Post on 18-Jan-2019
transcript
April 29, 2014
Global Economics Paper: 224
Economics Research
What makes a monetary union work?
As European policymakers attempt to construct the Euro area’s post-crisis steady state, we seek to
provide investors with a framework to assess whether the institutional changes and structural reforms
being implemented will be sufficient to ensure EMU’s long-term survival.
We argue that establishing the right fiscal and financial institutional framework – to deal with problems
when they occur – is more important than trying to ensure that the economic conditions for a monetary
union are ideal.
While real wage flexibility and – to a lesser extent – labour mobility have important roles to play in the
adjustment to regional shocks, business cycle synchronisation and trade integration appear less important.
The negotiations taking place to develop the Euro area’s steady-state framework have developed along
two separate institutional dimensions: one seeks closer fiscal integration, while the other seeks closer
financial/banking integration. Because fiscal and financial integration can act as substitutes for each other,
neither approach is necessarily ‘correct’.
Progress along the fiscal dimension has largely stalled. While the European Stability Mechanism (ESM)
and the ‘enhanced surveillance’ procedure have altered the Euro area’s fiscal framework, there appears
little prospect of more radical options – such as common Eurobond issuance – being implemented.
Progress along the financial dimension remains ongoing, through efforts to establish a Euro area banking
union. But the more far-reaching options along this dimension – such as the introduction of common
deposit insurance – also appear unlikely to be implemented.
Whether the changes that are being implemented will make the Euro area’s institutional structure
sufficiently robust to deal with a future crisis is questionable.
Investors should consider this report as only a single factor in making their investment decision. For Reg AC certification and other important disclosures, see the Disclosure Appendix, or go to www.gs.com/research/hedge.html.
Kevin Daly
+44(20)7774-5908 kevin.daly@gs.com Goldman Sachs International
The Goldman Sachs Group, Inc. Global Investment Research
Kevin Daly and Simon Wan would like to thank Huw Pill, Jan Hatzius, Dom Wilson and Dirk Schumacher for their advice and
detailed comments on previous drafts of this paper. Simon was an intern with the European Economics team.
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 2
Contents
Overview: What makes a monetary union work? 3
Chapter 1: Introduction – A factor-by-factor analysis of what makes a monetary union work 7
Chapter 2: The role of ‘economic’ factors – synchronisation of business cycles, trade integration, labour mobility and
wage flexibility 9
The synchronisation of business cycles 9
A high degree of trade integration 13
A high degree of labour mobility 15
A high degree of wage flexibility 19
Chapter 3: The role of ‘institutional’ factors – fiscal transfers, financial integration and political union 21
Mechanisms for fiscal transfers 21
Financial integration and banking union 23
Political integration 26
Chapter 4: Conclusions: Institutional reform plans are not yet sufficiently robust 28
Bibliography 31
Disclosure Appendix 34
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 3
Overview: What makes a monetary union work?
Making ‘EMU 2.0’ workable
Market concerns over the survival of the Euro have subsided since ECB President Draghi
made his pledge to do “whatever it takes” to preserve monetary union in July 2012. Yet the
quest to make the Euro area a more ‘workable’ monetary union still faces significant
challenges. At the area-wide level, institutional reform is needed to improve Euro area
governance; and, at the national level, the painful processes of economic restructuring,
fiscal consolidation and private-sector deleveraging all have some way to go before
financial and macroeconomic imbalances are unwound and the conditions for sustainable
economic growth restored.
As European policymakers negotiate the structure of ‘EMU 2.0’, this paper seeks to offer
investors a framework to assess institutional changes and structural reforms and their
implications for the Euro area’s long-term survival. We do not assess the factors underlying
the Euro area’s broader economic performance – many economies that have low GDP per
capita levels and/or growth rates still function well as a monetary union. Rather, our focus
is on establishing the criteria required to ensure that internal adjustments do not prompt
periodic existential crises.
A natural starting point for such an analysis is the (pre-crisis) Optimal Currency Area (OCA)
literature. This proposes a list of at least seven criteria that determine the optimality of a
currency union:
1. Synchronisation of business cycles, so that one monetary policy can fit all.
2. A high degree of goods market (trade) integration between participating states,
to maximise the benefit of sharing a single currency.
3. A high degree of inter-regional labour mobility, to aid in the adjustment to
region-specific shocks.
4. A high degree of wage flexibility, to allow real exchange rate adjustments to play
out more easily in the absence of nominal exchange rate flexibility.
5. Mechanisms for fiscal transfers – such as fiscal federalism – to offset the
negative consequences of region-specific shocks.
6. Financial integration – either via a unified banking system or via greater capital
market integration – to enable greater risk-sharing across the monetary union.
7. A high degree of political and institutional integration, to promote the
acceptance of region-specific shocks (among the electorates of participating states)
and the irrevocability of monetary union (in financial markets).
It is a demanding list. And, were it the case that a high degree of each of these criteria was
required to ensure the Euro area’s long-term survival, then the Euro area would be unlikely
to survive in the long term. But few economies fully satisfy each of the criteria set out
above, yet many function effectively as monetary unions. So, rather than focus on what is
required to make a currency union optimal, we focus on what makes it workable – i.e., we
attempt to separate the necessary from the ‘nice to have’.
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 4
Institutions matter more than ‘economic’ factors
Our analysis draws on the experience of the Euro area and that of the US – a similarly-sized
economy, with diverse states, which (self-evidently) functions as a monetary union.
The seven criteria set out above can be broadly grouped into ‘economic’ factors
(synchronisation of business cycles, the degree of trade integration, labour mobility and
wage/price flexibility) and ‘institutional’ factors (mechanisms for fiscal transfers, financial
integration and political union). One key finding from our analysis is that establishing the
right institutions – to deal with problems when they occur and to help ensure that the
monetary union is credible – is more important than trying to ensure that each of the
‘economic’ conditions for a monetary union are met. While real wage flexibility and – to a
lesser extent – labour mobility have important roles to play in the adjustment to regional
shocks, we argue that business cycle synchronisation and trade integration are less
important than implied by the pre-crisis Optimal Currency Area literature.
Of course, the various characteristics identified by the OCA literature interact with one
another: on some dimensions, they may be complements; on other dimensions,
substitutes. Moreover, the distinction between economic and institutional aspects is – in
some respects – somewhat artificial: for example, wage flexibility and labour mobility
reflect the institutional structure of the labour market in the same way that the extent of
financial integration dictates the mobility of capital across intra-Euro area borders.
That said, in drawing some main messages from our analysis of the seven characteristics
of a currency area listed above, we find the following:
A high degree of business-cycle synchronisation does not appear to be a
necessary condition in making a monetary union work. First, Euro area countries
have exhibited more business-cycle synchronisation than US states on most
measures in the 15 years since EMU began and yet the US functions as a
monetary union while the Euro area currently does not. Second, looking at the
performance of US states over a longer timeframe, we find that severe and
persistent state-specific economic shocks are commonplace, yet this has not
prevented the US from continuing to function as a monetary union.
A high degree of trade integration does not appear to be a determining factor of
workability: the Euro area exhibits a high degree of trade integration but has not
functioned well as a monetary union.
A high degree of labour mobility does not appear to be a sufficient condition of
workability. While labour mobility is higher in the US in general, the response of
net migration to unemployment differences does not appear significantly different
in the Euro area from that in the US, so it is difficult to argue that labour mobility
represents the key distinction between the US’s functioning monetary union and
the Euro area’s malfunctioning union.
A high degree of real wage flexibility – in allowing easier real exchange rate
adjustment – appears to play an important role in offsetting the effects of lost
nominal exchange rate flexibility. Real wage flexibility is higher in the US than in
the Euro area and this greater degree of wage flexibility appears to play an
important role in US regional adjustment.
Fiscal transfers play an important role in offsetting region-specific shocks in the
US but not in the Euro area. We find that the US’s federal fiscal system directly
offsets around 25-30% of the initial effect on income from state-specific economic
shocks, but that only half of this is due to inter-regional insurance. The other half is
due to inter-temporal smoothing, which can be provided by national tax systems.
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 5
Integrated financial and banking markets provide an important means of risk-
sharing in the US and other monetary unions. To the extent that private financial
markets facilitate the sharing of region-specific shocks, this can offset the need for
fiscal risk-sharing (all else equal). That said, one lesson from the Euro area crisis is
that, without the appropriate regulatory structures, greater financial market
integration can also result in increasing overall risks, rather than greater risk-
sharing.
The importance of greater institutional integration is reinforced by the self-
fulfilling aspect of the perception of irrevocability (among investors and
electorates). An important reason why the ‘US Dollar zone’ functions as a
monetary union – despite the existence of frequent and substantial state-specific
shocks – is because few question its existence. When a US state is hit by a
negative shock, it does not face the additional burden of a rise in funding costs
resulting from the perceived risk that it may leave the Dollar zone. By contrast, the
defining characteristic of the Euro crisis has been the emergence of what Mr.
Draghi has labelled “convertibility risk”. As concerns about a possible exit from
the Euro area emerged in some countries, the (natural) financial market response
(of widening spreads and increasing risk premia) exacerbated those concerns
rather than dampened them and the Euro area was placed on a destructive path.
Institutional reform plans are not yet sufficiently robust
The conclusion that ‘institutional’ factors (fiscal transfers, financial integration and political
union) and wage flexibility are more important than other ‘economic’ factors (cyclical
symmetry, the degree of trade integration and labour mobility) is – tentatively – positive for
the Euro area. This is because European policymakers have it within their power to adjust
institutional factors and to implement reforms that would increase wage flexibility,
whereas there is little they can do to adjust deeper economic relationships (such as
business cycle synchronisation and trade integration). This is not to suggest that adjusting
Euro area institutions or implementing labour market reforms to make EMU work is easy –
as it clearly isn’t – but it is at least possible.
In determining which criteria are necessary and which factors a monetary union can work
without, we also need to recognise that there is more than one way to make a monetary
union workable. As Huw Pill has argued in previous research, progress towards deeper
integration in one dimension can be a substitute for making progress along another.1 For
instance, the size of cross-country fiscal transfers in the Euro area may never match inter-
state transfers in the US, but a greater degree of risk-sharing through a more integrated
financial system could be sufficient to make EMU work.
The negotiations taking place to develop the Euro area’s steady-state framework have
developed along the two institutional dimensions discussed here: fiscal/political and
financial/banking. As the Euro area’s principal creditor country, Germany has fought to
ensure that fiscal and financial risks will not be shared across either dimension without
countries first displaying a willingness to consolidate and reform. It has also been one of
the strongest advocates for greater fiscal/political integration, while displaying markedly
less enthusiasm for financial/banking integration. France, by contrast, displays greater
enthusiasm for financial/banking integration, while being reluctant to relinquish
fiscal/political sovereignty.
