WHEN CAN A SINGLE CURRENCY EVER WORK - FLAWED FROM THE START - Summary and Analysis of The Euro: How a Common Currency Threatens the Future of Europe - Joseph E. Stiglitz

Summary and Analysis of The Euro: How a Common Currency Threatens the Future of Europe - Joseph E. Stiglitz (2016)



B.E.: before the euro. A world where you traveled from one European country to another, from France to Germany to the Netherlands, constantly changing currencies—from the franc to the deutsche mark to the guilder. The arrival of the euro simplified the life of a traveler. It has replaced 19 different currencies in 19 different countries with a single currency.1

Creating a single currency involved, however, not just a change in the pieces of paper that are used to buy goods. It involved creating a central bank, the European Central Bank, for the entire eurozone. The ECB determines the interest rates that prevail throughout the area and acts as a lender of last resort to the banks in the eurozone—providing money that banks may need, for instance, as depositors withdraw funds, even when no one else will provide those funds. The central bank sets interest rates and can buy and sell foreign exchange. Both of these affect the exchange rate between the euro and other currencies, such as the dollar.2 (The exchange rate specifies how many units of one currency trade for another. Thus, at the start of 2016, the dollar/euro exchange rate was 0.92, meaning if an American went to Europe with $100, he would get back €92.) A higher interest rate leads to a higher demand for euros, and thus a higher exchange rate. Monetary policy is one of the most important instruments in a government’s economic toolkit. When the central bank wants to stimulate the economy, it lowers interest rates and makes credit more available.3 And lowering interest rates leads to a lower exchange rate, making exports more competitive and discouraging imports. On both accounts, the economy is strengthened.

Markets on their own don’t ensure full employment or financial and economic stability. All countries engage in some form of public action, intervening in the market to promote macro-stability. Today, except among a lunatic fringe, the question is not whether there should be government intervention but how and where the government should act, taking account of market imperfections.

When two countries (or 19 of them) join together in a single-currency union, each cedes control over their interest rate. Because they are using the same currency, there is no exchange rate, no way that by adjusting their exchange rate they can make their goods cheaper and more attractive. Since adjustments in interest rates and exchange rates are among the most important ways that economies adjust to maintain full employment, the formation of the euro took away two of the most important instruments for ensuring that.

Because using a common currency took away the ability to use the exchange rate to adjust exports and imports, the obvious symptom that something was wrong—that some adjustment which should have occurred wasn’t happening—was a persistent disparity between the country’s imports and exports. Normally, if there is an excess of imports, the exchange rate falls, making imports more expensive and exports more attractive. If, for instance, Greece imported more than it exported, the imbalance could and would be corrected by making the value of its currency weaken; that would make exports more attractive, imports less so. With fixed exchange rates, this can’t happen. An excess of imports has to be financed, and if the country with the trade deficit can’t borrow the money to finance it, there will be a problem.

Ceding control over exchange and interest rates can be very costly for a country. There will be enormous problems unless something else is done. The magnitude of the costs and the nature of that “something else” depends on a number of factors, among which the most important is how similar the countries are. Given the diversity of Europe, the “something else” that needed to be done was much larger than what Europe did or is likely to do.

If countries with a shared currency are sufficiently similar, of course, all of the countries will be hit by the same shocks—an increase or decrease in the demand for their products by China, for example. In the same way, the responses that are appropriate for one country are appropriate for all. In such a situation, the cost of forming a currency union may be low. Economist Robert Mundell, my colleague, received a Nobel Prize for asking and answering the question, what are the conditions in which a group of countries can easily share the same currency?4 His analysis made clear that the countries of the euro are too diverse to easily share a common currency.

The founders of the eurozone worried about these differences. The 1992 Maastricht Treaty creating the euro required that those joining the eurozone satisfy so-called convergence criteria—intended to ensure that the countries would in fact converge. Governments joining the euro had to limit their deficits (the annual amount by which their revenues fell short of expenditures) and debts (the cumulative amounts they owed). They were required to have deficits less than 3 percent of GDP and debts less than 60 percent of GDP.5 Subsequently, all the countries of the EU (and not just those within the eurozone) reinforced their commitment to these deficit and debt constraints, in what was called the Growth and Stability Pact.6

It made sense that the countries of the eurozone worried about being too disparate. They understood excessive disparities would put strains on the euro: one part could be facing deep recessions and another inflation, one part could have a large trade surplus, another a large trade deficit. Any interest rate that worked for one part would exacerbate the problems of another. Somehow, they seemed to believe that, in the absence of excessive government deficits and debts, these disparities would miraculously not arise and there would be growth and stability throughout the eurozone; somehow they believed that trade imbalances would not be a problem so long as there were not government imbalances.

The evidence is now all too apparent. Even countries with no government deficits and low public debt (like Spain and Ireland) had crises. Many countries without large government deficits or debts have had large trade deficits. Many of the countries of the eurozone did not—and still have not—adjusted well to the shock of the 2008 global financial crisis and its reverberations. There has been neither growth nor stability, and the countries of the eurozone have diverged rather than converged.

The previous chapter showed that the euro had failed on its promise of European prosperity. This chapter begins the explanation of why. After explaining how it is that the 50 diverse states of the United States can share a common currency, I show that the necessary conditions—in terms of the institutional arrangements and conditions—for success were missing in Europe. I explain, too, why sharing a common currency is so difficult—how in the absence of the requisite institutions it can lead to sustained high levels of unemployment, to current account deficits (where countries persistently import more than they export), and, even worse, to crises. These are not just theoretical possibilities: these concerns have played out with a vengeance in Europe.

A natural question is, should the founders of the euro have imposed other conditions—for instance, not just on fiscal deficits but on current account deficits—to have ensured convergence? Would these conditions have ensured growth and stability? Later in this chapter we will explain why the obvious additional stricture—on current account deficits—simply wouldn’t work. There are a set of rules for countries, combined with a set of institutions for the eurozone, that could make a single currency work for even a diverse set of countries. In chapter 9 I will describe one such set of rules, applying not just to countries with trade deficits but also to those with trade surpluses. As we shall see, those rules and institutions are markedly different from those in place in the eurozone today.

As we have noted, this book is not only about events, about what has been happening in Europe, but about ideas, about the role of ideology in shaping the construction of the eurozone. The founders of the euro seemed to believe that satisfying the convergence criteria was key in ensuring the viability of the euro. They were obviously wrong. In this chapter, we also attempt to explain how neoliberalism (market fundamentalism) led them astray.


Many Europeans look at the United States and ask, if the 50 states of the United States can all share the dollar, why can’t the 19 states of the eurozone share a currency? We noted above that if the constituent states were similar enough, then whatever monetary policy was correct for one, would work for the others. But the individual US states differ markedly—there are agricultural states and industrial ones; there are those that are helped by a fall in oil prices and those that are hurt; there are states that are chronically borrowing from others and some that are net lenders. These differences give rise to major differences in perspectives about economic policy: states that are net debtors and in which manufacturing is important typically argue for low interest rates, while the creditor states, with a large financial sector, argue for high interest rates. There is a long-standing distinction between Wall Street and Main Street. So while similarity might be sufficient for the success of a currency area, it is hardly necessary.

There are three important adjustment mechanisms within the United States that enable the single-currency system to work. Unfortunately, as we will see, none of them are present within Europe—or at least present in sufficient strength to make the eurozone work.

