rapcer 01-18 - stefancollignon.de · stefan collignon, gianluca cubadda, giuseppe de arcangelis,...

Rapporto CER

Re-inventing Europe

n.1 2018

CENTRO EUROPA RICERCHE

ER produces short and medium-term forecast of the Italian economy, evalua-

tions on economic policy, reports on public finance, fiscal, monetary and industrial policy.

CER is regularly invited to auditions by the Italian Parliament on the economic outlook and

public finance trends. CER prepares a «consensus forecast» for the Italian Ministry of the

Economy jointly with other research institutes.

CER's forecasting and simulation expertise is embodied in its econometric models, which

are continuously updated to take into account structural changes in the national and in-

ternational economy. The econometric models are used to test the impact of policy

measures as well as provide forecasts of economic and financial variables.

The micro-simulation model, using data on wages and consumer expenditure, is used to

evaluate the distribution impact of tax and tariff measures on Italian households.

CER's reports are available to subscribers as are presentations and workshops on the reports

organised and sponsored by CER and attended to by experts and leading personalities

and policy makers.

Centro Europa Ricerche S.r.l.

Via G. Carissimi, 41 - 00198 Roma

Tel. (0039) 06 8081304

E-mail: [email protected]

www.centroeuroparicerche.it

Presidente Onorario: Giorgio Ruffolo

Presidente: Vladimiro Giacché

Vicepresidenti: Claudio Levorato, Gennaro Mariconda

Direttore della ricerca: Stefano Fantacone

Comitato scientifico: Paolo Guerrieri Paleotti (presidente), Barbara Annicchiarico, Pierluigi Ciocca,

Stefan Collignon, Gianluca Cubadda, Giuseppe De Arcangelis, Flavio Delbono, Francesco Ferrante,

Giovanni Ferri, Luca Fiorito, Sergio Ginebri, Paolo Giordani, Anna Giunta, Pasquale Lelio Iapadre, Mas-

similiano Marzo, Valentina Meliciani, Salvatore Nisticò, Antonio Pedone, Andrea Cesare Resti, Gio-

vanna Vallanti

Rapporto CER: pubblicazione periodica a carattere economico. Anno XXXV

Direttore responsabile: Jacopo Tondelli

Iscrizione n. 59/2016 del 5 aprile 2016 del Registro della Stampa del Tribunale di Roma

Proprietario della testata: Centro Europa Ricerche S.r.l.

C.C.I.A.A. Roma: R.E.A. 480286

Edizione: Centro Europa Ricerche S.r.l.

Printed out at CER – May, 2018

Re- invent ing Europe

The fol lowing authors have contributed to this report: Stefan Col l ignon,

Stefano Fantacone, Antonio Forte, Chiara Guerel lo.

R A P P O R T O C E R

5 5

Executive summary 7

The return of economic growth in Europe

11

UNEMPLOYMENT 13

PRODUCTIVITY GAPS 15

WAGE COST COMPETITIVENESS 17

Box 1. Non-TFP growth and output-gap 20

Policy mix

21

THE INTERACTION OF MONETARY AND FISCAL POLICY 21

MONETARY POLICY 25

FISCAL POLICY 28

EMPIRICAL EVIDENCE ON THE INTERACTION OF MONETARY AND FISCAL POLICY 32

Box 2. Fiscal-monetary policy mix in theory 36

Box 3. The ECB unconventional monetary stance 38

Box 4. Glossary for fiscal policy terms 39

European Fiscal Union

41

FISCAL POLICY COORDINATION IN THE EURO AREA 41

THE FLOW OF FUNDS ANALYSIS 43

Box 5. The current debate on fiscal spill-overs in the Euro Area 50

Towards a European Fiscal Union

51

THE EUROPEAN UNION BUDGET 52

A EUROPEAN FISCAL UNION FOR MACROECONOMIC STABILITY 55

References

57


7 7

Executive summary

President Macron has called for the re-invention of Europe. “Faced with the great chal-

lenges of our times, such as defence and security, great migrations, development, climate

change, the digital revolution and regulation of a globalized economy”, he asked, “have

European countries found means to defend their interests and values, and to guarantee

and adapt their democratic and social model that is unique worldwide? Can they address

each of these challenges alone? We cannot afford to keep the same policies, the same

habits, the same procedures and the same budget. No more can we choose to turn in-

wards within national borders. The only way to ensure our future, is the rebuilding of a sov-

ereign, united and democratic Europe” (1).

This year’s Rapporto Europa will focus on the economic context for re-inventing European

model. We give particular attention to the details of a European Fiscal Union. The Maas-

tricht Treaty did not incorporate any form of Fiscal Union, except in the very narrow sense

of fiscal discipline: each member state committed to maintain sound finances and the

Stability and Growth Pact (SGP) detailed the operationalisation of fiscal discipline. A com-

mon European budget exists, but it is small and has no specific functions for the Euro Area.

The Euro crisis has revealed that this institutional set-up is not optimal.

In June 2015, the so-called Five Presidents’ Report noted that “all mature Monetary Unions

have put in place a common macroeconomic stabilisation function to better deal with

shocks that cannot be managed at the national level alone” (p. 14). It proposed a “fiscal

union” that would “improve the cushioning of large macroeconomic shocks and thereby

make EMU over all more resilient” (p.14) (2).

Since then, many improvements have happened:

1. By the European Semester of economic policy coordination has been given clearer

guidance for the Euro Area as a whole and a stronger focus on social aspects.

2. European Fiscal Boards and National Productivity Boards have been set up.

3. Technical assistance to Member States was boosted with the creation of the Structural

Reform Support Service.

4. Important steps towards completing the Banking Union and Capital Markets Union

have been taken, notably by advancing in parallel on risk-reduction and risk-sharing

measures in the banking sector.

(1) http://www.elysee.fr/assets/Initiative-for-Europe-a-sovereign-united-democratic-Europe-Emmanuel-

Macron.pdf

(2) Juncker J‐C., D. Tusk, J. Dijsselbloem, M. Draghi and M. Schulz (2015): Completing Europe’s Economic

and Monetary Union, Five Presidents’ Report, 22 June. https://ec.europa.eu/commission/publica-

tions/five-presidents-report-completing-europes-economic-and-monetary-union_en

http://www.elysee.fr/assets/Initiative-for-Europe-a-sovereign-united-democratic-Europe-Emmanuel-Macron.pdf

http://www.elysee.fr/assets/Initiative-for-Europe-a-sovereign-united-democratic-Europe-Emmanuel-Macron.pdf

https://ec.europa.eu/commission/publications/five-presidents-report-completing-europes-economic-and-monetary-union_en

https://ec.europa.eu/commission/publications/five-presidents-report-completing-europes-economic-and-monetary-union_en

N. 1 - 2 0 1 8

8

In December 2017, the European Commission has presented new proposals and initiatives

that echoed President Macron’s discourse and Commission President Junker’s State of the

Union speech:

1. A European Monetary Fund (EMF) is to safeguard the financial stability of the Euro

Area, as well as the financial stability of the 'participating member states'

2. The so-called Fiscal compact (Treaty on Stability, Coordination and Governance) is to

be integrated into the Union legal framework, taking into account the appropriate

flexibility.

3. New budgetary instruments for the Euro Area are to be developed within the Union

framework.

4. For the period 2018-2020, EU funds will be mobilised in support of national reforms and

the Structural Reform Support Programme is to be strengthened.

5. Finally, reflection has started whether the Euro Area needs a European Minister of

Economy and Finance.

The interaction between sovereign member states’ fiscal policy and the unified monetary

policy is not optimal in the European Monetary Union (EMU). Therefore, the European Cen-

tral Bank is overcharged and the policy mix in the Euro Area has been suboptimal for a

long time. We believe that a European Fiscal Union will help to improve the policy mix and,

therefore, the long run development of the member states in the Euro Area. We develop

the analysis in this year’s Rapporto Europa starting from an overview of the main develop-

ments of the Euro Area economies. We then look at the economic governance of the Euro

Area, especially the coordination of fiscal policy between member states and the interac-

tion with monetary policy.

Economic activity is now well above the previous peak attained in 2008: it is 7.5% higher in

the Euro Area as a whole. Not only actual, but also potential output growth has improved

since the ECB has started its unconventional monetary policy in 2012. While this was a nec-

essary policy response to the crisis, the situation now calls for a cautious normalization of

monetary policy and a careful evaluation of fiscal policy.

The closing of the output gaps may explain why potential output is improving again across

Europe. Although the impact of closing the output gap on non-TFP growth differentials is

small, the indirect effect that output gaps exert on potential output growth through the

growth of factors of production is large. However, as the gaps disappear, one must con-

sider that in the long-run economic growth is determined by total factor productivity (TFP).

For stagnating countries like Italy, the challenge does not so much consist in stimulating the

economy, which could of course contribute to an increase in the use of capital and labour,

but more importantly structural reforms are urgently needed to improve the efficiency of

resource use and raise TFP. This is also necessary in order to improve competitiveness and

raise the purchasing power of wages.


9 9

Between 2011 and 2014, fiscal consolidation made the policy mix excessively tight; after

2014, monetary policy has contributed to a significant softening (real long-term interest

rates fell), but fiscal policy remained stuck at an aggregated cyclically adjusted deficit of

-1%. This aggregate development is not fully reflected in the levels of individual member

states’ real long-term interest rates: while in Italy nominal interest rates strongly declined,

the deflationary pressures shrank the gap between nominal and real interest rates, but in

Germany this gap widened, driving real interest rates into negative territory for several

years. The divergence in the real long-term interest rates’ dynamic observed in the last

years might be attributable to country-specific policy, and in particular to heterogeneous

fiscal stance. Before the crisis the relationship between long-term interest rate and the fiscal

stance was linear and negative across the countries, but during the crises and even more

recently this relationship has become highly non-linear.

We show that the government budget position does not explain the output gap, nor is the

volatility in government budgets affected by output gap volatility over the medium run. By

contrast, monetary policy is highly significant, confirming the excessive burden for

monetary policy in the European policy mix. However, monetary policy affects real long

run interest rates in the short-run, although in the medium run their dynamics depends also

on government budget positions. This highlights that the effectiveness of monetary policy

is highly dependent on the fiscal policy stance in shaping the dynamic of the long-run rate.

Since fiscal policies are heterogenous among the countries, the effectiveness of monetary

policy in stimulating the economy is highly uneven across countries.

Fiscal policies by member states are not neutral in the European monetary union, because

the changes of fiscal stances will generate spillover effects in neighbouring member states.

Our estimations suggest that signs and dimensions of spillovers are strongly hetoregeneous

among member countries so that a different form of fiscal policies coordination within the

Euro Area, defined by a strategic interaction between member countries on the timing of

consolidation, would be preferable to just a common set of rules, which might induce all

countries to consolidate at the same moment, making austerity unsustainable. However,

since the cross-country heterogeneity of attitude towards a joint consolidation across

country is a matter of fact, supranational coordination by means of a European Minister of

Finance might mediate this position, increasing the sustainability of fiscal consolidation in

Europe.

The budget of the European Union pays for policies carried out at the European level. The

total budget amounted to € 136 bln in 2016, which is less than 1% of gross national income

(GNI) of the European Union. Of this amount, € 117.9 bln are spend inside the EU but only €

77.2 bln in the Euro Area. There are therefore good reasons for increasing the budget for

the monetary union, as President Macron has demanded.

The experience has proven that the voluntary cooperation among the EU member states

is no longer sufficient to generate and allocate European public goods efficiently. Instead

N. 1 - 2 0 1 8

10

of intergovernmental cooperation, it may be a better solution to provide certain public

goods by a federal agency, which is acting on the basis of competing legislation. With

these elements, a re-invented European Union would be able to keep its founding promise:

namely to preserve peace, increase welfare and bring people together.


11 11

The return of economic growth in Europe

After a decade of economic recession, stagnation and crisis, Europe has finally returned

to solid economic growth. The recovery is part of a worldwide trend (IMF, 2018). Global

output has grown by 3.7% in 2017, and global growth forecasts for 2018 and 2019 have

been revised upward to 3.9%. The pickup in growth is broad-based, reflecting also upside

surprises in Europe and Asia and the expected impact of the recently approved U.S. tax

policy changes.

Economic activity, measured by GDP at constant prices, is now well above the previous

peak attained in 2008: it is 7.5% higher in the Euro Area as a whole. Contrary to the short-

lived export led economic revival in 2011, the present European recovery is fed by domes-

tic demand and has spread across countries and sectors. Figure 1 shows in the left panel

that in the Euro Area output is getting close to its potential and in the right panel that the

output gap has closed. This means that the recessionary tendencies and deflationary risks

are subsiding, while there are still no excessive demand pressures which could ignite a

wage-price inflation. The right-hand panel in Figure 1 also shows that not only actual, but

also potential output growth has improved since the European Central Bank has started its

unconventional monetary policy in 2012. While this was a necessary policy response to the

crisis, the situation now calls for a cautious normalization of monetary policy and a careful

evaluation of fiscal policy. However, the recent economic progress is not equally distrib-

uted. In Germany, GDP is 13.4% above its 2008 level, in France 8.6%, in Spain 4.3%, but it is

still -3.2% below that level in Italy and even -23.0% in Greece.

Figure 1. Economic activity and growth in the Euro Area

Source: European Commission AMECO database.

5000

5500

6000

6500

7000

7500

8000

8500

9000

9500

10000

10500

2000 2002 2004 2006 2008 2010 2012 2014 2016

-4

-2

0

2

4

6

8

Euro Area

Output Gap Real GDP Potential GDP

2000 2002 2004 2006 2008 2010 2012 2014 2016

-5%

-4%

-3%

-2%

-1%

0%

1%

2%

3%

4%

5%

Euro Area

Real GDP Growth Rate Potential GDP Growth Rate

N. 1 - 2 0 1 8

12

Figure 2 shows that Germany quickly closed the output gap after 2009, but potential

growth only improved after 2012, the year of aggressive monetary loosening. In France,

the output gap is still negative and potential output growth has nearly halved from 2% in

2000 to 1.2% in 2018, although the tendency is rising again. In Italy, the situation is far worse.

The output gap has been nearly double of France, and potential growth has been nega-

tive in every year from 2009 to 2016. In 2017 it turned positive with 0.3%. Thus, Italian living

conditions have deteriorated in absolute and relative terms. Nevertheless, Spain is an ex-

ample that not all of the South is doom and gloom: the output gap was larger than in Italy,

but potential output is now growing again and with 2.3% this is faster than in 2008. The worst

case is Greece, where the output gap is still large and potential output still falling.

