Chapter 9. Whither EMU?
- Tamim Bayoumi
- Published Date:
- October 2017
Flaws in the design of the European Monetary Union (EMU) helped generate the North Atlantic crisis as well as amplify and elongate its costs. For almost a decade after the introduction of the Euro in January, 1999, a prolonged expansion eased the task of the European Central Bank (ECB). Indeed, the new currency union was viewed as a success that was promoting growing prosperity and a convergence in income between the less wealthy peripheral countries and the more prosperous core. In reality, these trends largely reflected unsustainable financial booms in the periphery, much of it financed by loans from undercapitalized mega-banks in the Euro area core. These growing imbalances were missed because policymakers overestimated the level of integration of the Euro area and hence its ability to cope with shocks to members. The financial chaos after the Lehman Brothers bankruptcy created deep problems for the undercapitalized core mega-banks system that were subsequently exacerbated by fiscal crises in the periphery, where Greece, Ireland, and Portugal were forced into economic recovery programs, Spain suffered a house price collapse that required major support from the government, and the sustainability of high Italian government debt was called into question.
The currency union so carefully negotiated in the Maastricht Treaty provided an inadequate response. The independent central bank lowered interest rates but its charter hampered it from providing more direct support to crisis-hit banks or governments. Rather, bank support was the province of national governments, putting pressure on public finances particularly in Greece, Ireland, Portugal, and Spain, the countries most affected by asset price collapses. Fiscal support for these crisis countries from the rest of the currency union was minimal given rules that explicitly excluded direct support from other governments or from the small federal budget.
The result was a Euro area depression. This was particularly evident in the periphery, as doubts about the viability of major banks and the sustainability of government finances fed on each other to drive an upward spiral in borrowing costs. Growth in the core also suffered a prolonged slowdown as weak banks pulled back from lending and recessions in the periphery crimped demand for exports. The downturn in output was made worse by the importance given to a rapid return to fiscal probity in the absence of a fiscal backstop as well as hints that weaker members might have to leave the union.1 These largely self-inflicted wounds came to a head in 2012 as an upward spiral in borrowing costs in the periphery, most notably in Greece, put the viability of the currency union into question. In the end, the decision to move to a banking union provided cover for the president of the ECB to vow to do “whatever it takes” to preserve the Euro area, an announcement that rapidly calmed markets.
As discussed in Chapter 4, the deficiencies in the design of the Euro area reflected underlying tensions between the German and French concepts of the role of the currency union in economic integration. The German vision was that the currency union would come about only after successful economic and political integration, implying that the union would require minimal macroeconomic support beyond that provided by existing arrangements. By contrast, the French advocated an early currency union in order to promote economic convergence, combined with monetary and fiscal macroeconomic support to help members faced with recessions. The eventual unhappy compromise involved a “French” early move to a single money with a “German” independent central bank and minimal provisions for support for weaker members.
A crucial issue is whether the changes to Euro area design as a result of the crisis will allow it to respond better to future macroeconomic shocks or if more institutional surgery is needed. In other words, is the Euro area moving toward the smoothly functioning currency union that the French assumed would occur? Or is it likely to remain brittle and subject to major shocks and economic strife that will harm the future prospects of members? To answer these questions, it is first necessary to assess how much the architecture of the Euro area has improved.
* * *
The Institutional Response to the Euro Area Crisis
In the wake of the crisis, the Euro area has been made more responsive to large economic shocks by adding to the already extensive alphabet soup of European institutions. Rules about how to support countries facing a crisis have been expanded and codified through the creation of the European Stability Mechanism (ESM) and the ECB’s Outright Monetary Transaction program (OMT). In addition, financial supervision has been centralized at the European Central Bank (ECB). At the same time, the fiscal rules in the Stability and Growth Pact (SGP) have been tightened in an attempt to avoid a repeat of the lax fiscal policies in Greece that helped to precipitate the crisis.
The European Stability Mechanism, activated in September 2012, is a permanent Euro-area-wide facility that can lend to members facing an economic or financial crisis. The ESM merged two temporary funds used to provide money for the crisis programs with Greece, Ireland, and Portugal—the European Financial Stability Facility and the European Financial Stabilisation Mechanism. The ESM’s lending procedures are closely modeled on those of the International Monetary Fund. It can only provide a Euro area member with loans or precautionary funds (i.e., an overdraft facility in which money is only borrowed if it is needed) if the loans are accompanied by a Memorandum of Understanding outlining a program to restore financial health that has been vetted by the IMF, the European Commission, and the ECB (the European troika that organized the bailouts of Greece, Ireland, and Portugal over the crisis). The ESM has so far been used to support bank recapitalization by the Spanish government and a crisis program for Cyprus. In addition, although it has not done so, the ESM has some funds to directly recapitalize banks and take over their operations. The implications of this option will be discussed later in the context of the move to a single bank supervisor.