When, say, South Dakota faces an adverse shock, people move out. Because English is the nation’s common language, and because many key programs, like Social Security and Medicare, are national programs, migration is relatively easy. There are, of course, costs of moving, and in some professions, like law, licensing serves as a barrier. Still, there is little comparison with Europe: while in principle there is free migration in the EU, there are still large linguistic and cultural barriers, and even licensing differences. Americans are used to moving from one region to another; Europeans are not.

There is another problem: Few Americans in other states worry about, say, South Dakota becoming depopulated.7 But Greece does care if most Greeks, or even most talented young Greeks, leave the country. And it should care. In the United States, it makes little difference, in the larger scheme of things, whether people move to the jobs or the jobs move to the people.8 In Europe, the Greeks and Estonians want to be sure that enough jobs move to these relatively small countries to preserve their economy, culture, and identity.9

Another big difference is that South Dakotans think of themselves first and foremost as Americans, and that identity is unchanged as they move. A South Dakotan is not in California as a “guest” but as a right, a right the revocation of which is unimaginable. After a very short period, he has full voting rights and rights to all the benefits extended to those living in California. There is no distinction between a native Californian and an “immigrant” from South Dakota. Both are “Californians.” Recent European debates about how many years an immigrant has to live in his new country before he obtains certain welfare rights show that the same is not true in Europe. Despite the ease of working in the new location, a Pole in Ireland is still fundamentally a guest. Further, his political and cultural identity—and hopes for the future—will more than likely continue to be Polish.

There is another important adjustment mechanism in the United States: After a shock, South Dakota will receive financial support from the federal government in one way or another. Some of this support is automatic: with an economic downturn, many people will turn more to national welfare programs, to Medicaid (the national program that provides health care for the indigent, which is locally administered but largely federally financed), Medicare (the national program for health care benefits for the aged), Social Security (the national program for retirement for the aged), the Supplemental Nutrition Assistance Program (or SNAP, the national program that provides food-purchasing assistance for the poor), etc. In deep downturns, the federal government picks up much of the tab for unemployment insurance. In Europe, each of these programs is financed by national governments, so if Greece has a crisis, its government has to cover the increased welfare payments—at precisely the time when government revenues are falling.

Beyond these automatic programs, the federal government can use discretionary powers to support states in difficulty. If California is in recession, more military money can be spent in that state, in an attempt to resuscitate it. But Europe’s federal budget is, as we have noted, miniscule. There is simply little discretionary money that can be used in a countercyclical way.

In the United States, there is a third source of shared support in the event of an adverse shock: the banking system is, to a large extent, a national banking system. If any bank runs into a severe problem (as happened to many a bank in 2008), the institution is bailed out not by the individual state but by a federal agency (the Federal Deposit Insurance Corporation, or FDIC). If the state of Washington had been forced to bail out Washington Mutual, the country’s largest bank that failed in the financial crisis, it couldn’t have done it—it would have been similar to Iceland trying to bail out banks ten times bigger than the country’s GDP.10 Similarly, California might have had a hard time dealing with the problems posed by the failure of Countrywide Financial (the largest mortgage company, other than Fannie Mae and Freddie Mac).

Again—up until now—each country within Europe has been responsible for its own banks. And as we shall comment in the next chapter, this contributes to a downward vicious cycle: weak banks lead to the government’s fiscal position worsening, and that in turn weakens banks further.

Some of the reforms in the design of the eurozone that we suggest later attempt to move Europe toward the American model, to a system where a single interest rate and a single exchange rate are consistent with shared prosperity—to move the eurozone enough in this direction to make the prospects reasonable for the euro to work.


Even at the founding of the euro, most realized that differences among the countries of the eurozone were large and that the union lacked the type of institutional arrangements which would allow disparate economic entities to share a common currency. There were huge differences even in the beginning—between, say, Portugal with a GDP per capita of about 57 percent of Germany’s, similar to the differences among the US states, where Mississippi’s GDP per capita is 48 percent of Connecticut’s. Later entrants, though, included countries that were much poorer: Latvia, which joined in 2014, has a GDP per capita just 31 percent of Germany’s.11

The hope was that the countries could converge, that over time, they could become more similar, and with sufficient convergence, it could become a currency area that would work reasonably well. But in many cases this failed to materialize. For example, by 2015, not only had Portugal and Germany not converged, but Portugal had actually fallen further behind—its GDP per capita is now estimated at just 49 percent of Germany’s. As our discussion about the United States made clear, more than just “enough similarity” would be required: no matter how close they converged, there would remain enough differences that additional institutions would be required, such as America’s common banking system. Apparently, many didn’t grasp this, and those that did simply assumed that those institutions would arise as they were needed. The eurozone would supply the institutions that were needed as they were demanded. The political momentum created by the euro, it was hoped, would be strong enough to accomplish this.

There were two key challenges in making a single-currency area (like the eurozone) work: how to ensure that all of the countries can maintain full employment and that none of the countries has persistent trade imbalances, with imports exceeding exports year after year.

The problem with a common interest and exchange rate is simple: if different countries are in different situations, they ideally want different interest rates to maintain macroeconomic balance and different exchange rates to attain a balance of trade. If Germany is overheated and facing inflation, it might want the interest rate to rise, but if Greece is facing a recession and high unemployment, it wants the interest rate to fall. The monetary union makes this impossible. (We noted that for the United States, if there are enough other ways of adjusting, enough other institutions that can help in such a situation, then the disadvantage of not being able to set different interest rates can be overcome. There’s the rub: Europe didn’t pay attention to the necessary complementary institutions.)

It should be obvious that there are myriad shocks that would have different effects on different countries. A country that imports more oil will be more adversely affected by an increase in the oil price. A country that imports more gas from Russia will be more adversely affected by an increase in the price for gas that Russia charges. These differences are a result of the structure of the economy in each country. No matter how much countries converge in terms of deficits and debts, there will continue to exist large differences in structure.

Differences in debts and deficits, of course, also matter. But deficits and debts matter in the corporate and household sector as well as in the public sector. A country where households and firms borrow heavily abroad will be hurt by an increase in global interest rates, and conversely one that lends will be helped. The impacts may be more limited if borrowing takes the form of long-term bonds, so that what it pays only increases gradually, as old bonds are replaced by new. If it borrows short-term, then the borrower may immediately face a problem: if it is the government, it confronts a fiscal deficit, as the interest rate it has to pay for funds increases immediately. Highly indebted households and firms can go bankrupt, because they cannot meet their debt obligations.

The countries within Europe differed in these and other ways, implying that it was virtually impossible for them all to attain full employment and external balance simultaneously, in the absence of other institutional arrangements of the kind found in the United States. The eurozone failed to put into place these institutional arrangements.


An economy facing an economic slump has three primary mechanisms to restore full employment: lower interest rates, to stimulate consumption and investment; lower exchange rates, to stimulate exports; or use fiscal policy—increasing spending or decreasing taxes. The common currency eliminated the first two mechanisms, but then the convergence criteria effectively eliminated the use of fiscal policy. Worse, in many places it forces countries to do just the opposite, cutting back expenditures and raising taxes in a recession, just when they should be increasing expenditures and cutting taxes.

