asymmetric shocks? That if the Greeks and Germans became equally productive then the Greeks would start building BMWs and the Germans would start growing olives, so that any shocks would hit both economies more equally? But if that's the case, then the relation between productivity and OCAs could go either way. That's because the main benefit of a common currency is that it's supposed to make trade easier, and if two countries became more alike they would tend to benefit less from trade.
Maybe there's some sort of link between permanent productivity differentials, balance of payments deficits, and exchange rate regimes? But I can't figure it out. The link between permanent productivity and the balance of payments isn't obvious, and the link between those two and the exchange rate regime is even less obvious.
Anyway. I just want to try to make sense of this argument. There are lots of other arguments against common currencies that do make sense to me. But this one doesn't. What am I missing?
[This post is an attempt to be clearer than I was in my previous post on the same subject, where lots of good commenters didn't get my point, which means I wasn't making it clearly enough.]
Worthwhile Canadian Initiative 
Maybe a downturn in a big and rich part of the economy have a big effect on GDP and inflation -> responce from CB and effects on exchange rates.
Maybe a downturn in a smal and poor part of the economy have a smal effect on GDP and inflation -> no responce from CB and no effects on exchange rates.
The average af a big and a smal number will be more responsive to percentual changes in the big number.
Their implicit assumption is in line with
“Maybe they are assuming that permanent productivity differentials cause asymmetric shocks? That if the Greeks and Germans became equally productive then the Greeks would start building BMWs and the Germans would start growing olives, so that any shocks would hit both economies more equally?”
PK and SJ are assuming that permanent productivity differentials, especially when those differentials are driven by capital rich/capital poor, will imply capital flows to poor, which is equivalent to a CA deficit. That means that the capital poor country is getting over-leveraged over time.
It’s not the productivity, its the leverage.
Without the adjustment mechanism of the exchange rate a sudden stop is inevitable. Unless the capital poor and capital rich areas can be treated as having a consolidated balance sheet.
That’s my reading, at any rate… so, no, its not productivity that’s the problem, exactly.
A simple assumption that seems sufficient is that groups with different productivity levels are more likely to have lower correlation of productivity changes. Your 1/2 as productive example is implicitly assuming tight correlation.
Nick, when you put it this way, I’m equally confused. I don’t see a problem with long-run productivity differentials as such (although maybe they lead to problems, as argued in BSEconomist’s comment above). The problem is the combination of low labor mobility, lack of fiscal transfers, downward-sticky nominal wages, and heterogeneity (which leads to asymmetric shocks, but as you point out, it also underlies the justification for trade, and therefore for a common currency, in the first place). I think you would have the same problem even without a long-run productivity differential. (I’m not even sure how much sense it makes to compare productivity across regions that are producing different things. How productive are German factory workers at making olive oil? Or at accommodating tourists on the Mediterranean?) You would still get asymmetric shocks, and the results could still be disastrous.
A better high/low productivity shock example, I think, is the emergence of China and the developing world. Portugal and Greece generally compete directly with those countries in manufacturing, such as textiles. On the other hand, Germany exports complicated capital equipment that are in high demand as China expands factories and infrastructure.
I
Olives and BMWs are a function of climate and neighborhood effects, surely. Trade is still improving.
I read it as implicitly claiming permanent productivity differentials give regional governments different policy trade-offs over time, both ‘at equilibrium’ and in response to general recession. Simon Johnson emphasizes the diverging policy regimes of core and periphery, for example. But OCA requires a unitary monetary and fiscal response.
Addendum: thinking about it, the jump toward fiscal transfers is obviously an allusion to the standard OCA result. Krugman emphasizes this, but not Johnson. But it is easy to justify Krugman here: permanent productivity differentials induce different regional policy responses to a general shock; thus regional shocks then diverge anyway.
Take the twin deficit hypothesis, for example. High-productivity export-surplus “Germany” runs dual surpluses and low-productivity “Greece” runs dual deficits. When a crisis hits, “Greece” may undergo a solvency crisis whilst “Germany” does not, inducing different fiscal responses.
They are making no sense. Look at any country, you have college educated engineers living with high school drop outs. That’s a permanent productivity offset and under one currency. Krugman wants to claim this only works because of fiscal transfers? Arguably there is no mobility between these groups either.
That sounds like his politics creeping out again.
My guess is that they are giving an answer they like politically but learned as old keynesians. That is that these subsets have different levels unemployment and through some Philips curve magic he wants the CB to fix it.