1 “Creating a workable monetary union”, Huw Pill, European Economics Daily, October 23, 2012.
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 6
Because fiscal and financial integration can act as substitutes for each other, neither
approach is necessarily ‘correct’. Moreover, there is an element of complementarity
between them: banking integration may require a common area-wide fiscal backstop for
the financial system, which raises fundamentally fiscal issues. There is also a question of
sequencing: Germany may justifiably claim that sharing fiscal risks requires that its
partners have demonstrated a willingness and ability to make necessary but painful
adjustments, whereas France and countries in the periphery can equally justifiably argue
that, without explicit financial support from Germany, adjustment is infeasible in economic
and/or political terms. Finally, the choice of how to proceed across these dimensions of
reform and adjustment has important distributional consequences along national lines,
which can naturally complicate the ongoing negotiation process.
Partly as a result of these complications, there has been a marked reduction in the impetus
to implement changes across either dimension since market pressures in the Euro area
have abated following Mr. Draghi’s “whatever it takes” intervention:
In terms of fiscal/political integration, there appears to be a reluctance to move
beyond the ‘enhanced surveillance’ procedures set out in the revamped Stability
and Growth Pact. We are sceptical that the changes implemented to date imply a
sufficient degree of fiscal/political integration to ensure the Euro area’s long-term
survival (at least in the absence of much deeper financial/banking integration).
In terms of financial/banking integration, the introduction of a common
supervisory framework is a necessary – but not sufficient – step towards deeper
financial integration. However, there has been back-tracking on an earlier
commitment to introduce direct recapitalisation of peripheral banks from the ESM
bailout fund – precisely one of the banking issues that potentially has deep fiscal
implications.
Progress has been made but, in our view, Euro area institutions are not yet sufficiently
robust to ensure that the monetary union remains workable in the long run. While
institutional reforms that would represent a sufficient steady-state framework have been
discussed, they do not appear likely to be implemented to a sufficient degree. This is not to
suggest that an eventual break-up of the Euro area is likely, but the Euro area may ‘need’ a
renewed bout of market tension to provide policymakers with the impetus required to
complete the construction of a workable monetary union in order to overcome some of the
impediments noted above.
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 7
Chapter 1: Introduction – A factor-by-factor analysis of what makes
a monetary union work
As policymakers continue to grapple with the challenge of ensuring EMU’s long-run
survival and investors remain sceptical as to whether it will survive in the long run, in this
paper we ask: What makes a monetary union work?
The original Optimal Currency Area (OCA) literature, as first developed by Nobel prize-
winning economist Robert Mundell (1961) and advanced by the economist Ronald
McKinnon (1963), suggests that the benefit of maintaining a currency union depends
negatively on the degree of economic asymmetry (the preponderance of nation-specific
shocks) and positively on the share of output devoted to trade between the member states.
Over time, other studies have added to the list of criteria considered important in
determining whether a monetary union would be ‘optimal’ or not, with a greater focus on
the importance of mechanisms that enable participants in a monetary union to adjust to
region-specific shocks once they occur:
A high degree of inter-regional labour mobility (Feldstein (1998), Bonin et al (2008),
Zimmermann (2009)).
A high degree of wage flexibility (Berthold (1998), Krugman (2012)).
A high degree of fiscal/political integration (Kenen (1969), Sachs and Sala-i-Martin
(1992), Hishow (2007)).
A high degree of financial integration (Asdrubali, Sorensen and Yosha (1997)).
These criteria are likely to be substitutable to some extent – if a monetary union does not
have enough of one, it may be possible to substitute it with more of another. But, while a
number of previous studies have emphasised the importance of one factor over another,
no previous study that we know of has considered each of the factors collectively. Here, we
attempt to fill that gap: we consider the list of criteria proposed in the OCA literature and
identify the criteria that are necessary to make a monetary union work and those that are
simply ‘nice to have’.
Although our primary goal is to draw lessons for the Euro area, our empirical analysis
focuses as much on the US as it does on the Euro area. This is because we know that the
US ‘works’ as a monetary union. Moreover, while 15 years is arguably too short a time
period to judge the importance of each of the criteria in the context of the Euro area, the
US provides an example of a long-running monetary union that is of similar size to the
Euro area and combines a number of diverse states/regions. Trying to understand why the
US functions as a monetary union can help offer an insight into what could make the Euro
area work better.
What we mean by a ‘workable’ monetary union
One of the attractions of setting out a list of criteria that determine whether a monetary
union is ‘optimal’ or not is that it is relatively easy to define optimality. However, while an
optimal (or perfect) monetary union may be easy to define, in practice there is no such
thing as a perfect monetary union. A large number of monetary unions ‘work’ just fine,
even if they fall some way short of optimality. But trying to define precisely what we mean
by ‘work’ in this context is difficult.
We do not mean to assess the factors underlying an economy’s overall economic
performance – many economies that have low GDP per capita levels and/or growth rates
nevertheless function well as a monetary union.
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 8
Nor does our test of workability require that internal shocks within the monetary union
adjust quickly or painlessly – as we discuss in our results, many functioning monetary
unions do not satisfy this criterion. In the US, region-specific shocks often have permanent
effects on employment and income. And many European economies still suffer from the
legacy of regional shocks that pre-date the formation of EMU: parts of the UK are yet to
recover fully from the decline of heavy industry, the former German Democratic Republic
remains heavily dependent on large-scale fiscal transfers from western Germany, and
unemployment is persistently higher in southern Italy than in the north.
Rather, our focus is on establishing a set of criteria that is required to ensure that internal
adjustments – which we view as largely unavoidable – do not prompt periodic existential
crises. In other words, how institutional design ensures that shocks to the monetary union
do not trigger explosive dynamics in financial markets or the real economy that both create
and compound ‘convertibility risk’ leading to existential crises. Instead, we seek to identify
the institutional framework that would induce behaviour that supports the integrity and
viability of the monetary union.
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 9
Chapter 2: The role of ‘economic’ factors – synchronisation of
business cycles, trade integration, labour mobility and wage
flexibility
The synchronisation of business cycles
Why it is considered important
A key potential cost of forming a monetary union is that members lose two important
mechanisms for macroeconomic stabilisation. First, member exchange rates are no longer
freely floating against those of other members. Second, individual states can no longer set
monetary policy independently – they must adopt the policy stance set by the union-wide
monetary authority. A key tenet of the Optimal Currency Area (OCA) literature is that the
cost of losing monetary independence is limited if the countries/regions forming a
monetary union exhibit synchronised business cycles.
Who considers it important
The view that the frequency and severity of asymmetric shocks is an important
consideration in evaluating the costs and benefits of forming a currency union is discussed
in much of the OCA literature:
In Robert Mundell’s seminal “A Theory of Optimum Currency Areas” (1961), he
suggests that “if the case for flexible exchange rates is a strong one, it is, in logic,
a case for flexible exchange rates based on regional currencies, not on national
currencies”. And, for Mundell, the “region” that defines an optimal currency area
is one in which similar industries operate so that responses to economic shocks
are uniform throughout the area.
Eichengreen (1991): “An optimum currency area (OCA) is an economic unit
composed of regions affected symmetrically by disturbances.”
Bayoumi and Eichengreen (1992): “EMU involves a sacrifice of monetary
autonomy...If disturbances are distributed symmetrically across countries,
symmetrical policy responses will suffice...Only if disturbances are distributed
asymmetrically across countries will there be occasion for an asymmetric policy
response and may the constraints of monetary union bind.”
Furceri and Karras (2008): “The higher the correlation of business cycles, the lower
the stabilization cost of giving up an independent monetary policy...[I]f a member
economy’s business cycle is very highly correlated with the union-wide cyclical
output, then monetary policy conducted by the common central bank will be a
very close substitute for the country’s own independent monetary policy.”
Since the crisis, other authors have questioned whether similarity in economic structures
and business cycles is either a necessary or sufficient condition for a functioning monetary
union:
Carney (2014) shows that the industrial structures of the Euro area core and
periphery are more similar than the regional economies of the US and Canada:
“Theory notwithstanding, being similar doesn’t necessarily help and being
different doesn’t necessarily hinder. This suggests we should look elsewhere for
the ingredients of a successful union.”
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 10
Exhibit 1: Correlations of growth between Euro members
and the Euro area as a whole over the 1999-2012 period
are higher than…
Exhibit 2: …the correlations of growth between US
states and the US as a whole over the same period
Source: Eurostat, National Statistics Offices, Goldman Sachs Global Investment Research
Source: US BEA, Goldman Sachs Global Investment Research
What we find
To explore the importance of business cycle synchronisation further, we compare a
number of different measures of cross-country symmetry for the Euro area and of cross-
state symmetry for the US.
In a result that we believe may be contrary to many people’s priors, we find that
business cycles within the Euro area have been more synchronised than those within
the US across many of the measures that we consider. This is true of the pre-crisis
period – which suggests that the asynchronous nature of Euro area cycles was not a
primary cause of the crisis. But, on most measures, it even remains true when the pre- and
post-crisis periods are included together in the comparison.
The most commonly used measure of business cycle synchronisation in the pre-crisis
literature was simply to compare the correlations of annual GDP growth of participating
economies within the monetary union. Admittedly, this is a reduced-from exercise: it does
not distinguish the underlying fundamental shocks that drive divergent behaviour across
regions from the policy and behavioural responses to those shocks (that may dampen or
amplify their impact over time). Nevertheless, this exercise can give us an initial view of
cross-country variation.
Exhibits 1 and 2 plot the correlations of annual GDP growth rates of individual Euro area
countries with growth in the Euro area as a whole and similarly for the individual states of
the US, for the period 1999-2012. Over this timeframe, growth rates have been more
correlated within EMU than within the US (with an average correlation coefficient of 0.84 vs.
0.65 in the US). US states also exhibit greater variation in growth correlations, with four
states (Alaska -0.55, North Dakota -0.20, Louisiana -0.08 and Wyoming -0.08) exhibiting a
negative correlation with the US as a whole over the period 1999-2012.