Figure 2. Economic activity and growth in selected Member States

1000

1200

1400

1600

1800

2000

2200

2400

2600

2800

3000

2000 2002 2004 2006 2008 2010 2012 2014 2016

-6

-4

-2

0

2

4

6

Germany


2000 2002 2004 2006 2008 2010 2012 2014 2016

-8%

-6%

-4%

-2%

0%

2%

4%

6%

Germany


1000

1200

1400

1600

1800

2000

2200

2000 2002 2004 2006 2008 2010 2012 2014 2016

-4

-2

0

2

4

6

8

France


2000 2002 2004 2006 2008 2010 2012 2014 2016

-4%

-3%

-2%

-1%

0%

1%

2%

3%

4%

5%

France


1000

1100

1200

1300

1400

1500

1600

1700

2000 2002 2004 2006 2008 2010 2012 2014 2016

-6

-4

-2

0

2

4

6

Italy


2000 2002 2004 2006 2008 2010 2012 2014 2016

-6%

-4%

-2%

0%

2%

4%

6%

Italy



13 13

cont. Figure 2. Economic activity and growth in selected Member States


UNEMPLOYMENT

Unemployment is also coming down in Europe, although more slowly than in the US where

it is now at 4.3% - the lowest level since 2000 and the second lowest since 1970. As shown

in Figure 3, in the Euro Area, unemployment stood at 8.6% in 2018, still higher than at the

beginning of the Global Financial Crisis (7.5%), but below the level (9.7%) when monetary

union started in 1999. Thus, despite the severe crisis, it would be hard to argue that Europe

was better off without the euro.

Labour markets are improving in all member states (in the core as well as in the southern

periphery), although the levels are still diverging widely (see Figure 4). In Germany unem-

ployment stands at 3.5% of the labour force, in Ireland it has fallen from a peak of 14.5 to

5.5% and in Spain from 25.1 to 15.6%. Even in Greece, unemployment rates have come

down from 27.5 to 20.6%. Progress is slowest in Italy (10.9% down from 12.7%– still far above

the 6.8% before the crisis) and France (9.3% down from 10.4%). In the new member states,

unemployment rates stand between 7.7% (Latvia, Slovakia) and 2.9% (Czech Republic),

but in all Eastern member states they have come down by more than one half.

400

600

800

1000

1200

2000 2002 2004 2006 2008 2010 2012 2014 2016

-10

-5

0

5

10

15

Spain


2000 2002 2004 2006 2008 2010 2012 2014 2016

-5%

-4%

-3%

-2%

-1%

0%

1%

2%

3%

4%

5%

6%

Spain


0

50

100

150

200

250

2000 2002 2004 2006 2008 2010 2012 2014 2016

-18

-9

0

9

18

27

Greece


2000 2002 2004 2006 2008 2010 2012 2014 2016

-10%

-8%

-6%

-4%

-2%

0%

2%

4%

6%

8%

Greece


N. 1 - 2 0 1 8

14

Figure 3. Unemployment rates


Figure 4. Unemployment rates in selected member states


2

4

6

8

10

12

14

1980 1985 1990 1995 2000 2005 2010 2015

Euro Area (EA 12) United States

Ge

rma

n U

nif

ica

tio

n

Eu

ro

Le

hm

an

De

fau

lt

3

4

5

6

7

8

9

10

11

12

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Germany

7.0

7.5

8.0

8.5

9.0

9.5

10.0

10.5

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

France

6

7

8

9

10

11

12

13

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Italy

8

12

16

20

24

28

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Spain

4

8

12

16

20

24

28

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Greece

2

4

6

8

10

12

14

16

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Ireland

6

8

10

12

14

16

18

20

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Slovakia

4

5

6

7

8

9

10

11

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Slovenia

4

6

8

10

12

14

16

18

20

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Latvia

2

3

4

5

6

7

8

9

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Czech Republic

0

4

8

12

16

20

24

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Poland

3

4

5

6

7

8

9

10

11

12

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Hungary


15 15

PRODUCTIVITY GAPS

In previous Rapporti, we have emphasized the lack of demand which has aggravated the

crisis. We presented evidence (notably in Rapporto Europa 2016) that the length and

depth of output gaps have an impact on investment and potential output. The closing of

the output gaps may explain why potential output is improving again across Europe (see

Box 1). However, as the gaps disappear, one must consider that output gap is only a short-

run measure of the effective demand, while in the long-run economic growth is deter-

mined by total factor productivity (TFP). A country’s standards of living depend upon

productivity (i.e. on the efficiency by which capital and labour are used in the production

process). Productivity determines income, wages, the capacity of fiscal space and, wel-

fare and social expenditure. The relative stagnation of TFP, hence, appears as a deteriora-

tion of living standards relative to other countries. This is a major source for Euroscepticism.

Figure 5 presents the cumulative effect of TFP in total economic growth. We distinguish the

first decade of monetary union (1999-2010) from the post-crisis period (2010-2018) and ob-

serve an extreme variety of performances between countries and over time. Not surpris-

ingly, overall growth was higher in the first decade in all countries, except in Ireland. For the

Euro Area in aggregate, the contribution of TFP and additional resource employment is

equally balanced. However, the contribution of TFP varies substantially between member

states.

On the one side, we find stagnating countries like Italy, where GDP grew by 7% between

2000 and 2010 (equivalent to 0.7% per year), and only by 0.8% between 2010 and 2018

(equivalent to 0.1% per year). In the early period, TFP fell by -3.3%, so that extensive growth

through expanding uses of factors of production was 10.3%. In other words, while the Italian

demand boom of the first decade put more people to work, their efficiency declined. Dur-

ing the crisis years, this has changed. Italian TFP has now improved by a meagre 1.3%

(equivalent to 0.16% per year), but the contribution of extensive use of labour and capital

has fallen by 0.5%, so that observed growth is only 0.8%. In comparison, French TFP has

increased by 3.3% since 2010, and German TFP by 7.5%. Extensive growth is more con-

strained in Germany, which may be a consequence of operating at full employment, and

this may have led to increased productivity.

On the other side, Ireland has dramatically improved its productivity since 2010 with TFP

improvements of 39.4%, equivalent to 4.2% per year. This performance is heavily influenced

by the Manufacturing and ICT sectors and in particular by FDI enterprises in these sectors.

However, the narrow base of enterprises in high value-added sectors (particularly in foreign

owned Pharma and ICT) disguises underperforming sectors and skews Ireland’s productiv-

ity level and growth rate (3). Similar discrepancies occur in the new member states in Cen-

tral and Eastern Europe. Finally, the new member states in Eastern Europe have also signif-

icantly improved total factor productivity (on average 17% since 2010, i.e. 2% per year),

which is the result of profound transformations in the formerly planned economies under So-

viet domination, but also of the sustained wage competitiveness (which we discuss below).

(3) See (Irish National Competitiveness Council, 2017.

N. 1 - 2 0 1 8

16

Figure 5. Economic growth and total factor productivity


Some of the new member states have expanded the resource use of their production in

addition to TFP, although there is no necessary link between these two developments as

demonstrated by Slovenia and Latvia, where TFP is the nearly exclusive driver of growth.

Although this is often underestimated and ignored, the integration of Central and Eastern

Europe is an enormous challenge to the old member states. Because their productivity in-

creases more rapidly, and wage competition is higher, these new countries attract invest-

ment to the detriment of the old. To keep living standards improving, the old countries have

to accelerate innovation and productivity. For stagnating countries like Italy, the challenge

6,5%

5,6%

-3,1%

-1,7%

1,3%

1,3%

2,2%

3,1%

3,3%

4,2%

4,6%

5,9%

6,2%

7,5%

7,5%

12,3%

12,7%

16,1%

17,4%

35,9%

39,4%

5,5%

5,8%

7,0%

7,0%

11,3%

13,1%

25,4%

6,9%

6,3%

4,8%

-20,0% 0,0% 20,0% 40,0% 60,0% 80,0%

EU (excl. UK)

Euro Area

Greece

Cyprus

Italy

Luxembourg

Belgium

Austria

France

Finland

Portugal

Spain

Netherlands

Germany

Estonia

Malta

Lithuania

Slovenia

Slovakia

Latvia

Ireland

Denmark

Sweden

Poland

Hungary

Czech Republic

Bulgaria

Romania

Japan

United States

United Kingdom

TFP (2010- 1999) TFP (2019- 2010)

Non TFP ( 2010- 1999) Non TFP ( 2019- 2010)


17 17

does not so much consist in stimulating the economy, which could of course contribute to

an increase in the use of capital and labour, but more importantly in improving the effi-

ciency of resource use and raise total factor productivity.

In the UK, productivity has slowed down. A major reason for this development is the uncer-

tainty for investors created by Brexit. A detailed study by the National Institute of Economic

and Social Research (Young, 2017) has shown that the UK’s standard of living are deterio-

rating because of higher cost of living due to the currency depreciation that weighs heavily

on the unemployed, single parents and pension holders. In addition, weak investment and

the delocalisation of businesses will have long run effects, which are already showing up

as a 20% reduction in the productivity trend.

WAGE COST COMPETITIVENESS

The unequal developments in productivity affect the competitiveness in the different re-

gions and member states of the Euro Area. In previous RAPPORTI, we have elaborated a

methodology for assessing the competitiveness of wage levels.

We have derived a benchmark wage level from the classic textbook assumption that equi-

librium in a single market would reflect equal returns on capital. Given the relative levels of

capital and labour productivity, we can derive from this assumption the equilibrium level

of wages at which this equal return would be realised. When the actual wage levels are

below the equilibrium, an economy is considered to be competitive, and inversely if actual

wages are higher, it lacks competitiveness. This means that the evolution of the equilibrium

wage is highly significant for the improvement of standards of living and this improvement

depends on the developments of productivity. Figure 6 shows the developments of actual

and equilibrium wages for selected Euro Area member states and Figure 7 shows the 2018

wage gaps for all member states in the European Union.

We find in Figure 6 that Germany has regained competitiveness since it undertook labour

market reforms in the mid-2000s, while France has persistently lost competitiveness with

wages today 10% above equilibrium. The loss was accelerated in 2015 when actual labour

compensation increased despite a stagnation in the equilibrium level. Since 2016 compet-

itiveness has improved in Italy and Spain. In Italy this seems to be the effect of the Job Act

of 2014-15. In Greece, the equilibrium wage has now started to rise after a sustained re-

duction before 2016, which was not compensated by the cuts in actual wage costs. In

Ireland, by contrast, we observe a significant leap in equilibrium wages, which is explained

by increases in productivity: “Across the OECD, labour productivity growth in manufactur-

ing has tended to outpace services sector growth. Between 2009 and 2015, Irish growth in

manufacturing was 16% compared with 3% in services and -2% in construction. Irish manu-

facturing productivity data has been significantly influenced by corporate restructuring

[…], including the relocation of firms with significant IP assets and aircraft leasing, [leading]

to noteworthy increases in labour productivity, particularly in 2015 (22.5%)” (Irish National

Competitiveness Council, 2017, pp. 58-62).

N. 1 - 2 0 1 8

18

Figure 6. Equilibrium wages

Source: European Commission AMECO database and CER computations.

Figure 7 shows the gaps between actual labour compensation and the equilibrium level in

2018, measured in euros per month. A gap indicates how much average labour compen-

sation could be increased in competitive member states to reach equilibrium, or how

much the equilibrium wage would have to be brought up in non-competitive economies.

Ireland is an outlier, due to the jump in productivity in 2016, which has not been matched

by actual wage increases. This is specific to the Irish economic structure given that the het-

erogeneity of firms and productivities is large, but social cohesion requires wage differen-

tials not to be excessive. Hence, the competitive advantage translates into high returns on

capital in certain industries. A similar argument applies to Luxembourg, which is dominated

by the financial industry and a supporting ICT sector. However, in Central and Eastern Eu-

rope, wages are more broadly lagging behind productivity improvements. The northern

0.94

0.96

0.98

1.00

1.02

1.04

1.0628

32

36

40

44

48

96 98 00 02 04 06 08 10 12 14 16 18

Germany

Eu

ro 1

00

0

Co

mp

eti

tiv

en

ess in

de

x (

1=

eq

uili

bri

um

)

0.98

1.00

1.02

1.04

1.06

1.08

1.10

1.12

25

30

35

40

45

50

96 98 00 02 04 06 08 10 12 14 16 18

France

Co

mp

eti

tiv

en

ess in

de

x (

1=

eq

uili

bri

um

)

Eu

ro 1

00

0

0.94

0.96

0.98

1.00

1.02

1.04

20.0

22.5

25.0

27.5

30.0

32.5

35.0

96 98 00 02 04 06 08 10 12 14 16 18

Co

mp

eti

tiv

en

ess

ind

ex

(1

= e

qu

ilib

riu

m)

Spain

0.88

0.92

0.96

1.00

1.04

20

24

28

32

36

40

96 98 00 02 04 06 08 10 12 14 16 18

Italy

Co

mp

eti

tiv

en

ess

ind

ex

(1

= e

qu

ilib

riu

m)

Euro

s 10

00

0.90

0.95

1.00

1.05

1.10

1.15

12

16

20

24

28

96 98 00 02 04 06 08 10 12 14 16 18

Co

mp

eti

tiv

en

ess

ind

ex

(1

= e

qu

ilib

riu

m)

Euro

10

00

Greece

0.5

0.6

0.7

0.8

0.9

1.0

20

40

60

80

100

96 98 00 02 04 06 08 10 12 14 16 18

Act ual w ages Equilibrium wages Comet it iveness index

Ireland

Co

mp

eti

tiv

en

ess

ind

ex

(1

= e

qu

ilib

riu

m)

Euro

10

00

Euro

10

00

0

50

100

150

200

250

300

Titolo del grafico

Actual wages Equilibrium wages Competitiveness index


19 19

core countries of the Euro Area are narrowly below equilibrium wage levels. This is particu-

larly interesting in Germany, where the wage gap is € 95 below equilibrium, representing

an undervaluation of 4%. Not a dramatic cost advantage. As we saw in Figure 6, Italy and

Spain have regained some advantage and, are nowadays marginally below equilibrium.

By contrast, France, Belgium and Greece face serious wage cost problems, which need to

be handled by improving productivity.

Figure 7. Wage gaps (euro per month)


-5217

-2724-1586

-805-550

-461-330

-200-124-111-102-100-95-64-44

50144148

392

-422-225-221-166

-59-25

7229

-853-220-184

104151

-3000 -2000 -1000 0 1000

European Union

Luxembourg

Slovakia

Cyprus

Portugal

Estonia

Germany

Italy

Greece

France

Poland

Romania

Denmark

Croatia

Norway

Japan

N. 1 - 2 0 1 8

20

BOX 1. NON-TFP GROWTH AND OUTPUT GAP

Total factor productivity is measured by mean of a modelized production function, where output

(GDP in constant prices) is produced by capital and labour. Hence, output increases when more

capital and more workers are employed. However, with technological progress, the same amount

of labour and capital will be able to produce more. TFP is the residual growth that is not explained by

larger amounts of production factors.

We have first performed a panel analysis to assess whether increasing output gaps boosts non-TFP

growth, namely the economic growth due to change in factors of production and not in productivity.