Witness also that Karl has been pounding the table about targeting unemployment.
I agree, I’m surprised anyone would claim permanent productivity differentials matter.
When Krugman refers to the widespread perception of “backward, semi-developed countries like Greece or Portugal (not my view, but what you often hear) awkwardly tied to powerhouses like Germany” he most likely has in mind the idea that we expect productivity growth in Germany to outstrip that in Greece.
Note that he explicitly says “not my view” before going on to say that the data is inconsistent with that view in any case. Also, he’s “in an undisclosed location actually being a human being for a few days” so I suspect this is not one of his more carefully considered posts.
But it is surely true that if differences in productivity growth rates are immutable, due to something in the water as you put it, that is one more reason why EMU isn’t going to work. Differences in productivity levels shouldn’t be a problem.
Kevin: it is possible that Paul Krugman was saying that permanent productivity differentials have nothing to do with OCA whether or not there are fiscal transfers. And the bit at the end about fiscal transfers was just a BTW (and an excuse to put the boot into the Republicans).
“But it is surely true that if differences in productivity growth rates are immutable, due to something in the water as you put it, that is one more reason why EMU isn’t going to work.”
I don’t get that either. If Americans have a productivity growth rate permanently 1% higher than Canadians, due to something in the water, then we could simply have nominal per capita income growth permanently 1% higher in the US than in Canada. Unless the North American inflation target was too low, so this would mean occasional negative nominal wage growth in Canada, and there’s absolute downward nominal wage rigidity? Even here, I would think the within-country variance would be very large relative to the cross-country variance. (Or am I committing some variant of the Lewontin Fallacy?)
I read the Krugman post as building on a previous post where he noted how large the automatic fiscal stabilizers are in a federation (using Florida as an example).
http://krugman.blogs.nytimes.com/2012/06/02/florida-versus-spain/
The implication being that sizable fiscal stabilization and equalization occurs in a federation (that has a large federal government), preventing financial and economic crises from becoming currency crises, regardless of productivity differences between regions.
“Unless the North American inflation target was too low….”
If the ECB was setting the target it would be under 2%. So the scenario Krugman is toying with is one where the ratio of Greek to German wages needs to fall steadily. That’s easily accomplished if the Germans are paid in D-marks and the Greeks are paid in drachmas. But if both are paid in Euros, and the Germans are content with very modest increases, then the Greeks must take regular pay cuts.
However he doesn’t endorse that worry since there’s not much evidence for it really. The big problem is that monetary union requires fiscal union, meaning outright transfers (not loans) on a large scale. That would require us to think of ourselves as citizens of the United States of Europe.
I think your first premise, “good economists,” should be examined. Both gentlemen have become staunch polemicists, breathlessly or furiously supporting positions above all. Your analysis is very mainstream and logical.
How succinct. BSEconomist, Fmb, and Andy have this all down. The reason productivity differences don’t matter within countries are those fiscal transfers and because because of geographic locality. The less productive specialize in non tradeables allowing increased productivity for them though still less than those in tradeables.
Eapen – then how do you explain:
Persistent Macroeconomic Imbalances in the Euro Area: Causes and Consequences, Nils Holinski, Clemens Kool, and Joan Muysken, Federal Reserve St. Louis:
“In this paper, the authors document a growing divergence between current account imbalances in
northern and southern euro area countries from 1992 to 2007. The imbalance occurred without a concomitant rise in productivity and growth in the southern (deficit) countries”
or
PRODUCT MARKET POLICIES, ALLOCATIVE EFFICIENCY AND PRODUCTIVITY: A CROSS-COUNTRY ANALYSIS, Jens Arnold, Giuseppe Nicoletti and Stefano Scarpetta, OECD,2008
or
Diverging competitiveness among EU nations: Constraining wages is the key, Mickey Levy
19 January 2012, VoxEU
Within IMF, Eurostats, ECB, OECD and BIS (if I recall), there are economists making this claim in a number of studies in the published research.
“If two countries have a permanent difference in productivity levels, (or a permanent difference in productivity growth rates, or do not converge in productivity over time), then that is one reason why those two countries do not belong in the same Optimal Currency Area. Unless there are fiscal transfers from the high productivity country to the low productivity country.”
This is not what Johnson or Krugman said at all. I was commenting on this the last couple of days and then noticed that Krugman’s latest post seemed to borrow what I was saying almost word for word. (Maybe I’m just getting paranoid thinking Krugman is actually reading my comments). Let me explain.