In an economic slowdown such as 2008-2009, tax revenues plummet and expenditures for unemployment and welfare soar, so deficits rise. As we have noted, the convergence criteria require that countries limit their deficits to 3 percent of GDP, and almost all of the eurozone countries exceeded that limit at one time or another, when they went into recession. Between 2009 and 2014, for instance, all but Luxembourg exceeded the 3 percent deficit limit at least once.12 Countries that exceeded the limit were required to increase taxes or lower spending, weakening total demand. These austerity policies weakened the European economies further.


I have just described the standard Keynesian theory on economic downturns. Seemingly, behind the founding of the eurozone were alternative hypotheses about how an economy with high unemployment could get back to prosperity without increasing government spending and without the flexibility afforded by having its own exchange and interest rate: (a) cutting the government deficits would restore confidence, which would increase investment; and (b) markets by themselves would adjust, to restore full employment.

The confidence theory dates back to Herbert Hoover and his secretary of the Treasury, Andrew Mellon, and it has become a staple among financiers. How this happens has never been explained. Out in the real world, the confidence theory has been repeatedly tested and failed. Paul Krugman has coined the term confidence fairy in response.

When Hoover tried to reduce the deficit in the years after the 1929 stock market crash, he didn’t restore confidence; he simply converted a stock market crash into the Great Depression. When the IMF forcibly made the countries under its programs—in East Asia, Latin America, and Africa—reduce their deficits, it, too, converted downturns into recessions, and recessions into depressions.

No serious macroeconomic model, not even those employed by the most neoliberal central banks, embraces this theory in the models they use to predict GDP.

A few economists have nonetheless put forward the seemingly contradictory notion of “expansionary contractions.” Closer examination of the alleged instances shows that what happened was that a few countries had extraordinarily good luck. Just as they cut back on government spending, their neighbors started going through a boom, so increased exports to their neighbors more than filled the vacuum left by the reduced government spending. Canada in the early 1990s provides an instance. Today, even the IMF recognizes that austerity hurts the economy—and in doing so, even hurts confidence.13

I have often referred to the obsession with the debt and deficits as deficit fetishism. This does not mean that governments can run as large a deficit or debt as they might like. It just means that simplistic rules, such as those embedded in the convergence criteria, are indeed simplistic and do not provide a basis of good policy. It may be necessary to impose some constraints, but the constraints have to be carefully and thoughtfully designed—for example, taking into account the state of the business cycle and the uses to which the funds are being put. For instance, rather than focusing on the deficit, the designers of the euro should have focused on structural deficit—what the deficit would have been had the country been at full employment.

As it is, the convergence criteria not only prevented a country from responding to a downturn but created a built-in mechanism for deepening it. As GDP went down, say, because the market for the country’s exports diminished, its tax revenues went down. Under the convergence criteria, it would be forced to cut back expenditures or raise tax rates, both of which would lead to a still weaker economy. Economists refer to such provisions as “built-in destabilizers.” Well-designed economic systems have built-in stabilizers, not destabilizers.

One of the economic victories of the Clinton administration while I served as a member of the Council of Economic Advisers was on precisely this issue: Republicans wanted to pass a constitutional amendment that would have limited deficits, just as the convergence criteria did. While we were in a period of prosperity—so great that eventually the federal government ran surpluses—we knew that market economies were volatile. If we hit a bump, and we did hit a big bump in 2008, it would be important to be able to stimulate the economy. We correctly assessed that other mechanisms, like monetary policy, would not suffice. Had such a constitutional amendment passed, in 2009 the recession would have been much deeper.

The irony is that while the convergence criteria were intended to help the countries converge, and the austerity imposed was intended to reduce the fiscal deficit, typically, the effects were just the opposite. At best, the magnitude of the reduction in the deficit was far less than hoped, simply because the cutbacks led to an economic slowdown—and that in turn led to reduced tax revenues and increased expenditure on unemployment benefits and welfare.


The previous discussion explained how, even if the ECB set interest rates in the interests of the eurozone as a whole, the weakest economies would be left facing unacceptable levels of unemployment. Had they not been in the eurozone, these countries could have lowered their interest rates.

But as we observed in chapter 1, the mandate of the ECB is to focus on inflation, not unemployment. As long as there is inflation for Europe as a whole, and especially Germany, the ECB pays not only little attention to the plight of countries suffering with high unemployment but even to the average unemployment rate. That could be high—the eurozone would need a rate cut to restore the average unemployment rate to a reasonable level—but under the ECB mandate, that may not occur. Indeed, as we have already noted, twice in 2011 (in April and July) the bank raised interest rates despite the euro crisis.


The second problem posed by a single-currency area relates to external imbalances—where imports persistently exceed exports, requiring the country to borrow to finance the difference. Such borrowing, as we shall see, is often problematic and exposes countries to the risk of a crisis. When exchange rates can adjust, a devaluation (a decrease in the value of one currency relative to another) makes that country’s goods cheaper and imports more expensive, reducing imports and increasing exports. This is the market mechanism for correcting external imbalances. This mechanism is short-circuited in a currency area.

We observed earlier that when two countries decide to share the same currency, one can’t make its products more competitive relative to the other by adjusting the exchange rate. But if prices in the country decline relative to the other, then the real exchange rate changes, and its goods become more competitive—both relative to goods from other countries within the eurozone and relative to other countries in the world.14 This alternative adjustment mechanism is referred to as internal devaluation—and those who believe in the euro have put their faith in it. Indeed, one can see austerity policies as facilitating this adjustment process. The greater the weakness in the economy, the greater the shortfall in aggregate demand, the more the downward pressure on prices, and, therefore, the stronger the forces for adjustment. The advocates of the euro thought that this would be the mechanism that would eliminate external imbalances—if only government doesn’t short-circuit this mechanism by maintaining an excessively strong economy.

As curious as it may seem, the neoliberal advocates of the euro thought that in some ways unemployment was a good thing. Consider a country exporting shoes, which suddenly finds the demand for its products diminished (for example, because of Chinese competition). It has simultaneously an unemployment problem and a trade-balance problem. The notion was that if only one let nature take its course, both problems would self-correct. The unemployment would lead to lower wages, lower wages would lead to lower prices, exports would then increase, imports decrease, to the point where imports and exports were in balance. Meanwhile, the increased demand for exports would help restore the economy to full employment. Admittedly, the adjustment process could be painful—none of this would happen overnight. In the meanwhile, families would suffer from unemployment; children’s lives would be ruined, as a result of lack of nutrition or access to health. Humanitarian versions of these “tough” policies emphasized the importance of a safety net; but, of course, the convergence criteria—and the absence of help from others in Europe—meant that increased expenditures on such assistance had to come out of somewhere else in the budget, like public investment, hurting the country’s future growth. In practice, safety nets always proved inadequate, and, as we shall see in later chapters, the suffering was enormous.

Still, the neoliberal proponents of the euro would argue that perseverance pays: there is redemption through pain. Interfering in this natural market process might reduce unemployment today, but it only extends the period of adjustment.