European Central Bank support to Euro area members in crisis has also been expanded and codified through its Outright Monetary Transactions program (OMT). The OMT program was announced in September 2012 on the heels of the speech by President Mario Draghi to do “whatever it takes” to preserve the Euro. It allows the ECB to purchase unlimited amounts of government bonds of a member that has an ESM program, bond yields at stressed levels, and full access to market financing, thereby providing a monetary backstop for members facing a crisis. Together, the ESM and OMT facilities provide an emergency safety net for members that is separate from the international one provided by the IMF (while still utilizing the IMF’s expertise) and is more effective since it includes unlimited central bank support.
Another far-reaching institutional improvement brought about by the crisis was to switch the supervision of banks from national regulators to the ECB. Moving supervision of banks from competing national regulators to the ECB provides the potential for a true banking union in the Euro area. So far, however, the process is only half complete. While Euro area financial supervision has been centralized, the costs associated with bank support remain largely national. This is clearly true for retail deposit insurance, which remains the responsibility of member governments. The situation with regard to recapitalizing a troubled bank is more complex. A Euro-area-wide resolution fund to support insolvent banks has been agreed and is being gradually built using charges on banks. However, even when it is complete the fund will only comprise €55 billion (less than ½ percent of future Euro area output). This is too small a sum to provide meaningful guarantees for an industry dominated by mega-banks, even taking account of new rules to lower the cost of government rescues by requiring that uninsured lenders such as bond holders contribute to any recapitalization (called “bail-in” rules because private creditors participate in the rescue rather than being “bailed-out”). Once the Euro-area-wide fund is exhausted, the backstop remains individual Euro area members. There is also an option to provide centralized ESM funds to directly recapitalize banks, but the funds available to cope with such a crisis are again quite limited. The result is a half-formed banking union in which supervision is centralized but the costs of deposit insurance and of bank recapitalization basically remain with national governments.
Turning to fiscal arrangements, the Stability and Growth Pact was designed to prevent a crisis triggered by individual members using the credibility of the Euro area to borrow excessively. Despite the pact, however, the Greek government succeeded in doing exactly that in the run-up to the crisis. The October 2009 revelation that the Greek fiscal deficit was actually 12.5 percent of output (later revised up to 15.8 percent) rather than the official target of 3.7 percent triggered an abrupt rise in sovereign borrowing costs in the periphery. This was exacerbated by statements from the German government that high Greek debt put the Euro at risk and that Greece might have to leave the single currency. This drove interest rates even higher in Greece and other crisis countries as investors sought compensation for the risk that Greece and other crisis countries might leave the Euro area and repay their debts using a highly depreciated new currency (so-called denomination risk).2 The rise in government borrowing costs fed through to bank funding, which had already spiked after the Lehman Brothers bankruptcy in September 2008 had raised the costs of intra-bank lending on concerns that European banks could meet a similar fate. This created an upward spiral in the sovereign-bank interest rates that almost tore the currency union apart, before President Draghi announced that the ECB was prepared to do whatever it took to preserve the currency union.
In response to the Greek debacle, new and stronger rules have been brought in to bolster the fiscal rules embodied in the Stability and Growth Pact. On the preventative side, governments are now under stricter surveillance rules including (for example) a requirement to follow better fiscal procedures, create independent national monitoring agencies and economic forecasts, and provide early submission of budget plans to the Commission and the European finance ministers. Rules on excessive deficits have also been tightened by making it more difficult for ministers to overrule any decision by the European Commission on sanctions for a country that is running an excessive deficit. In addition, the rate at which debt above 60 percent of output should be reduced has been codified. However no fiscal sanctions have ever actually been applied, suggesting that enforcement of fiscal rules continues to be problematic.
The key issue looking forward is whether the new SGP rules will be sufficiently tough to prevent a future fiscal crisis. This is because once a crisis has occurred, rules that require a return to fiscal stability are generally counterproductive. For example, after the Greek revelations in 2009, the SGP excess deficits procedures led to a rapid tightening of fiscal policy that worsened the Euro area recession. Between 2010 and 2013, the cyclically adjusted fiscal deficit was reduced by over 5 percentage points of output in Greece, Ireland, Portugal and Spain, and by 3 percent in France and Italy. To put this in context, the 3¼ percent consolidation for the entire Euro area over this three-year period was larger than the tightening required in the four years before Euro area entry in 1999, when governments were scrambling to achieve the Maastricht criteria. It was also tougher. Fiscal belt tightening in the run-up to the creation of the Euro was aided by accounting tricks and lower nominal interest rates. By contrast, over the Euro area crisis budgets were closely monitored while the increase in interest costs made it more difficult to lower the deficit.