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Fin
lan
d
Fra
nce
Italy
Au
stria
Bel
giu
m
N'la
nds
L'bo
urg
Slo
veni
a
Ge
rman
y
Spa
in
Ave
rag
e
Por
tug
al
Irel
an
d
Cyp
rus
Est
oni
a
Slo
vaki
a
Mal
ta
Gre
ece
Average
Correlation coefficient of annual GDP growth with Euro area as a whole (1999-2012)
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
GA
CA NJ
FL
NV
AL
MO
MD
AR VT
CO
ME ID MI
NY HI
IA MS
UT
TX
KY
OK
DC
DE
WY
ND
Average
Correlation coefficient of annual GDP growth with US as a whole (1999-2012)
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 11
Exhibit 3: Growth correlations of US states over the
1978-2012 period were relatively low on average and very
low in some cases
Exhibit 4: Euro area dispersion has been relatively low
Negative divergence
Source: US BEA, Goldman Sachs Global Investment Research
Source: US BEA, National Statistics Offices, Goldman Sachs Global Investment Research
One can argue that 14 years of data is too short a period to judge the frequency and
severity of asymmetric shocks in the Euro area. Imbalances in the Euro area periphery
developed over a period that extended beyond the typical business cycle. In this respect,
the experience of US states may be more instructive since they have a longer history of
being members of a monetary union. Exhibit 3 plots the correlations of annual GDP growth
rates of individual states with the US as a whole (the same as Exhibits 1 and 2) over a
longer period, from 1978 to 2012. The average growth correlation over this period is very
similar to the 1999-2012 period (0.66 vs. 0.65) and the range of correlations is wide,
indicating that growth rates are reasonably dispersed. A number of states exhibited a low
correlation with the US over the period as a whole, and frequent periods of negative
correlation. Yet, despite this low degree of business cycle synchronisation, the US clearly
functioned as a currency union over this period.
The message presented by the static analysis above is reinforced by various dynamic
measures of business cycle synchronisation. Exhibit 4 displays a measure of growth
dispersion, calculated as the cross-country/state standard deviation of GDP growth rates in
the monetary union. The dispersion of growth rates rose with the onset of the sovereign
crisis but, even at its peak (in 2011), the level of growth dispersion did not appear
particularly acute relative to US norms.2
2 Previous studies have used a number of other dynamic measures of business cycle synchronisation that we do not reproduce here, largely because they provide similar results. These include measures of the absolute value of the difference in GDP growth between individual members and the monetary union as a whole (Giannone, Lenza and Reichlin (2009) and Kalemli-Ozcan and Papaioannou (2009)) and measures that capture underlying similarities in growth by comparing the residuals from regressions that strip out area-wide and time effects (Morgan, Rime and Strahan (2004)).
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
ILM
N WI
SC
CA
MO AZ
MA
MD FL
KY
ME
CT
AR
KS
MS
CO
OR TX
NE
DE
MT
OK
NM
WY
AK
Average
Correlation coefficient of annual GDP growth with US as a whole (1978-2012)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
78 80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12
EMU
US states
Cross-country standard deviation of growth rates
More dispersion
Less dispersion
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 12
Exhibit 5: The dispersion of country unemployment
rates in the Euro area has increased significantly since
the crisis
Exhibit 6: Goods trade with Euro area countries as a
share of total trade* is high * = Goods exports to and imports from other Euro area countries
as a share of total trade (1999-2012)
Source: Goldman Sachs Global Investment Research.
Source: IMF, Goldman Sachs Global Investment Research.
Measures of cumulative divergence suggest the Euro area’s post-crisis divergence
has been more extreme than in the US. One criticism of the static and dynamic
measures of business cycle synchronisation we have presented so far is that they do not
capture the cumulative divergence generated by prolonged economic booms and slumps.
One means of doing this is to compare the dispersion of output gap or labour gap-type
measures. Exhibit 5 plots the cross-country standard deviation of unemployment rates for
the Euro area and the US (i.e., it doesn’t strip out differences in structural unemployment,
which tend to be larger across Euro area countries than US states3). There has been a sharp
increase in the dispersion of Euro area unemployment rates since the onset of the
sovereign crisis. However, this appears to have been a consequence of the Euro area’s
difficulties in adjusting to the crisis, rather than a cause of the crisis.
Indeed, one of the paradoxes of the Euro area crisis was that its catalyst was a shock that
affected the monetary union as a whole (the 2007/08 financial crisis), whereas, prior to the
crisis, it was envisaged that a major challenge for the Euro area would be its ability to deal
with shocks that directly affected some parts of the monetary union but not others. In the
years leading up to the crisis, the dispersion of unemployment rates across the Euro area
had been decreasing but that dispersion rose sharply in response to a shock that initially
affected the Euro area as a whole.
We will argue that one of the most important lessons from the crisis has been that
weaknesses in the Euro area’s institutional structure meant that a common shock could
affect different parts of the Euro area in very different ways, and that the correlation of
shocks itself is a relatively unimportant factor in determining the workability of a monetary
union.
If the Euro area countries have exhibited a high degree of symmetry (but the currency
union does not currently function well), while US states often exhibit a high degree of
asymmetry (yet the US does function well as a currency union), this suggests that
symmetry among member states is not a key determinant of whether a currency area is
workable or not.
3 See, for example, Layard, Nickell and Jackman (1991).
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
78 80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12
EMU
US states
Cross-country standard deviation of unemployment rates
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
L'b
our
g
Por
tug
al
Aus
tria
Be
lgiu
m
Slo
ven
ia
Fra
nce
N'la
nds
Spa
in
Mal
ta
Italy
Slo
vaki
a
Ge
rman
y
Gre
ece
Est
oni
a
Fin
lan
d
Irel
an
d
AverageGoods trade with other Euro countries as share of total trade
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 13
A high degree of trade integration
Why it is considered important
A key potential benefit of forming a monetary union is that the adoption of a common
currency eliminates costs related to foreign exchange transactions between participating
states. The elimination of transaction costs provides a lump-sum benefit and is likely to
encourage greater micro-efficiency as economic resources are allocated within the region
more efficiently. The OCA literature thus proposes that regions that are suitable for forming
monetary unions exhibit a high degree of inter-regional trade, since the greater the degree
of trade that takes place between regions within the currency area, the greater the potential
benefit from eliminating foreign exchange costs related to these inter-regional transactions.
While the initial gain from reducing transaction costs in trade should be roughly
proportionate to the amount of existing trade that already takes place between regions
forming a monetary union, more recent OCA studies have also emphasised the dynamic
effect that reducing transactions costs can have on trade. States that share a common
currency are likely to trade more with each other because they share a common currency
relative to those that do not, creating a self-fulfilling element to the trade integration aspect
of OCA theory.
In the context of the analysis in this paper, one key issue is whether the gains in terms of
deepening trade integration via sharing a common currency can bolster the economic and
political workability of the monetary union as a whole. In turn, this raises questions about
the visibility and distribution of the economic gains from the resulting increases in trade,
thereby creating further complementarities and substitutabilities with other aspects of the
analysis, e.g., the flexibility of economies and the scope for offsetting fiscal redistribution
measures (both within and across countries).
Who considers it important
Multiple studies examining whether regions (EMU, the CFA Franc zone, or the states of the
Gulf Cooperation Council) should form a monetary union have stressed an increase in
inter-regional trade as one of the key potential benefits from forming a union:
Frankel and Rose (1997): “The benefits of being a member include a reduction in
the transaction costs associated with trading goods and services between
countries with different moneys. Countries with close trade links to EMU members
will benefit more from monetary union.”
Frankel (1999): “[one of the] big advantages of a fixed exchange rate, for any
country, is that it reduces transaction costs and exchange-rate risk, which can
discourage trade and investment.”
In a study of trade flows between Canadian provinces and US states, McCallum
(1995) provides evidence that trade between countries tends to be significantly
more limited than trade within countries.
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 14
Exhibit 7: Goods trade with Euro area countries as a
share of GDP is also high* * = 0.5 x Goods exports to and imports from other Euro area
countries as a share of GDP (1999-2012)
Exhibit 8: Intra-Euro area trade increased after 1999, but
at a slower rate than the growth of trade globally
Source: IMF, Goldman Sachs Global Investment Research.
Source: IMF, Goldman Sachs Global Investment Research.
What we find
The Euro area economies exhibit a high degree of trade integration. Exhibit 6 displays the
average of intra-Euro area merchandise exports and imports as a share of total exports and
imports over the period 1999-2012 for individual Euro area members. On average, half of all
goods trade conducted by Euro area economies has been conducted with other Euro area
economies. Over the same period, intra-EMU goods trade has represented around 45% of
GDP for Euro area countries, on average (Exhibit 7). Over time, trade within the Euro area
has also increased, including since the introduction of the Euro in 1999 (Exhibit 8).
It is difficult to say whether a high degree of trade integration is a necessary condition for a
monetary union to work, since there is no monetary union – that we know of – that does
not exhibit a high degree of trade integration.4 What we can say is that a high degree of
trade integration does not appear to be a sufficient condition for a monetary union, on the
basis that the Euro area exhibits a high degree of trade integration but has not functioned
well as a monetary union.
The evidence as to whether monetary union has boosted trade integration – viewed as a
key potential benefit of EMU prior to commencement – is inconclusive. Using gravity-type
models to examine whether EMU membership has had a significant impact on European
trade flows, studies find that EMU had a significant impact on trade in the early-to-mid
2000s but this effect appears to have died out by the late 2000s.5 While trade between Euro
area economies has risen as a share of GDP since the introduction of the Euro, it has risen
no faster than trade elsewhere in the world.
4 One would ideally be able to consider the performance of a selection of monetary unions with varying degrees of trade integration and functionality. We don’t have this luxury: while there is detailed data for trade between Euro area member states, there is no equivalent data on inter-state trade flows between US states. There is, however, data on trade between Canadian provinces, which suggests that their goods markets are highly integrated.
5 Broadbent and Bahaj (2008). One explanation for why the estimated effect of EMU on intra-Euro area trade may have declined during the mid-2000s is that the Euro real trade-weighted exchange rate appreciated significantly between 2002 and 2008, reducing the competitiveness of Euro area producers versus non-Euro area producers.
0.00
0.20
0.40
0.60
0.80
1.00
1.20
Be
lgiu
m
L'bo
urg
Slo
vaki
a
Slo
veni
a
N'la
nds
Mal
ta
Aus
tria
Est
oni
a
Po
rtug
al
Irel
and
Ge
rman
y
Fra
nce
Fin
lan
d
Spa
in
Italy
Gre
ece
Average
0.5 * (goods exports and imports to and from other Euro area countries) as share of GDP
0%
5%
10%
15%
20%
10%
15%
20%
25%
30%
91 93 95 97 99 01 03 05 07 09 11 13
World exports as% of world GDP(lhs)
Intra-EMU tradeas % of EMUGDP (rhs)
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 15
A high degree of labour mobility
Why it is considered important
In the absence of a floating exchange rate and independent monetary policy to carry out
macroeconomic stabilisation, the OCA literature suggests that factor mobility becomes an
important instrument for internal adjustment. In place of nominal adjustment through
independent monetary policy and flexible exchange rates, members of a monetary union
must rely on real adjustment to avoid overheating or stagnation. One means of adjusting in
this way is for idle factors of production from one part of the union to be redeployed to
other parts of the union, where they are needed more. If the monetary union is affected by
asymmetric shocks, high labour mobility may help correct internal imbalances by
redirecting unemployed workers from stagnating regions to growing ones within the union.