There has been considerable internal devaluation among European countries. Prices in crisis countries have on the average been increasing less than most elsewhere in the eurozone.15 But this internal devaluation has not worked—or at least not worked fast enough to quickly restore the economies to full employment. In some countries, such as Finland, low inflation has not been enough even to restore exports of goods and services to the levels before the crisis. In some of the crisis countries, Troika policies may not only have had adverse effects on demand but even on supply. In other countries, exports did not grow in the way that had been hoped—they did not grow enough to offset the adverse effects on the nontraded sector.16

If exports had grown at healthier rates, that would have stimulated the economy and helped restore full employment. But there is another, less healthy way of reducing a trade deficit. Imports fall when incomes plummet: one can achieve a current account balance by strangulating the economy. And that, not internal devaluation, was at the heart of the success that the eurozone achieved in getting trade balance. Even Greece had achieved close to trade balance by 2015. But most of the reduction in trade deficit came from a reduction in imports.17


If internal devaluation were an effective substitute for exchange-rate adjustment, then the gold standard would not have been a problem in the Great Depression. Yet most scholars believe that the gold standard was a major problem—some going so far as to place much of the onus for the Great Depression on the gold standard.18 By the same token, if internal devaluations were an effective substitute for exchange rate adjustments, Argentina’s fixing its exchange rate with the dollar prior to 2001 would not have been a problem. As that country’s unemployment increased—exceeding 20 percent—prices within its borders did fall, but again, not enough to restore the country to full employment, especially as in those years the dollar was strengthening.


There are several reasons that internal devaluations might not work: wages may not fall; the fall in wages may not lead to a fall in the price of export goods—or at least not enough of a decline; and the fall in prices may not lead to an increase in exports—or at least enough of an increase. Each of these elements has played out in the euro crisis.

Slow wage adjustments: blaming the victim

There are many in Europe (including Jean-Claude Trichet, head of the ECB from 2003 to 2011, and thus at the helm in the run-up to the crisis and in its first years) who blame the failure of the internal devaluation mechanism on wage rigidities—on the failure of wages to fall even in the presence of high unemployment. They believe that markets on their own, in the absence of unions and government intervention, would be flexible. In their attempts to blame the victim—workers are to blame for their own unemployment because they’ve demanded too high wages and too many job protections—these critics focus on constraints imposed by government and unions. But across Europe, and around the world, one can see high unemployment rates with little adjustment in wages in economies with weak unions and without government constraints.

Market rigidities

This “puzzle” of why wages in market economies often don’t decline in the presence of even high unemployment motivated my research, along with that of many others, into wage rigidities. The theory that I developed, called the efficiency wage theory, focused on the well-documented fact that cutting wages undermines worker productivity. It hurts workers’ morale, especially when they become distracted about their ability to keep their home; it weakens loyalty to the firm; firms worry about workers looking for better paying jobs, increasing turnover costs; it hurts firms’ ability to recruit especially good workers; and it weakens workers’ incentives.19 Given the magnitude of these effects, I have been more surprised at the magnitude of the wage decreases, especially in Greece, and I have not been surprised at the countervailing decreases in productivity noted in chapter 3.20

Why internal devaluations often lead to large decreases in GDP

Internal devaluations were seen, as we have noted, both as a way of correcting an external imbalance and of supporting a weak macroeconomy, for as exports increased, the economy would grow. The Troika has been consistently disappointed: the growth in exports has been smaller than expected (as we have just observed), but the decrease in GDP has been much larger than they expected—even larger than can be accounted for by the disappointing performance of exports.

Their analyses made two crucial mistakes. First, they paid insufficient attention to what would happen to the large and important nontraded goods sector, which includes everything from restaurants and haircuts to doctors and teachers, and typically amounts to something something like two-thirds of GDP. (By contrast, manufactured goods, such as textiles and cars, are called “traded” goods.) The contraction in demand and output in the nontraded sectors outpaced the slow response in the export sector—explaining the large decreases in GDP.

When countries (or firms and households within a country) are indebted in euros (or in a foreign currency), an internal devaluation increases leverage, or the ratio of what households, corporations, and even governments owe relative to their (nominal) income. High leverage is widely viewed to have been critical in bringing on the Great Recession. Internal devaluation increases economic fragility by bringing more households and firms to the brink of bankruptcy. Inevitably, they cut back spending on everything. The cutbacks in imports were one reason that trade balance was improved; the cutbacks in domestically produced goods is one reason that GDP declined so much.

The multiple consequences and manifestations of this fragility were apparent in the East Asia crisis. As their exchange rate fell, many firms and households simply couldn’t repay what they owed in foreign currencies. Defaults and bankruptcies soared. Compounding this problem were strategic defaults by households and firms that might have repaid their debts at great cost. They took advantage of the confusion surrounding mass foreclosures and bankruptcies to attempt to renegotiate their own debts.

This then had follow-on effects. First, lenders suffered. Their balance sheets deteriorated as defaults mounted. Their ability and willingness to make loans decreased. In the economic turmoil following wage cuts and the other economic changes associated with crises, risk increased, and this, too, discouraged lending.21

In the euro crisis, the exchange rate (of one country within the eurozone relative to others) couldn’t change, but internal devaluation, attempting a real devaluation, had exactly the same effects, with all the consequences noted here.

One of the consequences was to the banking system, and the failure to take adequate account of these was the second major reason that the Troika underestimated the magnitude of the adverse effects of internal devaluation. And they should have been aware that these effects were likely to be worse in Europe than in East Asia: as the banks weakened as a result of defaults and bankruptcies, money could easily move out of the banks in the weak countries to those in the strong within the eurozone. This in turn would lead to further decreases in lending and further decreases in GDP.

Explaining the disappointing performance of exports

All of this helps explain the disappointing performance of exports. First, even when wages fell, firms often didn’t pass on the wage cuts to prices. They were worried. They knew that if they needed funds, they couldn’t turn to the banks. It became imperative for them to build up their balance sheets—and for many of them, the crisis itself had done marked damage to their net worth. Firms operating simultaneously in both trade and nontraded sectors might be especially affected, because even if exports had remained strong, domestic sales declined. One of the few ways that firms have to strengthen their balance sheet is to maintain prices. Sure, there is a long-run cost of keeping prices high—but the short-run benefits in the world of austerity and tight finance that the eurozone had created in the crisis countries outweighed those costs. We see this reluctance to pass on wage cuts in the form of lower prices dramatically in the data. If wage cuts were fully passed on, real wages (wages adjusted for inflation) would have remained constant; prices would have fallen in line with the reduction in labor costs. In fact, as we noted earlier, unit labor costs decreased by about 16 percent from 2008 to 2014 in Greece. While nominal wages and labor costs were thus plummeting, prices continued to rise, albeit at a slower rate than in the rest of the eurozone.

This was not the only unintended but important supply-side effect of the eurozone reliance on internal devaluation. Earlier, I described the impact on the financial system. One of the most important costs of firms is the cost of capital. While eurozone leaders were preaching about the importance of the restoration of competitiveness in the crisis countries, they were actually undermining competitiveness—for both the structure of the eurozone and its policies led to a higher cost of capital, as banks were weakened and money fled the crisis countries.

Ironically, the “tough love” that was partly built into the structure of the eurozone, partly as a matter of policy choice, while intended to help exports (by driving down wages), hurt them in still another way. As small enterprises faced an increasing risk of bankruptcy, foreigners who might have bought their goods shied away, worried that when the time came for delivery, the companies would be unable to do so. They worried that before that date, they might be forced into bankruptcy. A retail outlet needed to be sure that the goods would be in the store in time for Christmas. A missed delivery could represent a massive loss of profits. Thus, again, demand and supply were intertwined.22The failure to ensure an adequate supply of capital and the increasing risk of bankruptcy made it even more difficult for the crisis countries to export.