The changes in design have made the Euro area better at combating future crises. In particular, access to centralized fiscal support through the ESM and to central bank funds through the OMT provide a convincing financial backstop. However, with the important exception of centralized banking supervision, little has been done to strengthen buffers to members that are not in crisis. Fiscal policy remains decentralized, and loans to support countries in difficulties are limited to members who are prepared to accept a program with the troika. The existence of this safety net provides some benefits even to governments that are not prepared to enter into a program, since it reduces the risk of members in crisis being forced to leave the Euro area involuntarily. In addition, the tougher approach to SGP surveillance makes it less likely that the Greek fiscal shenanigans in the run-up to the crisis will be repeated. However, it is also true that differences in government borrowing rates are higher than their (unsustainably) low level before the crisis and that fiscal flexibility in response to economic disturbances remains constrained. Similarly, in banking the glass is only half full since centralized supervision has not been supported by effective centralized support for failing banks.
The likelihood of further major changes to the underlying design of the Euro area is fading along with the immediate risks to the region. The history of the European Union is that major alterations occur at times of distress. As Jean Monnet, one of the architects of the European Union, said: “People only accept change when they are faced with necessity, and only recognize necessity when a crisis is upon them.” It is extremely difficult to tinker with the design of the currency union in normal times, particularly since many of the rules are embodied in the Maastricht Treaty. Indeed, it is notable that none of the recent changes to the structure of EMU have involved major amendments to the Maastricht Treaty, which would involve reopening long-running Franco-German disagreements about the purpose of the single currency.
The difficulty in changing the rules also reflects the European Union’s focus on consensus across countries.3 The key driver of the European project is the 1957 Treaty of Rome pledge to work towards “an ever closer union of the European peoples”. The “ever closer union” part commits members of the European Union to move towards a federated state with a single money, an arrangement that is usually backed by a political structure based on one-person-one-vote and majority rule so as to be able to adapt to changing circumstances. The mention of the “European peoples”, on the other hand, committed the Union to widening its membership. A wide union involving diverse countries typically requires a confederated structure, where individual countries are protected against unwelcome changes by rules emphasizing one-state-one-vote and consensus. In practice, the power center of European Union—the Commission—has a confederated structure, which makes significant changes to Euro area arrangements difficult to achieve.
This suggests that the current design of EMU is likely to remain in place for some time, although as the book went to press the election of President Macron in France may yield further major changes. In any case, the key issue for the future of EMU is the validity of the key assumption in the French/Monetarist view of monetary union, namely that a single currency can create the economic integration needed for a smoothly functioning currency area. If this is correct, then the North Atlantic crisis was simply a major hiccup along the road to ever closer union. If this assumption is incorrect and the German/Economist view that a single currency will not generate rapid integration and hence that the currency union should have been delayed until after the union had become more economically and politically integrated, then the North Atlantic crisis may be a harbinger of future economic doldrums. Answering this question requires assessing the properties of an integrated and smoothly functioning currency union.
* * *
What Makes a Good Currency Union?
The theory of optimum currency areas is the branch of economics that examines the suitability of a region for a single currency. Adopting a single money has both benefits and costs. The obvious benefit is that it makes it easier and cheaper to transact with others within the union. A person in Chicago may find it easier and simpler to buy a widget from a firm in distant Los Angeles rather than one in closer Toronto because the former also uses the US dollar while the latter purchase involves the uncertainty and expense of paying in Canadian dollars. In short, a single currency promotes closer economic ties. This is the essence of the French/Monetarist argument that the Euro will create an integrated economy.
The cost of a single currency is that it limits monetary flexibility in response to shocks. A single money requires a single monetary policy. If Toronto is booming and Chicago is in an economic slump, then the Bank of Canada can tighten its policy in order to cool the situation in Toronto while the Federal Reserve can loosen policy to give support to Chicago. By contrast, if Los Angeles is booming and Chicago is in a slump then the Federal Reserve faces a dilemma. Should it tighten policies to cool the LA economy or loosen them to support Chicago? In practice, it would presumably split the difference and leave policies unchanged. This means that buffers other than monetary policy are needed to cushion shocks that create divergent economic conditions across a currency union.
This implies three approaches to assessing the suitability of a region for a single money. The first is the degree to which the members face similar economic shocks. If Los Angeles and Chicago tend to be hit by similar shocks, so that when LA is booming Chicago is likely also booming, then tighter Fed policies will suit both places. It is only when the shocks diverge significantly than a single currency creates issues.4 The first criterion for a smooth functioning monetary union is therefore that economic disturbances across the members are similar—that they are coherent rather than incoherent.
A second criterion is the degree to which divergent shocks can be absorbed by other economic buffers that obviate the need for monetary support. The initial work on optimum currency areas by Robert Mundell in the 1960s focused on the importance of labor mobility in cushioning differing shocks.5 If labor can move easily between Chicago and Los Angeles, then differences in underlying shocks would be solved by workers moving out of slumping Chicago and into booming LA. This is also linked with the ease of firms to move and hire and fire, since labor mobility only works if firms are sufficiently flexible to create new jobs.
