The prolonged slump in the Eurozone since 2008 reminds us that asymmetric shocks in a currency area matter substantially – as the classical on optimal currency areas literature pointed out more than a half century ago (Mundell 1961, Kenen 1969). However, while the original shock triggering the crisis was a sudden stop in capital flows to the Periphery (Baldwin et al 2015), the fact that it happened in an area of permanently fixed exchange rates was unprecedented, and magnified the scale of the problem.
This raises a new (and EZ-specific) set of policy questions. The first is: How can we prevent and tackle sudden capital flow stops in a currency area?
More than any other developed economy in the world, the Eurozone currently looks prone to slide into a Japanese-style prolonged malaise. There are similar symptoms, such as the fragility of the banking system and the burden of accumulated sovereign debt in several countries. But, relative to Japan in the nineties, there are, as well, new specificities.
- First, the apparent re-emergence of hysteresis in unemployment (Blanchard and Summers 1985, Galí 2015).
- Second, a pronounced and persistent asymmetry in the unemployment rates between the Core and the Periphery.
The latter feature is illustrated in the left-hand panel of Figure 1, which plots the persistent gap in the unemployment rates between the so-called GIIPS countries (Greece, Ireland, Italy, Portugal and Spain) and Germany since 2008. This suggests a second EZ-specific policy problem – how can we tackle hysteresis in the presence of a persistent regional asymmetry in the unemployment rates – especially in light of a binding zero lower bound (ZLB) constraint on monetary policy.
Figure 1. Unemployment rate (left) , CA balance in GIIPS and Germany (central), dispersion in unemployment rates in all EA countries (right)
Note: dispersion is computed as cross-country standard deviation normalised by cross-country mean.
The central panel of Figure 1 depicts the evolution of current account imbalances between Germany and the GIIPS; whereas the right-hand panel depicts the evolution in the dispersion of unemployment across all EZ members since 2000. Notice that the 2000-08 period of widening current account imbalances corresponds to a narrowing of unemployment dispersion, with the reverse pattern emerging with the onset of the crisis. In other words, when asymmetries vanish in one dimension, they re-emerge in another.
The above evidence illustrates clearly that asymmetries constitute a central problem for EZ policymaking. The boom-bust cycle between 2000 and 2008 is essentially a story of cyclical asymmetries between the Core and the Periphery – initially, widening relative competitiveness and current account imbalances; then, deeply asymmetric recoveries.
I will start discussing the role of asymmetries from the perspective of monetary policy, but will suggest that the same arguments might naturally lead to a reconsideration of how fiscal policy should be conducted within the EZ.
Monetary policy and cyclical asymmetries
A central policy question is whether asymmetries should be viewed as a natural feature of a currency area, or as a structural problem. Broadly speaking, asymmetries can be of three (not mutually exclusive) types:
- First, asymmetries in shocks (i.e., country-specific disturbances);
- Second, asymmetries in structural features of the economy (e.g., price and/or wage rigidity, or frictions in labour/goods markets); and,
- Third, asymmetries in business cycles, due to the combined effect of the first two asymmetries.
It is important to emphasise that asymmetries in a currency area should not be prevented per se. Asymmetries should be relevant for monetary policy decision-making to the extent that asymmetries in business cycles are due to asymmetries in shocks and structures. This is what should be meant by cyclical (and inefficient) asymmetries.
- Under these conditions, asymmetries generate an inefficient degree of divergence in business cycles (a ‘business cycle-divergence gap’); that is, an excess cross-country dispersion relative to what should be deemed natural as a mere result of country-specific disturbances.
Simply put, monetary policy in the EZ does not address cyclical asymmetries. In fact, complacency on this matter has persisted during the ‘EZ Great Moderation phase’ (2000-08), probably due to the optical illusion that asymmetries among member countries were gradually vanishing. Even worse, the ECB conducted policy ‘as if’ those asymmetries did not exist. Two prominent examples suffice. First, the ECB accepting sovereign bonds as collateral in Repo transactions as if those bonds were perfectly substitutable. Second, the ECB inflation target being expressed as an average inflation rate as if the EZ was a relatively homogenous (and closed) economy.