Not only was the euro structure and policy bad in the short run: it was bad in the long run. The increase in risk meant that firms were less willing to undertake investments or even increase employment at any interest rate.

So once again, internal devaluation, while intended to increase competitiveness, restore external balance, and promote employment, had just the opposite effect—not just in the short run but even in the long.

Imbalances within Europe: competitive devaluations

Because wage flexibility differed among the countries of the eurozone, simply leaving it to the markets to adjust meant that, at least in the short run, trade imbalances could even grow. Those with more flexible wages and prices got a competitive advantage over their neighbors. Germany had even managed to get its workers to agree to lowering wages and benefits in the 1990s. But what matters more often is how effective unions are in just keeping up with inflation and getting a share of increased productivity. Such differences arise as a result of institutional arrangements (for instance, the way wages get bargained—in some countries, wages are bargained at the national level, in others at the sectoral level, in still others at the level of the firm or even of a production unit), cultural differences (German workers seemed more willing to accept wage cuts than workers in other countries), and structural differences (German households were less indebted than in other countries, so the adverse effects of increased leverage on households from wage cuts were much smaller).

Of course, these differences wouldn’t have mattered that much if the countries had not joined together to share a single currency. Then countries with more wage rigidities could have compensated by lowering their exchange rate. They could have sustained their economy by lowering interest rates. By joining the eurozone, they had given up these options.

In a world in which countries can set their exchange rate (the value of their currency, say, relative to the dollar), when a country lowers its exchange rate to get a competitive advantage over others—for instance, to help restore its economy to full employment—we say it has engaged in a competitive devaluation. It is a form of beggar-thy-neighbor policy: one country gains at the expense of its trading partners. Beggar-thy-neighbor policies marked the Great Depression, and one of the reasons for the founding of the International Monetary Fund was to discourage such competitive devaluations. Within a currency area, such as the eurozone, countries obviously can’t engage in the traditional form of competitive devaluation. But what we have just described is another form of competitive devaluation—where wages are suppressed so that the real exchange rate is lowered relative to one’s neighbors. And this form of competitive devaluation is just as much a beggar-thy-neighbor policy but even more onerous: the burden of the invidious policy is placed on the workers in the country engaging in the policy.

A bias toward unemployment

There was one more aspect of the design of the eurozone that made matters even worse. As we observed in chapter 1, the European Central Bank is required to focus on inflation. In effect, the eurozone, in its very construction, was biased toward having higher unemployment—toward having a more poorly performing macroeconomy on average than would have been the case had it had a more balanced mandate like that of the United States. And that simply exacerbated the problem of unemployment in the crisis countries.

Zero lower bound

But then, even the stronger countries came to be affected. The EU had succeeded in making Europe more economically integrated. A large fraction of each country’s trade is with other countries within Europe (close to two-thirds of exported goods and services by value go to other EU member states).23 Weaknesses, not just in the crisis countries but in France and Italy, ricochet back on the stronger, especially so when some of the stronger countries—like Germany—suffer from deficit fetishism, and so maintain austerity, even when they can easily access funds. The “old” doctrines had it, don’t worry: even if a country can’t use fiscal policy, monetary policy can stimulate the economy. If Germany should get weak, clearly, the ECB would come to the rescue by lowering interest rates.

But at the time of the construction of the eurozone, no one conceived that interest rates would ever reach the point where they couldn’t be lowered—that they would hit the zero lower bound.24 Once that happened, the adverse effects of its neighbors’ slowdown couldn’t be offset by monetary policy. No wonder that German growth has been so anemic, as we saw in chapter 3.


The discussion in this section—and even more so in the next, where I explain how persistent trade deficits often lead to crises—poses a puzzle. It should have been expected that taking away the ability to adjust the exchange rate could lead to trade deficits, and that trade deficits would put at risk the stability of the eurozone. Why didn’t those constructing the eurozone focus on these trade deficits? Instead, attention was centered on fiscal deficits.

To understand why they might have done this, one has to go back in time. The idea that government deficits and trade deficits were closely linked was very popular in the early 1990s when the eurozone was founded. It found its way into standard textbooks under the title “the twin deficits.” The reasoning was simple: If government increases spending and nothing else changes, aggregate demand will increase. If the economy is at full employment, that increase in aggregate demand can only be satisfied by increasing imports. If exports don’t change (say, because the exchange rate doesn’t change), then there necessarily must be an increase in the trade deficit. The italicized phrases show the critical assumptions in the reasoning, and subsequent research showed that those assumptions are often not true.

We now understand that trade deficits are often not caused by government profligacy but by private sector excesses, and that accordingly curbing government profligacy—as the convergence criteria attempt to do—won’t necessarily prevent large and persistent trade deficits.

In the United States, for instance, the government deficit came down dramatically during the Clinton administration—by the end of Clinton’s term, there was a surplus—and yet the trade deficit continued to grow. Investment increased, replacing government spending.

In fact, even the causal relationship between the trade deficit and government spending, to the extent that there is such a relationship, has been questioned. When a country experiences a trade deficit, perhaps as a result of a sudden decrease in its exports because of a downturn in a trading partner, unemployment will normally increase as a result of the deficiency in aggregate demand. But since Keynes, democratic governments tend to respond to such increases in unemployment by increasing government spending. Thus, it is not the increase in government spending that causes the increase in trade deficit, but the increase in the trade deficit that causes the government spending.25

What has happened in Europe under the euro has confirmed these insights. Only in Greece was government spending a substantial source of the country’s trade deficit.

There was another reason why many of the founders of the eurozone may have been less worried about trade deficits caused by the private sector. They believed that if that ever happened, it would not be a problem: in their ideology, the private sector could do no wrong. So if there was a trade deficit, it was a good thing. It meant, for instance, that firms were borrowing for productive investments that would yield a return more than enough to offset the debt required to finance it.

That they believed this was testimony to their faith in markets: there is a long history of market excesses, to which Spain and Ireland’s housing bubbles (not to mention that of the United States) add but further examples.


While some in Europe, and especially in Germany, are in denial—they continue to believe, in the face of overwhelming evidence to the contrary, that if one controls government deficits, one can manage trade deficits—others are more realistic. They believe that the problem of trade deficits has to be addressed head-on, demanding even stronger macroeconomic strictures: not only budgetary discipline but also other conditions, such as those associated with current account deficits (gaps between imports and exports).

Proposals to expand the list of requirements have a problem (besides the obvious political one, of how the eurozone could be brought to agree on an expanded set of rules): any such strictures run counter to the euro’s neoliberal architecture. How can one distinguish between a “good” trade deficit—one which is the natural consequence of economic forces—and a “bad” deficit? After the fact, it is clear that the Spanish deficits were “bad.” The funds flowing into the country were used to finance a real estate bubble. That was a mistake. But it was a private sector mistake, just as the tech bubble and the housing bubble in the United States were private sector mistakes. But beforehand, it is hard to tell good investments from those that are part of a wave of irrational exuberance. Market fundamentalists could never tolerate government even attempting to make such a distinction. That would imply relying on the judgment of some government bureaucrat over that of a businessman who presumably was putting his own money and that of his investors at stake. Indeed, even as it seemed obvious that the market had gone awry, with the real estate bubbles in Spain and Ireland, the eurozone’s neoliberal economic leaders were waxing poetic about the wonders of the market. When I said that government might want to tame such excesses, the notion was met with horror: Was I suggesting that government bureaucrats were wiser than the private sector?