We employ the common correlated effects estimator of Chudik e Pesaran (2015) a panel of 30 coun-

tries (Europe and US) for the period 1996-2019. We estimate so the correlation of the distance be-

tween country-specific non-TFP growth and the cross-countries average (csa), and output gaps. The

pooled estimates of the model are:

Eq. 1 𝑛𝑜𝑛𝑇𝐹𝑃𝑖,𝑡 = 0.09 𝑛𝑜𝑛𝑇𝐹𝑃𝑖,𝑡−1 + 0.09 𝑜𝑢𝑡𝑝𝑢𝑡𝑔𝑎𝑝𝑖,𝑡−1 + 0.95 𝑛𝑜𝑛𝑇𝐹𝑃_𝑐𝑠𝑎𝑡 + 𝜀𝑖,𝑡

We find a very strong cross-country dependence as shown by the high correlation with the aggre-

gate European non-TFP growth (cross-country average): the elasticity is 95%. This means that country-

specific factors play only a marginal role in determining the divergence of non-TFP from the European

trend. On average a 1% reduction of negative output gaps drives an increase of 9 basis points in the

non-TFP growth rates, over and above the European stance. However, although the dynamics of

non-TFP growth is highly interconnected, only in few countries does the distance of a single country’s

non-TFP growth to the European trend significantly respond to the output gaps of that country. Typi-

cally, large open economies, like US, Germany, France, Italy and Spain, show a positive correlation

between non-TFP growth and national output gap. For few other smaller economies, the same sig-

nificant association holds: Bulgaria, Denmark and Portugal inside the EU and, also, Switzerland.

Although the impact of closing the output gap on non-TFP growth differentials is small, the indirect

effect that output gaps exert on potential output growth through the growth of factors of production

is large. To show this stylized fact, we regressed the growth of real potential output on its determinants

(TFP and non-TFP growth rates), disentangling the non-TFP growth differentials due to increasing out-

put gap (𝑛𝑜𝑛𝑇𝐹𝑃̂𝑖,𝑡, the predicted values from Eq. 1) from the residual components 𝜀𝑖,𝑡 (inclusive of

the European average). The estimates are as follows:

Eq. 2 𝑃𝑜𝑡𝑒𝑛𝑡𝑖𝑎𝑙 𝑂𝑢𝑡𝑝𝑢𝑡 𝐺𝑟𝑜𝑤𝑡ℎ𝑖,𝑡 = 0.21 𝑇𝐹𝑃𝑖,𝑡−1 + 0.28 𝑛𝑜𝑛𝑇𝐹𝑃̂𝑖,𝑡−1 +

+0.38( 𝑛𝑜𝑛𝑇𝐹𝑃_𝑐𝑠𝑎𝑡−1 + 𝜀𝑖,𝑡−1) + 0.83 𝑃𝑜𝑡𝑒𝑛𝑡𝑖𝑎𝑙 𝑂𝑢𝑡𝑝𝑢𝑡 𝐺𝑟𝑜𝑤𝑡ℎ_𝑐𝑠𝑎𝑡 + 𝜐𝑖,𝑡

We observe that country-specific non-TFP growth due to changes in output gaps explains the diver-

gence of potential output growth from European trend as much as TFP growth does.


21 21

Policy mix

While long-run standards of living depend on productivity and supply side conditions, the

management of aggregate demand generates incentives for private investment and

growth. If these incentives are sustained over time, productivity and potential output will

rise. This is why a stability oriented macroeconomic policy framework is crucial for the per-

formance of the Euro Area economy. However, if different regions and sectors of produc-

tion respond with different elasticities to demand impulses, economic growth will diverge,

generating inequalities which might cause political backlashes. This is an important argu-

ment in favour of more coordinated economic policies in Europe.

The Euro Area has a unique set of policy rules and institutions, which makes policy coordi-

nation difficult. Monetary policy is centralized by the independent European Central Bank,

while fiscal and economic policy is decentralized in the hands of sovereign member states,

which are, however, constrained by a set of rules such as the Stability and Growth Pact

and the new Fiscal Compact. These rules do not apply to the same extend to member

states outside the Euro Area. However, experience of the Euro crisis has shown that the

interaction between monetary policy and fiscal policy, i.e. the policy mix, has not been

satisfactory. Fiscal policy was excessively rigid, while the European Central Bank has done

its utmost to save the euro and stimulate the euro economy. This sub-optimal performance,

which risks undermining the credibility of the ECB, has been widely recognized. It has given

rise to calls for a European Fiscal Union. Before analysing the effects of fiscal policy in detail,

we will first discuss the experience of the policy mix during the crisis.

THE INTERACTION OF MONETARY AND FISCAL POLICY

About the interaction of monetary and fiscal policies there is an abundance of literature

(4). Monetary policy refers to the central bank’s control of the availability of credit in the

economy to achieve the broad objectives of economic policy. Fiscal policy refers to the

government’s choice regarding the use of taxation and public spending to regulate the

aggregate level of economic activity. This is also called the stabilisation function of public

finance. Its purpose is to avoid excess demand pressures, which would cause inflation, and

lack of aggregate demand which would cause recessions and unemployment. Thus, the

output gap is a good proxy for measuring whether the stability objective is attained.

Because total public expenditure in the Euro Area is around 45% of GDP, the role of public

spending is of prime importance for the performance of aggregate demand. Yet, given

the independence of sovereign member states, each government can decide its own

budget, so that the aggregate budget position for the Euro Area is the random outcome

of national decisions. This means fiscal policy cannot be used as a policy instrument, and

as a consequence the full burden of managing the macroeconomic conditions in the Euro

Area falls on the European Central Bank. However, while the ECB can respond to an ag-

(4) For a review see (Hilbers, 2005) and (Fores, 2018).

N. 1 - 2 0 1 8

22

gregate shock to the Euro Area as a whole, it does not have instruments to manage asym-

metric shocks which translate into unequal output gaps between member states.

This is true for all integrated monetary economies, but in traditional nation states govern-

ments can use public finances to avoid or correct excessive deviations. Of course, eco-

nomic liberals are opposed to the interference by governments in the market, because

they believe the “invisible hand of the market” mechanism will restore equilibrium. How-

ever, our analysis of wage competitiveness shows that this adjustment process may take

longer than is politically acceptable. The populist alternative is to exit monetary union and

the single market. However, even if it were possible to do so without serious welfare losses

(which is not true) (5), this does not solve the fundamental economic problem in any inte-

grated currency union. Because substantial economic heterogeneities prevail, logically

northern Italy should also break away from monetary union with the Mezzogiorno, West

Germany from East Germany, Catalonia from Spain, and Flanders from Wallonia. Robert

Mundell, who was awarded the Nobel Prize for his work on optimum currency areas, has

therefore always argued that monetary union is a political project and not an economic

optimum. But this implies that a well-working monetary union requires a fiscal union with

additional policy tools and ultimately also a political union.

Thus, although monetary and fiscal policy are implemented by different bodies, these pol-

icies are far from independent. As explained in Box 2, a change in one will influence the

effectiveness of the other and thereby the overall impact of any policy change. Finding

the right policy mix is not always easy, but in the European Monetary Union the difficulty is

compounded by the fact that there is no federal authority to determine the aggregate

fiscal policy stance for the Euro Area.

Figure 9 gives an idea of the policy trade-off in the Euro Area. We have proxied monetary

policy by the long run real interest rates and fiscal policy by the cyclically adjusted struc-

tural deficits. The regression line, which stands for our efficiency line in Figure B2.1 of Box 2,

indicates that the long run real interest at balanced budgets would be 1.9%, which - in

“normal times” and given the inflation target of the ECB - is equivalent to a nominal long-

term rate of 3.9%. Given the historic record, this seems reasonable (see Figure 11).

Defining an optimal policy mix in the Euro Area poses a problem, because the aggregate

budget position which corresponds to the unified monetary policy stance is the sum of

different national government positions. Each national government has its own welfare

function and chooses its budget position in response to its national constituency. But if each

country is allowed to set its fiscal policy stance independently of the others, the conse-

quence will be that high deficits in some countries would push interest rates up for member

states with low deficits. Hence, national fiscal policies create externalities for all citizens in the

Euro Area. The resultant policy mix cannot be optimal. On the other hand, imposing a unique

policy stance for every government would create welfare losses because the policy prefer-

ences resulting from different welfare functions would be overwritten by the unique policy

rule. The Stability and Growth Pact effectively imposes a balanced budget at point E of Fig-

ure B2.1, where the indifference curve does not correspond with the policy preference of R.

(5) The National Institute (Young, 2017) has estimated that the first 18 months since the Brexit vote have

created costs “worth over £600 per annum to the average household” in the UK.


23 23

Figure 9. Policy mix in the Euro Area


The problem is partly economic, partly political. It is economic if the response of the eco-

nomic system to a fiscal deficit is significantly different from the Euro Area average. There

is evidence, discussed below, that the single currency has endogenously generated con-

vergence to a more unified economic response. However, as shown in Box 2, the political

problem remains unsolved, for example, at the balanced budget point E (Figure B2.1),

countries with a preference for point R remain on a lower level of utility. In a federal de-

mocracy the policy mix is chosen by all citizens through the democratic procedures of

electing a government. This means, there is only one collectively chosen welfare function,

which represents a democratic consensus. The point where such indifference curve

touches the efficiency line would therefore correspond to a welfare optimum. For exam-

ple, in the American context of the 1980s, the point R (for Republican) reflected the Volker-

Reagan policy mix, while the point E was closer to the Clinton Greenspan policy mix of the

1990s. In the European Union, where sovereign nation states are in command, such collec-

tive choice is not possible, and this undermines the perception of welfare gains in monetary

union. To overcome this problem, it would be desirable to find a form of policy coordina-

tion, which allows, on the one hand, the designation of a collective fiscal position that cor-

responds to the integrated monetary policy, and, on the other hand, that enables citizens’

representation in the conduct of fiscal policy. This would be the European equivalent of

the American slogan: “no taxation without representation”.

The most remarkable feature in Figure 9 is the complete absence of fiscal policy during the

Euro crisis. Fiscal consolidation between 2011 and 2014 made the policy mix excessively

tight: the points for those years are well above the equilibrium line. Finally, after 2014, mon-

etary policy contributed to a significant softening (real long-term interest rates fell verti-

cally), but fiscal policy remained stuck at an aggregated cyclically adjusted deficit of -1%.

No doubt, this was due to the fact that fiscal policy in the Euro Area is constrained by the

Stability and Growth Pact and the Fiscal Compact, which impose the identical rule of bal-

ancing cyclically adjusted deficits. We shall discuss this further below. Furthermore, in recent

years, budget positions have been quite heterogeneous across member states, while the

long-term interest rates had a tendency to decline. Figure 10 indicates the spread of na-

tional long- term interest rates in relation to the national deviation from the Euro Area fiscal

2010

2011

2012

2013

2014

2015

2016

y = -0,8631x - 0,1872R² = 0,7444

-0,5

0,0

0,5

1,0

1,5

2,0

2,5

3,0

3,5

4,0

-4,5 -4,0 -3,5 -3,0 -2,5 -2,0 -1,5 -1,0 -0,5 0,0

Rea

l Lon

g-Te

rm In

tere

ts R

ate

Structural Government Balance (% of potential GDP)

N. 1 - 2 0 1 8

24

stance. The scatterplot shows country-specific cyclically adjusted government budgets net

of interest (i.e. the distance from EA average) and the associated 10-year government

bond yield (distance from EA average) as the mean through the period 1999-2008 or 2009-

2017.

Figure 10. Correlation between fiscal stance and long-term interest rates

Source: European Commission AMECO database and ECB statistical data warehouse (SDW).

Before the financial crisis and the European sovereign crisis the relationship between long-

term interest rate and the fiscal stance was linear and negative. That means that if a

country under-consolidates (primary fiscal deficit higher than the Euro Area average), then

it would have paid a positive spread between the country’s long-term interest rate and the

Euro average one. This mechanism was called the market discipline, and it worked even if

it did not prevent the Euro crisis. By contrast, during the crises and even more recently this

relationship has become highly non-linear. This seems to be the result of the emergence of

four clusters. The spread of long-term interest rates over the average Euro rate was

negative for countries with fiscal consolidation, with the exception of Italy and Greece

(Greece being a large outlier, the data are not shown). These two countries have the

highest debt-GDP ratios in Europe and this explains why long-term bond yields contain

significant risk permia, so that even the strong fiscal consolidation does not bring the spread

down. However, another explanation would be that excessive austerity is

counterproductive because it increases market worries and uncertainty. This interpretation

is supported by the two other clusters in Figure 10. In Portugal, Cyprus, Latvia, and Lithuania

interest spreads remain high against a moderate fiscal losening. But in Spain, Slovakia,

Slovenia and France, the fiscal expansion goes together with a fall of long-term rates below

the Euro average, which makes sense if markets expect that the looser fiscal stance will

generate growth. Nevertheless, from the point of view of the Euro Area one could also

argue that this last cluster is free riding on the consolidation efforts by Germany, Luxemburg

and Italy.

BE

DE

FR

LV

LT

LU

MT

NLAT

SI

SK

FI

ES IT

CY

PT

BEDE IEES

FR

ITCYLV

LT

LU

MTNLAT

PT

SI

SK

y = 0,3135x3 - 0,4454x2 - 1,2962xR² = 0,4249

y = 0,0296x2 - 0,2722xR² = 0,4914

-2,5

-2,0

-1,5

-1,0

-0,5

0,0

0,5

1,0

1,5

2,0

2,5

-4,0 -3,0 -2,0 -1,0 0,0 1,0 2,0 3,0 4,0 5,0 6,0

Lo

ng

-te

rm In

tere

st R

ate

(M

S-E

A)

Fiscal Stance (MS-EA)

2009-2017 1999-2008 Trend (09-17) Trend (99-08)


25 25

MONETARY POLICY

Monetary policy has saved the euro. During the crisis the European Central Bank has re-

mained “the only game in town” for stabilizing the Euro economy. The rigid rules of the

Stability and Growth Pact and governments’ concern with consolidating deficits (which

was very different from the policy mix in the United States) has put an unprecedented onus

on monetary policy to support aggregate demand. The ECB has responded to this chal-

lenge by acting decisively to secure price stability in the face of an unparalleled economic

and financial crisis. However, experience after nearly one decade of monetary policy has

revealed that unconventional monetary policy loses efficiency over time. Thus, it is clear

the central bank cannot remain the only game in town indefinitely. This requires further

structural and institutional reforms in order to improve the policy mix for the Euro Area as a

whole and to ensure balanced growth for member states. With the closing of output-gaps

the time has now come to exit the unconventionally loose monetary policies.

Monetary policy during the crisis has gradually softened as economic conditions deterio-

rated. This development is not fully reflected in the levels of real long-term interest rates

shown in Figure 11. In fact, when the room for negative short-run rates is constrained, other

unconventional measures had to supplement nominal long-term rate cuts. In a liquidity

trap, when the nominal short-run interest rate approaches the zero-lower-bound (in Figure

11 this is summarized by the grey areas indicating when the rate are below 1%), the effec-

tiveness of monetary policy sharply declines. In this case, the central bank can control nei-

ther the dynamic of the long-term interest rate nor inflation and runs the risk of de-anchor-

ing inflation expectations from the 2% target.