Johnson said the following:
“This amounted to a very big bet that their economies would converge in productivity – that the Greeks (and others in what we now call the “periphery”) would, in effect, become more like the Germans.”
“In fact, the opposite happened. The gap between German and Greek (and other peripheral country) productivity increased, rather than decreased, over the last decade.”
This is unforgivable sloppiness on the part of Simon Johnson. In many peripheral euro countries, or countries pegged to the euro, where there was huge capital inflows, namely Bulgaria, Estonia, Greece, Ireland, Latvia, Lithuania and Spain, productivity (GDP per hour worked) actually grew faster than in Germany over 1999-2007 (in some cases much faster). What Johnson obviously meant to say is that unit labor costs (ULC) did not grow as fast in Germany. That’s partly because of German wage supression and partly just an empirical regularity.
Current account generated investment booms normally raise the relative cost of labor. The problem however is when monetary policy swings from relative laxity to relative tightness the drop in aggregate demand hits those countries/states with capital inflows the hardest. See for example the Southwest and Florida in the case of the US.
This distinction between productivity and ULC is important. If permanent differences in productivity mattered in currency unions then the US should not even exist. Productivity in Delaware is roughly double that of Mississippi. It is 50% higher in New York than in Michigan.
But asymmetric shocks can produce temporary differences in ULC, as they have between the core and periphery in the eurozone (and the BELLs) and in the core and periphery in the dollarzone. The main reason why the US has not had the same kind of crises as the eurozone is fiscal transfers. Florida would be our Spain right now if it weren’t for the flow of Treasury funds. It’s got absolutely nothing to do with permanent productivity differences. It has to do with temporary differences in ULC.
Nick – most recent study I can find from IMF
Fostering Growth in Europe Now, IMF, June 18, 2012
Staff Discussion Note SDN/12/07
Slower productivity growth explains the decline in euro area potential GDP growth,
while lower labor utilization is behind the lower GDP level with respect to peers. The
sharp decline in total factor productivity growth in the euro area (Figure 2) had the largest
contribution to the trend growth decline observed in the last three decades. While potential
growth rates have come down significantly, the slowdown is more pronounced in the
Southern euro area countries. But productivity is not the entire story: Mourre (2009) shows
that lower labor utilization explains two-thirds of the differential in the GDP per capita level
between the euro area and the United States in 2006. [page 6]
During the last decade, dissimilar patterns of growth across countries and increasing
competitiveness differentials exacerbated each other. Exports drove growth in Northern
euro area countries, while Southern countries relied on domestic demand with a large share
of the employment created in the cyclical and credit-dependent non-tradable sectors, e.g. real
estate. Much of the foreign capital that flew into the Southern euro area during the last
decade was in the form of debt while the tradable sector limped, creating brittle fundamentals
for growth and resource generation for servicing this debt. Relative prices, including nominal
unit labor costs, diverged rendering Southern euro area countries uncompetitive (Figure 3).
One additional clarification. The BELLs (Bulgaria, Estonia, Latvia and Lithuania) are all either pegged to the euro or have been admitted to the eurozone (Estonia). So they are more or less part of the same currency zone.
Mark – your post was extraordinarily helpful. It seems that Johnson (and to confess myself as well) were treating productivity and ULC as synonyms, when as you note, they are not.
Nick,
I think you are confused about Johnson and Krugman, because they are confused. Therefore a healthy reaction : – )
Both do not spell out that the problem is productivity, or better: unit labor cost, “CHANGES”. Johnson simply doesn’t get it, that one currency means indeed, no exchange rate changes. This has nothing to do with any productivity changes here and there. Something he just doesn’t get.
And the implicit assumption was also not that any problems would be solved by mobility, “Greek workers would go to Germany”. Nonsense! It would just aggravate the problem, a brain drain! Somebody working in Germany pays ALL their taxes in Germany. A very typical example of US loud mouth myopia, very little knowledge about the situation, not thinking things through, but lots of opinion.
The assumption was also not, that this would be an adaption / convergence process from a state out of equilibrium.
No, we had the currency snake since 1973 (please don’t get pious, who calls what from when on), many European countries were pegging then their currencies against the deutschmark in the 80ties, a convergence period after the Maastricht treaty 1992, and final exchange rate fixing in 1999 (maybe 1998). That was supposed to be a “fair level” starting point.
Actually there was a little bias built in against Germany, based on expected productivity gains in eastern Germany, which didn’t peter out.