Who considers it important
A number of OCA studies have cited labour mobility as an important factor in determining
whether a currency union is optimal or not. Moreover, several studies (including Feldstein
(1998) and Bonin et al (2008)) have suggested that levels of labour mobility in the Euro area
may be too low for EMU to function as a currency area.
Mundell (1961): “The argument [for an optimum currency area] works best if each
nation (and currency) has internal factor mobility and external factor immobility.
But if labour and capital are insufficiently mobile within a country then...one could
expect varying rates of unemployment or inflation in the different regions.”
Blanchard and Katz (1992) found that inter-state migration played an important
role in correcting internal imbalances between US states: “A (US) state typically
returns to normal after an adverse shock not because employment picks up, but
because workers leave the state.”
Feldstein (1998): “a decline in demand need not raise unemployment if workers
are geographically mobile and move to places where jobs are available. But
although the legal barriers to labour mobility within the European Union have
been eliminated, language and custom impede both temporary and long-term
movement within Europe...the American heritage of immigration and national
settlement makes Americans much more willing to move internally than their
European counterparts.”
Bonin et al (2008): “Geographic mobility serves as an equilibrating factor between
regional labour markets. To the extent that mobility of capital and goods does not
achieve convergence of employment and real wages in open or integrated
economies, mobility of labour may help balancing labour market outcomes.”
Set against this view, other authors dispute the extent to which labour mobility is a
substitute for exchange rate flexibility:
Bean (1992) expresses scepticism that factors of production can migrate between
regions over a shorter timescale than wages and prices adjust: “The Mundellian
argument (that factor mobility is a key condition for optimality) rests on the
premise that factors move faster than prices, something which I find
implausible...Certainly, some regions will prosper and others will decline, but this
will happen independently of whether there is a single currency or not.”
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 16
Exhibit 9: After the oil shocks, payroll levels diverged
significantly, suggesting a highly asymmetric response
Exhibit 10: The difference between unemployment in
Michigan and Texas was high and persistent after the oil
shocks
Source: US BLS
Source: US BLS
What we find
There are marked contrasts in the performance of employment and unemployment across
US states over time. In Exhibits 9 and 10, we compare the performance of employment and
unemployment in Michigan (a major car-producing state) and Texas (a major oil-producing
state) in the aftermath the oil shocks of the 1970s and the period following the Plaza Accord
in 1985 (when oil prices and the Dollar both fell sharply, for reasons that were not driven by
initial cross-state differences in the US).6
Employment fell sharply in Michigan and rose sharply in Texas following the 1970s oil shocks,
and never returned to its pre-crisis levels (Exhibit 9). The respective trends in employment in
these states that began in the 1970s were only broken by the Plaza Accord in the mid-1980s.
In other words, it took a major shock in the opposite direction to stabilise the labour market
divergence. Meanwhile, unemployment in Michigan and Texas also diverged significantly
over this period but ultimately re-converged in the aftermath of the Plaza Accord (Exhibit 10).
To consider the persistence of employment and unemployment shocks in the US and Euro
area more formally, we employ a similar methodology to Blanchard and Katz (1992). Our
results are based on annual data for all 50 US states and the original Euro area economies
and Greece (12 countries in total), using panel data estimations allowing for fixed state
effects.7
6 The US was hit by two oil shocks in the 1970s. In October 1973, members of the Organization of Arab Petroleum Exporting Countries (OAPEC) enforced an embargo, resulting in a fourfold increase in the price of oil by early 1974. The second oil crisis started from November 1978, when political tensions in Iran reduced supply and raised oil prices again. The rise in oil prices as a result of these crises was beneficial to oil-producing areas of the US, such as Texas, but naturally hurt areas associated with automotive and other manufacturing production, such as Michigan. The Plaza Accord (September 1985) was an agreement between the US and other advanced economies to deliberately devalue the US Dollar. From 1985 to 1987, the US Dollar depreciated 51% against the Yen. At the same time, oil prices also fell sharply from late 1985 onwards. This shock had the opposite effect to that of the oil shocks of the 1970s: oil-producing Texas suffered relative to manufacturing states in the Midwest, which benefited more directly from the devaluation of the US Dollar.
7 We use annual data from 1976-2012 for the US and 1993-2012 for the Euro area. However, Euro area migration data are only available for the period 1999-2010.
80
85
90
95
100
105
110
115
120
125
130
73 74 75 76 77 78 79 80 81 82 83 84 85 86 87
Non-farm payrolls (Sep 1973 = 100)
MichiganNew YorkOhioPennsylvaniaTexas
Plaza accord
2nd oil shock
-2
0
2
4
6
8
10
78 79 80 81 82 83 84 85 86 87 88
pp
Difference betweenMichigan & Texasunemployment
2nd oil shock
Plaza accord
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 17
Exhibit 11: State-specific employment shocks in the US
are generally not reversed Impulse response of 1pp state-specific employment shock
Exhibit 12: State-specific unemployment shocks in the
US are persistent but are ultimately reversed Impulse response of 1pp state-specific unemployment shock
Source: BEA, Blanchard & Katz (1992), Goldman Sachs Global Investment Research
Source: BEA, Blanchard & Katz (1992), Goldman Sachs Global Investment Research]
The variables we consider are: (i) the difference between state (country) and US (Euro area)
employment growth; (ii) the difference between state (country) and US (Euro area)
unemployment; (iii) the difference between state (country) and US (Euro area) wage
growth; (iv) the level of state (country) wages versus US (Euro area) wages; and (v) the
difference between state (country) net inward migration and US (Euro area) net inward
migration (as a share of the population).
In common with Blanchard and Katz, we find that state-specific employment shocks in the
US tend not to be reversed over time. Indeed, if employment rises/falls in one state relative
to the rest of the US, it continues to rise/fall over the following 5-6 years, significantly
extending the initial ‘shock’ (Exhibit 11). By contrast, state-specific unemployment shocks
do tend to be reversed over time but they are nevertheless quite persistent, with a ‘half-life’
of around five years (Exhibit 12). The principal reason why employment shocks in the US
are persistent but unemployment shocks are not is that, over time, workers leave the state
that has been negatively affected by the shock.8 While the US clearly functions better than
the Euro area as a monetary union, the process of regional adjustment in the US is lengthy
and involves the de-population of stagnant regions over time.
One of the complications of applying the same analysis to Euro area countries is that the
Euro area has been in existence for a relatively short period of time and the estimated
response of country-specific unemployment is sensitive to the time period chosen (Exhibit
13).9 Nevertheless, the adjustment of country-specific unemployment in the Euro area is
clearly slower than the adjustment of state-specific unemployment in the US.
Comparing Euro area countries with US states over the period 1993-2008, we find that full
adjustment takes around 9-10 years in total (vs. 6-7 years for the US) and that the ‘half-life’ of
country-specific unemployment shocks is around 6-7 years (vs. 5 years for the US; Exhibit 14).
If we include the post-crisis labour market performance in our analysis, country-specific
unemployment shocks in the Euro area appear to have even more persistent effects.
8 In principle, the difference between the performance of employment and unemployment could also be due to variations in state labour force participation but, in practice, this has not been quantitatively important.
9 To focus on the adjustment process with a monetary union, one would ideally want to restrict the analysis to the period since EMU began. However, the sample would then be relatively short and the results highly sensitive to whether the post-crisis period is included or not. Our justification for including the period from 1993 to 1999 in our main analysis – even though EMU didn’t begin until 1999 – is that the Euro area countries operated a fixed exchange rate regime during this period with no major realignments.
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
years
GS (76-12)
Blanchard & Katz (52-90)
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1 2 3 4 5 6 7 8 9 10
years
GS (76-12)
Blanchard & Katz (52-90)
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 18
Exhibit 13: Response of country-specific unemployment
in Euro area sensitive to period chosen Impulse response to 1pp country-specific unemployment
shock
Exhibit 14: Country-specific unemployment shocks more
persistent in Euro area than US Impulse response to 1pp state/country-specific
unemployment shock (1993-08)
Source: Eurostat, OECD, Goldman Sachs Global Investment Research
Source: BEA, Eurostat, OECD, Goldman Sachs Global Investment Research
Do differences in labour migration flows play an important role in accounting for the US’s
faster adjustment to unemployment shocks? It is certainly the case that inter-state
migration represents an important part of the US’s regional adjustment mechanism in the
long run. In the past 40 years, there has been a steady flow of people migrating from the
‘rust belt’ states of the north to the ‘sun belt’ states of the south, resulting in cumulatively
large differences in the relative size of state population and employment. Michigan’s share
of the total US population has declined from 4.4% to 3.1% since 1970, while the population
of Detroit – the largest city in the state of Michigan – has declined from a peak of 1.8 million
people in the 1950s to just 0.7 million people today.
However, while net migration flows in the US are cumulatively important over time, we
find that the response of net migration to differences in unemployment is quantitatively
small in the short run (Exhibit 15). Moreover, while labour mobility in the US is higher in
general, we do not find migration in response to unemployment differences to be
significantly higher in the US than in the Euro area (Exhibit 16). This finding needs to be
treated with some caution, not least because the Euro area results are based on a more
limited data set than for the US.10 But we nevertheless find it difficult to attribute a major
role in the higher persistence of country-specific unemployment in the Euro area to lower
migration. In addition, intra-Euro area mobility – and also extra-Euro area mobility –
appears to be rising over time (although whether this is a structural change or a response
to current incentives remains unclear).
In summary, while inter-state labour mobility appears to be an important adjustment
mechanism of internal imbalances between US states, these adjustments take time and are
far from costless. Moreover, the response of net migration to unemployment differences
does not appear significantly different in the Euro area from that in the US, so it is difficult
to argue that differences in labour mobility represent the key distinction between the US’s
functioning monetary union and the Euro area’s malfunctioning union.
10 One reason why this result is surprising it that it suggests that the various language and other differences that exist between Euro area countries has not impeded net migration relative to US states. While we are comfortable that our estimates are accurate given the data available, migration data for Euro area countries are only available for the period 1999-2010 and not for every country. It is also difficult to rule out the possibility that US state data fail to capture inter-state migration to the same degree that European country data capture international migration.
