reassessing the fiscal monetary framework of emu in 2018 … · 2019. 6. 10. · reassessing the...

ReassessingtheFiscalMonetaryFrameworkofEMUin2018

RamonMarimonADEMUScientificCoordinator

EuropeanUniversityInstituteandUPF– BarcelonaGSE

The consortium

EuropeanUniversityInstituteUniversityofCambridgeUniversityofBonnToulouseSchoolofEconomicsUniversityCollegeLondonCatholicUniversityofPortugalBarcelonaGraduateSchoolofEconomics

http://ademu-project.eu

https://voxeu.org/The-emu-after-the-crisis

In the aftermath of the euro crisis In the aftermath of the euro crisis


The need to strengthen the resilience of the EA


The need to strengthen the resilience of the EA

Two proposals based on ADEMU research

Proposal I

A European Stability Fund

Based “On the Optimal Design of Financial Stability Fund” by

Arpad Abraham, Eva Carceles-Poveda Yan Liu

and Ramon Marimon∗

∗European University Institute; SUNY at Stony Brook, Wuhan University and European University

Institute, UPF - Barcelona GSE & CEPR.

Strengthening the EA: 4 related themes

I. Risk-sharing and stabilization policies in normal times.

II. Dealing with severe crises (“a robust crisis management

mechanism”).

III. Resolving a debt crisis (the euro ‘debt overhang’).

IV. Developing ‘safe assets’.

Strengthening the EA: Current Proposals


A “Stabilisation Function” (EC) & “Risk-sharing” once we

get more convergence (Presidents’ Reports)

II. Dealing with severe crises.

Transforming the ESM into a European Monetary Fund (EC)

Strengthening the EA: Our approach

Concentrate on


In itself, not an easy problem.

Strengthening the EA: Our approach

Interesting finding: Solving for a

I. Stability Fund as a ‘constrained efficient risk-sharing

mechanism’ also helps to:

II. deal with severe crises,

III. resolve a debt crisis, and

IV. develop ‘safe assets’.

A basic principle:

The EU is a long-term self-enforcing partnership!

A basic principle:

The EU is a long-term self-enforcing partnership,

but not a Federal State!

A basic principle:

The EU is a long-term self-enforcing partnership,

but not a Federal State!

The solution should not bea European IMF or a Central Fiscal Authority!

Designing a European Stability Fund

• The ECB to address the problem of time-inconsistency

in monetary policy (mainly: ‘competitive devaluations’).

• The ESF to address the problem of time-inconsistency

in fiscal policy (to follow pro-cyclical fiscal policies, i.e.

primary surplus/GDP not sufficiently procyclical).

Designing a Stability Fund – say, the ESF –

• Key element of The ESF long-term contract between the

ESF and an EU country: it is based on a risk-assessment of

the country.

• The ESF long-term contract defines state-contingent

transfers, in contrast to unconditional debt contracts.

• A ESF contract induces countercyclical fiscal (primary

deficit) policies (it is not a crisis resolution debt contract)

Designing a Stability Fund – say, the ESF –

• Contingency is ex-post, in contrast with ex-ante eligibility

conditions (e.g. ‘austerity program’conditions).

• Normal-times-transfers ‘build trust’, in contrast to crisis-

relief-transfers which tend to create ‘stigma & resentment’.

Designing the Fund accounting for 3+2 constraints:

The sovereignty constraint: The country is sovereign and

can always ‘exit’ (although it may be costly)

The redistribution constraint: risk-sharing transfers should

not become ex-post permanent (Hayek’s problem).

The moral hazard constraint: the severity of shocks may

depend on which policies and reforms are implemented.

Designing the Fund accounting for 3+2 constraints:

The asymmetry constraint: there is no ex-ante ‘veil of

ignorance’ and countries may start with large (debt)

liabilities.

The funding constraint: the fund should be (mostly) self-

funded (it can create a safe asset).

The ESF as a ‘constrained efficient

risk-sharing mechanism’ means

its contracts are optimal, subject to the 3+2 constraints!