Krugman has a point, that the US considered themselves a “Nation” only after the Civil War.
Well, we tried this in Europe for 2000 years, unification by military force, didn’t work, despite so many attempts, and so high the cost. We will NOT try this again.
The rest of Johnson and Krugman are just head fakes, based on the multiple false assumptions, mentioned above.
Now let’s have a look how that was supposed to work, based on Ordnungspolitik.
I am shocked by all the weight put by some commenters in this thread to the role of fiscal transfers.
There are certainly some large transfer payments made in the US; I imagine Canada too. Where did people get the idea these had any macroeconomic function and weren’t more than a reflection of comity and social values?
Not all policies of the government need to be justified as growth measures. Sheesh.
ianlee: thanks for bringing empirical research to the discussion. I always feel I learn more when people do that.
What I think I’m learning from this analysis is that many southern (I think “southern” means “poor” in this context) EU members ran big trade deficits with the rest of the EU. The IMF now says this “created brittle fundamentals”. We’ve seen this movie; the IMF offered similarly insightful analysis after the onset of the Asia crisis in 1997 when some of the worst affected countries were blamed for running ongoing current account deficits.
This kind of analysis runs into some problems when you think about it. In neoclassical models of growth, you want to move capital from capital-endowed, richer countries with aging populations (“Northern” countries) to poorer, younger (“Southern”) countries where capital is relatively more scarce. If you think that model applies, that means you want “Nothern” countries to run trade surpluses while “Southern” countries run trade deficits. That’s what we saw prior to the crisis in Asia in the 90s and in Europe more recently (and don’t get me started on Latin America in the 80s….)
In hindsight, however, a crisis causes many to change the ways they thought about those international capital flows. Rather than helping reallocate capital efficiently across nations, they are perceived to be symptomatic of excess consumption rather than productive investment. Many nations stung by such changes in attitudes (think Thailand, Spain or Ireland) are particularly frustrated by the selective nature of the analysis; how much fuss is made about the ongoing current account deficits in the US?
I think story about capital flows is important in understanding the debate about exchange rates. Nick has had some good posts lately about the nature of causality in economics, stressing that sme things are determined simultaneously. That sounds like good advice in thinking about exchange rates and trade deficits. It is often misleading to think that a trade deficit is caused by an exchange rate; the opposite may in fact be true.
So what happens in a currency union like Canada or the EU? What happens when you want to move capital from North to South (or East to West) but don’t have an exchange rate to help adjust relative prices? It means the price adjustment has to happen by changing nominal prices. If prices are sticky, they move more slowly so the capital does not get transferred as efficiently. Meanwhile, we see relative ULC rise in the capital-importing area (i.e. the South or West) as wages get bid up, land prices there rise, construction trades are in high demand, etc. None of those things are bad; they are all signs that the marginal product of labour and capital are high. They all help the transfer of capital (and other resources, like labour.)
(Hmm….I thought I was talking about Europe, but I seem to have been also talking about Dutch Disease.)
Jon: Is it that you don’t think automatic stabilizers are fiscal transfers or that they have no macro basis?
@Jon,
You wrote:
“I am shocked by all the weight put by some commenters in this thread to the role of fiscal transfers.
There are certainly some large transfer payments made in the US; I imagine Canada too. Where did people get the idea these had any macroeconomic function and weren’t more than a reflection of comity and social values?”
From OCA.
Optimum currency area (OCA) theory represents a systematic way of deciding whether it makes sense for a geographical region to share a common currency. One often cited criteria for a successful currency union is a risk sharing system such as an automatic fiscal transfer mechanism to redistribute money to regions which have been adversely affected by asymmetric shocks. The European Union budget represents only about 1% of GDP so meaningful fiscal transfers are negligible in practice. Furthermore, Europe had a no bail-out clause in the Stability and Growth Pact, meaning that fiscal transfers were not allowed even in the event of a crisis.
Ah, yes, what Kevin said. In general, the lower is the target inflation rate relative to the productivity growth rate differential, the more vulnerable the currency union will be to asymmetric shocks (when those shocks happen to hit in the same direction as the productivity growth rate differential). If the productivity growth rate differential exceeds the inflation rate, then even a shock of size zero will be sufficient to destabilize the union.
Simon van Norden,
You arote:
“What I think I’m learning from this analysis is that many southern (I think “southern” means “poor” in this context) EU members ran big trade deficits with the rest of the EU.”