In most of the advanced economies, central banks continue to exert some control over

inflation, mostly by relying on forward guidance, namely the promise to keep the short-term

nominal rate to very low values for longer than necessary. They are thereby generating

inflation expectations at longer-term horizons. However, they have also attempted to cut

long-term interest rates by means of unconventional monetary policy (see Box 3). However,

while these policies were initially very effective in deeply troubled financial markets, they

have lost their power over the real long-term rate during the last few years. In 2015, real

long-term interest rates increased and eventually turned positive even though nominal

rates fell, because the inflation rate became negative in most countries. Nowadays, only

Germany and Sweden are facing negative real interest rates. In Germany, this phenome-

non is mostly attributable to very low (zero) long-term nominal interest rates coupled with

rising prices, while in Sweden it is a consequence of the ultra-loose monetary policy driving

the nominal short-term rates into negative territories as well. In Eastern European countries

with local currency, all interest rates were generally higher than in the Euro Area.

Outside the Euro Area most of the difference in the long-term interest rates is reflected in

the exchange rates dynamic, reported in Figure 12. For instance, after the Brexit vote (2016-

2017), sterling strongly depreciated against the Euro, and real interest rates in the UK en-

tered negative territories, although short-term interest rates were kept constant, or even

increased in the later quarters of 2017. However, within the European Monetary Union, the

divergence in the real long-term interest rates’ dynamic observed in the last years might

be attributable to country-specific policy, and in particular to heterogeneous fiscal stance.

N. 1 - 2 0 1 8

26

Figure 11. Long-term and short-term interest rates

Source: ECB Statistical Data Warehouse (SDW) and European Commission AMECO database.

Figure 12. Exchange rates against the Euro (% deviation from long run trend)

Source: ECB Statistical Data Warehouse (SDW).

-20%

-15%

-10%

-5%

0%

5%

10%

15%

20%

25%

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

Polish zloty Swedish krona UK pound sterling US dollar


27 27

As summarized in Box 3, the ECB has softened monetary policy by undertaking unconven-

tional measures and combining MPOs (Main Policy Operations) with long-term refinancing

operations (LTROs). This changed in 2015 when it started its Assets Purchase Programmes.

Other central banks, like the Fed and the Bank of England, had taken a more aggressive

approach already in 2009 (see Figure 13). Although long term interest rates in Europe

started to decline already in 2012, the Euro Area real interest rate turned negative only in

2015, when the assets purchasing program was announced. However, the combined ef-

fects of conventional and unconventional monetary instruments have been uneven across

European countries. While in Italy nominal interest rates strongly declined, the deflationary

pressures shrank the gap between nominal and real interest rates, in Germany this gap

widened, driving real interest rates into negative territory for several years.

Figure 13. QE in advanced economies

Source: ECB (EA), Board of Governors of the Federal Reserve Bank (US), Bank of England (UK).

Figure 14. QE and Long-Term (LT) rate in the Euro Area

Source: ECB Statistical Data Warehouse (SDW) and ECB balance sheet.

The main driver of the expansions of ECB assets during the last years has been the Assets

Purchase Programme, which has been implemented through the Public- Sector Purchase

Programme (PSPP). This quantitative easing measures allow the Euro-system to buy mem-

ber states government bonds, with the scope to control nominal long-term rates by miti-

gating sovereign risk premia. However, the implementation of these measures has differed

between member states and the ratio of the bonds purchased under the PSPP to the total

amount of outstanding government debt spans in January 2018 from 10% to 30%.

As reported in Figure 14 left panel, the countries with the highest share of debt held at ECB

0

500

1000

1500

2000

2500

3000

3500

4000

4500

2009

Q1

2009

Q3

2010

Q1

2010

Q3

2011

Q1

2011

Q3

2012

Q1

2012

Q3

2013

Q1

2013

Q3

2014

Q1

2014

Q3

2015

Q1

2015

Q3

2016

Q1

2016

Q3

2017

Q1

2017

Q3

Assets Purchasing Programs (€ bln)

EA US UK

0

50

100

150

200

250

300

2009

Q1

2009

Q3

2010

Q1

2010

Q3

2011

Q1

2011

Q3

2012

Q1

2012

Q3

2013

Q1

2013

Q3

2014

Q1

2014

Q3

2015

Q1

2015

Q3

2016

Q1

2016

Q3

2017

Q1

2017

Q3

Assets Purchasing Programs (% GDP)

EA US UK

N. 1 - 2 0 1 8

28

(cumulated net acquisition of bonds) at beginning of 2018, are typically the countries with

the lowest nominal long term-interest rate today. However, if we contrast the acquisition of

assets under the PSPP and the change in the long-term rate between 2015 and today, the

effect of the QE has been quite uneven. In several countries with relative large scale PSPP,

like Netherlands and Belgium, the change in the long-term rate before and after the im-

plementation of the policy measure has been almost null. For some others, like Italy, the

relatively smaller PSPP is reflected in even smaller changes in the long-term interest rate.

Actually, interest rates started declining in 2012 and most of the drop in the long-run rate

was motivated to the signalling effects of the “whatever it takes” speech of ECB president

Mario Draghi in 2012.

FISCAL POLICY

During the Euro crisis, fiscal policy became the driver of austerity, because the rules of the

Stability and Growth Pact inhibited the kind of anti-cyclical policies which were pursued in

the US. Of course, after the collapse of Lehman-Brothers in 2008 and the ensuing global

financial crisis, governments were quick to respond by active fiscal stimulations. However,

as soon as economic growth returned in 2011, European authorities started to implement

the 3%-deficit rule of the Stability and Growth Pact. As we document, this was a mistake,

because fiscal consolidation came too early and broke the economic upswing dynamic

so that output gaps deteriorated again.

Figure 15 provides an overview of some member states’ selected budget positions. The

Table 15. Budget positions in Europe


-7

-6

-5

-4

-3

-2

-1

0

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Euro Area

-5

-4

-3

-2

-1

0

1

2

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Germany

-8

-7

-6

-5

-4

-3

-2

-1

0

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

France

-6

-5

-4

-3

-2

-1

0

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Italy

-12

-10

-8

-6

-4

-2

0

2

4

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Spain

-35

-30

-25

-20

-15

-10

-5

0

5

10

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Ireland

-16

-12

-8

-4

0

4

8

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Greece

-4

-3

-2

-1

0

1

2

3

4

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Structural Actual

Sweden

-12

-10

-8

-6

-4

-2

0

2

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

United Kingdom


29 29

Euro Area operated in a rage of budget deficits between zero and -3% of GDP before the

crisis; immediately after the Lehman Brothers bankruptcy the actual deficit fell below 6%,

but consolidation after 2010 was steady and sharp. The rising gap between the structural

and actual deficit after 2012 is an indicator of the excessively tight fiscal consolidation in

the Euro Area. See Box 4 for an explanation of the terms used in this report. The policies

pursued by individual member states vary significantly. Germany only marginally overshot

the 3% deficit criterion in 2010 and moved into a structural and actual budget surplus after

2014. But most Euro Area members remained significantly below the 3% deficit ratio.

France, Spain and Greece only got there in 2017. In Italy fiscal consolidation was radical

in 2011-13, because financial markets where close to a melt-down, given the high debt-

GDP ratio. Outside the Euro Area some countries like Sweden also consolidated their budg-

ets and generated surpluses, while others like the UK used fiscal policy to stimulate demand

during the crisis.

These budget positions must be seen against the background of very diverse debt posi-

tions, as shown in Table 1. The debt to GDP ratio for the Euro Area is 88.1%. In Greece, it is

more than double and in Italy nearly 50% higher. However, Italy’s share in the total public

debt is the highest of all Euro Area member states, while that of Greece is only 3.3%.

Table 1. Public debt in the Euro Area


Euro Area Countries Non-Euro Area Countries

Debt/GDP (%) Share of EA debt

Debt/GDP (%)

Euro area (EA12) 88.1 100.0%

Greece 177.8 3.3% Bulgaria 24.3

Italy 130.8 22.8% Czech Republic 33.3

Portugal 124.1 2.5% Denmark 35.5

Belgium 102.5 4.6% Sweden 36.6

Cyprus 98.3 0.2% Romania 39.1

Spain 96.9 11.6% Poland 53.0

France 96.9 22.6% Hungary 71.5

Austria 76.2 2.9% Croatia 77.4

Slovenia 74.1 0.3% United Kingdom 85.3

Ireland 69.1 2.1% United States 108.4

Finland 62.1 1.4% Japan 239.1

Germany 61.2 20.6%

Netherlands 54.9 4.2%

Malta 51.6 0.1%

Slovakia 49.9 0.4%

Lithuania 37.9 0.2%

Latvia 35.5 0.1%

Luxembourg 23.0 0.1%

Estonia 9.1 0.0%

N. 1 - 2 0 1 8

30

Figure 16. Annual Standard Deviation of budget positions


Despite the constraining fiscal rules, budget positions have been more volatile in the Euro

Area during the crisis then outside. Figure 16 shows that the cross-country standard devia-

tion of actual budget deficits within the EMU has been significantly higher between 2008

and 2014 than in the Non-Euro Area countries. However, the standard deviation of the

structural deficits has been in excess of the Non-Euro Area only immediately after the Leh-

man-Brothers bankruptcy and in the depth of the Euro-crisis.

The excessively tight fiscal stance, especially between 2011 and 2014, is also documented

by Figure 17. The premature tightening is clear from the movements of the fiscal stance

Figure 17. Euro Area fiscal stance

1

2

3

4

5

6

7

2000

2002

2004

2006

2008

2010

2012

2014

2016

2018

Le

hm

an

De

fau

lt

Structural Budget Position

1

2

3

4

5

6

7

2000

2002

2004

2006

2008

2010

2012

2014

2016

2018

Euro Area St. Dev . Non-Euro Area St. Dev .

Le

hm

an

De

fau

lt

Actual Budget Position


31 31

cont. Figure 17. Euro Area fiscal stance


between 2009 and 2011. Only in 2015-16 did the Euro Area take a weak anti-cyclical and

stimulating position that is now turning into pro-cyclical loosening. This was different in Non-

Euro Area member states, which generally followed anti-cyclical policies.

The fiscal policy stance is measured by the variation of the cyclically adjusted structural

deficit net of interest rates. The Euro Area’s restrictive policy stance reflected the require-

ments in Germany, where – take in isolation - the relatively expansive boom would have

justified anti-cyclical tightening in 2011 and in 2012 (see Figure 12 and Figure 17), but it did

not take into consideration the more adverse conditions in some of the other Euro Area

economies. The alternating policy mix in Greece between tightening and loosening re-

flects the confused approach to the sovereign debt crisis in Greece. It is, however, interest-

ing that the policy mix in the Non-Euro Area countries was generally more anti-cyclical and

the fiscal policy stance was more stable: changes in the cyclically adjusted primary budget

20072008

20092010

2011

2012

2013

2014

2015

2016

2017

2018

2019

y = -0.1622x - 0.3699R² = 0.0778

-6

-4

-2

0

2

4

6

8

10

-20 -15 -10 -5 0 5 10

Tit

le

Title

Greece

anticyclical tightening

anticyclical losening

procyclical tightening

procyclical losening

N. 1 - 2 0 1 8

32

position varied between +1.6 and -1.4 percentage points, but in the out-countries only be-

tween +1.03 and -0.96 points. Thus, fiscal policy was actually more active in the Euro Area

than outside, but the crisis was also deeper here.

EMPIRICAL EVIDENCE ON THE INTERACTION OF MONETARY AND FISCAL POLICY

We will now investigate the relationship between fiscal policy and monetary policy. We

quantify the direct impact of monetary policy by a change in the short-run interest rate or

in the stock of assets held by the ECB (two proxies for conventional and unconventional

monetary stance) and fiscal policy by a change in the fiscal stance. We also look at the

indirect impact transmitted by the business cycle (measured by the output gap) on these

monetary variables. We also consider the loopback response of monetary policy to the

effects of the fiscal reaction. For this purpose, we have constructed a Panel VAR model,

estimating contemporaneously the correlation between the fiscal stance, the output gap,

short-term money market rates, 10-year government bond yields and the change in ECB

assets. The model has the form:

Eq. 3 𝒚𝒊,𝒕 = 𝜌𝒚𝒊,𝒕−𝟏 + 𝛽𝒙𝒕 + 𝛼𝑖 + 𝜀𝑖,𝑡

Where 𝒚𝒊,𝒕 is a vector composed (in order) of the output gaps (AMECO), the cyclically ad-

justed government surplus/deficits net of interests as a percentage of GDP (AMECO), the

Euro Area 3-months money market rate (Eurostat) and the long-run interest rates as the 10-

year government bond yields (Eurostat). Furthermore, the eventuality of common factors

affecting both the European business-cycle is collected by including in 𝒙𝒕 the first three lags

of the cross-sectional average of the output gaps (6). In order to capture the heterogeneity

between member states (MS), we look at MS output gaps and fiscal position as the differ-

ence from the aggregate Euro Area (EA) values by including the cross-sectional average

fiscal stance in 𝒙𝒕.

We ran the estimates for the full monetary union period 1998-2017 and the coefficients

represent the average elasticities across the 18 Euro Area countries (Estonia excl.)

considered. The variables named MS are the deviation of each member state from the

integrated Euro Area performance. We catch the average cross-country heterogeneity in

the business cycle and in the fiscal policy stance, which are due to structural differences,

by estimating the fixed effects of the VAR model. However, the model does not tell us much

about the heterogeneity in the elasticities between countries. Despite this limitation, we are

able to distinguish between country-specific and aggregate Euro Area shocks. Hence, the

coefficient for the (lagged) member state (MS) budget reflects the average elasticity of

individual countries and the Euro Area (EA) government budget indicates the change in

the dependent variables due to the aggregate fiscal policy stance of the Euro Area. For

instance, an increase in a member state’s government deficit may be due to a national

policy of increasing public expenditure or it may be induced by more flexible fiscal rules at

the European level (say because the Stability Pact is less strictly applied).

Figure 18 reports the contribution of fiscal and monetary policy to the forecasted volatility

(6) This method, proposed by (Chudik & Pesaran, 2015) allows to consider implicitly the cross-states depend-

ence of output gaps due to business cycle synchronization.


33 33

of selected variables; Figure 19 shows the Impulse Response Functions (IRF) after a negative

shock to the short term interest rate or after an improvement in the fiscal stance. Those

results jointly show how the output gap, fiscal policy and monetary policy (i.e. short and

long term interst rates) respond to the various shocks.

Figure 18. Fiscal and monetary policy contribution to the volatility of selected variables

Source: ECB Statistical Data Warehouse (SDW), EC AMECO database and CER computations.

Figure 19. Panel VAR on the monetary and fiscal policy mix in the Euro Area

IRFs to a shock to either the short-term rate (-1 p.p.) or to the government surplus-to-GDP (+1 p.p.)


0%

5%

10%

15%

20%

25%

30%

35%

40%

Output

Gap

ECB

assets

Long-term

rate

Short-term

rate

Gov.

Budget

Output

Gap

ECB

assets

Long-term

rate

Short-term

rate

Gov.