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
1 2 3 4 5 6 7 8 9 10
years
Unem vs. EA (93-12)
Unem vs. EA (93-08)
Unem vs. EA (99-08)
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1 2 3 4 5 6 7 8 9 10
years
Unem vs. EA (93-08)
Unem vs. US (93-08)
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 19
Exhibit 15: Migration response is quantitatively small but
cumulatively important over time Impulse response to 1pp state-specific unemployment shock
(2001-12)
Exhibit 16: Greater persistence of Euro area
unemployment shocks not due to migration Impulse response to 1pp country-specific unemployment
shock
Source: BEA, US Census Bureau, Goldman Sachs Global Investment Research
Source: Eurostat, OECD, Goldman Sachs Global Investment Research
A high degree of wage flexibility
Why it is considered important
A high degree of nominal wage flexibility can offset the need for nominal exchange rate
flexibility – if a region experiences a negative demand shock that leads to higher
unemployment, then a timely and extensive downward adjustment of real wages would
reduce the cost to employers of hiring and retaining staff, and thus would place downward
pressure on unemployment. Indeed, in the limit, if wages and prices can adjust
instantaneously to shocks, wage and price flexibility are perfect substitutes for exchange
rate flexibility, and there are no costs associated with the loss of monetary independence.
Who considers it important
The original OCA literature of the 1960s, in keeping with the dominant Keynesian
orthodoxy of the time, assumed that prices and nominal wages were largely sticky and
persistent even following shocks, and so did not focus on wage flexibility as a relevant
criterion for optimal currency areas. More recently, however, the academic literature has
focused more on the role of real wage flexibility in determining the performance of
employment subsequent to slumps.
Berthold (1998): “Real wage flexibility has been an issue in Europe at least since
the late 1980s when persistent unemployment became the number one topic of
economic policy in Europe.”
Krugman (2012): if “workers [from a stagnating state] can’t or won’t leave the state,
the only way to restore full employment is to regain the lost jobs, which will
probably require a large fall in relative wages to make the state more competitive,
a fall in wages that is much more easily accomplished if you have your own
currency to devalue”...“the Euro experience strongly suggests that downward
nominal wage rigidity is a big issue. This means that ‘internal devaluation’ via
deflation is extremely difficult, and likely to fail politically if not economically”.
Schmitt-Grohe and Uribe (2012): “The observed failure of nominal wages to adjust
downward after 2008 despite sizable increases in unemployment motivates a
model in which the combination of downward nominal wage rigidity and a fixed
exchange rate lie at the heart of the current unemployment crisis.”
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
1 2 3 4 5 6 7 8 9 10
years
Unemployment (vs. US)
Total net migration (vs. US)
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
1 2 3 4 5 6 7 8 9 10
years
Unemployment (vs. EA) (99-10)
Total net migration (vs. EA) (99-10)
Equiv. US response
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 20
What we find
Using the same panel data approach applied in the previous section on labour mobility, we
find significant differences between the degrees of real wage flexibility in US states and
Euro area countries. State-specific wage levels respond more rapidly to state-specific
unemployment shocks in the US than is the case for Euro area economies (Exhibit 17).11
However, while wages respond more flexibly in US states in the short run, wage shocks are
also much less persistent in the US than country-specific wage shocks are in the Euro area.
This is not surprising. If there is little labour market flexibility – in quantities or prices – it is
not the case that the relative adjustment that is required will simply be avoided. Rather, the
adjustment is likely to take longer and be more painful (in the sense that unemployment
will have to rise by more to achieve the same decline in real wages) than would have been
the case under the flexible labour market scenario.
Painful as the price of adjustment may be in the Euro area, it is a price that peripheral
countries have so far been prepared to pay. Unemployment has risen sharply in the
periphery, but clear progress is being made in regaining competitiveness. Exhibit 18
displays whole economy ULCs relative to the Euro area for Germany and the peripheral
states from the start of the crisis (2008) onwards. With the exception of Italy, all of the
peripheral states have made significant progress in regaining competitiveness.
Lastly, it is worth emphasising that, of the various ‘economic’ factors highlighted by
Optimal Currency Area theory, only real wage flexibility can credibly claim to play an
important role in the US’s greater workability as a monetary union. Limited wage flexibility
appears to be the ‘Achilles Heel’ of the Euro area adjustment process.
Exhibit 17: Wage response in Euro area countries takes
longer to come through Impulse response to 1pp country/state-specific
unemployment shock
Exhibit 18: Most peripheral economies have gained
competitiveness since 2008 Real unit labour costs relative to the Euro area (2008=100)
Source: BEA, OECD, Goldman Sachs Global Investment Research
Source: EU Commission
11 Other authors have also found that real wages in Europe respond less to unemployment than in the US. See Grubb, Jackman and Layard (1983), and Bruno and Sachs (1985).
-3.0
-2.0
-1.0
0.0
1.0
2.0
3.0
4.0
1 2 3 4 5 6 7 8 9 10
years
Unemployment (vs. EA)
Country wage response
Equiv. US response
86
88
90
92
94
96
98
100
102
104
08 09 10 11 12 13
Germany
Greece
Ireland
Italy
Portugal
Spain
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 21
Chapter 3: The role of ‘institutional’ factors – fiscal transfers,
financial integration and political union
Mechanisms for fiscal transfers
Why it is considered important
As we discussed in the previous chapter, OCA theory posits that the business cycles of the
economies within a monetary union should be reasonably correlated, so that one monetary
policy can fit all, and so that high levels of factor mobility and price flexibility can reduce
the adjustment cost to asymmetric shocks once they occur. A third means of improving the
functionality of a monetary union is for participants to insure against the risk of an
asymmetric shock, either through increased fiscal integration or through increased
integration of capital markets and banking systems.
The important role that a more unified fiscal system can play in reducing the costs
associated with the loss of monetary independence, by levying taxes on cyclically-strong
nations/regions and transferring those resources to cyclically-weak nations/regions, is one
that has been emphasised by a number of OCA studies, both before and since the crisis. It
is also commonly noted that the US – through its federal fiscal system – and other well-
functioning monetary unions all possess such a mechanism, but the Euro area does not.
One can argue further that the example of German unification – which brought together
two very different economies – illustrates that there is always a level of transfers that can
make every monetary union work.
Who considers it important
A number of studies focus on fiscal federalism as a relevant criterion for optimal currency
areas, starting with Kenen (1969). Some notable examples include:
Sachs and Sala-i-Martin (1992), in a study of the US, estimate that federal fiscal
taxes and transfers offset between 30% and 40% of a fall in state income: “One of
the reasons why the US exchange rate system has held up reasonably well is the
existence of a ‘Federal Fiscal Authority’ which insures states against regional
shocks.”
Mongelli (2002): “Countries sharing a supra-national fiscal transfer system that
would allow them to redistribute funds to a member country affected by an
adverse asymmetric shock would also be facilitated in the adjustment to such
shocks and might require less nominal exchange rate adjustments.”
Hirshow (2007): “...Europe will not turn into an OCA soon but it could finally
achieve similar quality. What must come to enable the economic and monetary
union to better absorb asymmetric shocks?...A highly efficient fiscal response.”
Set against this view, other authors downplay the role that fiscal federalism plays in
ensuring that the US functions well as a monetary union:
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 22
Exhibit 19: The US federal budget is much larger than its
EU equivalent Federal budget revenues as % of GDP
Exhibit 20: US federal fiscal policy offsets 25% of income
shocks, but only half is due to inter-regional insurance Estimated response of federal fiscal policy to GDP shock
Source: US Office of Management and Budget, Eurostat
Source: Goldman Sachs Global Investment Research
Fatas (1998) argues that Sachs and Sala-i-Martin (1992) overestimate the amount
of interstate insurance provided by the US federal system by a factor of three
because they fail to exclude the effects of inter-temporal income smoothing (that is
provided by the US federal system but which could just as easily be provided by
national tax systems). In other words, there is a need to distinguish between (i)
cross-country risk-sharing within the monetary union; and (ii) inter-temporal risk-
sharing with individual countries (fiscal jurisdictions) within the monetary union. In
the face of temporary asymmetric shocks, either mechanism could, at least in
principle, be used to smooth the profile of economic activity. He concludes that
“the benefits of a European fiscal federation would be modest”.
What we find
One basic means of gauging the relative scope for area-wide fiscal stabilisation in the US
versus the Euro area is to compare the respective size of their area-wide budgets. US
federal budget expenditure has averaged around 20% of US GDP in the past 20 years
(1993-2013), but the budget of the European Union budget has represented only 1% of EU
GDP (there is no budget that is specifically designated for the Euro area) (Exhibit 19).
In addition to being much larger than the EU budget, the means by which US federal taxes
are gathered and expenditure is allocated provides a significant degree of automatic
stabilisation of state-specific shocks (because tax revenues fall and budgetary expenditure
rises automatically in response to negative region-specific shocks). This is less true of the
EU budget, where country contributions are voted on over multi-year periods and where
much of the expenditure is in areas – such as the Common Agricultural Policy – that are not
designed to be counter-cyclical.
0
5
10
15
20
25
90 92 94 96 98 00 02 04 06 08 10 12 14
US
EU
0.0
0.2
0.4
0.6
0.8
1.0
1.2
GDP shock Federal fiscal offset Net federal fiscal vs.US
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 23
More formally, to estimate the role that federal fiscal transfers and taxes play in offsetting
state-specific shocks in the US, we use the same panel-data approach that we applied in
the previous sections. To construct a measure of net federal transfers into and taxes from a
state, we use: (a) data on federal taxes, broken down by state of collection, published by
the Internal Revenue Service; (b) data on grants from the federal government to state
governments, from the US Census Bureau; and (c) data on federal transfers to individuals
(e.g., various benefit payments such as Social Security), broken down by state, from the
Survey of Current Business (published by the US Bureau of Economic Analysis). Our
measure of net federal transfers/taxes is calculated as the difference between the transfers
into a state and taxes out of the state (i.e., (b) + (c) – (a)). Our estimates are based on annual
data spanning the period 1990-2010.
We first consider the importance of federal fiscal policy in offsetting the impact of gross
state product (GDP) shocks to US states. We find that US federal fiscal policy offsets
around 25-30% of the initial effect on income from state-specific economic shocks, with
around two-thirds of this effect occurring in the same year as the shock and the remainder
of the effect occurring in the subsequent year. This estimated response is a little smaller
than the results obtained by Sachs and Sala-i-Martin (1992) – who estimated that federal
fiscal taxes and transfers offset between 30% and 40% of a fall in state income – but not
dramatically so.