I think it is incorrect to generalize this as merely “southern” in nature. The nations most affected by the current account reversals were in fact not the the GIIPS (which also includes a “northern” nation) but the BELLS (Bulgaria, Estonia, Latvia and Lithuania). It is true those nations were not on the euro during the crisis but they were pegged to the euro, and they may not be suffering from sovereign debt crises, but I would argue the macroeconomic shock was perhaps even more severe. They are on the road to recovery but Latvia, for example, is not forecast by the IMF to get back to previous peak GDP until 2017.
And:
“Rather than helping reallocate capital efficiently across nations, they are perceived to be symptomatic of excess consumption rather than productive investment.”
It is important to underscore the fact that this is more a perception than a reality. Most of the flow of capital to these nations did in fact result in productive investment. The problem is nominal in nature, not real. In my opinion a good dose of aggregate demand simulus to the eurozone would cure most of the current macroeconomic imbalances.
Very good comments on this post. I’m just letting the conversation run, since I have little useful to add to what others are saying.
Lord: Jon: Is it that you don’t think automatic stabilizers are fiscal transfers or that they have no macro basis?
Correct, pure transfer payments do not contribute to output therefore automatic stabilizers do not take the form of transfer payments. Second, automatic stabilizers only function in the short run with respect to shocks. I think my response to Andy will help understand further.
Andy: I don’t know how you can claim to compare the inflation rate to the size of the differential shock. If you have two populations with ibr currency but otherwise in isolation your assumption that the inflation rate is uniform is false. There is no quantity against which your comparison makes sense.
Thought experiment: inflation is fixed by averaging over the two populations–initially the rate is equal in both populations. There is a differential shock. Prices drop in one region, the CB must loosen. An inflation gap will open, inflation will push up in one region at a rate above target even as prices are pushed at a rate below in the other region.
Now the point if a common currency is trade. Even with a permanent productivity differential, the higher rate of inflation in one region is going to drive production to the shocked region until the rates equalize. The productivity of one worker is irrelevant: Th law of one price tells you that one unit of output must be priced the same in both countries. Therefore inflation must be the same. Therefore productivity is irrelevant.
Now the point if a common currency is trade. Even with a permanent productivity differential, the higher rate of inflation in one region is going to drive production to the shocked region until the rates equalize.
The main point of the ECB arrangement was to create common nominal risk-free rates, not trade or a common currency per se. The friction due to currency exchange is negligible. That is not a reason to adopt a common currency. What you are adopting is a common central bank that sets the risk free zero day funding costs for the entire region. It is the convergence of nominal interest rates that you are adopting, which causes a divergence of real rates.
The productivity of one worker is irrelevant: Th law of one price tells you that one unit of output must be priced the same in both countries. Therefore inflation must be the same. Therefore productivity is irrelevant.
What is this “Law of one price”? Who enforces it? Do you really believe prices are the same in Alabama as in San Francisco? Really? Offices that have regional COLA adjustments are just wasting their time? The BLS measures regional inflation rates for no reason? Land prices are not the same, and that means that the prices of any goods that require land to be sold are not the same. Even tradeable goods cost more because the rent must be paid by the retailer, and the retailer must pay a higher wage to staff their store. But of course the bulk of the CPI basket is locally produced services.
And the part about capital moving is funny. I am sure that banks in Manhattan and Firms in Silicon Valley are going to be relocating to Detroit any time now. The firms are where they are because of the productive eco-system. Networks of highly skilled workers combined with networks of highly skilled managers. Credit and supplier relationships. It is because of this ecosystem that the firm moves to the area, not because of the price of yogurt.
But there is a law of one price vis-a-vis the nominal risk-free rate.
Correct, pure transfer payments do not contribute to output therefore automatic stabilizers do not take the form of transfer payments. Second, automatic stabilizers only function in the short run with respect to shocks
From the point of view of the Federal government, California sending money to Florida is a transfer payment and not fiscal policy. But from the point of view of the two states, it is fiscal policy. It is a type of re-cycling similar to California creditors lending money to Florida debtors in order to enable them to purchase californian output (the state runs a trade surplus vis-a-vis the rest of the nation).
And this does effect output, in the sense that if these payments were not made, output would be lower.
Similarly, euro-wide, taxing germans to give money to Greeks — say under the rubric of a common unemployment insurance policy — might be fiscally neutral in the long run, but it would certainly have positive stabilizing effects and increase real output.
RSJ:
Is that supposed to convince me that creates output? You’re just restating the original claim with a thin of veneer of a story.