In addition, fiscal policy can help to cushion the shock. Indeed, a federal fiscal system automatically provides more demand to regions in a downswing and withdraws it from regions in an upswing since federal tax revenues fall and spending rises in a slump while the opposite occurs in a boom. This is a regional equivalent to national automatic stabilizers. But there is a crucial difference, namely that the federal deficit remains unchanged because the higher deficit in Chicago is cancelled out by the lower one in LA. In essence, the federal tax system transfers money from Los Angeles to Chicago as their economic conditions diverge. Since the level of federal debt is unchanged, there is no reason for taxpayers in (say) Chicago to partially offset the boost to demand coming from the federal government in anticipation of higher taxes in the future. This provides a more effective stimulus than if Chicago tries to use its local taxes and spending to boost the economy, since the stimulus will be muted as taxpayers save part of the tax cut in anticipation of higher taxes down the road. However, a federal tax system also implies long-term transfers from rich to poor regions, which requires a high level of political cohesion that the Euro area has never achieved, which is why fiscal policy remains a national responsibility.
An integrated banking system can also play a role in offsetting divergent economic shocks, although this mechanism has been given little attention in the literature. If banking is regional, then the downturn in Chicago will put stress on local banks even as the boom in Los Angeles leaves its banks flush with cash. This will tend to loosen financial conditions in booming Los Angeles while tightening them in Chicago, exacerbating the initial divergent shock. While in theory LA banks could lend money to people in Chicago, in practice it is difficult for an LA bank to assess the risk of a loan to a person or firm in unfamiliar Chicago. Similarly, uncertainty about the financial health of Chicago banks may make their LA cousins reluctant to loan them funds through the interbank market. National supervision can help reduce these uncertainties, as it makes it easier for LA banks to assess the creditworthiness of individual Chicago banks, allowing them to channel their funds to sound institutions. National banks with operations in both cities can also provide support as such banks have the expertise to assess risks in both markets. However, national banks are a double-edged sword. If loans in Chicago are relatively profitable because the economy is slumping but the value of collateral such as houses remain strong, then a national bank will push resources into Chicago. If, however, the main problem is a slump in the value of collateral then a national bank may pull loans out of slumping Chicago and toward Los Angeles.6
A third criterion for assessing whether the Euro area is a natural currency union is the speed with which the union is becoming more economically integrated over time. Nobody believed in the early 1990s that the European economy was sufficiently integrated to comprise a natural currency union. Rather, the French assumption was that the single money would create the closer economic bonds needed to make a union work smoothly and generate the convergence that would justify its existence. Assessing the validity of this assumption requires a dynamic assessment of the rate at which the Euro has led to greater integration of the participants.
* * *
How Fast Is EMU Integrating?
How does the current Euro area stack up against these various criteria for a functioning currency union able to smoothly absorb regional shocks? The earlier discussion suggests three ways of assessing this issue. How far did the EMU increase economic integration across the participants, how much did it promote more similar economic shocks, and how far did it enhance the banking, fiscal, labor market, and corporate shock absorbers? In answering these question, I focus on the initial eleven members who joined the union in 1999 plus Greece, which adopted the Euro in 2001, the same countries that are examined in the first section of this book on the long-term origins of the crisis.7
How Far Did the Euro Boost Economic Integration?
How far did the creation of the Euro promote economic integration across the initial participants? A natural measure is the impact of the introduction of the Euro on international trade. There are several reasons for using trade as a measure of economic integration. It tracks a basic economic link across countries, namely the degree to which goods produced in a country cross international borders rather than staying within them. In addition, trade has the advantage of being well tracked by statistical agencies and of having a successful empirical literature explaining the level of trade including, as discussed below, the impact of a shared currency. On the other hand, it is difficult to do a comparison of intra-Euro area trade with intra-US trade, although the limited information available suggests that US trade is much higher. Researchers have also found that price movements across US regions are much more similar than across Euro area countries, suggesting that the United States is a more integrated union.8
Figure 53 shows trade of the initial Euro area members with each other, with other major advanced economies, with eastern Europe, with China, and with the rest of the world since 1980, all measured as a ratio of the output of the initial Euro members. Both the initial Euro area members and the other advanced economies were slow growing economies that were already highly integrated into the global trading system by 1980. By contrast, eastern Europe and China were much more dynamic economies that grew much faster by opening to the rest of the world and using low wage costs to develop manufacturing (the rest of the world largely comprises commodity exporters).
Figure 53.Trade between initial EMU entrants expanded moderately.
Source: IMF Direction of Trade Statistics and IFS.
Trade across the early EMU members (the light gray dotted area) has risen gradually since 1980, from slightly under 20 percent of output in 1980 to its current level of just under 30 percent of output. Most of this expansion occurred between the agreement to move to a single currency in the early 1990s and the North Atlantic crisis of 2008 (with a notable fillip in 2000 with the advent of EMU) followed by stasis since the crisis. This basic pattern is repeated as trade with other advanced countries (such as the United States, United Kingdom, and Switzerland) also grew slowly. By contrast, the much smaller level of trade between the initial Euro members and eastern Europe and China expanded much faster in every time period. This spectacular rate of increase was boosted by rapid growth and pro-trade policies—the eastern European countries first left the Soviet sphere in the late 1980s and applied for EU membership in the 1990s while China entered the World Trade Organization in 2001.9 Indeed, one of the surprises about the European monetary union is that the German supply chain did not move south to the periphery of the (then) Euro area but rather to lower wage economies in eastern Europe. Trade with eastern Europe (as a ratio to the initial Euro area entrants output) rose almost ten-fold between the early 1990s, as did trade with China. The fact that trade expanded much faster with eastern Europe and China than across other early members of the currency union implies that monetary union supported trade but was not a game changer.
