University of Verona A regional version of Mirage model and a trade policy simulation PhD Student Gabriele Standardi Doctorate in Economics & Finance XXI.

University of Verona

A regional version of Mirage model and a trade policy

simulation

PhD Student Gabriele Standardi

Doctorate in Economics & Finance

XXI Cycle

September 8 2008

• Baseline of Mirage model

• Regional Model

• Trade policy simulation

Mirage model

Database

Mirage stands for Modelling International Relationship in Applied General Equilibrium

Mirage is a multi-region, multi-sector computable general equilibrium model, developed by CEPII (Centre d’études prospectives et d’informations internationales) and devoted to trade policy analysis

Mirage uses the GTAP 6.x databaseSAM for 87 countries or regions and 57 sectors with 5 production factors (capital, skilled labour , unskilled labour, land and natural resources)

Trade barriers are described by MacMap database, (developed by CEPII) « ad valorem » tariffs at the bilateral level for 137 countries 

Mirage Model

Mirage Structure

• The demand side

• The supply side

• Imperfect competition

• Capital, investment and macroeconomic closure

• Labour market

• Dynamic set-up

The demand side

Each region has a representative agent, whose utility function is static

Saving rate is exogenous

Total demand is made up of final consumption, intermediate consumption and capital goods

Product differentiation according to geographical origin (the so-called Armington assumption)

The regional representative agent includes the government, he therefore pays and earns taxes, so the public budget constraint is implicit to meeting the representative agent’s budget constraints.

Mirage structure

The supply side

5 production factors:skilled labour (H), unskilled labour (L), capital (K), land (TE), natural resources (RN)

Factor endowments are assumed to be fully employed Their growth rates are exogenous for natural resources (set at zero) and for labour based on demographic forecast provided by World Bank

Land supply: CET function that takes account of land’s imperfect mobility across uses

Capital supply : putty-clay hypothesis

Unskilled labour supply and skilled labour supply are perfectly mobile across sectors

Production : Leontief function between the value added (VA) and the intermediate consumption (Cnter)Value added : CES function of land, natural resources, unskilled labour and a CES bundle of capital and skilled labour (fictive factor Q)

Mirage structure

Imperfect competition

Horizontal differentiation of products and increasing returns to scale

Firms compete in a Cournot-Nash way

Mirage structure

Mirage structure

Capital, investment and macroeconomic closure

In many models , among which the GTAP one, international financial flows result from the assumption of perfect capital mobility across countries and sectors. This modelling is micro-founded, but it induces unplausibly high cross-border flows. On the other hand, directly using the results of econometric estimates would give more realistic results, but it would lack theoretical consistency.

So installed capital is assumed immobile across sectors and regions (putty-clay hypothesis) and the investment become the only adjustment device for capital stock. The investment can vary across sectors and regions. This modelling is a good compromise between theoretical consistency and empirical realism. The domestic investment’s setting is consistent with FDI’s one and it depends on FDI determinants, such as market size, growth rate or market potential, that influence the rate of return to capital.

Mirage structure

Labour market

Introduction of the Lewis hypothesis to take account of unskilled workers flows between urban areas and rural areas: useful to study the employment dynamic in some developing countries

Mirage structure

Dynamic set-up

The Mirage dynamic set-up takes place through the investment, the rate growth of production factors and the technological progress (exogenous).

Regional Model

Objective: analysing trade policy effects on the different regions of Europe

Regional Model Structure

• Micro-regions and macro-regions

• Sectors

• Demand side

• Supply side

• Imperfect competition

• Capital, investment, macroeconomic closure and dynamic set-up

• Labour market

Regional Model

Regional Model Structure

Macro-regional levelThree macro-regions  EU15 France, Germany, Belgium, Luxembourg, Netherlands, United Kingdom, Italy, Ireland, Denmark, Spain, Portugal, Greece, Finland, Sweden, AustriaNMB rest of EuropeROW rest of the world