.

The euro area ‘stressed countries’ with the ESF...

Comparing two alternative borrowing & lending regimes

Incomplete markets with default (IMD) and a risk-free

rate r: 1/(1 + r) ≥ β

• countries smooth shocks, and borrow and lend, with long-

term non-contingent debt;

• there can be default (full, in our case);

• default is costly and the country has no access to

international financial markets, temporarily;

• by exerting costly effort, they can reduce the probability of

adverse (government expenditure) shocks.

Comparing two alternative borrowing & lending regimes

Financial Stability Fund (Fund) as a risk-neutral agent

with discount 1/(1 + r) ≥ β.

• a country could leave the Fund at any time, in which case it

is like a country which defaults in an IMD regime;

• persistent transfers are limited by the amount of

redistribution that is mutually accepted;

• there are incentives for countries to apply policies which

reduce risks.

The ESF contract has the following 5 properties:

Consumption smoothing: consumption is less volatile and

less procyclical.

Countercyclical fiscal policies: primary surpluses are highly

procyclical.

Government bond spreads are very low (& negative): the

real spreads of ESF contracts (debts) are very low (& possibly

negative).

The ESF contract has the following 5 properties:

High capacity to absorb severe shocks (& existing debts):

in a severe shock (a rare event) a country with an ESF

contract disposes of a large line of credit.

An ESF contract for the 5 EA ‘stressed’ countries

(1980 - 2015)

1st Moments Data IMD Fund

Mean

Debt to GDP ratio 77.29% 78.35% 175.2%

Real bond spread 3.88% 3.61% −0.03%

G to GDP ratio 20.18% 19.34% 19.10%

Primary surplus to GDP ratio −0.78% 1.70% 3.79%

Fraction of working hours 36.74% 37.23% 38.10%

Maturity 5.38 5.38 5.38

2ndMoments Data IMD Fund

Volatility

σ(C)/σ(Y ) 1.49 1.46 0.33

σ(N)/σ(Y ) 0.92 0.70 0.62

σ(G)/σ(Y ) 0.91 0.96 0.51

σ(PS/Y )/σ(Y ) 0.65 0.81 0.94

σ(real spread) 1.53% 0.97% 0.01%

Correlation

ρ(C, Y ) 0.88 0.74 0.59

ρ(N, Y ) 0.67 −0.10 0.94

ρ(PS/Y, Y ) −0.29 0.13 0.95

ρ(G, Y ) 0.35 0.08 0.03

ρ(real spread, Y ) −0.35 −0.30 0.24

ρ(e, Y ) n.a. 0.00 -0.20

Comparing the economies in normal times

IMD vs. Fund Business Cycle Paths: shocks and allocations

0 50 100

0.1

0.2

0.3

0.4

Shocks

3

G(#3)

0 50 100

0.1

0.15

0.2

y

0 50 100

0.1

0.12

0.14

0.16

c

0 50 100

0.3

0.35

0.4

n

FSF-MH

IMD-MH

IMD vs. Fund Business Cycle Paths: shocks and assets

0 50 1000.15

0.2

0.25

0.3

0.35

0.4

e

0 50 100

-40

-20

0

20

==y (%)

0 50 100

-200

-150

-100

-50

0

a0=y (%)

0 50 100

0

2

4

6

rf ! r (%)

FSF-MH

IMD-MH

Comparing the economies in times of crisis

IMD vs. Fund: combined shock impulse-responses: allocations

0 50 100

0.1

0.15

0.2

0.25

0.3

Shocks

3

G(#3)

0 50 1000.12

0.13

0.14

0.15

0.16

0.17

y

0 50 100

0.11

0.12

0.13

c

0 50 1000.3

0.32

0.34

0.36

0.38

n

FSF-MH

IMD-MH

IMD vs. Fund: combined shock impulse-responses: assets and effort

0 50 100

0.3

0.32

0.34

0.36

0.38e

0 50 100-40

-30

-20

-10

0

==y (%)