Comparing trade across the initial members of EMU and other advanced countries reinforces the view that the single currency provided only a modest boost to economic integration. The rise in the ratio of intra-EMU trade to output is around 50 percent, somewhat faster than the 30 percent seen for other advanced economies, with most of this divergence occurring in the decade from the adoption of the single currency in 1999 to the North Atlantic financial crisis of 2008. Within the initial Euro area members, the boost to trade seems to have been largest in the more export-orientated hard currency countries (Germany, the Netherlands, Belgium, Austria, and Finland) than in the soft currency remainder, suggesting that the benefits of the single currency were largest for those countries that were already closely integrated into the global trading system.
These results are consistent with more formal analysis looking at the impact of currency unions on trade. These studies use the empirically successful gravity model of trade in which exports and imports between pairs of countries are linked with the size of the two economies and the distance between them (the name comes from the similarity with Newton’s law in which gravitational force depends on the mass of the two objects and the distance between them). The results find that a single currency provides a moderate boost to trade that ranges from insignificant to 50 percent.10 Comparing the increase in trade within early Euro adopters (50 percent) with the increase to other advanced economies (30 percent) suggests an impact somewhere around the middle of this range. As to timing, a study using the gravity framework to examine the impact of EMU over time confirms that most of the trade boost happened around adoption, with only limited evidence for continuing long-term benefits. Consistent with the fact that the European supply chain anchored around Germany expanded into the east rather than the south, the boost to trade from joining EMU seems to be larger for the newer eastern European members.11 All of this suggests that the boost to trade integration from the introduction of the Euro has been modest. But did EMU make shocks more similar?
Did the Euro Make Shocks More Similar?
In the early 1990s, when a European currency union was in the offing, Professor Barry Eichengreen and I reported a simple approach to assessing the suitability of European countries for a single currency based on the similarity of their macroeconomic shocks.12 The first part of our procedure identified macroeconomic shocks hitting different countries from the 1960s to the late 1980s. Next, we calculated the correlation between the shocks of potential EMU members and Germany, the anchor of any future monetary union. Finally, the results across European countries were compared with correlations between US regions and the Mideast, the anchor region of US monetary union that includes the financial hub of New York and the political capital, Washington DC. This approach remains widely accepted in the literature.13
The results indicated that the European countries were less well suited for a currency union than the United States. Both regions had a core with relatively similar underlying shocks and a periphery with less similar ones, but in both cases the European shocks were less coherent that those in the United States. The European core comprised the countries clustered around Germany and hence with closer economic ties—France, the Netherlands, and Belgium—coincidentally the members of the Euro area core identified in the first section of this book. Similarly, the currency union periphery we identified largely comprises the countries labeled earlier in this book as part of the Euro area periphery, including Italy, Spain, Greece, Ireland, and Portugal. In the United States, the core comprised a cluster around the Mideast—the Great Lakes, the Southeast, and New England—plus the Far West, dominated by the diversified economy of the California. The periphery comprised of the Rocky Mountains and the Southwest regions that specialized in mining and oil production, respectively.
Updating these results since the 1990s allows an assessment of whether the Euro area is now a better candidate for a currency union. In particular, by comparing the changes in correlations across the Euro area with those for US regions, the role of the introduction of the Euro can be assessed. Slightly ironically, the most important complication in this analysis is the North Atlantic financial crisis. This generates exceptionally large negative shocks across virtually all countries in 2009, including those in the Euro area. When included, the increase in the correlations of the shocks makes the Euro area appear to be a better candidate for a currency union even though the crisis almost destroyed it. Since 2009 involved a large global shock that has little to do with the suitability of the Euro area for a currency union, it is excluded from the analysis. For the US regions (where the exclusion of 2009 makes less of a difference) the recent correlations are similar to the 1960–88 results, a sensible baseline given that the US currency union has remained relatively stable over time.
Comparing the Euro area and US results for the recent period suggests that the Euro area remains less well suited for a currency union than the United States (Figure 54 shows correlations of aggregate supply shocks on the vertical axis and aggregate demand ones on the horizontal axis). The US continues to divide into a core with relatively high correlations of shocks with the anchor and a periphery with less similar shocks. Furthermore, the two groups have similar memberships to the earlier analysis.
Figure 54.The US remains a more coherent currency union than the Euro area. Correlations with anchor region (Mideast in US, Germany in EMU).
Source: Bayoumi and Eichengreen (2017).