Micro-regional levelNine micro-regions

EU15 is made up of 5 micro-regions:DEU Germany, Austria FRA France, Belgium, Luxembourg, NetherlandSCD Denmark, Finland, SwedenENG United Kingdom, IrelandMED Italy, Spain, Portugal, Greece

ROW is made up of  3 micro-regions:OEC Australia, New Zealand, Japan, South Korea, Canada, Mexico, Turkey, Switzerland, Rest of EFTAUSAROW Rest of macro-region ROW

NMB remains the same set: NMB Bulgaria, Cyprus, Czech Republic, Hungary, Malta, Poland, Romania, Slovakia, Estonia, Lithuania, Latvia, Slovenia

USAROW (Reste du macro région ROW)  EU15 est divise en :DEU Allemagne, Autriche FRA France, Belgique, Luxembourg, Pays BasSCD Danemark, Finlande, SuèdeENG Royaume Uni, IrlandeMED Italie, Espagne, Portugal, Grèce NMB reste le même ensemble : NMB (Bulgarie, Cipre, République Tchèque, Hongrie, Malte, Pologne, Roumanie, Slovaquie, Estonie, Lituanie, Lettonie, Slovénie).

Regional Model Structure

4 Sectors AGM AgriculturePRM Primary products IND Manufactures SERV Services

USAROW (Reste du macro région ROW)  EU15 est divise en :DEU Allemagne, Autriche FRA France, Belgique, Luxembourg, Pays BasSCD Danemark, Finlande, SuèdeENG Royaume Uni, IrlandeMED Italie, Espagne, Portugal, Grèce NMB reste le même ensemble : NMB (Bulgarie, Cipre, République Tchèque, Hongrie, Malte, Pologne, Roumanie, Slovaquie, Estonie, Lituanie, Lettonie, Slovénie).

Regional Model Structure

Demand

The structure of demand side remains exactly the same, defined at macro-regional level

Regional Model Structure

Supply side

The most important change with respect to Mirage: production is specified at micro-regional level

5 production factors:skilled labour (H), unskilled labour (L), capital (K), land (TE), natural resources (RN)

Factor endowments are assumed to be fully employed

Simplification with respect to Mirage : land supply is exogenous (there is indeed only one agricultural sector, so the CET function for land is useless)

The supply of factors is at micro-regional level except capital supply that is at macro-regional level: capital is perfectly mobile across the different micro-regions of the same macro-region

All the factors are perfectly mobile across sectors except natural resources

Regional Model Structure

Imperfect competition

Removal of the imperfect competition hypothesis

Regional Model Structure

Capital, investment, macroeconomic closure and dynamic set-up Removal of putty-clay hypothesis : the installed capital in each sector and in each micro-region is not immobile, but determined by the FOCs for profit maximization with respect to capital

No dynamic

Regional Model Structure

Labour market

Removal of Lewis hypothesis

Introduction of the hypothesis of unskilled labour imperfect mobility across the micro-regions of the same macro-region on the basis of relative wages (implicit emigration cost)

Unskilled labour supply: macro-regional wage income’s maximisation subject to a CET constraint

Trade policy simulation

Shock definition

Bilateral shock: Eu15 and the rest of Europe (NMB) suppress their tariff barriers in all the sectors toward the rest of the world (ROW) and the rest of the world does the same thing toward EU15 and the rest of Europe (NMB)

Trade policy simulation

ROW EU15 NMBAGM ROW 0,1473329AGM EU15 0,1036747AGM NMB 0,0495082 0,0610791PRM ROW 0,0164278PRM EU15PRM NMB 0,0087803IND ROW 0,0510492IND EU15 0,0338301IND NMB 0,0076003 0,0396776

ROW EU15 NMBAGM ROW 0,1473329 0,0526993 0,0565393AGM EU15 0,1069915 0,1036747AGM NMB 0,1268727 0,0495082 0,0610791PRM ROW 0,0164278 0,001957PRM EU15PRM NMB 0,0030439 0,0087803IND ROW 0,0510492 0,0282547 0,0685235IND EU15 0,0609952 0,0338301IND NMB 0,0696799 0,0076003 0,0396776