Budget

2-year 5-year

FE

VD

Response Varibale and time horizon

Output Gap ECB assets Long-term rate Short-term rate Gov. Budget

0,00

0,10

0,20

0,30

0,40

0,50

0,60

0,70

0,80

0,90

1,00

0 1 2 3 4 5

Output Gap

Short-Term Interest Rate Government Surplus/Deficit

-1,20

-1,00

-0,80

-0,60

-0,40

-0,20

0,00

0,20

0,40

0 1 2 3 4 5

Short-Term Interest Rate


-0,35

-0,30

-0,25

-0,20

-0,15

-0,10

-0,05

0,00

0 1 2 3 4 5

Long-Term Interest Rate


-0,20

0,00

0,20

0,40

0,60

0,80

1,00

1,20

0 1 2 3 4 5

Government Surplus/Deficit


N. 1 - 2 0 1 8

34

We start with the determinants of the forecasted volatility of selected variables in Figure 18.

The output gap is not explained by the government budget position, nor is the government

budget volatility affected by output gap volatility in the medium run. We interpret this as a

sign for the inconsistent fiscal policy stances shown in Figure 17. Hence, better policy

coordination would improve the economic performance of the Euro Area. Furthermore,

monetary policy is highly significant. The contribution to output gap volatility by short-term

interest rate changes, which are controlled by the ECB, is larger than for long-term rates;

the impact by unconventional monetary policy measures, which have expanded the ECB

balance sheet, is even stronger. This confirms the excessive burden for monetary policy in

the European policy mix.

Furthermore, we turn to the analysis of the dynamic response to either a monetary or a

fiscal policy shock. We start by looking at the impact on the output gap in the top-left panel

of Figure 19. First of all, we observe that output gaps respond quickly to monetary policy,

but are rather persistent in the long run. The response to a fiscal policy shock is slow and

weak. Another striking result, although not reported here, is that it is positively correlated

with the aggregate output gap of the Euro Area. This clearly means that by improving the

policy mix for the Euro Area as a whole, the economic conditions in each member state

will improve. This is an important argument in favour of better macroeconomic policy

coordination at the European level.

Second, this latter result is confirmed by the dynamics of the member states fiscal stance

after a monetary shock, because fiscal policy does not respond to monetary policy,

supporting the view that fiscal stance is an independent and autonomous decision.

Third, monetary policy (proxied by short-term interest rate) responds weakly to country-

specific conditions, as shown in Figure 19, but strongly to the Euro Area aggregate fiscal

stance. Hence, monetary policy does not respond to fiscal policy in member states, but in

fact, the ECB lowers short-term rates exclusively in response to fiscal consolidation in the

aggregate Euro Area stance while increases them when the output gap turns positive and

risks generating inflation. This is in line with our discussion of the policy mix (see Box 2).

Finally, we found that monetary policy does affect real long run interest rate in Figure 18 in

the short run, but in the medium run the dynamics also depends on government budget

positions. A similar conclusion can be drawn from the IRFs of Figure 19. In fact, from the

second year onwards there is no difference between the IRFs after an expansionary

monetary or a contractionary fiscal policy shock. This highlights that the effectiveness of

monetary policy in shaping the dynamic of the long-term rate, is highly dependent on the

fiscal policy stance.

After a cut in the short-term interest rate, the government balance deteriorates (weakly),

causing an increase in the sovereign spread, hence in the long-term interest rate, which

bounds back towards the equilibrium. Therefore, since fiscal policies are highly

heterogenous among the countries, the effectiveness of monetary policy in stimulating the

economy is highly uneven across countries. As previously shown in Figure 10, in the last years

the correlation between fiscal stance and long-term rates has turned from clearly negative

trade-off to a highly non-linear relation. It follows that the fiscal expansions before the crisis

were counterproductive because they increased the long-term rates, while today with

higher fiscal policy heterogeneity, monetary policy might have become more supportive.


35 35

In the decade since 2008, monetary policy was constrained by the zero-lower-bound of

nominal interest rates, but by using a combination of long run refinancing operations and

quantitative easing, the ECB was able to exert some direct control on the long-term rate,

thereby increasing its effectiveness. We have run estimations for different measures of ECB

balance sheets for the unconventional monetary policy stance over the period 2009-2017:

Long Term Refinancing Operations (LTRO, TLTRO, etc.), and total securities and government

lending, or the sum of the two. In Figure 20 we show the responses of the output gap, the

money market rate, government stance and 10-year yield to an increase of 1 p.p. of GDP

in either the securities held for policy purposes (blue line) or in the overall assets purchased

for unconventional monetary policy (LTRO plus APP). We do not report the responses after

an increase in the LTRO’s assets because those policies had heterogenous effects over the

period considered, since target operations were much more effective than initial LTRO. The

monetary-fiscal policy interaction was even stronger during those years than before the

crisis, because after an increase in ECB assets for monetary purposes, either just in the form

of securities or combined with Long-Term Refinancing Operations, short-term interest rates

dropped, while in the first-year long-term interest rates rose because the government bal-

ances deteriorated, pushing up interest spreads. However, and this is important, the large

increase in the long-term interest rate is the driver of the small widening in the output gap.

Therefore, most of the relative ineffectiveness of quantitative easing measures is attributa-

ble to fiscal policy moral hazard.

Figure 20. Panel VAR on the monetary and fiscal policy mix in the Euro Area

IRFs to a shock to the ECB assets (+ 1mln)


-1,4

-1,2

-1,0

-0,8

-0,6

-0,4

-0,2

0,0

0 1 2 3 4 5

Output Gap

ECB Assets EBC Securities

-1,8

-1,6

-1,4

-1,2

-1,0

-0,8

-0,6

-0,4

-0,2

0,0

0 1 2 3 4 5

Short-Term Interest Rate


-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0 1 2 3 4 5

Long-Term Interest Rate


-0,8

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0 1 2 3 4 5

Government Surplus/Deficit

ECB assets EBC Securities

N. 1 - 2 0 1 8

36

BOX 2. FISCAL-MONETARY POLICY MIX IN THEORY

The relation between stability, fiscal and monetary policy, and democracy can be described by a

set of four simple equations:

Eq. 4 Stability condition: 𝑦 – 𝑦∗ = 𝜑0 − 𝜑1𝑟 − 𝜑2𝑠

Eq. 5 Monetary policy rule: 𝑟 = 𝑟∗ + 𝑓(𝜋 − 𝜋∗)

Eq. 6 Efficiency frontier: 𝑟∗ = 𝜑0

𝜑1−

𝜑2

𝜑1𝑠

Eq. 7 Welfare function: 𝑈 = 𝑟𝛼𝑠1−𝛼

Equation 3 gives the stability condition whereby the output gap is a function of interest rates are and

the primary surplus of the government. Equation 4 gives the monetary public policy rule, whereby the

central bank sets the interest rate as a mark-up over the long run natural rate at which price stability

is ensured. The efficiency frontier (Eq. 5) will then establish the trade-off between equilibrium interest

rates and fiscal positions. Finally, equation 6 is the welfare function by which the utility of a society is

determined by the relative weight given to interest rates or fiscal policy. Democracy is modelled as

the shifting of these consensual weights through elections.

Figure B2.1. Policy Mix

The relation between monetary and fiscal policy is expressed by Figure B2.1, which is derived from

the equations above. It shows the monetary proxy r (for real interest rate) on the vertical axis and the

fiscal position (cyclically adjusted primary surplus, s) on the horizontal axis. On the right of the vertical

zero-line, the budget is in surplus, to the left it is in deficit. The downward sloping line, which goes

through the points R and E is the efficiency line that models the trade-off between higher interest

rates and a negative budget balance versus lower interest rates and a budget surplus. Any point on

the efficiency line is efficient in the sense that output gaps are zero. This means that the efficiency line

represents cyclically adjusted budget positions. Above the efficiency line the policy mix is too restric-

tive and therefore causing depressions, below the efficiency line, it is too lose creating inflation.

Hence, the efficiency line is the locus of multiple equilibria. In order to choose an optimal policy- mix,

we need to add the welfare function. At point R the deficit is relatively high and so are interest rates;

at point E the budget is balanced, and interest rates are low.


37 37

All points above the efficiency line reflect an excessively tight policy mix, so that negative output

gaps are on the right side above the equilibrium, while points below this line signal a stimulating policy

mix. An excessively tight policy mix can be adjusted by easier monetary policy (lower interest rates)

or by a looser fiscal policy stance (higher deficits).

The so-called automatic stabilizers of fiscal policy imply that the cyclically adjusted “structural” deficit

reflects the position on the efficiency line, while actual deficits vary during the business cycle. How-

ever, fiscal policy requires taking a pro-active stance with respect to the interaction between fiscal

and monetary policy.

N. 1 - 2 0 1 8

38

BOX 3. THE ECB UNCONVENTIONAL MONETARY STANCE

The relation between stability, fiscal and monetary policy, and democracy can be described by a

set of four simple equations:

Phases of The Crisis Problems Remedies

I. Liquidity crisis

after Lehman

bankruptcy

2008 - 2010

Excessive lending in boom

• Uncertainty about scale of

exposure to subprime

products

• Sudden stop of money

market funding

• Rush to cash

• Lower main refinancing rate

to 1%

• Longer duration for liquidity

provisions

• Enlarged ECB balance sheet

II. Sovereign

Debt Crisis

2011-2014

Bank-sovereign nexus

• Risk of bank exposure to

government defaults

• Risk of assets fire sales by

banks

• Disruption of monetary

transmission channel

• Sovereign spread expressing

redenomination risk, distorted

asset prices and credit

conditions

• Lengthening maturity

operations

• Expanding range of eligible

collateral

• Outright monetary

transactions

III. Return to

Equilibrium

2014-2018

Balance sheet recession

• Corporate deleveraging led

to reduce lending (credit

crunch)

• Risk of deflation due to

savings-investment

imbalances

• Very negative equilibrium

interest rates

• November 2013: rate cuts to

0.25%

• Introduced forward

guidance

• Negative interest rates on

ECB deposit facility

• Flattening short end of yield

curve

• Asset Purchase Program

(APP): covered bonds, asset-

backed securities, sovereign

and corporate bonds

• Target long term refinancing

operations (TLTRO)


39 39

BOX 4. GLOSSARY FOR FISCAL POLICY TERMS

The budget position of general government is the difference between government revenue and ex-

penditure (T-G). A negative position is it deficit, a positive position is a surplus. Surpluses are also called

net lending, or net government savings; a deficit is public sector net borrowing. The total budget

position is the sum of the structural and cyclical component of the budget.

The cyclical component of the budget position reflects short-term shocks to the income or expendi-

ture of the government. Because a negative income shock which reduces tax income and forces

the government to borrow, the cyclical component of the budget position is also often called auto-

matic stabilizer.

◼ This structural or cyclically adjusted budget position indicates the underlying balance between

long-term government revenues and expenditure, while removing factors that are attributable to

shocks or the business cycle.

◼ The primary budget position is the difference of revenue minus expenditure net of interest charges.

It indicates the discretionary room for governments to adjust their net spending.

◼ The fiscal stance is the change in the cyclically adjusted primary budget position which represents

a change in policy orientation

The methodological problems of measuring cyclical and structural budget positions are substantial.

For details and discussion see:

https://www.bundesbank.de/Redaktion/EN/Downloads/Publications/Monthly_Report_Arti-

cles/1997/1997_04_budget_deficits.pdf?__blob=publicationFile

https://www.bundesbank.de/Redaktion/EN/Downloads/Publications/Monthly_Report_Articles/1997/1997_04_budget_deficits.pdf?__blob=publicationFile

https://www.bundesbank.de/Redaktion/EN/Downloads/Publications/Monthly_Report_Articles/1997/1997_04_budget_deficits.pdf?__blob=publicationFile


41 41

European Fiscal Union

Economic theory assigns three distinct functions to public finance: the stabilisation function,

the distribution function, and the allocation function. European Fiscal Union is about the

stabilisation of macroeconomic aggregates by an appropriate policy mix. Because effec-

tive demand is monetary purchasing power, the management of macroeconomic stabili-

sation must be centralized at the level of the monetary union. A transfer union is dealing

with the redistribution of income between rich and poor regions and income groups. Such

re-distribution requires a high degree of solidarity, which is often assumed to be absent in

the European Union. However, the budget of the European Union is already a small instru-

ment for making such transfers. Finally, the allocation function involves principles of fiscal

federalism, i.e. the allocation of responsibilities for making decisions on the provision of pub-

lic goods. When President Macron spoke of “the great challenges of our times, such as

defence and security, great migrations, development, climate change, the digital revolu-

tion and regulation of a globalized economy”, he effectively described European public

goods, which are at the moment allocated inefficiently by national governments and not

by a European authority.

In this chapter we deal with the flow of funds in a European fiscal union. In the next chapter

we will look at the European budget and conclude discussing some issues of fiscal federalism.

FISCAL POLICY COORDINATION IN THE EURO AREA

During the Euro crisis it became popular to ascribe the debt problems of the Southern coun-

tries by the accumulation of current account deficits. The logic behind this argument is that

excessive borrowing by governments (the sovereign debt crisis in Greece) or by private

investors (the real estate crisis in Spain and Ireland) had led to unsustainable “foreign” debt

levels. According to national income identities, these imbalances are reflected in current

account positions (see Eq. 8 below), and they add up to the external debt position of a

country. The European Commission even built its new Macroeconomic Imbalance Proce-

dure on this reasoning. However, the focus on current account deficits is misleading: first,

such deficits mean that the country is running out of foreign exchange and will ultimately

be forced to depreciate its currency. However, within a currency union there are no for-

eign exchange reserves and a country cannot run out of money. The money to make a

payment in another member state comes from the ECB and is distributed through the

banking system. The ECB steers money supply in line with its primary objective of preserving

price stability in the Euro Area as a whole, but whether debt is sustainable or not depends

on credit status of the debtor and not on the debtor’s nationality. Hence, member states

in the monetary union are not “foreign” countries, but regions of the currency union. Nev-

ertheless, within the Eurozone context debt financed fiscal expansions increase trade def-

icits (7), and this implies significant spill-overs to the private sector of other member states,

as we will see below.

(7) See (Beetsma, et al., 2008).

N. 1 - 2 0 1 8

42

Second, the current accounts of Euro member states mirror trade balances, which reflect

partly the flow of domestic euros, and partly the flow of external currency balances. For

example, Figure 21 shows that the surplus in Germany’s trade balance in the first period of

monetary union was caused by exports to neighbours in the European Union, but when

demand from EU neighbours collapsed under the austerity program after the global finan-

cial crisis, Germany shifted its exports towards the rest of the world. In Italy, on the other

side, the trade balance with the rest of the world – but not with the EU - turned negative in

2003. But this did not lead to a loss of foreign exchange reserves. In that period, being in

monetary union has softened the external constraints for the Italian economy.

Figure 21. Trade balance in Germany and Italy

Source: European Commission AMECO Database.

Third, in the Italian debate it is often argued that during the Euro crisis this country was less

vulnerable than other southern countries because public debt was largely held by Italians.

This interpretation is wrong. Home bias is a tendency by local investors and banks to keep

a considerable share of assets in domestic equities despite the well-known benefits from

diversifying into foreign equities. Local banks often hold debt issued by own country’s gov-

ernments, because they act as governments’ agents or simply because they know the

local habitat best. However, home bias generates regional financial market segmentation.