The Euro area results are less easy to interpret. The shocks remain less coherent than in the United States, insofar as they are further away from the top right-hand corner. In addition, the dots are further from the lead diagonal, implying that some regions have significantly higher correlations in one shock than the other, a pattern also seen in the earlier time period. The most surprising result, however, is that the countries closest to the “good” upper right-hand corner comprise Italy, Portugal, Spain, Ireland, and (possibly) Greece—in other worlds the crisis countries. Their limited integration with Germany makes them implausible candidates for a Euro area “core”, suggesting that the outcomes reflect distortions in the Euro area.
Consistent with the discussion in the first part of this book, our analysis interprets the results as reflecting a Euro area banking super-cycle, as lax supervision of internal risk models drove a massive pre-crisis boom and a post-crisis slump in the periphery, linked to the fortunes of the export-led German economy.14 These results suggest that rather than creating similar economic shocks to which a single monetary policy could respond, the creation of the Euro was accompanied by centrifugal financial forces that pulled regions apart, and were ill-suited to a single monetary policy. This implies that optimum currency theorists should have focused much more on the role of financial integration for a smoothly functioning currency union. In particular, it also underlines the potential for stronger and more centralized Euro area financial regulation to support the currency union. This brings us to the issue of the extent that non-monetary buffers can compensate for the loss of monetary autonomy as a result of a monetary union.
Is the Euro Promoting Better Shock Absorbers?
How far can non-monetary buffers to shocks make up for loss in national monetary policy? An obvious place to start is banking. While flawed financial integration helped to drive the 2008–2012 banking cum fiscal crises, the recent switch to ECB supervision of banks is a major change to the Euro area structure. Centralized bank regulation could impact a currency union in two ways. The first is by encouraging the creation of region-wide banks. European banking remained relatively splintered after the introduction of the Euro because supervision stayed in the hands of national regulators who showed more interest in promoting their own banks than in creating a truly integrated financial system.15 Centralized supervision of the major banks by the ECB is likely to erode this resistance. In addition, centralizing the costs of bank rescues would eliminate the feedback from banking problems onto national debt that generated the sovereign-bank spiral featuring so prominently in the European half of the North Atlantic crisis.
The history of the US monetary union provides a way of examining the effects of the creation of national banks. Through the late 1970s, US banks were constrained to operate in one state and sometimes only in one branch within that state. This situation changed as earlier restrictions were lifted, and by the late 1990s national banks had appeared. Indeed, this transformed the geographic range of US banks more dramatically than any change as a result of centralized ECB supervision, since EU banks were not banned from cross-border mergers in the way regulated US banks were confined to operate in a single state, they were merely discouraged by local regulators.
The literature does indeed suggest that the spread of interstate banking made US regional business cycles smaller and more correlated, even if the similarity of pre- and post-1990 correlations of shocks suggests the impact was limited.16 The theoretical impact on output volatility is ambiguous, as the benefits from having access to out-of-state loans over a credit crunch has to be set against the loss of local loans when there is a fall in collateral values. Empirical results, however, find that fluctuations in output in US states became smaller and more coherent as US bank integration increased.17 Strikingly, this improvement occurred even as supervision of the regulated banks was being tightened as a result of the introduction of prompt corrective action in the early 1990s. This suggests that part of the benefits came from the fact that banks were constrained from excessive speculation. Consistent with this interpretation, when the same framework is applied across countries rather than US states there are few benefits and even possibly some increase in volatility from bank integration, a more negative pattern that is consistent with the fact the international debt flows are more prone to speculation and contagion.
A strong case can also be made that centralized support for banks would improve the ability of the Euro area to respond to shocks. The United States has had federal bank support since the Great Depression, when the 1933 Banking Act created federal deposit insurance and the Federal Deposit Insurance Corporation (FDIC). The FDIC oversees insured depositors but also has a much broader mandate to resolve failing banks, for example by arranging the sale of a failing bank to a competitor (US bank rescues often involve a healthy bank buying the insolvent one for a low price).18 The FDIC is funded by levies on the banking industry, but also has a line of credit with the federal government so that in practice bank deposits are guaranteed by the full faith and credit of the US government.
Federal insurance greatly limited the feedback from earlier regional banking problems on the local and national economy. For example, the US experienced three significant housing busts from the late 1980s through the mid-1990s. Real house prices fell by one-third in New England between 1988 and 1997, with similar outcomes in California (1989–97) and Texas (1987–97).19 Hence, from the late 1980s through the mid-1990s about a quarter of the US economy was experiencing major falls in real housing prices. While the economies of the three regions suffered (their growth rates dipped some 1½ percent below their long-term average) the rest of the US economy continued to grow at its long-term rate. And while these regional slowdowns were accompanied by financial disruption, most notably in New England, the housing slumps did not generate the kind of negative fiscal-banking feedback loops that happened in the Euro area from 2009 to 2012 for the simple reason that the costs of cleaning-up banking were borne by the federal government. The advantages of federal funding of bank rescues was also be seen over the North Atlantic financial crisis, when the United States was able to organize a coherent response to the banking crisis involving federal support along with credible stress tests that forced banks to re-capitalize. In the Euro area, the response was much less effective in large part because member governments had to worry much more about the potential costs to their own budgets.