Tariff barriers before the shock

Tariff barriers after the shock

Trade policy simulation

ROW USA OEC totAGM 0,15% 0,38% 0,24% 0,20%PRM -0,04% -0,17% 0,09% -0,04%IND -0,53% -5,55% 5,52% -0,17%SERV 0,10% 1,09% -1,72% 0,10%

FRA DEU SCD ENG MED totAGM 0,03% 0,17% -0,28% 0,18% -0,05% 0,02%PRM -4,08% -3,44% -3,64% -3,57% -3,03% -3,61%IND 7,31% -3,58% 3,51% -7,55% 7,32% 1,39%

SERV -2,59% 1,96% -1,20% 2,35% -2,42% -0,26%

NMB totAGM 0,10% 0,10%PRM -2,40% -2,40%IND 0,61% 0,61%

SERV -0,35% -0,35%

Production with unskilled labour immobility ( = 0)

Trade policy simulation

Production value at the reference year (2001) for each micro-region and for each sector with respect to factor and intermediate consumption

ROW NMB USA FRA DEU OEC SCD ENG MEDL 18,13% 13,45% 24,92% 16,27% 15,33% 22,96% 20,60% 19,78% 16,27%K 26,17% 24,20% 17,33% 29,15% 29,31% 24,47% 18,91% 20,87% 31,39%H 11,40% 9,06% 20,22% 13,63% 11,90% 14,51% 17,83% 16,29% 14,28%

Cnter 44,30% 53,29% 37,53% 40,95% 43,46% 38,07% 42,66% 43,06% 38,05%

Services

Manufactures

ROW NMB USA FRA DEU OEC SCD ENG MEDL 9,36% 10,61% 15,66% 12,86% 16,06% 10,45% 14,08% 16,40% 11,86%K 14,04% 11,70% 11,06% 11,96% 9,10% 15,99% 12,18% 13,07% 14,11%H 2,07% 2,28% 8,81% 5,73% 7,76% 5,18% 6,81% 7,60% 5,19%

Cnter 74,52% 75,40% 64,46% 69,45% 67,08% 68,38% 66,93% 62,92% 68,85%

Trade policy simulation

Production value at the reference year (2001) for each micro-region and for each sector with respect to factor and intermediate consumption

Primary products

Agriculture

ROW NMB USA FRA DEU OEC SCD ENG MEDL 27,02% 16,55% 15,26% 27,23% 25,97% 22,61% 19,52% 25,80% 30,89%K 12,29% 5,67% 14,79% 12,27% 10,58% 16,79% 22,65% 14,09% 12,88%H 0,48% 0,42% 1,80% 2,38% 2,15% 0,99% 1,88% 2,10% 2,55%TE 12,78% 9,57% 14,70% 4,34% 5,84% 10,88% 4,36% 5,24% 3,78%RN 2,34% 1,21% 1,18% 1,21% 2,92% 2,95% 4,37% 2,01% 5,43%

Cnter 45,09% 66,57% 52,26% 52,56% 52,54% 45,79% 47,21% 50,77% 44,47%

ROW NMB USA FRA DEU OEC SCD ENG MEDL 5,35% 19,33% 9,14% 0,64% 16,80% 4,21% 1,28% 2,85% 10,88%K 37,79% 11,27% 26,29% 60,15% 8,60% 25,43% 33,18% 35,04% 19,95%H 1,03% 1,62% 4,20% 1,20% 7,54% 1,80% 2,14% 3,53% 6,49%

RN 24,71% 22,49% 24,51% 27,02% 23,09% 22,54% 23,75% 24,94% 23,56%Cnter 31,12% 45,29% 35,86% 11,00% 43,97% 46,00% 39,65% 33,63% 39,13%

Trade policy simulation

FRA DEU SCD ENG MED totAGM 2,36% 1,19% 2,34% 1,73% -0,32% 1,09%PRM -3,20% -4,47% -2,90% -2,98% -3,50% -3,25%IND 32,54% -9,15% -6,43% -43,71% 19,85% 1,95%