Thus, the euro is caught between a rock and a hard place: Markets by themselves can’t do the adjustments necessary to ensure full employment and external balance. Internal devaluations don’t work. And simply regulating fiscal deficits won’t suffice either: it makes ensuring full employment more difficult and does little to ensure trade balance. Regulating the economy in other ways, to ensure that imports are in line with exports, would require interventions in the market economy that would be intolerable, at least from the perspective of the neoliberal ideology that underlies the euro. There are other institutional developments, entailing further economic integration—moving in directions that has enabled a single currency to work in the diverse United States—that might enable the euro to work. We discuss these later in the book. But first, we need to understand more fully the dangers of not correcting current account deficits.


Countries that try to fix their exchange rate relative to others face not only protracted periods of unemployment, because they lack the instruments to return to full employment when they are hit by an adverse shock, but are prone to crises. The same is true of countries within a currency area; they, too, cannot adjust their exchange rates. The economic costs of these crises are enormous; they are felt not only in the high unemployment and lost output today but in lower growth for years—in some cases, decades.

Such crises have happened repeatedly, with the euro crisis being only the most recent and worst example. In chapter 2, for instance, I described the Argentine crisis in 2001-2002.

It is easy to understand why such crises are so common with currency pegs. If somehow the exchange rate becomes too high, there will be a trade deficit, with imports exceeding exports. This deficit has to be financed somehow, offset by what are called capital inflows. These can take the form of debt or direct investments. The problems posed by debt are most obvious: eventually the debt reaches so high a level that creditors’ sentiments begin to change. They worry that they will not be repaid. These sentiment changes can occur gradually or rapidly. Surprisingly—again testimony to market irrationality—the changes often happen suddenly, and not in response to any real “news.” The mystery then is often not that the lenders are pessimistic, but why weren’t they more pessimistic earlier. Such changes have been called “sudden stops”:26 the flow of money into a country suddenly halts and the country goes into crisis.

It is when sentiments change quickly that a debt crisis is likely to follow, as lenders refuse to roll over their debt as it becomes due, and the country can’t find anyone willing to lend. The country’s choices are then limited: it either defaults, or goes to the IMF for a rescue package, accepting the loss of its economic sovereignty with a loan accompanied by strong conditions.27

Typically, when there are private excesses leading to a trade deficit, what is borrowed abroad, say by households or firms, is intermediated by domestic banks, who borrow, say, dollars from abroad and lend them to, say, consumers to buy cars (as in Iceland) or other consumer goods. If the foreign lenders demand repayment from the banks, and the banks can’t come up with the money, we then have a financial crisis.28

The trade deficit itself creates another problem. A high level of imports weakens domestic aggregate demand. Unless there is an investment boom—such as a real estate bubble—to sustain full employment, the government has to spend more.29 In this case, the capital inflows go to finance the resulting government debt. But unless the government spends the money wisely, for instance, on public investments that increase the country’s productivity and output, eventually, the creditors’ sentiments will change.

Again there is a debt crisis: the government can’t roll over its debt, if the debt is short-term. But even if there is no rollover problem, because the debt is long-term, the country suddenly can’t finance its fiscal deficit. If it doesn’t want to finance its spending by printing money, with the possible resulting inflation, it has to cut back on spending, plunging the economy into a downturn.30

Banking and financial crises more generally are associated with economic crises, as aggregate demand and output fall and unemployment soars. In the case of a financial or debt crisis, banks can no longer lend, and firms and households whose spending depends on borrowing have to cut back.

In the case of the euro, of course, there was no option of the exchange rate, say, for Greece or Spain, falling to correct a trade imbalance. This made the economic, debt, and financial crises even worse. As we shall see, to bring back a semblance of “balance,” those in the crisis countries are being sacrificed. With a big enough recession or depression, imports are brought into alignment with exports.

In each of those instances where a crisis emerged, one might well ask, why didn’t the lenders see that the country was getting overindebted? How could they let the situation arise? There are two answers. The first is that financial markets are short sighted. They often can’t see or understand what is going on in their own backyard, as the US 2008 financial crisis demonstrated. Seeing and understanding what is going on elsewhere is even more difficult. Markets get caught up in fads and fashions—lending to Latin America in the 1970s and to East Asia in the early ’90s, for example, which eventually resulted in the Latin American and East Asian crises.

Of course, the bankers don’t like to blame themselves; they blame the borrower. But these loans are voluntary transactions. If there is an irresponsible borrower, it means at a minimum that there is an irresponsible lender, who has not done due diligence. The reality is worse. Lenders are supposed to be experts in risk management. And in many cases, they take advantage of inherent political problems:31 governments are easily induced to borrow excessively. The current government benefits from the increased spending, and the prosperity that results, while costs are borne by some future government.

The interaction between shortsighted financial markets—with incentive structures that encourage reckless lending—and shortsighted governments creates a potentially explosive mix, which, when combined with fixed exchange rates, is almost sure to explode. These problems arise within honest governments, who are simply obeying natural incentives. But obviously, corruption can make matters worse, with a corrupt financial sector paying kickbacks in one form or another to corrupt politicians or their political parties.

The second reason countries get overly indebted is that lenders are regularly bailed out—by the IMF, by the European Central Bank, and by governments. With losses thus socialized, there are especially perverse incentives to engage in excessive lending abroad. What happened in the eurozone is again only the latest of a long string of such bailouts. (These bailouts typically have names of countries, like the Mexican, Korean, Argentinean, Brazilian, Indonesian, and Thai bailouts. But in each instance, the bailout is really just a bailout of Western banks.)

Of course, some observers in Europe have recognized this. That is why they have demanded what has euphemistically been called private sector involvement (which has its own acronym, psi) or bail-ins. Private sector lenders too have to bear losses. But too often, these demands come too late, after the smart short-term money had had a chance to pull out. This places the burden on those who made longer-term loans. (Of course, these were partially to blame for the crisis as a result of their overlending; but the short-term lenders were also to blame—indeed in some cases more so. The long-term loans may have been made in a time where the prospects of the country looked great. The impending problems should have been more apparent to the short-term lenders.) And often, bail-in demands are imposed on the wrong parties. In 2012-2013, Europe demanded that ordinary depositors in Cyprus take a haircut—that is, a write-down in the value of their deposits. The eurozone authorities thought that by forcing depositors to take a bigger loss, they would reduce the amount of money that they would have to fork over.32 Until then, it was assumed that when banks went through restructuring, depositors would be protected. The possibility that depositors could take a loss has undermined confidence in banks throughout the weaker countries of the eurozone and contributed to funds leaving these banks, thereby contributing to Europe’s malaise.33


The role of the single currency in creating crises is greater than this discussion suggests. We began with the assumption that somehow, the exchange rate had gotten misaligned—some countries had too high of an exchange rate. As countries entered the eurozone, there was an attempt to ensure that the exchange rate was “right”—that is, the conversion from the old currency to the euro was done correctly.