0 50 100

-150

-100

-50

a0=y (%)

0 50 100

0

0.01

0.02

0.03

0.04

rf ! r (%)

FSF-MH

IMD-MH

Welfare gains and absorbing capacity

Welfare gains and absorbing capacity

Shocks (θ,Gc) Welfare Gain (b′/y)max: M (b′/y)max: F

(θl, Gh) = (0.148, 0.038) 7.37 1.71 78.02

(θm, Gh) = (0.299, 0.038) 6.35 107.61 170.05

(θh, Gh) = (0.456, 0.038) 4.32 215.15 318.32

(θl, Gl) = (0.148, 0.025) 6.51 1.84 78.81

(θm, Gl) = (0.299, 0.025) 5.90 111.47 170.23

(θh, Gl) = (0.456, 0.025) 4.12 214.78 316.75

Average 5.84

• Welfare gains are expressed in consumption equivalent terms at b = 0 (%).

• bmax is the maximum level of country indebtedness expressed as the percentage of

GDP in a given financial environment (Markets or Fund).

Strengthening the EA: as we have seen...

Solving for a

I. Stability Fund as a ‘constrained efficient risk-sharing

mechanism’ also helps to:

II. deal with severe crises,

III. resolve a debt crisis, and

IV. develop ‘safe assets’.

Implementing the ESF

Light version

• ESM implements ESF contingent contract on crisis resolution

contracts (existent and new)

• ESF conditionality becomes ex-post

• ESF is allowed to provide risk-sharing contracts (e.g.

countries that may fail SGP conditions)

• (The Macroeconomic Imbalance Procedure with an

opportunity instead of a non-credible threat!)

Implementing the ESF

Full version

• Full risk-sharing, stabilization capacity integrated to its crisis-

resolution capacity

• ESF contracts become ‘safe assets’ in the ESF balance sheet

• ESF becomes a proper Fund that can accommodate other

EMU needs (Backstop for SRM, EUIS Fund,...?)

• ESF allowed to issue a safe bond backed by its safe assets.

Proposal II

A European Unemployment InsuranceSystem (EUIS)

Arpad Abraham Joao Brogueira de SousaRamon Marimon Lukas Mayr

European University Institute

A European Unemployment Insurance System?

• Several policy proposals then and now:◦ �e Marjolin Report (1975),◦ Dullien (2014), Beblavy et al. (2015), Beblavy and

Lenaerts (2017), Dullien et al. (2018).

• High unemployment + low de�cit requirements: costlydelivery of unemployment insurance during recessions.

• Business cycles are not perfectly correlated across EUcountries: risk sharing.

◦ Social and political cohesion, labour marketintegration, worker mobility and EU identity.

• Large di�erences in unemployment rates and labourmarket �ows: a system of transfers across countries? .

A European Unemployment Insurance System?

• Several policy proposals then and now:◦ �e Marjolin Report (1975),◦ Dullien (2014), Beblavy et al. (2015), Beblavy and

Lenaerts (2017), Dullien et al. (2018).

• High unemployment + low de�cit requirements: costlydelivery of unemployment insurance during recessions.

• Business cycles are not perfectly correlated across EUcountries: risk sharing.◦ Social and political cohesion, labour market

integration, worker mobility and EU identity.

• Large di�erences in unemployment rates and labourmarket �ows: a system of transfers across countries? .

Unemployment in Europe (2010Q1-2017Q3)

�ree questions

• What could be the (net) bene�ts of a EUIS?

• Can all the EU countries involved bene�t from a commonand simple change to their current UI systems?

• Can unanimous agreement for this change be achievedwhen permanent transfers across countries are avoided?

? �e answers depend on how the EUIS is designed.

A step back

We �rst develop a �exible quantitative dynamic modelto

? Describe the observed heterogeneity of labour markets inthe EU (job creation and destruction, search costs, and av.wage di�erentials).

? Evaluate the impact of di�erent types of EUIS’s.