It is therefore natural to ask what the cost might have been (and still could be) from the ECB’s neglect of asymmetries. In the presence of cyclical asymmetries of the third type, it is conceivable that, for example, targeting a simple average inflation rate might have entailed large welfare costs.
What does economic theory suggest?
The recent literature suggests a consistent message when it comes to the role that asymmetries should play in monetary policy making. Unlike a homogenous, closed economy and/or a multiple country setting with flexible exchange rates, the optimal monetary policy objective function in a currency area should feature the following two targets (Benigno 2004, Corsetti et al 2011):
- A nominal rigidity-weighted (rather than GDP-weighted) average of the inflation rates of the member countries (i.e., with weights reflecting the respective, country-specific, structural frictions, such as the degree of price or wage stickiness, and/or the degree of persistence in inflation ).
Therefore the inflation rate of the Periphery – comprising the so-called GIIPS countries where nominal rigidities are structurally higher – should hold a higher weight than the inflation rate of the Core.
- A within-Eurozone terms-of-trade gap (i.e., the deviation of the current within-Eurozone terms-of-trade from its natural value; the one prevailing in the absence of nominal rigidities).
Variations in relative competitiveness across regions (of which we had a vigorous manifestation during 2000-08) should be adequately addressed, rather than neglected.
Targeting the arithmetic (or GDP-weighted) average inflation rate in the EZ is hardly justifiable in terms of efficiency. More broadly, what matters in a currency area is not only the average level of inflation, but also its composition.
To better appreciate this point, and as a mere illustration, we conduct the following exercise. We first draw information on the average frequency of price changes for EZ countries from the micro-data based study by Dhyne et al. (2005). These data are reported in Table 1. The higher the value of each entry, the higher the degree of price flexibility. Next, we compute an index of price rigidity for each country as the inverse of each entry, and take the average of that index for the Core and the Periphery countries.1 Finally, we compute an artificial nominal-rigidity weighted HICP inflation rate, weighing Core and Periphery according to their relative degree of price rigidity. Noticeably, the outcome of this exercise is that, in computing the weighted average inflation rate for the EZ, the relative weight of the Periphery should be higher than the one of the Core (respectively 0.557 vs 0.443).
Table 1. Percentage of prices changing in a given month
Source: Dhyne et al. (2005).
Figure 2 displays the HICP official inflation rate for the EZ against the nominal-rigidity weighted HICP inflation series obtained with the above procedure.
Figure 2. HICP inflation rate: official vs. nominal-rigidity weighted
Two results stand out. First, in the 2000-06 “moderation phase” the two measures of inflation diverge significantly, with the nominal-rigidity weighted measure consistently exceeding the official measure. The gap between the two series is a first rough measure of the inefficiency stemming from the “neglect attitude” toward imbalances that characterized this phase of the Ecb policy. Even more importantly, the nominal-rigidity weighted index would have pointed out, in that period, a more dramatic (excess) deviation from the official target of 2 percent. Second, and interestingly, the gap between the two measures almost disappears in the post-2006 phase, when overall inflation becomes more unstable.
Hence we see that inefficient imbalances in the relative inflation rates between Core and Periphery tend to emerge, somewhat subtly, during periods of stability of the general level of inflation, when complacency about the overall inflation performance might lead monetary policy to neglect them.
Consider now the current state of the EZ economy. While raising the average inflation rate back to its target is certainly desirable, it should not be neglected that what the EZ needed since 2008 (and still needs) is an adjustment in the Core-Periphery real exchange rate. Against this background, and to the extent that it characterises itself as relative deflation in the Periphery, the current ‘deflation (or low inflation) evil’ might appear under a different light. Notice, though, that ‘relative deflation’ in the Periphery might result not only from absolute deflation in the Periphery, but also from inflation in the Core (or both).