However, as Fatas (1992) has argued, part of the offset provided by US federal fiscal policy
in response to economic shocks affecting states reflects the inter-temporal smoothing
provided by fiscal policy in response to common shocks affecting the US as a whole.
Distinguishing between the smoothing provided by inter-temporal transfers and inter-
regional insurance is important because inter-temporal transfers can be provided by
countercyclical budgets at a national level (i.e., they don’t require a common federal
budget).
To estimate the genuine inter-regional insurance provided by the US’s federal system, we
consider the difference between the federal fiscal response to state-specific shocks at a
state level and at a national level. On this basis, we find that the amount of inter-state
insurance provided by the US federal fiscal system drops by a factor of two, to around 14%.
This is a slightly higher estimate than that obtained by Fatas (1992). There is little or no
equivalent smoothing provided by EU fiscal policy.
It is difficult to gauge how critical the degree of inter-state insurance provided by the
federal fiscal system is in ensuring the US exchange rate system’s long-term survival. Our
own assessment is that the size of the transfers is material without being critical and could
be substituted by effective counter-cyclical policy at a national level and/or the increased
integration of capital markets and banking systems.12
Financial integration and banking union
Why it is considered important
Integrated financial and banking markets can provide an alternative means of risk-sharing
to fiscal federalism in a monetary union, operating via two principal channels: via the
capital markets channel (through increased portfolio diversification) and via the credit
channel (through increased cross-country borrowing and lending).
12 Dirk Schumacher has argued that there is sufficient flexibility within national fiscal policy to adjust to ‘normal’ shocks, which he defines as 2% drop in national GDP (European Economics Analyst, 2013/11).
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 24
To the extent that private financial markets facilitate the sharing of region-specific shocks,
this can – all else equal – offset the need for inter-country/state fiscal risk-sharing. That said,
one lesson from the Euro area crisis is that, without the appropriate regulatory structures,
greater financial market integration can also result in risks being pooled rather than shared
– bank failures and recapitalisation requirements have severely worsened government debt
dynamics in a number of peripheral Euro area economies (most notably, Ireland and Spain).
By comparison, the losses suffered at a regional level during the US Savings and Loan
crisis of the late 1980s were largely borne by the US federal government.
Who considers it important
Asdrubali, Sorensen and Yosha (1997) emphasise the importance of integrated
capital markets as means of risk-sharing within the US. They estimate that 39% of
the impact of an asymmetric shock to gross state product on state consumption is
smoothed by capital markets (risky investments), 23% by credit markets
(borrowing and lending), 13% by the federal government and the remaining 25% is
not smoothed.
Daly (2001) argues that the introduction of the Euro, in reducing home bias in
portfolio allocation through the elimination of exchange rate risk, combined with
the secular increase of privately-held pensions, could increase the importance of
risk-sharing via portfolio diversification in Europe.
Furceri and Zdzienicka (2013), who apply Asdrubali et al’s (1997) framework to
European economies, find that there is negligible inter-country sharing of risk in
Europe via the capital markets and inter-country fiscal risk-sharing channels.
What we find
Asdrubali, Sorensen and Yosha (1997) identify three broad channels through which the
impact of an asymmetric shock to gross state product on state consumption can be
smoothed in the presence of a unified fiscal regime and integrated capital markets: first,
risk can be shared via the cross-ownership of assets (the capital markets channel); second,
the federal tax-transfer system can smooth income (the fiscal channel); and, third,
consumption can be smoothed through borrowing and lending from an integrated banking
system (the credit markets channel). There will also be a residual degree of idiosyncratic
volatility in state consumer spending that remains uninsured.
The authors argue that it is possible to estimate the relative importance of each of these
channels by comparing the performance of gross state product, state income, disposable
state income and state consumption: if state income is more highly correlated with national
output than gross state product, it implies that some of the idiosyncratic risk implied by
state output has been diversified away through the cross-ownership of assets (i.e., the
capital markets channel); if state disposable income is more highly correlated with national
output than state income, it implies that some of the idiosyncratic risk implied by state
income has been diversified away through the fiscal channel; and so forth. With full risk-
sharing, state consumption should be a fixed proportion of US output.
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 25
Although conceptually attractive, one problem with the approach taken by Asdrubali et al is
that it is unclear whether the state-level data are measured with sufficient accuracy to
isolate the three channels in the manner that the authors claim. While gross state product
and net fiscal transfers can both be measured with a significant degree of accuracy, the
split between company profits that are domestic to the state and those that come from
other states is not well measured, and there is no measure of consumption at a state level
(Asdrubali et al (1997) build a proxy of state consumption based on retail sales data). This
means that, while it is possible to identify the importance of the fiscal channel as a means
of risk-sharing among US states – as we did in the previous section13 – it is not possible to
separate out the effects of the capital markets and credit channels.14
This is not to suggest that the capital and credit channels are not important – our own view
is that the US’s integrated capital and credit markets play an important role in smoothing
the impact of region-specific shocks in that economy. Rather, we question whether it is
possible to quantify the relative importance of each of these channels for US states with
the degree of precision suggested by Asdrubali et al (1997).
The issue of measurement is less of a problem for European countries, as there is good
data at a national level for total consumption (household and government) and for factor
income flows from abroad. Using national income data, we apply the same approach as
Asdrubali et al (1997) to the original Euro area economies (plus Greece) in a panel data set-
up. Our results are displayed in Exhibits 21 and 22, and we compare our estimates with
those of Furceri and Zdzienicka (2013) for Europe and with those of Asdrubali et al (1997)
for the US in Exhibit 23.
Exhibit 21: There has been little/no cross-country fiscal or
capital market risk-sharing among Euro area countries...
Estimated channels of income smoothing
Exhibit 22: ...and the degree of risk-sharing has fallen
since EMU began
Estimated channels of income smoothing
Source: Goldman Sachs Global Investment Research.
Source: Goldman Sachs Global Investment Research.
13 It is notable in this respect that Asdrubali et al’s estimate of the degree of fiscal smoothing provided by the US federal system is very similar to our own.
14 Another problem with the approach adopted by Asdrubali et al is that it would have failed to capture the pooling of capital market and banking risks that took place in the Euro area prior to the crisis.
-0.20
0.00
0.20
0.40
0.60
0.80
1.00
1.20
Capitalmarkets
Net transfersPublic saving Privatesaving
Unsmoothed
1971-2012
-0.20
0.00
0.20
0.40
0.60
0.80
1.00
1.20
Capitalmarkets
Net transfersPublic saving Privatesaving
Unsmoothed
1999-2012
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 26
Exhibit 23: Country-specific shocks go mostly unsmoothed in the Euro area, in contrast to
state-specific shocks in the US Estimated channels of income smoothing across countries/states (panel regression results)
Source: Goldman Sachs Global Investment Research, Furceri and Zdzienicka (2013) and Asdrubali et al (1997).
Political integration
Why it is considered important
The importance of greater institutional integration is reinforced by the self-fulfilling aspect
of the perception of irrevocability, both among investors and electorates. The perception
that the ‘US Dollar zone’ is irrevocable is partly self-fulfilling: it functions as a monetary
union – despite the existence of frequent and substantial state-specific shocks – precisely
because few question its existence.
When a US state is hit by a negative shock, it does not face the additional burden of a
sharp rise in funding costs. By contrast, doubts over the long-run sustainability of the Euro
area have – at least until Mario Draghi’s “whatever it takes” commitment in July 2012 –
resulted in higher borrowing costs in peripheral Euro area countries, which, in turn, have
exacerbated the strains that pose an existential threat to the Euro area.
From the perspective of electorates (as opposed to investors), this distinction is also
important. When US states are affected by asymmetric shocks, the winners and losers do
not live and vote in separate political jurisdictions. This contributes to the acceptance of the
costs of high regional unemployment and the notion of irrevocability.
Who considers it important
The role that the perception of irrevocability plays in making a monetary union work is not
one that was widely discussed in the (pre-crisis) OCA literature. However, it has been
viewed as increasingly important since the crisis:
De Grauwe (2011) argues that the weakness of the Euro area’s governing structure
gives rise to self-fulfilling crisis equilibria. He contrasts the performance of Spanish
and UK bond yields arguing that, despite a better deficit and debt performance,
Spain’s borrowing costs rose much more than the UK’s because of the perceived
fragility of the Euro area’s institutional structure and the existence of convertibility
risk.
De Grauwe and Ji (2012) provide evidence that a large part of the increase in
sovereign bond spreads in the Euro area periphery in 2011 and 2012 were
unrelated to conventional bond market fundamentals (such as government deficit
and debt levels).
1 2 3 4
US
1971-2012 1999-2012 F&Z (79-10) ASY (63-90)
Capital markets -0.06 -0.01 -0.01 0.39
0.01 0.01 0.04 0.13
Saving/credit markets 0.44 0.23 0.31 0.23
of which Public 0.08 -0.01 0.09
Private 0.37 0.23 0.22
Unsmoothed 0.61 0.78 0.66 0.25
Inter-country/ state fiscal transfers
Euro area
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 27
Carney (2014) argues that, more than other factors, “a durable, successful currency
union requires some ceding of national sovereignty.”
What we find
The role that the perception of irrevocability plays in making a monetary union work is
largely intangible and, for this reason, has to be indirectly inferred rather than directly
observed. However, one indication of its importance is the fact that the Euro area
descended into a sovereign debt crisis despite having government deficits and debts that,
in aggregate terms, were significantly less than in the US and most other advanced
economies. If the increase in Euro area periphery bond yields cannot be explained with
reference to commonly used bond market ‘fundamentals’, the most obvious explanation
for the rise is that they were singled out in this way because they were perceived as being
subject to ‘convertibility risk’ (i.e., the risk that one or more countries might leave the
monetary union). This is because, in contrast to the US, the Euro area remains a collection
of nation states whose commitment to the irrevocability of EMU remains in doubt.
One indication of the importance of convertibility risk in driving the high level of Euro area
periphery spreads in 2011H1/2012H2 is simply to observe the extent to which those
spreads narrowed once Mario Draghi made his “whatever it takes” commitment in July
2012 (because Mr. Draghi’s statement altered perceptions of convertibility risk but left bond
market fundamentals otherwise unchanged). Following DeGrauwe and Ji (2012), Exhibit 24
plots the change in Euro area spreads between June and December 2012 against the initial
level of those spreads. There is a strong link between the two, with an R-squared of 0.96
(falling to 0.93 if Greece is excluded).