? European UI systems are de�ned by replacement rateand duration of unemployment bene�ts and, usually, are�nanced by payroll taxes

Map of EU Labour Markets

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18

Job finding rate: Non-Active

0

0.05

0.1

0.15

0.2

0.25

Jo

b f

ind

ing

ra

te:

Un

em

plo

ye

d

Labour Market Institutions (model)Austria

Belgium Germany

EstoniaSpain

Finland

FranceGreece

IrelandItaly

LuxemburgLatvia

Netherlands

Portugal

SloveniaSlovakia

Map of EU Labor Markets

0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2

Average Job finding rate: Non-Employed

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

Jo

b S

ep

ara

tio

n r

ate

: E

mp

loye

d

Labour Market Institutions (model)

Austria

Belgium

Germany

Estonia

Spain

Finland

France

Greece

IrelandItaly

Luxemburg

Latvia

Netherlands

Portugal

Slovenia

Slovakia

Map of EU national UB policies

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Replacement rate

0

1

2

3

4

5

6

7

8

9

10

Ma

xim

um

du

ratio

n,

qu

art

ers

Unemployment benefit policy (data)

Germany

Austria

Estonia

SpainFinland

France

GreeceIreland

Italy

Luxemburg

Latvia

Netherlands

Portugal

SloveniaSlovakia

How can there be a unique recipe?

A ‘rainy day’ EUIS for large asymmetricnegative shocks?

Consider the following scenario:• Member states keep the existing national UB systems.

• EUIS provides support when large negative shockincreases unemployment.• A Member State enters unemployment recession alone

(the rest of the Union not a�ected).• Two scenarios:

◦ Without a EUIS: the country facing theunemployment crisis must balance the budget;◦ Under the EUIS: government receives a transfer

equivalent to the increase in national UB expendituresduring the recession.


Consider the following scenario:• Member states keep the existing national UB systems.• EUIS provides support when large negative shock

increases unemployment.

• A Member State enters unemployment recession alone(the rest of the Union not a�ected).• Two scenarios:





increases unemployment.• A Member State enters unemployment recession alone

(the rest of the Union not a�ected).

• Two scenarios:







◦ Without a EUIS: the country facing theunemployment crisis must balance the budget;

◦ Under the EUIS: government receives a transferequivalent to the increase in national UB expendituresduring the recession.

Labor market states in France during recession

0 5 10 15 20 25 30 35 4080

85

90E

mplo

yed in %

0 5 10 15 20 25 30 35 405

10

15

Unem

plo

yed in %

0 5 10 15 20 25 30 35 40

quarters

0

5

10

Inactive in %

France’s taxes during a recession, without andwith an EUIS

0 5 10 15 20 25 30 35 40

quarters

1

1.5

2

2.5

3

3.5

4

in %

Autarky

Union

�e bene�ts of sharing risk through an EUIS?

Employed Un. Eligible. Un. Non-Elig. Non-Active TotalAustria 0.05 0.04 0.04 0.03 0.04Belgium 0.18 0.14 0.17 0.10 0.15Germany 0.02 0.01 0.01 0.00 0.02

Spain 0.26 0.18 0.20 0.09 0.21Finland 0.27 0.20 0.27 0.16 0.24France 0.10 0.06 0.06 0.02 0.09Ireland 0.11 0.06 0.08 0.04 0.09

Italy 0.01 0.01 0.01 0.01 0.01Luxembourg 0.02 0.02 0.02 0.01 0.02Netherlands 0.04 0.02 0.02 0.01 0.04

Table: EUIS for large shocks: welfare gains in CEV% (Cons. Equiv. Value)

10 EA countries’ taxes during a recession

0 20 40 60 80 100

quarters

0

1

2

3

4

5

6

7

8

in %

Austria

Belgium

Germany

Spain

Finland

0 20 40 60 80 100

quarters

0

1

2

3

4

5

6

7

8

in %

France

Ireland

Italy

Luxemburg

Netherlands

10 EA countries’ taxes during a recessionwith EUIS insurance

What could be the bene�ts of sharing riskthrough an EUIS?