The policy lesson is that the Euro area banking system needs two further reforms to switch from the amplifier of shocks seen over the crisis to an effective shock absorber. The first it to reign in the internal risk models that encouraged the creation of thinly capitalized mega-banks that were prone to speculation. The second is to switch to Euro-area-wide bank support by centralizing the funding of deposit insurance and bank rescues. This would allow the creation of an FDIC-like organization to oversee bank rescues across the Euro area to complement the centralized ECB role in bank supervision.
A similar underlying story about the importance of central funding applies to fiscal policy, although for slightly different reasons. Centralized banking support avoids the amplification of cycles coming from the need for individual countries to foot the bill for banking rescues. By contrast, fiscal policy can directly offset a shock to demand in a country through higher spending and lower taxes. As discussed earlier, in the case of the United States and virtually every other country in the world, the federal budget provides automatic support in response to divergent shocks within the currency union. Research suggests that in the US this amounts to a 20 cent offset for every dollar of shock. Hence, individual US states get a significant fiscal support in a slump even though almost all have balanced budget amendments that constrain their ability to respond through their own budgets.20
The Euro area has much less effective fiscal buffers because of the need for countries to fund their own fiscal response. Any fiscal boost to cushion a downturn involves a rise in national debt that will have to be offset by tighter future policies. Anticipation of these future costs blunts the impact of national fiscal support compared to the equivalent support from a federal system. While the size of this “Ricardian offset” to the fiscal stimulus coming from the anticipation of tighter future policies is difficult to measure, most estimates suggest it approximately halves the support provided by any increase in the fiscal deficit.21 Hence, in order to provide the same support as the 20–cent fiscal boost coming from the US federal system, a Euro area member would have to provide a stimulus of 40 cents since the support to activity is only half as large, implying a need to deliberately augment automatic stabilizers. However, the European Stability and Growth Pact generally limits active policies in response to shocks.22 The pact is designed to minimize the risks from excess borrowing using the credibility of the Euro area by limiting any build-up of debt rather than to respond to shocks. Fiscal deficits are supposed to be capped at 3 percent of output and by a “debt break” that requires members with government debt of over the 60 percent of output ceiling to reduce their debt ratio at a specific rate. If deficits exceed either cap, an excessive deficit procedure defines the rate at which countries should rectify the situation. While this calculation adjusts for the impact of the cycle on the deficit, so as to avoid excessively tight policies in the face of a downturn, there is no provision for active policies to support the economy. If history is any guide, Euro area members will run deficits close to the 3 percent limit in good times, constraining the ability to provide active government fiscal support in slumps.
Another buffer that can reduce the costs of diverse regional shocks is to improve the flexibility of an economy. Rather than reduce the impact of shocks, as with centralized banking and fiscal policy, a more flexible economy reduces the costs of coping with the consequences of shocks as labor and/or firms move from the region in a slump to the region that is booming. The United States, a large continent-wide currency union encompassing many different regions and economic realities, has relatively flexible labor and product markets. Such markets have helped the economy to adapt to unexpected changes such as the rise in oil prices in the 1970s (which supported the oil-producing Southwest region while hurting the manufacturing core of the Great Lakes economy), through the slump in oil prices later in the 1980s, the rise in the information technology sector in the 1990s, and the more recent oil shale boom. While these shocks have all involved regional winners and losers, the costs have been reduced by the ability of the US economy to adapt.
An important expectation for European monetary union was that the constraints of a single currency would inspire deregulation of labor and product markets.23 This would increase the rate of growth of the region and make it more flexible and hence a better currency union. These incentives were expected to be particularly powerful on the more sclerotic southern members of the currency union that generally had the least flexible economies as a result of government rules and regulations. This process, which was supported by inter-governmental initiatives such as the Lisbon strategy launched in 2000 just after the introduction of the Euro, would accelerate overall growth and the convergence of incomes across the monetary union, making it more homogeneous in terms of underlying prosperity and flexibility.
There is indeed evidence that European labor markets are becoming more flexible over time. Recent analysis of European countries has found that the short-term role of migration in responding to labor market shocks has increased significantly between a pre-EMU sample (1977–99) and a largely post-EMU one (1990–2013).24 In the pre-EMU period, migration absorbed about 20 percent of labor market shocks in the first two years after an employment shock. In the more recent period this has risen to more like 30 percent. There has been a similar increase in the (larger) role played by migration in absorbing labor shocks within EU countries, from 40 to 50 percent. The gap between the role of migration in intra- and inter-country mobility suggests that impediments such as language and culture continue to limit labor market adjustment between initial Euro area members. Consistent with this, analysis suggests that EU enlargement in 2004 and 2007 that added lower-wage eastern European countries to the single market aided the speed of labor market adjustment as these workers proved more willing to migrate.25 Sadly, this has come at the cost of a political backlash against such migrants, as exemplified by Brexit and rising support for anti-immigrant parties elsewhere.