If those california creditors hadn’t lent money to the florida debtors, they would have bought the california output itself. If they hadn’t, inflation would have been lower, the CB would have eased, and then they would have done so. Ergo, the output existed either way.
What you need to show is why these transfer payments within the EU increase EU output. In order for that to be the case you need to show that transfer payments increase efficiency (i.e., that they shift the AS curve)–remember all you’ve done is change who is doing a particular round of purchasing.
If those california creditors hadn’t lent money to the florida debtors, they would have bought the california output itself. If they hadn’t, inflation would have been lower, the CB would have eased, and then they would have done so. Ergo, the output existed either way.
The CB would not have eased just for California. That is the whole problem. There are different states with different inflation rates. Manhattan is doing just fine. Detroit is not. The CB is going to respond to the average. Rates will be too low in Manhattan and too high in Detroit.
What you need to show is why these transfer payments within the EU increase EU output.
That is easy. Any form of insurance is going to reduce risk and increase consumption and investment demand. Particularly for those who are credit constrained. The jobless, as a class, are going to be more credit constrained. If the government can borrow at risk free rates (or just tax) and provide income to the unemployed, and then later on tax them when they regain their jobs, that is a much better deal than having the unemployed borrow at credit card interest rates and face fixed payment schedules.
The unemployed are either going to refrain from borrowing or they will borrow at much higher rates. In either case, they will be pushed off of their optimum consumption plans. The unemployment insurance (or income insurance, more generally) reduces income risk and allows them to be closer to their optimum consumption paths.
Nick – this blog (WCI) gets better and better and better. I learn more from this blog than any other source – because arguments herein forces me to re-examine what I am thinking or think that I know or confront assumptions that I did not know I was assuming.
And you are a rare economist – who thinks philosophically – like Deidre McCloskey or Hannah Arendt. From someone who studied political philosophy (Hobbes, Rousseau, Marx,) for his comps, you should be teaching pol theory. Restated, it seems to me that macroeconomics is a variant of political theory that is addressing (wittingly or not) the most important questions that can be or should be asked – and you are raising some of these questions here.
Now let’s have a look how that was supposed to work, based on Ordnungspolitik.
That claim would have some merit if the German Ambassador had not asked Canada to contribute money to the IMF to bail out Europe. The ambassador did ask.
ISTM that the Euro violated the key aspects of a “sovereign” when it came to borrowing and money.
I say the aspects are:
1) One top government for the currency area. Provinces, states and municipalities nice but not necessary.
2) That top government has unlimited power to levy taxes from its citizens in its own currency.
3) There is one central bank which sets interest rate policy for the country and which serves as fiscal agent to the government.
4) The central bank is responsible in law to the sovereign government, which is its only client government and which is both client and owner. (The Fed makes the cut because of Presidential appointment and Congressional oversight of Fed governors).
It is only under this conditions that a central bank can control interest rates and can use inflation to control its debt, debt to control inflation and the central bank has a powerful source of leverage in its control of the interest on government debt. Further it is only under these conditions that a sovereign cannot default through lack of funds.
Under this definition, no Euro country is a sovereign borrower, not even France and Germany. It should have been obvious to European governments that all Euro area countries were opening themselves the possibility of a hard default through lack of funds in a way that wasn’t true previously from the moment the Euro started to circulate.
rsj: You missed the point of my story response about California and Florida. Rather than continue with that story, let me tackle your second response as that might get to the bottom of the matter faster:
I feel like we’ve skipped a few steps here. We started with a very strong claim: a currency union must have internal transfer payments in order to be stable. I argued that claim was surprising because transfer payments do not change output, under the monetary policy regime of the EU. This must be so, because all fluctuations in output are linked to fluctuations in the inflation-rate which is the goal variable of the ECB.
Now you seem to be concerned about who is consuming output, I fail to see an efficiency connection between who does the consumption and how much is made. The same output is made in both cases. So whether one distribution of consumption is better than the other is a moral question.
So… consider, if there were no union and no transfer of consumption, your complain would remain. So no, it isn’t that fiscal transfers are required for a currency union to be stable; it is instead that you think fiscal transfers are required in themselves.
No doubt, you’re thinking now: but if it weren’t for the currency union, there wouldn’t be unemployment in that region over there. My concern with that claim is that if there is no excess demand for money, there is link between the common monetary policy and unemployment. So let’s assume that since the CB is going the best it can, any increase the provision of hot money will induce inflation. So no more hot money is provisioned.