On the other hand, it should be acknowledged that, currently, part of the overall EZ-wide low inflation symptom is the direct implication of a persistently binding ZLB constraint. Therefore:
- A further, currency area-specific policy problem consists of distinguishing between the benign relative deflation signal (i.e., the one fostering the required terms-of-trade adjustment between Core and Periphery) and the malign one (i.e., due to the occurrence of the ZLB).
Monetary policy and structural reforms
The way the ECB has so far addressed the asymmetry problem is in its enduring emphasis on ‘structural reforms’ (allegedly to be implemented in the Periphery). Very few central banks in the world (if any) make the reference to structural reforms such a persistent argument of their official policy statement. However, while the ECB acknowledges that structural asymmetries do exist (and are important), it fails to take them explicitly into account in its policy setting.
Put differently, the current policymaking environment in the EZ holds that:
- Structural asymmetries should be dealt with by fiscal policy; and,
- Monetary policy should be vocal about the importance of reducing structural asymmetries, but should refrain from taking them explicitly into account.
Invoking the reduction of structural asymmetries (the so-called ‘structural reforms’), however, without an explicit, country-specific, aggregate demand management, might even be counterproductive. Consider, for instance, the repeated emphasis, in the aftermath of the EZ crisis, on labour market reforms and enhanced wage flexibility (to be implemented in the Periphery). The argument holds that, in light of the asymmetric shock that hit the Periphery between 2008 and 2011, a wage-based ‘internal devaluation’ is warranted to offset the lack of exchange rate flexibility.
For the sake of argument, let’s think what happens, in an economy, if wages are ‘made more flexible’. For one, wages will be more unstable (which is a cost); but, on the other hand, employment will be more stable (a benefit). What determines the breakdown between ‘higher wage instability’ and ‘higher employment stability’ is the ability of monetary policy to manage aggregate demand – the stronger this ability, the larger the benefits in terms of employment stabilisation relative to the costs in terms of wage volatility. In the EZ, however, where monetary policy does not respond to asymmetric disturbances, a mere emphasis on higher wage flexibility in the Periphery might result, almost exclusively, in higher wage instability, with limited benefits in terms of employment stabilisation (Galí and Monacelli 2015). Therefore:
- In a currency area, policies addressing asymmetric aggregate demand deficiencies are complementary to structural (supply-side) reforms.
Importantly, this argument reinforces the idea that monetary policy should tackle asymmetries ex ante, by targeting inflation rates asymmetrically and/or by explicitly targeting the within-area terms of trade. It should be noted, however, that the same argument equally supports the view that fiscal policy (in its aggregate demand management dimension), alongside structural reforms, should be actively used to address asymmetries. We will return on this point below.
Current account asymmetries
How to optimally manage international capital flows is among the most prominent policy dilemmas of our times, and particularly so within an area of permanently fixed exchange rates. The magnitude of cross-border capital flows observed during the 2000-08 boom-bust EZ cycle, from the Core to the Periphery and vice versa, was a vigorous manifestation of the relevance of asymmetries within a currency area (see again Figure 1).
What role the management of capital flows should play in optimal policymaking, and especially within a currency area, remains a largely unexplored issue. Even less is known about the appropriate tools to tackle the problem. Certainly, capital controls as policy instruments are once again fashionable (IMF 2011).
The general, more recent, view is that capital controls should be used to address externalities. The latter are of two different types:
- Pecuniary externalities, deriving from asset valuations affecting the ability of private agents and financial institutions to borrow (in the presence of credit market imperfections – Lorenzoni 2011, Bianchi and Mendoza 2010, Bianchi 2011) ;
‘Financial’ externalities of this type typically arise from cross-border capital flows (already so prominent within the EZ), giving rise to current account imbalances.
- Aggregate demand externalities (deriving from the presence of nominal rigidities – Farhi and Werning 2012, Korinek and Simsek 2015).
This type of externality is a genuine feature of a currency union, due to the combined effect, in the presence of full capital mobility, of nominal rigidities and permanently fixed exchange rates (Schmitt-Grohe and Uribe 2015).
It is important to notice that the aggregate demand externality argument is exactly the one (expounded above) whereby optimal monetary policy in a currency area should target the within-area ‘terms of trade gap’.