SERV -8,78% 4,02% 1,66% 10,67% -4,94% 0,10%

ROW USA OEC totAGM 0,09% 0,63% 0,41% 0,23%PRM -0,24% -0,24% -0,04% -0,21%IND -2,51% -9,00% 11,14% -0,17%

SERV 0,54% 1,62% -2,98% 0,11%

NMB totAGM 0,20% 0,20%PRM -2,53% -2,53%IND 0,57% 0,57%

SERV -0,34% -0,34%

Production with unskilled labour mobility ( = 10)

Trade policy simulation

Production, manufactures (with respect to different values of )

-50%

-40%

-30%

-20%

-10%

0%

10%

20%

30%

40%

0,00 2,5 5,00 7,50 10,00

sigma

var%FRA

ENG

Trade policy simulation

Amplification effect with unskilled labour mobility  

No mobility of unskilled labour in the macro-region (σ = 0)   Unskilled labour supply doesn’t change with respect to the reference year after the shock  Mobility of unskilled labour in the macro-region (σ = 10)   Unskilled labour supply changes with respect to the reference year after the shock 

1,3 51,38

1,02 51,30

FRAFRAS

ENG ENGS

Lw

w L

1,02 53,46

1,01 48,87

FRAFRAS

ENG ENGS

Lw

w L

ROW EU15 NMBAGM 0,27% 2,49% 2,20%PRM 0,27% 2,49% 2,20%IND 0,27% 2,49% 2,20%

SERV 0,27% 2,49% 2,20%

ROW NMB USA FRA DEU OEC SCD ENG MEDAGM 0,70% 1,88% 1,01% 2,12% 2,16% 0,82% 1,91% 2,15% 2,00%

Trade policy simulation

Change in factor prices and intermediate consumption prices ( = 0)

Capital

Unskilled labour

Land

ROW NMB USA FRA DEU OEC SCD ENG MED0.41% 1,95% 0.41% 2.81% 2,10% 2.25% 2.37% 2.11% 2.67%

ROW NMB USA FRA DEU OEC SCD ENG MEDAGM 0,01% 0,86% -0,03% 1,33% 1,37% -0,03% 1,47% 1,36% 1,40%PRM -0,06% 0,81% -0,07% 1,43% 1,66% -0,04% 1,83% 1,60% 1,46%IND -0,12% 0,73% -0,12% 1,32% 1,38% -0,14% 1,30% 1,34% 1,31%

SERV -0,08% 1,14% -0,01% 1,74% 1,78% -0,04% 1,75% 1,81% 1,69%

Trade policy simulation

Change in factor prices and intermediate consumption prices ( = 0)

Intermediate consumption

ROW NMB USA FRA DEU OEC SCD ENG MED0,15% 2,01% 0,21% 1,64% 2,74% -0,22% 2,13% 2,95% 1,60%

Skilled labour

ROW NMB USA FRA DEU OEC SCD ENG MEDAGM 0,77% 1,91% 1,12% 2,62% 2,70% 0,90% 2,09% 2,70% 2,30%PRM 0,26% -1,93% 0,18% -2,95% -2,91% 0,47% -3,09% -2,78% -2,29%

Natural resources

ROW EU15 NMBAGM -0,20% 0,66% 0,24%PRM -0,17% 0,86% 0,52%IND 0,03% 0,60% 0,38%

SERV -0,12% 0,19% -0,05%

Trade policy simulation

Change in welfare measured in terms of consumption ( = 0)

Change in imports ( = 0)

ROW EU15 NMBAGM 2,7% 12,0% 10,9%PRM -1,3% 3,8% 2,4%IND 4,8% 6,9% 6,8%

SERV -2,7% 3,6% 1,6%

Further objectives of the regional version of Mirage model:

analysing trade policy effects on the different regions for each European nation (for example for Italy, Lombardia, Veneto, Lazio, etc..) in order to implement accurate support policies to the most penalized sectors and regions