By contrast, it is difficult to find a strong relation between the change in spreads and
conventional bond market fundamentals. Exhibit 25 plots the change in spreads against the
change in debt levels between 2011 and 2013. We have also compared the change in
spreads with the change in deficit levels, with similarly poor results.
That a commitment to the irrevocability of the Euro area from the ECB President could
have such a material effect on Euro area bond spreads – in the absence of any change in
bond market fundamentals or, indeed, in the absence of any actual policy action from the
ECB – underlines how important the perception of irrevocability is in ensuring a monetary
union’s long-term survival.
Exhibit 24: “Whatever it takes” triggered a large
compression in spreads…
Change in Euro area sovereign spreads vs. initial spreads
Exhibit 25: ...that was unrelated to fundamentals Change in Euro area sov. spreads vs. change in gross govt.
debt 2011-2013
Source: Goldman Sachs Global Investment Research.
Source: Goldman Sachs Global Investment Research.
R² = 0.9636
-16
-14
-12
-10
-8
-6
-4
-2
0
2
0 5 10 15 20 25
Change in spreads pp (6/12-12/12)
Initial level of spreads
R² = 0.0039
-16
-14
-12
-10
-8
-6
-4
-2
0
2
0 5 10 15 20 25
Change in gross dept pp
Change in spreads pp (6/12-12/12)
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 28
Chapter 4: Conclusions: Institutional reform plans are not yet
sufficiently robust
We have provided a factor-by-factor analysis of what makes a monetary union work,
discussing each of seven factors proposed by Optimal Currency Area (OCA) theory. The
most significant finding from our analysis, in our view, is that it is more important to
establish the right institutions and mechanisms to deal with problems when they occur and
to help ensure that the monetary union is credible than to try to ensure that each of the
‘economic’ conditions for a monetary union is met (and, in particular, we find that business
cycle synchronisation is less important than the original OCA literature proposed).
In addition, we have also argued that: (i) the adjustment to region-specific shocks is painful
even in successful monetary unions, such as the US; (ii) Europe’s lack of wage flexibility
represents its key ‘economic’ – as opposed to ‘institutional’ – weakness; (iii) the inter-
regional insurance provided by US federal fiscal policy directly offsets around 14% of the
initial effect on income from state-specific economic shocks; (iv) among Euro area
economies, there has been negligible cross-country fiscal or capital market risk-sharing and
the degree of country or state-specific risk that remains unsmoothed is much higher than in
the US; and (v) the perception of irrevocability – among electorates and investors – plays a
crucial role in ensuring that monetary unions work.
One difficulty in trying to evaluate the precise importance of each of the factors discussed
in this paper is that, because they operate as complements and substitutes for each other,
there is no single combination of factors that ensures the workability of a monetary union.
A separate difficulty is that each of the factors discussed operates across different
dimensions and across different timeframes (with fiscal and capital market risk-sharing
helping to smooth the effects of region-specific shocks in the short run, while wage
flexibility and labour migration provide more permanent adjustments).
Nevertheless, the broad conclusion that ‘institutional’ factors (fiscal federalism, financial
integration and political union) and wage flexibility are more important than other
‘economic’ factors (cyclical symmetry, the degree of trade integration and labour mobility)
is – tentatively – positive for the Euro area. This is because European policymakers have it
within their power to adjust institutional factors and to implement reforms that would
increase wage flexibility, whereas there is little they can do to adjust deeper economic
relationships (such as business cycle synchronisation, trade integration and cross-country
labour mobility). This is not to suggest that adjusting Euro area institutions or
implementing labour market reform to make EMU work will be easy – as it clearly isn’t –
but at least it is possible.
Labour market reform
Of the various ‘economic’ factors highlighted by Optimal Currency Area theory, only real
wage flexibility can credibly claim to play an important role in the US’s greater workability
as a monetary union.
Regaining external competitiveness through wage adjustment is typically costly because
the Phillips curve – which sets out the relationship between the size of the output gap (or
the level of unemployment) and inflation – tends to flatten as unemployment rises. While
wage growth and inflation tend to fall when an economy is cyclically weak (i.e., there is a
negative output gap), inflation and wage growth typically struggle to fall below zero
regardless of how weak the economy becomes.
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 29
Exhibit 26: Cyclical adjustment of prices alone is difficult
because Phillips curve is flat Stylised Phillips curve linking inflation to output gap
Exhibit 27: A less flat Phillips curve means less 'pain' is
required to regain competitiveness Stylised Phillips curve linking inflation to output gap
Source: Goldman Sachs Global Investment Research
Source: Goldman Sachs Global Investment Research
Starting from Point A in Exhibit 26, the decline in inflation implied by the transition to Point
B starts the process of regaining competitiveness. But, as the output gap turns increasingly
negative (Point C), the additional benefit in terms of lower inflation (and, therefore,
improved competitiveness) is small.
This downward price rigidity at large negative output gaps in part reflects the importance
of the 0% nominal threshold in the setting of prices and the fact that nominal wages, in
particular, are rigid downwards (i.e., workers are less prepared to accept a 1% decline in
nominal wages when inflation is zero than a 1% rise in wages when inflation is 2%). In
economies where there is a high degree of flexibility in the setting of wages and other
prices, the importance of the 0% threshold typically appears less pronounced.
Two factors can help to reduce the cost of labour market adjustment in the periphery:
Structural reforms that adjust the slope of the Phillips curve: Exhibit 27 shows
how the situation would differ if the Phillips curve did not flatten out as the output
gap turns negative. If wage growth is responsive to labour market weakness, then
unemployment will need to rise by less to bring about the required
competitiveness adjustment. (In Exhibit 27, as the economy weakens, it transitions
to Points B’ and C’, where inflation is lower than at B and C and, thus, the ‘pain’
required to regain competitiveness is less.) Labour market reform can improve the
unemployment-inflation trade-off in this way, if it increases flexibility. Spain has
introduced reforms that have contributed to a steepening of its Phillips curve but
progress elsewhere has been mixed (and, given that labour market legislation is
implemented at a national level, this is likely to remain the case).
Raising the average rate of inflation at which the adjustment takes place: The
adjustment of real wages is particularly difficult when average inflation rates are
low (because workers are typically more willing to accept nominal pay freezes than
nominal pay cuts). Germany will have to accept a period of above-average
inflation if the periphery is to complete the task of regaining lost competitiveness.
Both factors played a role in Germany’s adjustment from Europe’s ‘sick man’ to
‘powerhouse’ during the 2000s. Its transition was partly due to the implementation of the
important Hartz labour market reforms, but it was also facilitated by the relatively high
inflation rates that existed in the rest of the Euro area.
-2.0
-1.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
-8 -6 -4 -2 0 2 4
Inflation
Output Gap
ABC
-2.0
-1.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
-8 -6 -4 -2 0 2 4
Inflation
Output Gap
AB
CB'
C'
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 30
Institutional reform
In determining which institutional structures are necessary and which a monetary union
can do without, we need to recognise that there is more than one way to make a monetary
union workable. Progress towards deeper integration in one dimension can be a substitute
for making progress along another. For instance, the size of cross-country fiscal transfers in
the Euro area may never match inter-state transfers in the US, but a greater degree of risk-
sharing through a more integrated financial system could be sufficient to make EMU work.
The negotiations taking place to develop the Euro area’s steady-state framework are
developing along the two institutional dimensions discussed here: fiscal/political and
financial/banking. As the Euro area’s principal creditor country, Germany has fought to
ensure that fiscal and financial risks will not be shared across either dimension without
countries first displaying a willingness to consolidate and reform. It has also been one of
the strongest advocates for greater fiscal/political integration, while displaying markedly
less enthusiasm for financial/banking integration. France, by contrast, displays greater
enthusiasm for financial/banking integration, but is reluctant to relinquish fiscal/political
sovereignty.
Because fiscal and financial integration can act as substitutes for each other, neither
approach is necessarily ‘correct’. But there is a clear risk that the lack of agreement on
which approach to prioritise could result in neither being implemented adequately.
Worryingly, there has been a marked reduction in the impetus to implement changes
across either dimension since market pressures have eased:
In terms of fiscal/political integration, there appears to be a reluctance to move
beyond the ‘enhanced surveillance’ procedures set out in the revamped Stability
and Growth Pact. Adherence to the Stability and Growth Pact rules may be a
necessary condition for preventing a future crisis but it is clearly not a sufficient
condition – after all, some of the countries that were hit hardest by the crisis had
been scrupulous adherents to the rules (Spain, Ireland), while those who had
flouted the rules went relatively unscathed (Germany, France). In the future, it will
remain difficult to uphold fiscal commitments made – even when enshrined in
treaties – as long as the political legitimacy remains at the national level. In other
words, there may need to be genuine political union before a fiscal union can work.
We are therefore sceptical that the changes agreed imply a sufficient degree of
fiscal/political integration to ensure the Euro area’s long-term survival (at least in
the absence of much deeper financial/banking integration).
In terms of financial/banking integration, the introduction of a common
supervisory framework is a necessary – but not sufficient – step towards deeper
financial integration. However, there has been back-tracking on an earlier
commitment to introduce direct recapitalisation of peripheral banks from the ESM
bailout fund and there appears to be little prospect of a single Euro area deposit
guarantee.15
Progress has been made but, in our view, Euro area institutions are not yet sufficiently
robust to ensure that the monetary union remains workable in the long run. While
institutional reforms that would represent a sufficient steady-state framework have been
discussed, they are yet to be implemented to a sufficient degree. It may be that the Euro
area ‘needs’ a renewed bout of market tension to provide policymakers with the impetus
required to complete the construction of a workable monetary union.
15 The European Commission originally proposed, in June 2012, that there would be four pillars to Europe’s banking union. These included: (i) a single EU deposit guarantee scheme covering all EU banks; (ii) a common resolution authority and a common resolution fund for the resolution of, at least, systemic and cross-border banks; (iii) a single EU supervisor with ultimate decision-making powers, in relation to systemic and cross-border banks; and (iv) a uniform single rule book for the prudential supervision of all banks.
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 31
Bibliography
Asdrubali, P, Sorensen, B & O. Yosha (1996) “Channels of interstate risk sharing: United States 1963-1990”, Quarterly
Journal of Economics, Vol. 111, 1081-1110.
Bayoumi, T. & B. Eichengreen (1992) ‘Shocking aspects of European Monetary Unification’, NBER Working Papers 3949,
National Bureau of Economic Research, Inc., January.
Bean, C. (1992), “Economic and Monetary Union in Europe”, Journal of Economic Perspectives, Vol. 6, No. 4 (Autumn),
pp 31-52.
