SIMULATION MODELS 2nd Week

1
SIMULATION MODELS2nd Week

• INTRODUCTION
• Master économie et affaires internationales
• Univ. Paris Dauphine September 2010
• Dr. Rafael de Arce
• Professeur Univ. Autónoma de Madrid
• http//www.uam.es/rafael.dearce

2
WHY DO POLICY-MAKERS NEED MODELS? Reference
paper Piermartini, R. and Teh, R. (2005)
Demystifying Modelling Methods for Trade
Policy. WTO Discussion Papers, num. 10.
September, 2005.

• Economic models provide a theoretically
  consistent, rigorous and quantitative way of
  evaluating different policy analysis
• A simulation of the model can confirm that
  judgment and provide estimate of the likely gains
• Models can alert about effects in inter-related
  issues
• But, whats first? The tail shouldnt wag the
  dog

Master économie et affaires internationales,
Université Paris Dauphine
3
WHY DO POLICY-MAKERS NEED MODELS?
4
WHY DO POLICY-MAKERS NEED MODELS?
5
EXPONENTIAL GROWTH

• Advances in theoretical analysis
• Largeness and wideness Statistical collection
• Increase of Computational power

Master économie et affaires internationales,
Université Paris Dauphine
6
Moore Law
Sylicone chip procesor doubles its capacity in
18 months....
7
STRENGTHS

• Reflects Inter-dependency of economic variables
• Computer based models allow us to track all of
  these interactions
• Simulation of several scenarios is available
• Simulations are performed in a transparent
  hypothesis frame
• They discipline thinking about how economies
  actually work

Master économie et affaires internationales,
Université Paris Dauphine
8
LIMITATIONS

• Aggregations can obscure important underlying
  relationships
• Data are not always high quality
• Responsiveness of supply and demand to price
  changes are not necessarily accurate
• Choices among scenarios and model specification
  can imply very different results
• Technical comprehension is only available for
  experts

Master économie et affaires internationales,
Université Paris Dauphine
9
CHOICE THE MODEL

• COMPARATIVE STATIC AND DYNAMIC ANALYSIS
• Static changes in policies and direct
  transmission to endogenous variables of model
• Dynamic final and intermediate process of change
• PARTIAL OR GENERAL EQUILIBRIUM ANALYSES
• Ceteris Paribus criteria
• Whole economy linkages

Master économie et affaires internationales,
Université Paris Dauphine
10
Reference paper Piermartini, R. and Teh, R.
(2005) Demystifying Modelling Methods for Trade
Policy. WTO Discussion Papers, num. 10.
September, 2005.
Whole economy linkages
11
KIND OF THECNICAL STRATEGIES FREQUENTLY USED IN
INTERNATIONAL SIMULATIONS

• Traditional Econometric Multiequational Models
• Optimization Models
• Gravity Models
• Panel Data Models
• Leontief Based Models (I-O Models)
• Social Account Matrix Models

Master économie et affaires internationales,
Université Paris Dauphine
12
Exogenous ? Endogenous Relationship

• Optimization
• One target function
• Constraints to be accomplished
• Linear and non linear available linkages
• Sometimes, more than one solution (local optimum)

13
Exogenous ? Endogenous Relationship

• Gravity Models

Example Gravity Panel Data Model used in Arce
and Mahia (2008) DETERMINANTS OF BILATERAL
IMMIGRATION FLOWS BETWEEN THE EUROPEAN UNION AND
SOME MEDITERRANEAN PARTNER COUNTRIES ALGERIA,
EGYPT, MOROCCO, TUNISIA AND TURKEY. FEMISE 2008
14
Exogenous ? Endogenous Relationship Leontief
Based Models (I-O Models) and Computable General
Equilibrium (CGM)

• Actors
• Activities
• Commodities
• Factors
• Enterprises
• Households
• Government
• Capital Account
• Rest of the World

• Behavioral parameters
• Elasticity of producers and consumers to price
  changes
• Value Added
• Armington
• Demand/income

15
SAM Structure
16

• The most important thing between the screen and
  the keyboard is the person who is in the middle
  of them.