Even when they got the initial conditions right, however, the euro itself led to a misalignment—to a real exchange rate (taking account of local prices) that was too high in several countries, so that their imports systematically exceeded exports. Money, for instance, rushed into Spain and Ireland in the years after the start of the euro in 1999, and especially in the years preceding the 2008 crisis. Lenders somehow thought that the elimination of exchange-rate risk meant the elimination of all risk. This was reflected in the low interest rates that these countries had to pay to borrow. By 2005, the risk premium on Greek bonds (relative to German bonds—that is, the amount extra that had to be paid to compensate for market participants’ perception of the higher risk of Greek bonds) fell to a miniscule 0.2 percent; on Italian bonds, it also dropped to 0.2 percent; on Spanish bonds, to 0.001 percent.34

The low interest rates at which funds were available in Spain (combined with the euro-euphoria, the belief that the euro had opened up a new era of prosperity and stability) helped create a real estate bubble; in the years before the 2008 crisis, more homes were constructed in Spain than in France, Germany, and Ireland combined.35 The real estate bubble distorted the Spanish economy. Private sector irrational exuberance, not government spending, put the Spanish economy off-kilter. Indeed, some in the government were worried, and worked hard to diversify the economy, so that if it turned out that there was a real estate bubble, other sectors (in Spain, for instance, green energy and new knowledge sectors) could pick up the slack when it burst. The bubble broke before these efforts were fully successful.

In the absence of the eurozone, a country flooded with speculative money could have raised interest rates to dampen the real estate bubble. But being part of the euro made this impossible. A country like Spain and Ireland facing a real estate bubble could and should have imposed capital gains taxes or imposed restrictions in lending by local banks. But the market fundamentalist ideology dominant in the eurozone ruled these measures out.

The government could, alternatively, have cut back on its spending and allowed wasteful private investment to crowd out more productive public investments. But the governments of Spain and Ireland were flush with tax revenues generated by the boom, and they felt no need to do so. Both countries were in surplus. Managing inflation was the job of the ECB, not their job. Their job was to stick within the budget constraints demanded by the Maastricht Treaty and spend money well; and they were doing that. According to the convergence criteria—and the ideology of the time—everything was fine. In short, the constraints imposed by the eurozone, combined with irrational markets and neoliberal ideology framework created the overvalued euro exchange rate that in turn led to the crises.


The euro created crises for another reason, one that appears to have been largely unanticipated, partly because it created a situation that had never occurred before. Debt crises typically do not occur in countries that have borrowed in their own currency. They can at least meet the promises they made by printing more of their own money. The United States will never have a Greek-style crisis, simply because it can print the money that it owes.36

In the past, countries had a choice: issue debt in a foreign currency, and face the risks of owing more—in domestic terms—if the exchange rate falls, and the even worse risk of default if it falls too much; or issue debt in one’s own currency, and pay a higher interest rate. Of course, if markets worked well, the higher interest rate did nothing but compensate foreign lenders for the exchange-rate risk—the money they were repaid might be worth much less than at the time they lent it.37 But shortsighted governments (often egged on by international financiers and even the IMF) encouraged borrowing in foreign currency, because the impact on today’s budget was less, simply because the nominal interest payments were lower.

After the rash of crises in the 1990s and early 2000s, though, many governments had learned their lesson. They began to borrow in local currency, and put pressure on firms and households in their country to borrow in local currency,38 and local currency markets developed.

The eurozone created a new situation. Countries and firms and households within countries borrowed in euros. But though they were borrowing in the currency they used, it was a currency that they did not control.

Europe unwittingly created the familiar problem faced by highly indebted developing countries and emerging markets. Greece does not control the printing presses of the currency in which it has borrowed. They owe money in euros. Their creditors are not willing to roll over their debts. Their earnings of euros from exports are insufficient to pay what is owed. They simply can’t meet their obligations.

Markets should have recognized the new risky situation that the eurozone had created, and limited their lending in response. But as so often happens, they were caught up in another episode of irrational exuberance—euro-euphoria. They focused on the benefits of the elimination of exchange-rate risk, not the costs associated with an increased default risk, especially of government bonds. Many saw the large flows of funds as a sign of success for the eurozone. Having watched analogous instances of capital flowing across borders as restrictions on capital flows were brought down, especially from my perch at the World Bank, I was not so sanguine.

To avoid this situation, Europe as a whole could have borrowed in euros, on-lending the proceeds to the different countries, who would then be responsible for repayment of the money. But it chose not to centralize lending in that way. In making this choice, the eurozone leaders enhanced the likelihood of a debt crisis.

Not only did those countries signing up to the eurozone not fully realize the consequences of borrowing in a currency out of one’s control, they also didn’t realize the implications for their national sovereignty: a transfer of power had occurred that could be—and was—abused. When lenders wouldn’t lend to, say, Spain, the only recourse the country had was to turn to their partners in the eurozone, to get money through the European Central Bank or through some other mechanism.39 It was a fateful development.

Without access to funds, the country would careen toward bankruptcy. But there is no good international legal framework for dealing with the bankruptcies of countries, as Argentina and many other countries have learned to their chagrin. In the absence of such a framework, defaults can be very costly. Creditors have an incentive to scare the country at risk—to make them believe that the costs of default will be very high, far higher than knuckling under to the creditors’ demands.40

As soon as some of the countries in the eurozone owed money to other member countries, the currency union had changed: rather than a partnership of equals striving to adopt policies that benefit each other, the ECB and eurozone authorities have become credit collection agencies for the lender nations, with Germany particularly influential. Though this was never the intent when the eurozone was created, that this is what happened should have been expected. It is simply a reflection of the old adage “He who pays the piper calls the tune.” Germany had the money. The German parliament had to approve any significant new program. It became clear that their parliament would only approve these new programs if there were sufficient “conditions” imposed on the crisis countries. The Germans have repeatedly even challenged the constitutionality of the actions the ECB designed to help the crisis countries.

The power to withhold credit becomes the power to force a country to effectively cede its economic sovereignty, and that is precisely what the Troika, including the ECB, have done, most visibly to Greece and its banks, but to a lesser extent to the other crisis countries. They have imposed policies not designed to promote full employment and growth but to create surpluses that in principle might enable the debtor countries to repay what is owed.


I have explained why, with fixed exchange rates, countries may wind up with large trade deficits, and how a sudden stop in the willingness of others to finance that deficit can precipitate a crisis. The way Europe has chosen to get rid of trade deficits is to put the economy into depression. When the country is in a depression, it no longer buys goods from abroad. Imports are reduced. The “cure” is as bad as or worse than the disease.

But even trade surpluses are a problem. Germany has been running huge trade surpluses—consistently larger than China’s as a percentage of GDP, and in many years, even larger in dollar terms. Indeed, in recent years, its surpluses as a percentage of its GDP have been almost twice that of China’s.41 The United States has criticized China’s surpluses, saying they represented a risk to global stability. The reason is simple: over the entire world, the sum of the surpluses has to equal the sum of the deficits. If some country is running a surplus, exporting more than it is importing, other countries must be running a deficit, that is, importing more than they are exporting. So if deficits are a problem, so too for surpluses. If China’s surpluses are a global problem, so too are Germany’s.42

Similarly, if the exchange rate is set so that the eurozone as a whole has a trade balance, which it roughly did in the years before the crisis, and Germany has a surplus, that means the rest of the eurozone must have a deficit.