Employed Un. Eligible. Un. Non-Elig. Non-Active TotalAustria 0.05 0.04 0.04 0.03 0.04Belgium 0.18 0.14 0.17 0.10 0.15Germany 0.02 0.01 0.01 0.00 0.02

Spain 0.26 0.18 0.20 0.09 0.21Finland 0.27 0.20 0.27 0.16 0.24France 0.10 0.06 0.06 0.02 0.09Ireland 0.11 0.06 0.08 0.04 0.09

Italy 0.01 0.01 0.01 0.01 0.01Luxembourg 0.02 0.02 0.02 0.01 0.02Netherlands 0.04 0.02 0.02 0.01 0.04

Table: EUIS for large shocks: welfare gains in CEV% (Cons. Equiv. Value)

What role for an EUIS?

• Welfare gains of insuring large shocks are small, butpositive for all Members.

• An interesting �nding:

◦ All EA UB systems de�ned by: the replacement rateand the duration of eligibility.

• What would be the optimal UB policy for each Memberstate, given its existing labour market institutions?

◦ Assumption: small open economies⇒ constantprices

• Remarkably similar across countries:

◦ UB Replacement rate 20%-45% and very long durationof eligibility.



• An interesting �nding:◦ All EA UB systems de�ned by: the replacement rate

and the duration of eligibility.

• What would be the optimal UB policy for each Memberstate, given its existing labour market institutions?

◦ Assumption: small open economies⇒ constantprices







• What would be the optimal UB policy for each Memberstate, given its existing labour market institutions?◦ Assumption: small open economies⇒ constant

prices







• What would be the optimal UB policy for each Memberstate, given its existing labour market institutions?◦ Assumption: small open economies⇒ constant

prices

• Remarkably similar across countries:◦ UB Replacement rate 20%-45% and very long duration

of eligibility.

Why a common UB recipe?

• National governments do not internalize generalequilibrium e�ects of their reforms on citizens in othercountries.

• Finding: “Optimal” reforms are welfare worsening inseveral countries once General Equilibrium e�ects aretaken into account.

• Mechanism: Too generous bene�t policies⇒ Privatesavings ↓ ⇒ European capital stock ↓ ⇒ wages ↓

⇒ Lack of coordination across countries may result indetrimental reforms.

A fully harmonized EUIS?

• Replacement rate (15%) and unlimited duration of bene�ts.

• Financed with a common EUIS wage tax.

Employed Unemp. Inactive Total Tax rate TransferAustria -0.31 -0.05 -0.28 -0.29 1.47 -0.51Belgium 0.36 -1.01 0.18 0.25 1.47 -0.71Germany 0.16 0.79 0.14 0.20 1.47 -0.14Spain 1.51 1.42 1.31 1.45 1.47 0.74Finland 1.11 0.63 0.92 1.04 1.47 -0.87France -0.06 -0.02 -0.08 -0.06 1.47 -0.16Ireland 0.66 1.07 0.63 0.68 1.47 -0.09Italy 0.75 2.13 0.76 0.87 1.47 0.29Luxemb. -0.43 -0.20 -0.34 -0.40 1.47 -0.64Netherlds. 0.14 0.55 0.12 0.17 1.47 0.07

Table: Welfare gains (in % CEV), tax rate (in %) and transfers (in % of GDP)

Can all countries bene�t from an EUIS whencross-country transfers are neutralized?

• Common replacement rate (15%) and unlimited duration of bene�ts.