If a systemically relevant local debtor gets into trouble, especially when it is the govern-

ment, the risk of contagion is significantly higher than if portfolios are well-diversified. The

crisis has shown that insufficient integration of Europe’s financial markets has become a

source of increased financial vulnerability. Completing the Banking Union with a compre-

hensive deposit insurance would reduce it.

For these reasons, it is more important to analyse the flow of net lending and borrowing

between different sectors within the currency area rather than focus on current account

deficits. Such flow of fund analysis gives a more complete picture of the risks and opportu-

nities, of the gains and losses and potential imbalances that occur in a monetary union. It

will allow us to understand how the coordination of fiscal policy through a European fi-

nance minister could improve economic performance.

0

50

100

150

200

250

300

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

2016

2018

-40

-20

0

20

40

60

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

2016

2018

Extra EU Trade Balance Intra EU Trade Balance Total Trade Balance

Germany Italy


43 43

THE FLOW OF FUNDS ANALYSIS

In a monetary economy, goods are not exchanged but sold for money. This means for

every transfer of goods and services there is a counter-movement in the form of money

flows. The expenditure of the buyer is the income of the seller. The asset of the lender is the

liability of the borrower. Spending in excess of income requires borrowing financial surpluses

from other sources. A financial surplus is also called net lending, and a financial deficit net

borrowing (8). A financial surplus can be used either to build up financial assets or to repay

debt and deleverage balance sheets. Deleveraging may be desirable to reduce financial

risks, but in that case the money is not spent on goods and services so that it will lead to

recessions.

To analyse the flow of payments in the Euro Area, we distinguish between four institutional

sectors: households save and lend their money to corporations who borrow to invest. Gov-

ernments either save by running budget surpluses or borrow to finance deficits. Finally,

households, corporations and governments can borrow from or lend to the rest of the

world. The sum of all these net borrowings and lendings is always zero.

Flow of funds analysis reveals the flows of payments between economic sectors and re-

gions. We will first discuss the sectoral relations and then the regional implications. In a text

book setting, households save and lend to the corporate sector which invests these funds

and thereby generates income, while government budgets and the current account are

in balance. In reality, this is rarely the case. If the balance between household net savings

and the corporate sector’s net borrowing is positive (S – I > 0), this net lending of the private

sector must be absorbed by government net borrowing (T – G < 0) or by the rest of the

world (CA > 0), otherwise, a recession will occur. But if the private sector net balance is

negative (S – I < 0), the financing gap of the private sector must be closed by borrowing

from banks, the government (i.e. net tax reductions) or from abroad. If these excess savings

are not absorbed by borrowing, they become idle money and cause aggregate income

to shrink, thereby causing unemployment. On the other hand, if the demand for funds ex-

ceeds the sum of savings by households and by the government, foreign investors or banks

will have to close the gap (9). Yet, bank lending creates money supply (10), and if this

money creation is excessive it will create inflation. Hence, what determines the perfor-

mance of a monetary economy is the flow of net lending and net borrowing between

different sectors in the economy.

National income identities help us to identify the net flow of borrowing and lending. They

tell us that the current account balance (CA = X - M) is the sum of the net savings-invest-

ment balance (S-I) of the private sector and the government budget position (T-G):

Eq. 8 CA = (S – I) + (T- G)

(8) For simplicity we have aggregated Monetary and Financial Institutions and Non-financial corporations.

(9) The sum of private and public-sector saving is also called national saving:

Snat = S + (T-G) = (Y-T-C) + (T-G) = Y-(C+G)

(10) “Loans make deposits” is the formula by which modern monetary theory explains money creations.

See (McLeay, et al., 2014). In the old days, economists spoke about the government using the printing press.

N. 1 - 2 0 1 8

44

S stands here for household savings, I for investment, T for tax income, and G for govern-

ment spending. In a closed or balanced economy (CA = 0), national savings (i.e. the sum

of private household saving S and government saving (T-G) are always equal to invest-

ment). These are balance sheet identities, but the economic literature provides different

models for explaining the dynamics of net lending and borrowing between sectors and

the responses to fiscal policy impulses. We are particularly interested into the effect of fiscal

policy on the flow of funds within the monetary union.

A budget surplus (T > G) is government saving (net lending) and a budget deficit is net

public sector borrowing (see Box 4). Fiscal consolidation means the government lowers its

deficit or increases its surplus, which represents an increase in net government lending. A

fiscal expansion is the opposite. Standard macroeconomic theory focuses on the multiplier

effect of government budget positions on the private sector (11). However in monetary

union, the policy decisions by national governments and the reactions by the private sec-

tor have spillover effects into other member states, because if fiscal policy changes aggre-

gate demand in one country, consumers and investors from all countries will seek to exploit

the new opportunities (or avoid new risks). The spill-overs will then spill back into the original

member state, where it will affect government revenue and therefore the local fiscal

stance. Previous literature has already provided evidence of the large magnitude of those

spill-overs inside the Euro Area in terms of different measures of fiscal stance (revenues,

government consumption, debt-to-GDP, yield spread, etc.), as summarized in Box 5.

We distinguish five classical patterns for the interaction of fiscal policy with private sector

net lending.

1. In standard Keynesian models a fiscal consolidation (an increase in government sav-

ing) either will increase taxes, thereby lowering disposable income for consumption, or

will reduce public spending. Either way, the fiscal consolidation will depress consump-

tion and investment, so that private sector net lending (S-I) will increase. In this case

our estimated multiplier would be positive.

2. Crowding out occurs when the expansionary fiscal policy reduces investment, be-

cause the government’s demand for loanable funds pushes up interest rates. If the

productivity of private investment is higher than public spending, this will cause a slow-

down in economic growth. Crowding out is the limit for Keynesian policies, which

emerges when productive capacities are fully employed.

3. The opposite are so-called non-Keynesian effects when a fiscal consolidation reduces

the risk premium for unsustainable public debt and this stimulates investment and con-

sumption, so that a fiscal contraction will lead to an economic expansion. Here, the

fiscal multiplier is negative, as the lower deficit or higher budget surplus reduces the

risk premium for investors, so that the private surplus (S-I) shrinks. This is crowding in of

private investment by fiscal policy.

4. Ricardian Equivalence models argue that an increase in government savings trans-

lates one to one into a reduction in private savings, so that fiscal policy is neutral (an

(11) Frequently, the multiplier of fiscal policy is measured as the effect on GDP. We concentrate on the

saving or net lending in the private sector, which comes to the same. See (Giavazzi, et al., 1999) for a similar

approach.


45 45

increase in government savings generates a reduction in taxes, higher consumption

and investment and lower household savings). In this case, our estimated multiplier is

minus 1.

5. Finally, in the twin-deficit case, it is lending and borrowing from the rest of the world

that complement domestic funds. Hence, the current accounts are not necessarily

balanced, so that an increase in government savings (fiscal consolidation) will simul-

taneously decrease the current account deficit. However, this makes only sense if the

economy operates as in Keynesian models. In case of Ricardian Equivalence, there

should be no effect from fiscal policy to current accounts, and if public sector borrow-

ing crowds out private investment, the long-run effect would be zero, too. In the case

of non-Keynesian effects, a budget deficit reduction would even increase the current

account deficit.

So far, we have reasoned in terms of the two main institutional sectors (private and public)

of any economy. We will now add member states as geographical sectors of the Euro

Area. This renders the argument more complex but does not change the basic logic of

flows of funds. To show this logic while keeping things simple, we may assume that the Euro

Area as a whole has a balanced current account and is composed of two countries. We

can then decompose Eq. 9 into two sectors standing for region 1 and 2:

Eq. 9 CA= CA1 + CA2 = 0 => CA1 = - CA2

Eq. 10 (S1 – I1) + (T1- G1) = -(S2 – I2) - (T2- G2)

It is then obvious that the current account position, i.e. the flow of funds into one region, is

the mirror of the rest of the Euro area. If the savings in one region are insufficient to finance

local investment, savings from another region will have to provide the funds.

There are now two ways how this imbalance can be financed. First, in a properly function-

ing financial market (i.e. without home bias) the flow of lending and borrowing is uncon-

strained, so that household savings in one country will be lent to investors in other member

states – provided the risk-return conditions are attractive. For this reason, competitiveness,

as measured above, is a good indicator for where the flow of funds will go. Economic pol-

icies by national governments may improve these conditions, but if they fail and investment

is blocked by asymmetric investor expectations, it may be appropriate to unblock it at the

European level. This would be one of the functions of a European finance minister.

Second, it is often argued that the Euro Area does not function as a fully integrated cur-

rency union because it lacks fiscal transfers. A transfer union would imply that the budget

surplus of one government finances the deficit of another government:

Eq. 11 (T1- G1) = - (T2- G2) Transfer Union definition

Clearly, a transfer union is not a necessary requirement for transferring funds between

member states in monetary union, because the savings-investment balance in the private

sector can fulfil this function. In fact, in a monetary union with a single market and a fully

integrated banking union, the adjustment of shocks works primarily through the private sec-

tor channel and this explains the relative robustness of monetary unions, which fixed ex-

N. 1 - 2 0 1 8

46

change rate arrangements can never achieve. It is, however, true that the social conse-

quences of such adjustments can be harsh. In nation states fiscal policy serves as a redis-

tribution tool to cushion shocks, but in the European Union the use of fiscal policy for the

purpose of improving social welfare has not yet been accepted.

Nevertheless, fiscal policy in the European monetary union is not neutral, because the

changes of fiscal stances will generate spillover effects in neighbouring member states.

These effects will impact the balance of private net lending though changes in income,

consumption and investment, which spills back on neighbouring budget positions and ulti-

mately comes back to the original member state. Hence, even without a fiscal union, and

even without any form of fiscal coordination, national fiscal policies generate externalities

for all member states. These externalities are lower for countries outside the currency area,

because exchange rate adjustments inhibit the flow of funds.

How fiscal policy affects these different sectoral and regional flows of funds cannot be

decided on theoretical grounds, but an empirical matter. We have estimated the relevant

coefficients, as is common in the literature, in a global vector autoregressive model

(GVAR), which indicates the structure that holds the Euro Area together. Following

Dragomirescu-Gaina & Philippas (2015) we apply the GVAR estimator to a set of variables

composed of the net lending of the public and private sector, and we consider the

weighted trade flow variables as the main transmission channel, as trade has been shown

to be more relevant than financial flows or sovereign risk premia. The model has been esti-

mated on a sample of 15 European countries, among which 10 within the single currency

area and all of them belonging to the European Union. The period considered spans from

1999 to 2016 with quarterly frequency, covering different business cycle phases (i.e. expan-

sions and recessions) and period of large uncertainty from both the macroeconomic and

the policy points of view (i.e. financial crisis, sovereign crisis, etc).

Eq. 12 𝑃𝑟𝑖𝑣𝑖,𝑡

𝐺𝑜𝑣𝑖,𝑡= 𝛼0 + 𝛼1𝑡 + 𝜸𝟏

€𝑃𝑟𝑖𝑣𝑖,𝑡€ + 𝜸𝟐

€𝐺𝑜𝑣𝑖,𝑡€ + 𝜷𝟏

𝑃𝑟𝑖𝑣𝑖,𝑡−1

𝑃𝑟𝑖𝑣𝑖,𝑡−1€ + 𝜷𝟐

𝐺𝑜𝑣𝑖,𝑡−1

𝐺𝑜𝑣𝑖,𝑡−1€ =

= 𝛼0 + 𝛼1𝑡 +𝛾1,1

€

𝛾1,2€

𝑃𝑟𝑖𝑣𝑖,𝑡€ +

𝛾2,1€

𝛾2,2€

𝐺𝑜𝑣𝑖,𝑡€ + [

𝛽1,1 𝛽1,1€

𝛽1,2 𝛽1,2€

]𝑃𝑟𝑖𝑣𝑖,𝑡−1

𝑃𝑟𝑖𝑣𝑖,𝑡−1€ + [

𝛽2,1 𝛽2,1€

𝛽2,2 𝛽2,2€

]𝐺𝑜𝑣𝑖,𝑡−1

𝐺𝑜𝑣𝑖,𝑡−1€

𝑤𝑖𝑡ℎ 𝑃𝑟𝑖𝑣𝑖,𝑡€ = ∑(𝜔 𝑃𝑟𝑖𝑣𝑗,𝑡)

𝑗≠𝑖

𝑎𝑛𝑑 𝐺𝑜𝑣𝑖,𝑡€ = ∑(𝜔 𝐺𝑜𝑣𝑗,𝑡)

𝑗≠𝑖

Where 𝑃𝑟𝑖𝑣𝑖,𝑡 stands for the private sector net lending (S - I), and 𝐺𝑜𝑣𝑖,𝑡 for the budget posi-

tion (T-G) in country i ; while 𝑃𝑟𝑖𝑣𝑖,𝑡€ and 𝐺𝑜𝑣𝑖,𝑡

€ are the foreign counterparts. Those latter var-

iables are respectively the weighted averages of the other countries private and public

net lending, with the weights 𝜔 computed as the trade shares collecting for the main trans-

mission mechanism. For the purposes of our analysis of fiscal policy effects, we are inter-

ested in the parameters 𝜸𝟐€ and 𝜷𝟐. The first, 𝜸𝟐

€, measures the average contemporaneous

impact of a change in the fiscal stance in any member state on the private net lending

position 𝛾2,1€ or on the budget position 𝛾2,2

€ of other countries or in the Euro Area as a whole.

This elasticity measures the direct impact which does not take into account the intra-sec-

toral adjustments within a country. The coefficient 𝛾2,2€ then reflects the heterogeneity of

fiscal policy in the Euro Area. The second parameter, 𝜷𝟐€ , measures the effect of a change


47 47

in the aggregate fiscal stance of the Euro Area, due to a change in any other countries

fiscal stance, on private sector net lending 𝛽2,1€ and on budget positions 𝛽2,2

€ in the Euro

Area, in individual member states, and in the non-Euro member states. When the coeffi-

cient measures the effect of a given country’s budget position on its own private sector

net lending 𝛽2,1, it is a measure for the fiscal multiplier in the traditional sense of the literature.

If it measures the effect on another country’s private net lending 𝛽2,1€ , it approximates the

spillover effect.

Our estimates of the model identify a long-run relationship between private and public-

sector savings-investment balance (i.e. for almost all the countries one cointegration rela-

tionship exists). Hence, the effects of fiscal policy are very persistent and, although typically

smaller on impact, the responses are comparable in size between the 1-year and the 10-

year horizon (12). We then quantify the values of fiscal multipliers of investment and spill-

overs through the impulse response functions to a 1 percentage point reduction in the

public deficit-to-GDP ratio, both in the Euro area as a whole (common shock) and in the

single countries (idiosyncratic shock). Also note that the common shock is constructed as

a simultaneous fiscal restriction in all member countries so that aggregate European

budget consolidates for 1 point of European GDP. This replicates the operation of Stability

Pact which requires that fiscal policies of single countries move in the same direction, with

zero budget balance as a target. We are interested in detecting whether fiscal consolida-

tion is absorbed by crowding in or even non- Keynesian effects in the private sector or by

an increase in the current account, i.e. inducing a deflationary impulse abroad. The aver-

age values of the impulse responses for common and idiosyncratic shocks are reported in

Table 2, with the main diagonal representing the fiscal multipliers and the others coeffi-

cients measuring fiscal spillovers.