So, how can there be an excess demand for money in one country and not in the other? It seems to me that there is a contradiction. The fundamental characteristic of an excess demand for money is an inability to sell anything, but if when the CB creates more hot money and there is induced inflation then it must be the case that sellers cannot be found at the lower price.
There is plenty going wrong in Greece (and Spain) I just don’t agree you can claim those things are due there being a common currency per-se. Greece is particularly bad off because there is an expectation that obligations will soon become denominated in drachmas. Debtors want to settle in drachmas and there are no drachmas, so they aren’t settling, but this isn’t a problem that there is a currency union. It is a problem that one debtor (the Greek government) is perceived as having the right and the motivation to declare that all debts in that country are going to be paid in drachmas instead of euros.
I have a real problem with the claim that the instability here is arising from the lack of transfers to greece. The instability is arising because there isn’t a strong commitment to back the fiat currency by the force of law within Greece.
There is a missing feature of the Euro system; it is a believe that par will be maintained. Though clearly this is an accident as commentators are want to remind us that the Euro treaties contain no exit clauses–and of course they don’t. The implicit creation of an exit clause, by claiming that a grexit is anything other than banditry is precisely the issue which is crippling those countries now.
(There is also some evidence that HICP which the ECB is using as its inflation measure is unstable and wrong but that’s another layer of the problem).
Mark: We agree. When I wrote “southern”, my tongue was firmly planted in my cheek.
And you’re right to stress that I’m talking about perceptions and not necessarily reality.
A currency union must have fiscal transfers or substantial regional mobility of goods, labour and capital in order to avoid regional unemployment, which can threaten the unity of political federations.
Jon,
I think this is a strong claim:
I argued that claim was surprising because transfer payments do not change output, under the monetary policy regime of the EU. This must be so, because all fluctuations in output are linked to fluctuations in the inflation-rate which is the goal variable of the ECB.
It suggests
1) That output is independent of price volatility. Imagine an extreme scenario in which the nation is divided into two regions. Region 1 experiences a 20% increases in prices, and region 2 experiences a 20% decrease. The sum of inflation is zero. But the sum of total inflation volatility is not zero. Even according to standard sticky price models in which price stickiness is the only non-utopian force allowed by the modeller, still the aggregate output of the unified region will be less than what it would have been if both regions had stable prices. The region with 20% inflation does not produce enough “extra” output to make up for the output lost by the region with 20% deflation. But a central bank is forced between pushing the output loss onto region 1 or onto region 2, because it must choose which nominal rate to set — the appropriate rate for region 1 or that for region 2.
2) That monetary policy can adjust agg demand without transfers. If you truly believe that, then you must believe that if the U.S. eliminated unemployment insurance, deposit insurance, as well as food stamps, that our output would be the same in this recession than if we had not eliminated them. I can understand why some models might point to this outcome, but it is just a shortcoming of th e model. In our world, income insurance programs are critical to maintaining output, which is why we do not rely on the central bank alone to manage aggregate demand, but also have automatic stabilizers.
Now, given 1) and 2) for our mythical region. if the economy is overheating in one region, causing inflationary pressures, then a transfer program to the depressed region is what funds the automatic stabilizers that reduces the decline of output in the second region while also reducing the inflationary pressures of the first region. This is something that the central bank cannot address because it must pick one interest rate for all regions.
Studying OCAs, I never managed to convince myself that cities belonged in the same currency area as the countryside. While the two traded extensively, there are so many asymmetric shocks that they just don’t seem well suited to share a currency. I eventually convinced myself that currency unions mostly reflect political or other realities rather than Mundell’s theory of OCAs.
So I think David sums up the discussion in one sentence; fiscal transfers aren’t there for a fundamental economic purpose, but for a political one. They (may) make continued membership in a currency union politically sustainable even in the face of adverse and asymmetric shocks. In a frictionless world, those shocks just move activity (production, consumption, factors) from the less- to the more-favoured region. That’s “South” to “North” these days in Europe or “East” to “West” in Canada (yes, I’m including NL as part of that “West”.) In a world like that, fiscal transfers don’t do much.
But when prices are sticky, resources don’t move much because they don’t get the right price signals. And if some factors are more fixed than others, the more fixed factor takes the brunt of the adjustment. If the fixed factors vote, they will rationally seek ways to alter that outcome. Common reactions can include devaluations and/or the imposition of capital controls (to try to keep other factors from moving.)
Simon van Norden,
Exactly, ala recent debate on Dutch disease.