• Country-speci�c tax rates

Employed Unemp. Inactive Total Tax Rate Cur. TaxAustria 0.21 0.51 0.24 0.23 0.73 0.99Belgium 1.05 -0.24 0.81 0.92 0.45 2.14Germany 0.34 0.44 0.28 0.38 1.27 0.45Spain 0.65 0.65 0.58 0.62 2.53 4.33Finland 2.09 1.83 1.90 2.03 0.22 3.75France 0.11 0.18 0.09 0.11 1.23 1.90Ireland 0.77 1.31 0.73 0.79 1.34 1.97Italy 0.48 1.58 0.51 0.60 1.90 0.25Luxembourg 0.30 0.51 0.33 0.32 0.55 0.53Netherlands 0.06 0.13 0.03 0.08 1.57 1.01

Table: Welfare gains (in % CEV) and tax rates (in %)

Can all countries bene�t from an EUIS whencross-country transfers are neutralized?

• Common replacement rate (15%) and unlimited duration of bene�ts.• Country-speci�c tax rates

Employed Unemp. Inactive Total Tax Rate Cur. TaxAustria 0.21 0.51 0.24 0.23 0.73 0.99Belgium 1.05 -0.24 0.81 0.92 0.45 2.14Germany 0.34 0.44 0.28 0.38 1.27 0.45Spain 0.65 0.65 0.58 0.62 2.53 4.33Finland 2.09 1.83 1.90 2.03 0.22 3.75France 0.11 0.18 0.09 0.11 1.23 1.90Ireland 0.77 1.31 0.73 0.79 1.34 1.97Italy 0.48 1.58 0.51 0.60 1.90 0.25Luxembourg 0.30 0.51 0.33 0.32 0.55 0.53Netherlands 0.06 0.13 0.03 0.08 1.57 1.01

Table: Welfare gains (in % CEV) and tax rates (in %)

Wrapping up on EUIS

• With an EUIS for large asymmetric negative shocks

◦ Risk Sharing bene�ts are small and bene�t the employed more ifthere are no UB cuts without EUIS

• Substantial gains to be made withan EUIS with a common coverage. How?

◦ With national systems, if National EUIS Funds have enoughborrowing capacity�

◦ With a centralized EUIS Fund absorbing deviations fromsteady-state beyond certain threshold (|Ui − Ui| ≥ tr; Ui =steady-state unem. rate of country i; tr = 0→ full centralization)

◦ With central funding there is a need for monitoring, periodic riskassessments, and a proper contract between each participatingcountry and the EUIS Fund.

Wrapping up on EUIS

• With an EUIS for large asymmetric negative shocks◦ Risk Sharing bene�ts are small and bene�t the employed more if

there are no UB cuts without EUIS

• Substantial gains to be made withan EUIS with a common coverage. How?

◦ With national systems, if National EUIS Funds have enoughborrowing capacity�



Wrapping up on EUIS



• Substantial gains to be made withan EUIS with a common coverage. How?◦ With national systems, if National EUIS Funds have enough

borrowing capacity�



Wrapping up on EUIS



• Substantial gains to be made withan EUIS with a common coverage. How?◦ With national systems, if National EUIS Funds have enough

borrowing capacity�◦ With a centralized EUIS Fund absorbing deviations from

steady-state beyond certain threshold (|Ui − Ui| ≥ tr; Ui =steady-state unem. rate of country i; tr = 0→ full centralization)


Wrapping up on EUIS

• With an EUIS for large asymmetric negative shocks

◦ Risk Sharing bene�ts are small, although in times of crisis evensmall gains count!

• Substantial gains with common EUIS◦ 15 % replacement rate◦ unlimited duration

• “Experience rating taxes” provide incentives for higher taxcountries to improve their labour markets.

• An EUIS can also enhance labour cohesion and mobility acrossEU participating countries, and social identity with the EU!

Wrapping up on EUIS

• With an EUIS for large asymmetric negative shocks◦ Risk Sharing bene�ts are small, although in times of crisis even

small gains count!

• Substantial gains with common EUIS◦ 15 % replacement rate◦ unlimited duration

• “Experience rating taxes” provide incentives for higher taxcountries to improve their labour markets.

• An EUIS can also enhance labour cohesion and mobility acrossEU participating countries, and social identity with the EU!

Thanks

reassessing the fiscal monetary framework of emu in 2018 … · 2019. 6. 10. · reassessing the...

Documents