First, we observe that, in the common shock case, the impulse transmitted to private net

lending has negative and statistical significative signs in the member countries, while is near

null outside the Euro Area. This confirms that in the Euro Area fiscal consolidation generates

spill overs and possibly some crowding-in of private investment. However, European aggre-

gate fiscal multiplier reaches only 55 basis points, confirming that on average the common

fiscal consolidation generates deflationary effects, which translates into an improvement

of the European current account.

For single member countries, the evidence on spill-overs from the common shock is mixed.

In the medium run (after two years) the reported value of the response function is close to

1 in Germany, meaning that the private sector strongly reacts to aggregate fiscal consoli-

dation, absorbing much of the deflationary effect of the latter. And in Ireland the value is

higher than 1, possibly signaling the presence of a non- Keynesian mechanism. Spain, Por-

tugal, the Netherlands, Belgium, Finland and Greece have spill-overs value close to Euro

Area average multiplier, while France and Italy are the two economies for which the coun-

ter balancing response of private sector to aggregate fiscal consolidation is weaker. Only

20% of the original fiscal impulse is absorbed through a reduction in the private sector net

lending.

(12) See also Kempa & Khan (2017) and Belke & Osowski (2016).

N. 1 - 2 0 1 8

48

Table 2. Medium run multipliers and spillovers on private net lending as a response to a

shock in the government surplus (+1%)

Note: Values in Italics are not statistically significant different from zero at the 95% level; median estimates

of boostrapped Generalized IRFs, value for quarter 8 (2-years)

Source: CER elaborations on ECB Quarterly Sectoral Accounts.

We understand from such evidence why single member countries could have different

attitudes towards aggregate fiscal consolidation in the Euro Area and this should be a

strong argument for revising today rules of coordination. For better understanding this point,

we now turn to the domestic multipliers, considering how single countries’ private sector

reacts to an independent fiscal consolidation. Indeed, the difference in the fiscal multipliers

of investment after a common or an idiosyncratic shock are not as much homogenous as

one can expect looking at previous literature (see Box 5). On the contrary, own countries

multipliers are highly heterogenous across Euro Area countries, in the size and in the com-

parison with the common shock. We observe that in Germany, France, the Netherlands

Spain, Portugal, Ireland and Finland multipliers are higher than the spillover coefficients de-

rived from the common fiscal shock. It is thus interest of these countries to follow a domestic

consolidation strategy, that would stimulate a stronger crowding-in effect in the private

sector. In details, considering only the major countries, the balancing effect of private net

lending to a 1% fiscal shock rises from 0,2 to 0,8 in France, from 0,6 to 0,9 in Spain and from

0,9 to 1,1 in Germany. This also means that fiscal consolidation is absorbed to a lesser extent

through the current accounts, so that the negative spill-overs to the global European and

non-European demand are reduced. In the case of Germany (and also of Netherlands,

Portugal, Ireland and Finland) where the non- Keynesian mechanism would become prev-

alent, the ultimate effect is to narrow the current account, so that the rest of the world

seems to reap some benefits from domestic fiscal consolidation in these countries.

-0,06% 0,03% -0,03% -0,03% -0,15% -0,20% 0,05% 0,67% -0,03% 0,40% -0,02%

-0,55% -0,52% -0,32% 0,13% -0,14% -0,33% 0,16% 2,27% 0,88% 1,08% -0,02%

GERMANY -0,92% -1,08% -0,02% 0,19% 0,06% -0,28% 0,24% -5,77% 1,84% 1,36% 0,02%

FRANCE -0,21% -0,08% -0,83% 0,06% -0,02% -0,03% 0,03% -1,33% -1,68% 0,19% -0,01%

ITALY -0,21% -0,14% -0,06% -0,05% -0,09% -0,23% 0,01% -0,68% 0,70% 0,64% -0,06%

SPAIN -0,56% -0,26% -0,52% 0,40% -0,93% -0,31% 0,18% 0,54% 3,25% 1,92% -0,03%

NETHERL. -0,65% -0,54% -0,09% 0,06% -0,10% -1,41% 0,39% -9,73% 0,32% 1,92% -0,09%

BELGIUM -0,51% -0,51% -0,30% -0,01% -0,16% -0,62% 0,46% 0,49% 1,72% 1,86% 0,02%

IRELAND -1,69% -1,67% -0,92% 0,53% -0,38% -1,02% 0,09% 5,41% -0,60% 0,93% -0,11%

FINLAND -0,47% -0,49% -0,38% 0,14% 0,05% -0,20% 0,25% -6,17% 5,00% 0,59% -0,04%

PORTUGAL -0,43% -0,31% -0,40% 0,21% -0,35% -0,32% 0,15% -1,19% 0,44% 1,95% -0,03%

GREECE -0,64% -0,60% -0,27% 0,12% -0,11% -0,49% 0,22% -5,05% 1,32% 1,37% -0,04%

Rest of EU

Euro Area

EURO

AREABELGIUM IRELAND FINLAND PORTUGAL GREECE

Country

Response

Country

Impulse

GERMANY FRANCE ITALY SPAIN NETHERL.


49 49

In terms of policy strategy, our evidence suggests that general welfare gains can be ob-

tained if fiscal consolidation is pursued at single country level and not contemporaneously

by the Euro Area as a whole. This is not surprising because, with no exchange rates mech-

anism operating, deflationary effects cumulate in case of contemporaneous fiscal shocks.

A different view applies to Italy and Greece, where after an idiosyncratic fiscal shock we

observe multipliers which are closed to zero, much smaller than the responses after a com-

mon fiscal shock. Besides, independent fiscal consolidation in Italy translates into a large

increase in the current account surplus, that obviously comes at a cost for the other Euro

Area member countries. We read this evidence in face of the high level of the Italian (and

Greek) public debt. A fiscal consolidation at European level is conductive to a low interest

rates environment, that is strongly beneficial for the Italian budget, given that interest rate

expenditure amounts at nearly 5% of GDP (the highest level among major countries). Put it

in others words and given the past track record, Italian budget policy seems to face a

reputational problem, so that credibility premia can be extracted in the case of common

fiscal consolidation. Thus, a general orientation towards fiscal expansion could not be in

the interest of Italy, while the worst case remains the necessity to adopt an independent

budget consolidation.

All this considering, a new form of fiscal policies coordination within the Euro Area, defined

by a strategic interaction between member countries on the timing of consolidation,

would be preferable to just a common set of rules, which might induce all countries to

consolidate at the same moment, making austerity unsustainable. However, since the

cross-country heterogeneity of attitude towards a joint consolidation across country is a

matter of fact, supranational coordination by means of a European Minister of Finance

might mediate this position, increasing the sustainability of fiscal consolidation in Europe.

N. 1 - 2 0 1 8

50

BOX 5. THE CURRENT DEBATE ON FISCAL SPILL-OVERS IN THE EURO AREA

The measurement of fiscal multipliers has received large attention in the academic literature. After surveying the

vast literature on the topic, Ramey (2011) has concluded that fiscal multipliers of deficit-financed increases in

government spending lie between 0.8 and 1.5. During financial crisis, deep recessions or when the monetary policy

is constrained by the zero-lower bound of interest rate, the multiplier are larger and typically above one. However,

in “normal” times both government spending and tax cut multipliers are typically below one: by mean of the

National Institute Global Econometric Model (NiGEM), Carreras et al. (2016) quantify them for Euro Area countries

between 0.1 and 0.9 per cent depending on the instrument employed for fiscal consolidation.

A robust result is that fiscal multipliers depend on the country size and the degree of openness. Indeed, in more

open economies multipliers are smaller because much of the fiscal shock propagates to other economies via

trade linkages. The existence of international externalities implies, however, that fiscal multipliers will be larger if

fiscal policy is coordinated across countries, with this latter effect larger for more open economies, due to the

larger benefits of high imports in the trade partners, whereas smaller for countries more open to financial markets,

due to the impact on the long-run interest rates.

Besides several papers that analyse the international externalities of fiscal shocks by means of structural macro-

econometrics models, which however rely on strong behavioural assumptions, a large swath of empirical literature

has provided evidence of fiscal spillover relying on Global VAR. This large scale econometric estimator allows to

make inference on the effects of both domestic and foreign shocks, also fiscal, on domestic variables by means

of interrelated country VAR model.

Contrary to the results based on estimated theoretical models, empirical evaluations (see Hebous & Zimmerman

(2013) and Cavallo, Dallari, & Ribba (2018)) indicate that bilateral spillover tend to be small (one tenth of the

domestic effect) in the short run but after some years international fiscal multipliers are comparable in size with the

domestic multipliers (Eller, Martin, & Huber (2017) quantify them between two third and one half of the initial do-

mestic effect). Few contributes try to assess the spillover looking at alternative measure of fiscal policy that fiscal

balance, like debt-to-GDP ratio or disentangling between government consumption and receipts. According to

Belke & Osowski (2016) and Ricci-Risquete & Ramajo-Hernandez (2015) on impact the effects of tax variation are

similar to those of changes in government consumption, but it is sizably smaller in the following years. However, the

effect of increasing the debt-to-GDP tends to become permanent mainly in high-debt countries due to the long-

run effect on interest rates (Kempa & Khan (2017) and Belke & Osowski (2016)).

Most of this literature agree that the single currency amplifies the fiscal spillover in the Eurozone because empirical

evidence point out that spill-overs are on average larger among euro area countries than towards the other Eu-

ropean Union member states or the US. Furthermore, fiscal coordination amplifies fiscal shock to the extent that a

similar shock stemming from the monetary union as a whole, tends to give rise to larger fiscal spillover, although

less costly in terms of fiscal policy (Ricci-Risquete & Ramajo-Hernandez (2015) and Hebous & Zimmerman (2013)).

Supporting this findings, Ricci-Risquete & Ramajo-Hernandez (2015) and Dragomirescu-Gaina & Philippas (2015)

attempt to provide evidence on the extent of fiscal coordination in Europe. Looking at the elasticities on impact,

they point out a remarkable degree of synchronization among business cycles or private flow of funds but very

little for fiscal policy. The strongest positive co-movements between fiscal policy (either debt or deficit) are those

of highly indebted countries, whilst the impact elasticities of the public deficits of the other countries may eventu-

ally be negative. The same evidence applies also looking at financial variables like government bonds yields and

corporate yields (Caporale & Girardi (2013) and Nickel & Vansteenkiste (2013)). Furthermore, those investigations

show that if fiscal policy is to some extent coordinated, the high substitutability between domestic and foreign

private deficits to compensate external imbalances tend to vanish because partially compensated by the joint

movements in domestic and foreign public deficits. Furthermore, shocks to the private deficits cannot be easily

accommodated if the reaction of fiscal policy to foreign shock in the private sector is delayed and, hence, not

coordinated. This is particularly true for global shocks impacting all the countries of the currency union.


51 51

Towards a European Fiscal Union

In this last chapter, we will now look at the transfers channelled through the European

budget and then propose an alternative European Fiscal Union through a new form fiscal

coordination. Our analysis leads to a clear conclusion: the European economy would per-

form better with a clearly defined aggregate fiscal stance for the Euro Area that interacts

with monetary policy and an additional mechanism that, departing from today rules, also

considers country-specific developments. How this could be achieved is not obvious.

The purpose of fiscal policy is the provision with public goods. Public goods are defined by

their externalities, which occur when the production or consumption of a good or service

affects not only the producer or consumer, but also other people. Externalities are therefore

“extra” costs and benefits that arise when someone takes an action, but does not bear all

the cost (negative externality) or enjoys extra benefits (positive externality). Now, clearly

macroeconomic stability is an externality that affects all citizens living in the Euro Area; yet,

governments take policy decisions with respect to their own national constituencies. There

is a contradiction between the macroeconomic stabilisation function and the microeco-

nomic allocation of public goods. This issue is addressed by theories of fiscal federalism.

Fiscal federalism is a normative theory that articulates guidelines for how to match public

goods with the desires and preferences of people living in different geographic areas. It

responds to the question: who should pay for public goods? The answer is: in principle eve-

ryone who can enjoy their benefits. When the externalities can be clearly separated, na-

tional public goods must be funded from national governments’ budgets, while European

public goods ought to be paid for by revenue raised at the European level. With respect

to macroeconomic stability, this implies that the fiscal policy stance of the Euro Area ought

to be determined at the European aggregate level. Fiscal policy in the Euro Area is far from

this ideal norm. The experience over the last 30 years of fiscal decentralization all over the

world has shown that “ideal type” theories of fiscal federalism are rarely fully implemented

in practice. The balance between decentralisation and centralisation is a political deci-

sion.

The main reason for this discrepancy between norms and realities are political. National

governments are elected by national constituencies; they therefore service the prefer-

ences of their electorates. As seen in Box 2, this leads to heterogeneous preference curves

and suboptimal welfare. In practice, it means that governments resist making payments

for public goods that affect others, even if it has benefits for themselves. Consequently, the

budget of the European Union remains extremely limited and cannot play a significant

stabilizing function.

The lack of political solidarity also affects the traditional distribution function of public fi-

nances. Some regions are rich and could therefore contribute more to the financing of

N. 1 - 2 0 1 8

52

collective tasks; others are poor and have greater needs. However, when the cement of

cultural identity is weak, the richer do not feel obliged or even willing to pay for the poorer.

Hence, measures that contribute to building a European culture and identity have eco-

nomic value.

Yet, equal burden sharing could impose unsustainable burdens on the poorer regions. In

federations, the federal government provides the funds to close the gap between local

revenue and expenditure. In the United States, the federal government gives grants to the

states. The German cooperative model of federalism combines federal grants with re-

gional transfers between rich and poor states. In the European Union, wealthy states trans-

fer structural funds to poorer regions through the European Union budget, while the tradi-

tional own resource (TOR) are minimal with 14% of GDP.

THE EUROPEAN UNION BUDGET

The budget of the European Union pays for policies carried out at the European level. The

total budget amounted to € 136 bln in 2016, which is less than 1% of gross national income

(GNI) of the European Union. Of this amount, € 117.9 bln are spend inside the European

Union but only € 77.2 bln in the Euro Area. There are therefore good reasons for increasing

the budget for the monetary union, as President Macron has demanded.

As reported in Table 3, the two main expenditure items in the EU budget are cohesion and

agriculture. The title Smart and Inclusive Growth relates to competitiveness and the social

cohesion within the European Union and represents 41.2% of the budget. The largest

budget item is Sustainable Growth - Natural Resources which is agricultural policy rela-

belled. It represents 42.1% and less than 4‰ of GNI. Foreign policy stands at 7.5% of the

total budget and security and citizenship only at 2.3%. The administration of the European

Union costs 8.8% of the budget and less than 1‰ of gross national income in the Euro Area.

In a political federation, the budget of the federal government is funded by own resources,

usually taxes paid by citizens who have political representation through parliament. In the

European Union, these own resources amount to only 13.9% of total revenue, while 77.8%

are funded directly by national government budgets. Given that the two main expenditure

blocks are redistributing revenues for cohesion and agricultural purposes, is it fair to say that

the European Union is a transfer union. This raises the issue who is paying whom?