Presumably, each type of externality might require the design of a different policy tool. This is a central, but much less understood, question, which definitely warrants further research.
There is however a common element – both types of externalities give rise to misalignments in international relative prices, with potentially destabilising effects. Those effects can be dangerously exacerbated by the participation in a currency area, due to the impossibility of correcting them (either ex ante or ex post) via variations in nominal exchange rates.
This observation reinforces the argument, already suggested above, whereby optimal policy in a currency area should give a more prominent role to targeting the within-area terms of trade (or, more generally, the composition of inflation). Whether capital control tools should be used by monetary or fiscal authorities, and whether or not they should be used in a state-contingent fashion, still remain open issues. Fernandez et al (2015) show that, historically and in stark contrast to theory, capital controls are largely acyclical. In light of recent theory, there are therefore strong reasons for employing those tools in a more active fashion, especially within a currency area.
Macroprudential framework, asymmetries and coordination (or lack thereof)
With the onset of the 2011 crisis, the EZ has endowed itself with a new, and somewhat complicated, macroprudential framework (Gadatsch et al. 2015). In this framework, a macroprudential role played by the Ecb in the banking sector co-exists with country-specific institutions aiming at the implementation of regulatory tools such countercyclical capital buffers or loan-to-value ratios. Currently, all EZ countries have either legislation in force or are finalising laws to designate a single macroprudential authority. Although these are general steps in the right direction, they suffer of the same, and fundamental, flaw that affected the design of the fiscal policy framework at the onset of the EZ: a basic lack of coordination.
This lack of coordination in macroprudential policy is potentially highly problematic. For the externalities stemming from the coexistence, within a currency area, of nominal rigidities and fixed exchange rates generate cross-country spillovers similar in nature to those that justify the presence of a common set of fiscal rules, and currently incorporated in the so-called Fiscal Compact. For instance, overborrowing in the Periphery leads to an excess appreciation of the terms of trade in the Periphery (and, simmetrically, to an excess depreciation in the Core), increasing the likelihood of capital account reversals, which in turn bear (as the EZ crisis manifestly witnessed) area-wide aggregate implications.
Fiscal policy and asymmetries: An argument for coordination
The reasoning so far has focused on monetary policy. Most of the arguments, however, might well support the notion that fiscal policy, rather than monetary policy, should be used to address cyclical asymmetries. All arguments outlined above, in fact, can be interpreted as supportive of some form of fiscal union. However – what exactly should be understood by fiscal union?
Our previous analysis has highlighted three main economic arguments that support the desirability of a fiscal union:
- The presence of an aggregate demand externality (rooted in nominal rigidities coupled with fixed exchange rates), which implies that the within-area terms of trade adjust inefficiently;
- The presence of a pecuniary externality (rooted in financial imperfections), which gives rise to ‘excess’ within-area capital flows; and
- The notion that implementing region-specific supply side reforms (aimed at addressing structural asymmetries) should be complemented by a region-specific management of aggregate demand.
A fiscal union is therefore any fiscal policy arrangement aimed at addressing these. Whether a fiscal union should take the practical form of (state contingent) capital controls, taxes, government spending, or transfers is a fundamental issue which remains open to debate.
Here I wish to emphasise a different angle:
- In the EZ public discourse, ‘fiscal union’ invariably means fiscal policy simultaneity.
For instance, the so-called ‘Juncker plan’ of EZ-wide public investment in infrastructures. However, what a fiscal union should primarily evoke is a correct concept of policy coordination.
This means fostering the notion that, in the presence of cyclical asymmetries, an asymmetric fiscal policy response is desirable from a social welfare perspective. In practice, fiscal policy coordination means that, in light of a regional slump caused by an asymmetric sudden stop in the Periphery, a coordinated EZ fiscal stimulus might well contemplate cutting taxes (or increasing government spending) only in the Periphery (Galí and Monacelli 2008) – with the understanding that, to put it bluntly, this would be in the interest of the Core countries as well. Hence, and to summarise:
- In the presence of cyclical asymmetries (of the inefficient type discussed above), it is desirable from a social (i.e., union-wide) welfare perspective for fiscal policy to be asymmetric.
