Transcript Policy Modelling

An Overview of
Policy Modelling
Dr. Imtithal AL-Thumairi
Webpage:http://www-users.york.ac.uk/~iaat100/
Outline of the Talk
Need for applied Policy Modelling
An overview of policy modelling
Best strategy in modelling
Future directions
Issues and challenges
Discussion
Why do we need models?
Complex interactions
Policy matters
Policies have economy wide effects
Policies change behaviour
Policies may have international effects
Long-run effects may be larger
Monitoring and analyzing policies
Forecasting their impacts
Modelling helps us to understand complex issues
and to take better decisions
Increasing Demand & Shortage
Increasing demand for quantitive analysis
E.g. Kyoto agreement, WTO negotiations,
financial crisis,…have global implications and
potentially affect all sectors of the economy
Regional policies and governments gaining
influence and power
Shortage / lack of tools for quantitive economic
analysis
What is a Model?
A set of equations
Embody the history of theoretical
and empirical economic knowledge
Macro or/and detailed structure
Key bits
Identities
Behavioural equations
Exogenous inputs
Desiderata for policy models
Relevance
Link policy variables to outcomes
Linking policy variables to outcomes
Aggregate and structural indicators:
identifying winners and losers
Transparency
‘Black- Box’ Syndrome
Timeliness
Validation & estimation
Diversity of approaches
How to use Economics Models
Very carefully!
Can get both quantitive estimates
and new insights on complex issues
What Features are Important?
Does the model take into account the
specifications of the country?
Does the model explain anything we
observe today or in the recent past
(VALIDATION)?
Do the model results pass the test of
common sense?
What Features are important?
Are the mechanisms in the model
transparent to other trained
economists?
Is the model continually reviewed
and updated by experts who actually
use it; is it open to evaluation by
others?
Types of Models
Accounting & regression models
Input/output models
Computable general equilibrium models
Overlapping generations models
Macro-econometric models
Intertemporal macro-econometric models
Intertemporal general equilibrium models
Micro-simulation models
Optimization models
Regional models
Types of Models
Input/output models
Trace the flow of resources between sectors
Little role for relative price changes or
substitution of inputs or consumption bundles
Tend to be static
No allowance for capital accumulation or
international financial flows
Ignore the role of money and asset prices
Government behaviour not properly modelled
Tax structure not properly modelled
Types of Models
Computable General Equilibrium Models
Derived from microeconomics optimization
theory
Considerable attention to individual
behaviour
Considerable detail in the tax structure
Considerable disaggregation
Relatively easy to understand and results
given theoretical structure
Inadequate macroeconomic behaviour
Tend to be comparative static or recursive
dynamic
Types of Models
Computable General Equilibrium Models
Inadequate treatment of intertemporal
saving and investment decisions, capital
accumulation, financial capital flows
Ignore the role of money and asset
prices
Rarely validated with actual experience
either econometrically or through
forecasting or shock replication
Types of Models
Macroeconometric models
Rely on correlations in time series data
based on aggregate economic theory
Reasonably good for short term forecasting
(tend to be quarterly)
Difficult to understand results because of
lack of theoretical structure
Unclear long run properties
Government behaviour not properly
modelled
Tax structure not properly modelled
Types of Models
Intertemporal macroeconometric models
More tightly specified theory
Rational expectations in some markets
Short run data consistency with long run
theoretical properties
Tend to be quarterly
Government behaviour not properly modelled
Tax structure not properly modelled
Types of Models
Intertemporal GE models
Integrates the key features of the other
types of models
Mix of econometric estimation and
calibration
Annual frequency
Problem with large degree of
disaggreagtion because of complexity
of the numerical algorithms needs
Types of Models
Micro simulation models
Extremely detailed tax/benefit structure
Lack macro-economic linkages
No dynamics
Types of Models
Overlapping generations models
Very useful for pension reform studies
Usually closed economy
Often one or few sectors
Not detailed tax structure
Integrated use of models
CGE models can be used jointly with
macro models for forecasting and
projections
Or CGE and macro models can be
fully integrated
Best Strategy in Modelling
Start small! Go from small to large
Start simple! Go from simple to
complex
Develop several models
Have always two versions of the
model:
One operational
One in development
A Strategy for the Future
Short-term forecasting model
General equilibrium models
Sectoral disaggregation
Global model for trade, energy and
environment
Overlapping Generations
Long term work (regular updating)
Human resources
Step in Applied Modelling
1. specify dimensions of the model
Number of agents
Number of goods and factors
Number of countries/regions
Number of active markets
2. chose functional forms (behaviour on markets)
3.Construct micro-consistent data set (CGE)
4.Calibration & econometrics estimation
5. replication and debugging
6. testing and fine tuning
7. counter-factual experiments, scenario and impact
analysis
8. regular updating
Agents, Markets, Key Features
Agents
Households
Firms
Government
institutions
Markets
Goods & Services
Factors of
production
Money
Financial Assets
Foreign Exchange
Agents’ Behaviour & Markets
How do agents behave?
Microeconomics
Functional forms
How do markets operate?
Microeconomics
Industrial economics
Model Use
Policy, impact, scenario analysis
Forecasting
Projections
planning
Issues & Challenges
Data
Parameter specification
Systematic sensitivity analysis
Maximum entropy econometrics
Model preselection
Functional forms
Expectations
Capital and labour mobility
Alternative terminal conditions
Expectations
Market structure
Expectation
Myopic
Adaptive
Limited forward-looking expectations
Perfect foresight
Market Structure
Empirical evidence: Scale economies
and deviations from marginal cost
pricing
Models with IC allow to quantify
industrial organization effects, which
often dominate policy debates but are
not captured by PC models
Inclusion of industrial organization
effects may significantly affect
simulation results
Alternative Models of IC Firm Conduct
Specifications differ along several
dimensions:
Conduct:
Bertrand-Cournot-Conjectural
Variations- Monopolistic Competition
Market Regime:
Segmentation VS integration across regions
Product Substitutability:
Homogeneity VS differentiation across firms
Alternative demand nesting hierarchies
Alternative Models of IC Firm Conduct
Specifications differ along several
dimensions:
Conduct:
Bertrand-Cournot-Conjectural
Variations- Monopolistic Competition
Market Regime:
Segmentation VS integration across regions
Product Substitutability:
Homogeneity VS differentiation across firms
Alternative demand nesting hierarchies