Figure 22 shows the absolute amounts of transfers in billions of euros. A positive figure indi-

cates an inflow, negative figure an outflow. Germany, France, the UK and Italy were the

largest net contributors, Poland, Romania and Greece the largest receivers of transfers.

Given that the four largest net payers are also the largest and richest member states in the

European Union, this is not surprising, although Germany pays nearly three times as much

as Italy. In aggregate, the Euro Area member states make net transfers to the non-Euro

Area countries.


53 53

Table 3. European Union Budget

Source: European Commission budget (2016).

Nevertheless, the absolute amounts say nothing about the burden and fairness of these

transfers. Table 4 indicates that transfers as a percentage of Gross National Income and

per person are highest for France, Belgium, Germany and Austria. Italy is one of the least

burdened contributors from the Euro Area. Spain is the largest receiver with transfers per

person that are four times as high as in Greece. Hungary and the Czech Republic are the

largest receivers outside the Euro Area. It is clear from the data that Brexit will make a large

gap in this system of redistribution.

In the debate on the re-foundation of the European Union many voices have asked for the

creation of new European public goods: defence and security, great migrations, develop-

ment, climate change, the digital revolution and regulation of a globalized economy, as

we mentioned above. These public goods would be beneficial for all European citizens

and they need additional funding. There is nothing wrong with paying taxes for desirable

public goods, but there is everything wrong with paying taxes and not getting the benefits.

Given that these public goods benefit all European citizens, they ought to be financed by

direct taxes on all citizens. Yet, the logic of a transfer union with a fractured constituency

2016

EXPENDITURE mln € % of total % of GNI

SMART AND INCLUSIVE GROWTH 56,265.0 41.2% 0.38%

COMPETITIVENESS FOR GROWTH AND JOBS 18,461.2 13.5% 0.12%

ECONOMIC, SOCIAL AND TERRITORIAL COHESION 37,803.8 27.7% 0.26%

SUSTAINABLE GROWTH: NATURAL RESOURCES 57,411.8 42.1% 0.39%

SECURITY AND CITIZENSHIP 3,077.3 2.3% 0.02%

GLOBAL EUROPE 10,277.1 7.5% 0.07%

ADMINISTRATION 9,324.2 6.8% 0.06%

TOTAL EXPENDITURE 136,416.4 100.0% 0.92%

REVENUE mln € % of total % of GNI

VAT-BASED OWN RESOURCE 15,895.1 11.0% 0.11%

GNI-BASED OWN RESOURCE 95,578.4 66.3% 0.65%

UK CORRECTION 626.1 0.4% 0.00%

REBATES -19.5 0.0% 0.00%

TOTAL NATIONAL CONTRIBUTION 112,080.2 77.8% 0.76%

TRADITIONAL OWN RESOURCES (TOR) 20,094.1 13.9% 0.14%

SUGAR LEVIES (100%) 165.8 0.1% 0.00%

CUSTOMS DUTIES (100%) 24,951.9 17.3% 0.17%

TOR COLLECTION COSTS -5,023.5 -3.5% -0.03%

TOTAL OWN RESOURCES 132,174.3 91.7% 0.89%

SURPLUS FROM PREVIOUS YEAR 10,565.8 7.3% 0.07%

OTHER REVENUE 1,349.1 0.9% 0.01%

TOTAL REVENUE 144,089.2 100.0% 0.97% GROSS NATIONAL INCOME (GNI) 14,790,681

Source: European Commission 2016 budget

N. 1 - 2 0 1 8

54

has taught us that intergovernmental cooperation does not produce optimal outcomes.

Hence, a larger budget with special focus on the Euro Area must increase the own re-

sources through taxes and reduce the share of GNI-contributions from national budgets.

Figure 22. Net Transfers between EU member states in 2016 (€ bln)

Source: European Commission budget and CER computations.

-14343

14343

-10988

-9216

-6272

-3207

-1534

-968

-946

-639

-424

-309

7

21

119

180

181

479

504

516

1143

1678

1717

1946

1986

3222

3581

4286

5962

6973

-20000 -15000 -10000 -5000 0 5000 10000 15000 20000

Euro Area

Non-Euro Area

Germany

France

United Kingdom

Italy

Belgium

Austria

Sweden

Denmark

Finland

Netherlands

Luxembourg

Cyprus

Malta

Slovenia

Ireland

Estonia

Latvia

Croatia

Lithuania

Spain

Portugal

Bulgaria

Slovakia

Czech Republic

Hungary

Greece

Romania

Poland


55 55

Table 4. Transfers through the European Union Budget

Source: European Commission budget (2016).

A EUROPEAN FISCAL UNION FOR MACROECONOMIC STABILITY

A larger budget for the European Union does not solve the issue of better coordination of

fiscal policy in order to improve the macroeconomic stability of the Euro Area. An efficient

European budget arrangement should provide vertical flexibility in order to deal with mac-

roeconomic shocks affecting the whole Euro Area, and horizontal flexibility that allows the

stabilisation of asymmetric shocks in specific countries. In addition, it should have a mech-

anism whereby the European budget reflects the preferences of European citizens for the

public goods they share, including their views on stabilisation, redistribution and solidarity.

Exp. Transfers Exp. Transfers Exp. Transfers

Mln € Mln € Rank % of GNI

% of GNI

Rank per

person per

person Rank

EU-28 117,875 0.00 0.80% 0.00% 230.5 0.0

Euro Area 77,180 -14,342 0.72% -0.13% 226.8 -42.1 Non - EA 40,695 14,342 1.01% 0.36% 237.9 83.9

Net Contributors

France 11,275 -9,216 -2 0.50% -0.41% -1 168.6 -137.8 -1

Belgium 7,333 -1,534 -4 1.74% -0.36% -2 649.2 -135.8 -2

Germany 10,082 -10,988 -1 0.32% -0.34% -3 122.2 -133.2 -3

Austria 1,939 -968 -5 0.56% -0.28% -4 222.0 -110.7 -4

Finland 1,530 -424 -6 0.71% -0.20% -5 278.6 -77.2 -5

Italy 11,592 -3,207 -3 0.69% -0.19% -6 191.2 -52.9 -6

Netherlands 2,289 -309 -7 0.33% -0.04% -7 134.4 -18.1 -7

Net Recipients

Luxembourg 1,788 7 12 5.08% 0.02% 12 3060.4 12.8 12

Cyprus 185 21 11 1.06% 0.12% 10 217.3 24.9 11

Spain 11,593 1,678 4 1.04% 0.15% 9 249.6 36.1 10

Ireland 2,038 181 8 0.93% 0.08% 11 435.0 38.7 9

Slovenia 545 180 9 1.39% 0.46% 8 263.9 87.3 8

Portugal 3,384 1,717 3 1.87% 0.95% 7 327.8 166.3 7

Latvia 734 504 6 2.92% 2.01% 5 374.7 257.4 6

Malta 208 119 10 2.22% 1.27% 6 474.3 271.5 5

Estonia 674 479 7 3.29% 2.34% 4 512.0 363.9 4

Slovakia 2,663 1,986 2 3.34% 2.49% 2 490.3 365.7 3

Greece 5,850 4,286 1 3.32% 2.43% 3 542.5 397.5 2

Lithuania 1,477 1,143 5 3.99% 3.09% 1 515.1 398.4 1

Net Contributors

Denmark 1,431 -639 -1 0.50% -0.22% -2 249.8 -111.5 -1

UK 7,052 -6,272 -3 0.30% -0.27% -1 107.5 -95.7 -2

Sweden 1,712 -946 -2 0.36% -0.20% -3 172.5 -95.4 -3

Net Recipients

Croatia 921 516 1 2.09% 1.17% 6 220.9 123.8 6

Poland 10,638 6,973 6 2.61% 1.71% 5 276.8 181.5 5

Bulgaria 2,345 1,946 2 5.00% 4.15% 1 329.0 273.1 4

Romania 7,360 5,962 5 4.47% 3.62% 2 373.5 302.6 3

Czech Rep. 4,690 3,222 3 2.86% 1.96% 4 443.9 304.9 2

Hungary 4,546 3,581 4 4.19% 3.30% 3 463.2 364.9 1

N. 1 - 2 0 1 8

56

Given, that the bulk of expenditure in the EU is allocated by national governments, a mech-

anism is needed to define the desired aggregate fiscal position (total public expenditure

minus revenue). This could be accomplished by first agreeing the aggregate fiscal stance

for the Euro Area in response to the prevailing macroeconomic conditions and then by

allocating borrowing rights to each member state.

The aggregate fiscal policy stance for the Euro Area should reflect the economic condi-

tions of the whole of European Monetary Union, but also collective preferences for the

allocation of resources, including their distribution between national and European public

goods. This would create a larger degree of fiscal flexibility, which is not possible with the

present rigid fiscal rules. However, once the aggregate deficit is defined at the European

level, each jurisdiction must be assigned a share of this total fiscal stance for implementa-

tion. Within their quota national governments would then set the priorities for collective

goods that reflect their voters’ preferences. For example, one country may prefer a large

public sector and therefore higher taxes, while another may opt for small government and

low taxes, but both must stick to the authorised net borrowing requirements.

Technically the procedure of first defining the macroeconomic aggregate and then its

micro application in a second step is well known in certain member states. For example,

the budget process in Italy defines first the multi-annual macroeconomic framework law,

the Documento di programmazione Economica e Finanziaria (DEF), and then the legge

finanziaria, which implements the actual budget allocations. Defining the aggregate Euro

Area fiscal stance would function like a European “DEF”. Because it affects all European

citizens, the responsibility for defining this aggregate Euro Area policy stance should be the

responsibility of the European Finance Minister (13).

Assigning quota for public borrowing to member states would depend on the size of coun-

tries, their economic situation, competitiveness conditions and the need to reduce public

debt. In accordance with the Treaty on European Union, defining such quota should be

the joint task of the European Finance Minister, the European Council and the European

Parliament.

The experience over the last two decades has proven that the voluntary cooperation

among the twenty-seven member states is no longer sufficient to generate and allocate

European public goods efficiently. This is undermining the legitimacy of the European Un-

ion. It is also becoming increasingly clear that in several member states government failure

and corruption prevent the fair and efficient employment of public funds. For this reason,

it is time to rethink the governance of European public goods. If failing governments are

part of the decision-making process and have veto rights, the policy output is unlikely to

achieve desirable results. Instead of intergovernmental cooperation, it may be a better

solution to provide certain public goods by a federal agency, which is acting on the basis

of competing legislation.

With these elements, a re-invented European Union would be able to keep its founding

promise: namely to preserve peace, increase welfare and bring people together.

(13) For further details of this proposal, see http://www.stefancollignon.de/PDF/OENBank.pdf

http://www.stefancollignon.de/PDF/OENBank.pdf


57 57

REFERENCES

Beetsma, R., Klaassen, F. & Giuliodori, M., 2008. The effects of public spending shocks on

trade balances and budget deficits in the European Union. Journal of the European

Economic Association, 6(2/3), pp. 414--423.

Belke, A. & Osowski, T., 2016. Measuring fiscal spillovers in the EMU and beyond: a global

VAR approach, s.l.: CEPS.

Caporale, G. M. & Girardi, A., 2013. Fiscal spillovers in the Euro Area. Journal of International

Money and Finance, 38((2013)), pp. 84.e1-84.e16.

Carreras, O., Kirby, S., Liadze, I. & Piggott, R., 2016. Fiscal policy spillovers, s.l.: National

Institute of Economic and Social Research.

Cavallo, A., Dallari, P. & Ribba, A., 2018. Fiscal policies in high debt Euro-Area countries.

s.l.:Springer.

Chudik, A. & Pesaran, H. M., 2015. Common correlated effects estimation of heterogenous

dynamic panel data models with weakly exogenous regressors. Journal of

Econometrics, 188(2), pp. 393-420.

Dees, S., Di Mauro, F., Pesaran, H. M. & Smith, V. L., 2007. Exploring the international linkages

of the euro area: a global VAR analysis. Journal of Applied Econometrics, 22(1), pp. 1--

38.

Dragomirescu-Gaina, C. & Philippas, D., 2015. Strategic interactions of fiscal policies in

Europe: a global VAR perpective. Journal of International Money and Finance, 59(2015),

pp. 49--76.

Eller, M., Martin, F. & Huber, F., 2017. How would a fiscal shock in Germany affect other

European countries? Evidence from a Bayesian GVAR model with sign restrictions. Focus

on European Economic Integration, Volume 1, pp. 54--57.

Fores, P., 2018. Monetary and Fiscal Policies Interaction in Monetary Unions. Journal of

Economic Surveys, February, 32(1), p. 226–248.

Furceri, D. & Zdzienicka, A., 2013. The Euro Area crisis: need for a supranational fiscal risk

sharing mechanism?, s.l.: International Monetary Fund.

Giavazzi, F., Jappelli, T. & Pagano, M., 1999. Searching for Non-Keynesian Effects of Fiscal

Policy CSEF (Centre For Studies In Economics And Finance) Working Paper no. 16

February 1999, Salerno: CSEF (Centre For Studies In Economics And Finance).

Hilbers, P., 2005. Interaction of Monetary and Fiscal Policies: Why Central Bankers Worry

about Government Budgets. In: Current Developments in Monetary and Financial Law.

Washington DC: IMF.

IMF, 2018. World Economic Outlook Update, Washington DC: International Monetary Fund.

Irish National Competitiveness Council, 2017. Ireland's Competitiveness Scorecard 2017,

Dublin: NCC.

Kempa, B. & Khan, N. S., 2017. Spillover effects of debt and growth in the euro area:

evidence from a GVAR model. International Review of Economics and Finance,

N. 1 - 2 0 1 8

58

49(2017), pp. 102--111.

McLeay, M., Amar, R. & Ryland, T., 2014. Money creation in the modern economy. Bank of

England Quarterly Bulletin , Volume Q1.

Nickel, C. & Vansteenkiste, I., 2013. The international spillover of fiscal spending on financial

variables. In: The GVAR handbook: Structure and applications of a macro model of the

global economy for policy analysis . s.l.:Oxford University Press, pp. 182--193.

Ramey, V. A., 2011. Can government purchase stimulate the economy?. Journal of

Economic Literature, 49(3), pp. 673--685.

Ricci-Risquete, A. & Ramajo-Hernandez, J., 2015. Macroeconomics effects of fiscal policy

in the European Union: a GVAR model. Empirical Economics, 48(2015), pp. 1587--1617.

Young, G., 2017. Commentary: Monetary and fiscal policy normalisation as Brexit is

negotiated. National Institute Economic Review , 242(November).

Zimmermann, T. & Hebous, S., 2013. Estimating the effects of coordinated fiscal actions in

the euro area. European Economic Review, 58(2013), pp. 110--121.

Rapporto CER

Re-inventing Europe

n.1 2018

rapcer 01-18 - stefancollignon.de · stefan collignon, gianluca cubadda, giuseppe de arcangelis,...

Documents