Consider, for instance, the current debate on the desirability of using ‘fiscal space’ to address the weak recovery in the Periphery (particularly prominent in Italy). Presumably, it would make a substantial difference, especially in terms of market sentiment, whether a fiscal expansion (e.g., a tax cut) were implemented in the Periphery only as a form of coordinated fiscal response, as opposed to being the result of a lengthy and discretionary bargaining process unilaterally promoted by a single member country.
This is the correct logic of fiscal coordination (as far as stabilisation policy is concerned), which is still lacking in the current EZ version of fiscal coordination – the so-called Fiscal Compact. Notice that this argument does not necessarily imply that fiscal policy should be centralised in the hands of a single fiscal authority (a desirable outcome per se, but probably unrealistic, even in the medium run). What ‘fiscal coordination’ ought to mean, though, is that, in light of cyclical asymmetries, a cooperative (and union-wide optimal) fiscal policy response should be asymmetric in nature – that is, targeting country-specific demand deficiencies in the interest of stabilising area-wide macroeconomic conditions.
Asymmetries are a defining feature of the EZ. Many believe they should be dealt with by supply-side, long-run reforms since they are ‘structural’. In this mind set, the asymmetries are taken as given when it comes to monetary and fiscal stabilisation policy. It is however plausible that cyclical asymmetries continue to persist in the Eurozone precisely because they are left un-tackled by stabilisation policy. The boom-bust cycle of the Eurozone was a plain manifestation of the dire consequences of this neglectful attitude. It seems therefore of paramount importance that EZ policymaking, whether monetary or fiscal, re-considers cyclical asymmetries as a central target of its framework.
Benigno P.P and D. Lopez-Salido (2006), “Inflation Persistence and Optimal Monetary Policy in the Euro Area”, Journal of Money, Credit and Banking.
Bianchi J. (2011), Overborrowing and Systemic Externalities in the Business Cycle”, American Economic Review.
Bianchi J. and E. Mendoza (2010), “Overborrowing, Financial Crisis and Macroprudential Policy”, NBER wp
Blanchard, O. and Lawrence Summers (1986): "Hysteresis and the European Unemployment Problem," NBER Macroeconomics Annual.
Dhyne E., Luis J. Álvarez, Hervé Le Bihan, Giovanni Veronese, Daniel Dias, Johannes Hoffmann, Nicole Jonker, Patrick Lünnemann, Fabio Rumler and Jouko Vilmunen. (2005), “Price setting in the euro area: Some stylised facts from Individual Consumer Price Data”, ECB Working Paper No. 524.
Farhi, E., and I. Werning (2012). “Dealing with the Trilemma: Optimal Capital Controls with Fixed Exchange Rates”.
Gadatsch, T. Körner, I. Schnabel, B.Weigert (2015), “Macroprudential Supervision in the Eurozone: Beware of Excessive Expectations”, VoxEU.org
Galí J. (2015), “Insider-Outsider Labor Markets, Hysteresis and Monetary Policy”, Mimeo CREI.
Galí J. and T. Monacelli (2008), “Optimal Monetary and Fiscal Policy in a Currency Union”, Journal of International Economics.
Galí J. and T. Monacelli (2015), “Understanding the Gains from Wage Flexibility: The Exchange Rate Connection”, mimeo CREI and Bocconi University.
Korinek A. and A. Simsek (2015), “Liquidity Traps and Excessive Leverage”, American Economic Review forthcoming.
Lorenzoni G., (2008), “Inefficient Credit Booms”, Review of Economic Studies.
Schmitt-Grohe S. and M. Uribe (2015), “Downward Nominal Wage Rigidity, Currency Pegs, and Involuntary Unemployment, Journal of Political Economy forthcoming.
1 Based on Table 1, the Core countries are Belgium, Austria, Germany, Netherlands, France and Finland; the Periphery countries are Italy, Spain and Portugal.