In some respects, surpluses are an even bigger problem than deficits, as Keynes argued,43 because they contribute to a shortfall in global demand. These countries are producing more than they are buying and thus not spending all of their income. Of course, the deficit countries would like to buy more. But if no one is willing to lend to them, they can’t, or if they are under an IMF or Troika program that prevents them from spending more, they can’t. When there are large imbalances (such as associated with Germany’s trade surplus), the not-spending of the surplus countries is not fully compensated for by overspending in the deficit countries. The result is that overall global demand is weakened.

Before the crisis, Germany recycled its surpluses, in effect, lending them to the periphery countries, like Spain and Ireland, which allowed them to run deficits.44 But in doing so, it helped create the euro crisis and enhanced divergence within Europe. The periphery countries became debtors, with Germany as the great creditor. As we have noted, there is no greater divide than the one between creditors and debtors.

Our discussion here has focused on borrowing and lending by the country. The analysis does not distinguish between private and public borrowing. That’s important: trade imbalances can be a problem whether they originate in the public sector (as in Greece) or in the private (as in Spain and Ireland). The convergence criteria only focused on the public sector problem. Historically, private sector borrowing has been as or more important.


The countries in surplus often look at their trade surplus—not buying as much from abroad as they are getting from selling goods and services—and the associated savings as a badge of honor. Savings is a virtue. Germany says that all countries should imitate what it does. But Germany’s virtue is a peculiar one: as we’ve seen, by definition not all countries can run surpluses, and anything Germany does to create its surpluses is, in effect, increasing some other country’s deficit. The country with a deficit, in turn, is likely to face weak demand, and possibly even high unemployment and a crisis. If exports create jobs, then imports destroy them. It is hardly a virtue if one can only obtain it by forcing someone else to be a sinner—if one’s actions inevitably lead to problems in some other country.

Modern economics, beginning with Keynes, has explained that in a world of unemployment there is the paradox of thrift. If everyone tries to save more, but investment is fixed, all that happens is that incomes fall. Ironically, total savings is not increased. The reasoning is simple: In equilibrium, savings must equal investment. Thus, if investment does not change, savings cannot change. If individuals insist, say, on saving a higher fraction of their income, the only way that the level of savings can be changed is if the level of income is reduced.

Today the world is in this precise situation, with a deficiency of aggregate demand leading to slow growth and some 200 million unemployed around the world. This deficiency of aggregate demand is the cause of what many have referred to as global secular stagnation. (The term secular just means that it is long-term as opposed to cyclical, temporary slow growth that is part of recurrent business cycles.) The jobs “gap” has increased enormously since the onset of the Great Recession, with some 60 million fewer jobs in existence than what would have been expected if there had been no crisis.45

There is another reason that surpluses are particularly problematic. Typically, it is easier for the surplus country to deal with its surplus than it is for the deficit country to do something about its deficit. There is ample scope, for instance, for both Germany and China to increase wages, especially at the bottom. Until recently, Germany hasn’t even had a minimum wage, and even now, its minimum wage is only €8.50 an hour (which at 2015 average exchange rates was equivalent to $9.35),46 as compared to €9.47 (roughly $10.42) in France. If German workers’ incomes increased, they would buy more, including more imported goods.

By and large, the countries in crisis, and even France and Italy, have managed to reduce their current account deficits; with Germany not only having maintained but actually increased its surplus, the eurozone now has a large overall surplus estimated to be $452 billion in 2015.47 But by the basic arithmetic of global deficits and surpluses, this has been accomplished only by increasing deficits in the rest of the world. It will be hard, if not impossible, for the eurozone to maintain this surplus indefinitely without problems appearing in the rest of the world. Something will have to give.


When individuals, firms, or countries impose costs on others—which they do not themselves have to pay—economists say that there is an externality. The analysis above explained how German trade surpluses impose externalities on the rest of the world, including on its partners within the eurozone. A well-designed global economic system—and a well-designed eurozone—would have adopted measures to deal with this externality. Currently, at the global level, we have no way of inducing countries not to run surpluses.48 The G-20—the group of the 20 largest economies, which sees its role as helping coordinate global economic policies—attempts to cajole countries not to do so. Ironically, it has focused its attention on China, whose surpluses have been coming down, instead of Germany.

Within the eurozone, too, nothing has been done to curb surpluses. As we have noted, the focus of attention has been on government activities, and only on deficits. The mistaken presumption has been that government deficits are the main source of externalities, the main aspect of a country’s behavior that impinges on the well-being of others and the functioning of the economic system.


The founders of the euro knew that making a single currency work for a diverse set of countries would not be easy. But their analysis of what it would take was deeply flawed: the convergence criteria that they formulated, limiting public (fiscal) deficits and debts, made the task of achieving full employment throughout Europe even more difficult. Adding to the difficulty was the task of limiting trade deficits, the creation of which the euro itself was at least partially responsible. Persistent trade deficits set the scene for crises: the predictable and predicted crises emerged just a decade after the beginning of the euro.


Seeing how things were turning out so differently from what had been promised—crises and depressions rather than a new era of prosperity—one might have hoped that Europe’s leaders would have realized both that the economic analysis underlying the euro was flawed and that, if the euro was to work, the “incomplete project” needed to be completed, and in a hurry. They had to put into place the institutional arrangements required to make up for the loss of countries’ ability to use the interest and exchange rate to maintain full employment and to keep imports in line with exports.

Instead, Germany has tried to blame the euro crisis on failures to enforce budgetary discipline. Our analysis has argued otherwise: it is the very structure of the eurozone itself, not even the failings of the individual countries, that is to blame. Market mechanisms (internal devaluation) are not an adequate substitute for the loss of the ability to adjust interest and exchange rates. The euro created the euro crisis.

That the narrative blaming the euro crisis on fiscal deficits was wrong should have been obvious: several countries in Europe had maintained fiscal discipline and yet have been facing severe unemployment, even crises—not just the allegedly profligate countries of Europe’s south, but even the more responsible countries of the north, Ireland and Finland. Market economies can suffer from crises simply as a result of the dynamics of capitalism; in the absence of adequate regulation, there are often credit bubbles. But the ideology of neoliberalism ignored these sources of volatility—the source of the Great Recession itself as well as the East Asia crisis—as it conveniently focused on the budgetary failures in Greece, which were easier to understand, and excoriate.

This failure to diagnose the source of the eurozone’s problem was inevitably linked to the failure to take actions that would address those problems. Instead of correcting the underlying problem, they continued implementing policies based an obviously flawed theory. As the crisis unfolded, they renewed their commitment to the convergence criteria, binding themselves to stricter enforcement.

That they did so suggests one thing: it was a matter of ideology, not economic science. It was a willful failure not to look at the evidence. It may not have been in Germany’s interest to understand the failures, for that might have called upon it to do more than just lecture its partners.


Although the convergence criteria were supposed to foster convergence, the structure of the eurozone has increased differences among the countries of the region in fundamental ways—exacerbating the large differences that already existed when the eurozone was formed. Most importantly, it increased the divide between creditor and debtor countries. The next chapter explains the ways in which the current eurozone structure results in the stronger countries within the eurozone becoming stronger, and the weak weaker. The creditor countries become richer; the debtor countries, poorer. And a hoped-for convergence has transformed into divergence.