Berthold, N. (1998) ‘Real wage rigidities, fiscal policy, and the stability of EMU in the transition phase’, ZBW Papers No.
22.
Blanchard, O.J. & L. Katz (1992) “Regional evolutions”, Brookings Papers on Economic Activity.
Bonin, H. et al (2008) ‘Geographic mobility in the European Union: Optimising its economic and social benefits’, Final
Report conducted by the Institute for the Study of Labour, NIRAS Consultants and The Swedish National Labour Market
Board for the European Commission DG Employment, Social Affairs and Equal Opportunities.
Broadbent, D. & S. Bahaj (2008) “Trade and dispersion of growth: The optimal currency area discussion revisited”, in The
Euro at Ten (Chapter 1), Goldman Sachs Global Investment Research.
Bruno, M., and J. Sachs (1985) Economics of Worldwide Stagflation, Oxford: Basil Blackwell.
Carney, M., (2014) “The economics of currency unions” Speech by Governor of the Bank of England, January 29, 2014.
Daly, K. (2001) ‘EMU: Asymmetry and risk-sharing’, M.Sc. Dissertation, University College London.
De Grauwe, P, (2011), "The Governance of a Fragile Eurozone", CEPS Working Documents, Economic Policy, May.
De Grauwe, P and Ji Y, (2012), “Self-fulfilling crises in the Eurozone: an empirical test”, Journal of International Money
and Finance, vol. e-pub.
Eichengreen, B. (1991) ‘Is Europe an Optimum Currency Area?’, NBER Working Papers 3579, National Bureau of
Economic Research, Inc., June.
Fatas, A. (1998) ‘Does EMU need a fiscal federation?’, Economic Policy, vol 13, no. 26 pp. 163-203, Wiley-Blackwell: April.
Feldstein, M. (1998) ‘The political economy of the European Economic and Monetary Union: Political sources of an
economic liability’, NBER Working Papers 6150.
Frankel, J. (1999) ‘No single currency regime is right for all countries or at all times’, Essays in International Finance,
Princeton University, August.
Frankel, J. & A.K. Rose (1997) ‘Is EMU more justifiable ex post than ex ante?’, European Economic Review, vol 41: 753-
760.
Frankel, J. & A.K. Rose (1996) ‘The endogeneity of the optimum currency area criteria’, NBER Working Papers 5700,
Cambridge, MA: NBER.
Furceri, D. & G. Karras (2008) ‘Business-cycle synchronization in the EMU’, Applied Economics, 40:12, 1491-1501.
Furceri, D. & A. Zdzienicka (2013) ‘The Euro area crisis: Need for a supranational fiscal risk sharing mechanism’, IMF
Working Paper WP/13/198.
Giannone, D., M. Lenza & L. Reichlin (2009) ‘Did the Euro Imply More Correlation of Cycles?’, in Alesina, A. And Giavazzi,
F. (eds.) Europe and the Euro, NBER Volume, University of Chicago Press.
Grubb, D., Jackman, R. and R. Layard (1983) “Wage rigidity and unemployment in OECD countries”, European
Economics Review 21 (1-2): 11-39.
Hishow, O.N. ‘Fiscal Federalism in the EMU: What makes the eurozone an OCA?’, SWP Working Paper FG1, 2007/15,
August 2007, SWP Berlin.
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 32
Kalemli-Ozcan, S., E. Papaioannou & J.L. Peydro (2010) ‘Financial Regulation, Integration and Synchronization of
Economic Activity’. TUSIAD-Koc University Economic Research Forum Working Papers 1005, revised April 2010.
Kenen, P.B. (1969) “The Optimum Currency Area: An Eclectic View”, in Mundell and Swoboda (eds.), Monetary Problems
of the International Economy, Chicago: University of Chicago Press.
Krugman, P. (2012) ‘Revenge of the Optimum Currency Area’, Address to NBER Macroecnomics Annual Conference,
April 20-21 2012, forthcoming in D. Acemoglu, J. Parker and M. Woodford (eds.) NBER Macroeconomics Annual 2012,
vol. 27.
Layard, R., S.J. Nickell & R. Jackman (1991) Unemployment: Macroeconomic Performance and the Labour Market,
Oxford University Press, November.
McCallum, J. (1995), “National borders matter: Canada-US regional trade patterns”, American Economic Review, 85 (3),
615-623.
Mongelli, F.P. (2002) ’”New” views on the optimum currency area theory: what is EMU telling us?’, ECB Working Paper
Series 138, European Central Bank, April 2002.
Morgan, D.P., B. Rime & P.E. Strahan (2004), ‘Bank Integration and State Business Cycles’, Quarterly Journal of
Economics, 119(3): 1555-85.
Mundell, R.A. (1961) ‘A theory of optimum currency areas’, American Economic Review, Vol. 51, No. 4:657-665.
Pill, H. (2012) “Creating a workable monetary union”, European Economics Daily, October 23, 2012, Goldman Sachs
Global Investment Research.
Sachs, J. & X. Sala-i-Martin (1992) ‘Fiscal Federalism and Optimum Currency Areas: Evidence for Europe from the United
States’, CEPR Discussion Papers 3855.
Schmitt-Grohe, S. & M. Uribe (2012) “The Case for Temporary Inflation in the Eurozone”, Working paper.
Schumacher, D., (2013), “Does EMU need a ‘genuine’ fiscal union in order to survive?”, European Economics Analyst,
13/11 (March 14, 2013), Goldman Sachs Global Investment Research.
Zimmermann, K.F. (2009) ‘Labor Mobility and the Integration of European Labor Markets’, IZA Discussion Paper Series,
IZA.
April 29, 2014 Global Economics Paper
Goldman Sachs Global Investment Research 34
Disclosure Appendix
Reg AC
I, Kevin Daly, hereby certify that all of the views expressed in this report accurately reflect my personal views, which have not been influenced by
considerations of the firm's business or client relationships.
Disclosures
Global product; distributing entities
The Global Investment Research Division of Goldman Sachs produces and distributes research products for clients of Goldman Sachs on a global
basis. Analysts based in Goldman Sachs offices around the world produce equity research on industries and companies, and research on
macroeconomics, currencies, commodities and portfolio strategy. This research is disseminated in Australia by Goldman Sachs Australia Pty Ltd
(ABN 21 006 797 897); in Brazil by Goldman Sachs do Brasil Corretora de Títulos e Valores Mobiliários S.A.; in Canada by either Goldman Sachs
Canada Inc. or Goldman, Sachs & Co.; in Hong Kong by Goldman Sachs (Asia) L.L.C.; in India by Goldman Sachs (India) Securities Private Ltd.; in
Japan by Goldman Sachs Japan Co., Ltd.; in the Republic of Korea by Goldman Sachs (Asia) L.L.C., Seoul Branch; in New Zealand by Goldman Sachs
New Zealand Limited; in Russia by OOO Goldman Sachs; in Singapore by Goldman Sachs (Singapore) Pte. (Company Number: 198602165W); and in
the United States of America by Goldman, Sachs & Co. Goldman Sachs International has approved this research in connection with its distribution in
the United Kingdom and European Union.
European Union: Goldman Sachs International authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority
and the Prudential Regulation Authority, has approved this research in connection with its distribution in the European Union and United Kingdom;
Goldman Sachs AG and Goldman Sachs International Zweigniederlassung Frankfurt, regulated by the Bundesanstalt für
Finanzdienstleistungsaufsicht, may also distribute research in Germany.
General disclosures
This research is for our clients only. Other than disclosures relating to Goldman Sachs, this research is based on current public information that we
consider reliable, but we do not represent it is accurate or complete, and it should not be relied on as such. We seek to update our research as
appropriate, but various regulations may prevent us from doing so. Other than certain industry reports published on a periodic basis, the large
majority of reports are published at irregular intervals as appropriate in the analyst's judgment.
Goldman Sachs conducts a global full-service, integrated investment banking, investment management, and brokerage business. We have
investment banking and other business relationships with a substantial percentage of the companies covered by our Global Investment Research
Division. Goldman, Sachs & Co., the United States broker dealer, is a member of SIPC (http://www.sipc.org).
Our salespeople, traders, and other professionals may provide oral or written market commentary or trading strategies to our clients and our
proprietary trading desks that reflect opinions that are contrary to the opinions expressed in this research. Our asset management area, our
proprietary trading desks and investing businesses may make investment decisions that are inconsistent with the recommendations or views
expressed in this research.
The analysts named in this report may have from time to time discussed with our clients, including Goldman Sachs salespersons and traders, or may
discuss in this report, trading strategies that reference catalysts or events that may have a near-term impact on the market price of the equity
securities discussed in this report, which impact may be directionally counter to the analyst's published price target expectations for such stocks. Any
such trading strategies are distinct from and do not affect the analyst's fundamental equity rating for such stocks, which rating reflects a stock's
return potential relative to its coverage group as described herein.
We and our affiliates, officers, directors, and employees, excluding equity and credit analysts, will from time to time have long or short positions in,
act as principal in, and buy or sell, the securities or derivatives, if any, referred to in this research.
This research is not an offer to sell or the solicitation of an offer to buy any security in any jurisdiction where such an offer or solicitation would be
illegal. It does not constitute a personal recommendation or take into account the particular investment objectives, financial situations, or needs of
individual clients. Clients should consider whether any advice or recommendation in this research is suitable for their particular circumstances and, if
appropriate, seek professional advice, including tax advice. The price and value of investments referred to in this research and the income from them
may fluctuate. Past performance is not a guide to future performance, future returns are not guaranteed, and a loss of original capital may occur.
Fluctuations in exchange rates could have adverse effects on the value or price of, or income derived from, certain investments.
Certain transactions, including those involving futures, options, and other derivatives, give rise to substantial risk and are not suitable for all investors.
Investors should review current options disclosure documents which are available from Goldman Sachs sales representatives or at
http://www.theocc.com/about/publications/character-risks.jsp. Transaction costs may be significant in option strategies calling for multiple purchase
and sales of options such as spreads. Supporting documentation will be supplied upon request.
All research reports are disseminated and available to all clients simultaneously through electronic publication to our internal client websites. Not all
research content is redistributed to our clients or available to third-party aggregators, nor is Goldman Sachs responsible for the redistribution of our
research by third party aggregators. For research or data available on a particular security, please contact your sales representative or go to
http://360.gs.com.
Disclosure information is also available at http://www.gs.com/research/hedge.html or from Research Compliance, 200 West Street, New York, NY
10282.
© 2014 Goldman Sachs.
No part of this material may be (i) copied, photocopied or duplicated in any form by any means or (ii) redistributed without the prior written consent of The Goldman Sachs Group, Inc.