Despite these improvements, labor market flexibility across Euro area members remains significantly lower than within the United States even though the gap has narrowed. This reflects both the increase in labor mobility in the Euro area and a decrease in the United States. The US reduction may reflect a gradual lowering of regional divergences in labor market conditions across regions as the industrial structure has become more similar over time, which has reduced the incentives to migrate.26 This is much less true in the Euro area. Differences in regional unemployment have always been larger in Europe than in the United States and, although the gap started closing after the introduction of the Euro, it has widened again since the crisis, suggesting that much of the earlier progress reflected unsustainable booms in the periphery. All in all, a sensible overall conclusion is that labor market adjustment to shocks across Euro area countries continues to be only about half the size of the United States.
Product market flexibility has also shown some convergence between EMU and the United States, as well as convergence within the Euro area. The Organisation for Economic Cooperation and Development (OECD) produces an index of product market regulation that combines assessments of the role of the state, barriers to entrepreneurship, and barriers to trade and investment for 1998, 2003, 2008 and 2013, conveniently spanning the decade from the creation of EMU until the eve of the crisis (1998–2008) and the early response to the crisis itself (2008–13). The numbers suggest that in 1998 the initial members of EMU had significantly more regulated product markets compared to other advanced economies and a much larger gap with the United States and the United Kingdom, the traditional leaders in this area. The gap between Euro area countries and other advanced economies reversed by 2013, when a typical Euro area country had less regulated product markets than other advanced economies, although it remained some way from the United States and the United Kingdom. There were particularly large improvements after 2008 in the crisis countries, especially Greece, Italy, and Portugal. The major uncertainty is whether the post-crisis impetus for faster reforms will take on a momentum of its own, or fade with memories of the crisis and the end of economic programs backed by the European Commission, the ECB, and the IMF.
Overall, the results suggest that EMU did help to create a more flexible Euro area economy both in absolute terms and compared to other advanced economies. However, not entirely surprisingly, the region remains well behind the United States. This helps to explain why business cycles across Euro area members remain less coherent than across US regions, and hence that Euro area members are more susceptible to regional shocks that a single monetary policy cannot solve.27
* * *
The Future of EMU
EMU occurred in spite of long-standing disagreement over its role in the European integration process. Since at least the Werner Report in the early 1970s, the French and like-minded countries held that an early currency union was needed to create the integrated economy that would produce a smooth-functioning currency union, while the Germans and their supporters felt that EMU should occur only after such integration had already been achieved. While this disagreement went into relative abeyance in the decade after the Euro was launched, when rapid convergence appeared to be occurring, it has resurfaced since the crisis, recast as a discussion about whether to loosen existing policy constraints in order to provide support for countries in crisis or maintain the rules so as to avoid further moral hazard in the future. Tension over this issue is unlikely to be resolved soon given that it has been simmering for almost half a century.
Since the crisis, progress has been made in providing better responses to diverse shocks with the Euro area. Banking supervision has been centralized, providing the potential for a truly integrated banking system of the type taken for granted in most currency unions, including the United States. The issue here is whether internal risk models will be curbed and deposit insurance and other forms of banking support will also be centralized. Such moves have the potential to greatly improve the workings of the monetary union while failure to follow through will leave an unstable, half-formed banking union that will be a significantly less effective shock absorber. In addition, tighter fiscal rules might convince Euro area members to run smaller deficits, allowing more leeway to provide stimulus in the face of negative shocks. Finally, significant progress has been made recently on structural reforms, most notably in the periphery whose economies were the least flexible at the outset of the single currency. As with banking union, the main uncertainty is whether this progress will be maintained as memories of the existential crisis fade and weariness with economic reforms grow.
If the crisis generates a true banking union, lower government deficits and debt, and further progress on structural reforms, the Euro area would to look more and more like the United States, another massive continent-wide economy. EMU would then be moving into the type of smooth currency union that the French/Monetarists predicted that a common currency would create. Such a currency union would find it much easier to cope with unexpected shocks. In particular, banking problems could be approached in a more organized and effective manner, while more fiscal leeway and more economic flexibility would blunt the inevitable trauma for those whose economies are hit with negative shocks. It might also create momentum for further fiscal and political integration.
There is, however, a much less optimistic future if progress stalls on banking reform, government deficits and debt remain high, and structural reforms slow. Under this scenario, the Euro area would continue as an unsatisfactory currency union with limited economic integration. Such a union would remain susceptible to divergent shocks that leave individual members in the economic doldrums for long periods. Such centrifugal economic forces would gradually erode the support for the single money. In this context, it is worth remembering the demise of the nineteenth-century Latin currency union, formed in the 1860s and eventually dissolved by mutual agreement in the 1920s, rather than in the white heat of a crisis. The circumstances are in many ways quite different. The Latin union existed under the classical gold standard that invoked relatively permanent fixed exchange rates and limited monetary autonomy. That said, however, it is very possible that, just like old soldiers, badly designed currency unions do not die but just gradually fade away.