Also, I’m a different David to the Simon was referring to.
David writes: “A currency union must have fiscal transfers or substantial regional mobility of goods, labour and capital in order to avoid regional unemployment, which can threaten the unity of political federations.”
I agree with this statement, but I find it extremely broad so as to be meaningless. How can one have a common CB without then linking the capital markets of the regions?
I’m not sure that qualifies as a currency union though if you cannot take the currency past the border and cannot tried. I also agree that such an economy would suffer unmanaged boom/bust cycles due to AD shocks if there was no monetary policy (e.g., it was a pegged regime or the supply of currency was fixed in advance).
Another way of viewing the value of an automatic fiscal transfer mechanism is to see it as a form of insurance against the worst asymmetries when the monetary authority makes “mistakes.”
Prior to the crisis nominal GDP was growing more or less at a constant 4.3% annual rate in the eurozone. The economy may have been running a little hot on the periphery and a little cold in the core, but in aggregate things were pretty good with inflation close to the implicit target and the estimated output gaps by nation, whether positive or negative, of inconsequential magnitude. Now however, nominal GDP is some 12% below trend and the only nation that anyone can reasonably argue is performing alright is Germany. Consequently all of the economic performance variance is on the downside. Ireland for example is still some 20% below peak real GDP after four years.
Were nominal GDP brought back to trend by the ECB I suspect the old “not too hot, not too cold” situation would be restored. But given that the mistake has been made, the costs for the inhabitants of the GIIPS and BELLs are proving to be enourmous, and the situation is proving to be economically and politically destabilizing for the eurozone as a whole.
I agree with this statement, but I find it extremely broad so as to be meaningless. How can one have a common CB without then linking the capital markets of the regions?
I’m not sure that qualifies as a currency union though if you cannot take the currency past the border and cannot tried. I
I think the capital markets are unified, which is the problem. Anyone can own a share of BMW, regardless of where they are located. You can move currency and stock certificates with effectively zero friction from one nation to another.
But real capital, in this case the physical plants, the Bavarian industrial tradition, generations of skilled machinists, designers, relationships between engineering colleges and apprenticeship programs — that is not something that instantly moves from one place to another when a stock certificate is sold.
You have firms rooted in geographies whose shares are traded on the unified equities markets, and who must sell bonds into a unified financial market, with a common risk free reference point, irrespective of local savings demands. I don’t think the heterogeneities are properly accounted for.
I think one cannot have an OCA without a LOLR. While that may be considered monetary policy, it would as often be seen as fiscal transfers.
Nick, I think that introducing diferent demand policy because of the ECB’s incapability to manage internal demands optimaly, there iis a reason to say that two country with permanent diferencial in productivity must nave its own Central Bank? See
http://www.miguelnavascues.com/2012/06/un-debate.html
I wonder if the Balassa-Samuelson effect is relevant to this discussion. My understanding is that the productivity/real exchange rate relatioinship stems from the Balassa-Samuelson effect which explains the “fact that countries with higher per capita real incomes have a higher real exchange rate. ” The sequence is growth in producvitity in tradable sector -> rise in tradable sector wage -> rise in wage in non-tradable sector -> increase in relative prices of non-tradable -> increase in real exchange rate. I am not sure how “permament” productivity differences come into play and perhas that is the central question being asked.
I believe Prof. Choudri and others at Carleton have been looking at this subject e.g. PRODUCTIVITY, THE TERMS OF TRADE, AND THE REAL EXCHANGE RATE:
BALASSA-SAMUELSON HYPOTHESIS REVISITED with this conclusion: This result provides a potential explanation of the mixed empirical results that have been obtained on the relationship between productivity and the real exchange rate.
I think your confusion is based on the fact that you ignore two important facts. One is that if two countries have a productivity differential, all other things equal, it’s not that the exchange rate will settle at a level different from one, but rather will change at a rate different from zero. This is something you cannot of course do in a currency union. The other thing (probably more important in practice) is monetary policy, such as the inflation target and (base) interest rate. When countries have different productivity, they can stabilize their exchange rate by having different monetary policies, for instance by setting different interest rates. This is also not available in a currency unit, where monetary policy is set by a single central bank (ECB). Note that this is precisely what caused the large current account imbalances from the core to the periphery in Europe that were the real cause of today’s crisis (according to the economists that did not get it miserably wrong.) (Large) fiscal transfers can correct for those lacks of monetary differentials and exchange rate change differentials in the presence of productivity differentials. And the US states are a clear case study showing this.