GSIMv2
This version 25 April 2007
Instructions.
Overview: This spreadsheet implements a multi-region non-linear model of Armington-based international trade. It
supports up to 35 regions. Policy options include the implementation of changes in the bilateral tariff structure (as would
follow for example from imposition of the injury margin in antidumping cases) as well as the implementation of price
undertakings (defined as an agreed increase in price relative to benchmark prices) and domestic or export subsidies or taxes.
Data requirements: The following data are required to calibrate the model.
o The bilateral matrix of trade flows in value terms. These should be expressed c.i.f.
o The value of domestic shipments (entered in the trade matrix as own-trade).
o The vector of initial output taxes or subsidies, and also the matrix of bilateral trade taxes (import and export).
o Elasticities of supply, composite demand, and substitution. Composite demand means, for example, “all wine” where
the model then examines varieties of wine differentiated by country of origin, and where the elasticity of substitution
measures the degree of substitution between national varieties.
o Base employment for the protected and downstream industries.
o Employment share of value added for protected and downstream industries.
If no baseline policy data are available, one can approximate the marginal impact of new duties by setting all initial taxes to
zero.
If baseline employment is unknown, the value of 100 can be entered as an index. Value added shares are needed to calculate
employment effects.
Note that taxes are entered as T=1+t where t is the tax rate.
Experiment definition: The experiment is defined by supplying a set of counterfactual policy matrices. This includes the
option to specify price increases through a price undertaking, defined as the ratio of the agreed price to the benchmark price.
Spreadsheet Structure:
Tab 1: RESULTS summary: This worksheet summarizes the results of the model simulation experiment.
Tab 2: INPUT1 data and parameters: This is where the user specifies region names, elasticities, and the initial bilateral trade
matrix (including own shipments).
Tab 3: INPUT2 experiment definition: This is where the initial and counterfactual set of policy matrices are specified.
Tab 4: CALIBRATION: This is where CES share terms, supply and demand curve coefficients, benchmark prices and
quantities, and derivations of these values are calculated. This page should be left alone.
Tab 5: MODEL SOLUTION: After all data have been input, the user should invoke the Excel solver from this page and
solve the model. Note that the model may violate the default time limit or iteration limit on the solver a few times before
solving.
Underlying Theory: This model implements a non-linear, non-nested Armington model with up to 35 regions engaged in
bilateral trade. Supply is based on assumed constant-elasticity national industry supply curves, while demand is derived
from national constant elasticity composite demand schedules, combined with a CES-based aggregation technology for
composite goods. The details are spelled out in
J. Francois (2007), "An Extended Global Simulation Model: Analysis of Tariffs & Anti-Dumping Policy Impacts on Prices,
Output, Incomes, and Employment ," Tinbergen Institute mimeo, 2007.
© J. Francois 2004-2007
,XXXX
SUMMARY OF GLOBAL RESULTS
Produce Prices, Quantities, and Revenues -- change
A B C D E
% change in % change in % change in
% change in PRODUCER FOB (or ex- producer value of change in producer
OUTPUT PRICE factory) PRICE revenues revenues
EU -17.30 -12.69 -12.69 -27.80 -724,153,743,983.1
US -17.30 -12.69 -12.69 -27.80 -484,823,902,489.6
China -17.30 -12.69 -12.69 -27.80 -794,653,311,103.0
Singapore -17.30 -12.69 -12.69 -27.80 -110,274,096,719.1
ROW -17.30 -12.69 -12.69 -27.80 -4,325,727,708,023.0
… 0.00 0.00 0.00 0.00 0.0
… 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
, Producer Surplus Tax Revenue Effects
F G H I J
part 2, rents from change in output change in export
part 1, change in producer surplus under-takings taxes taxes change in import taxes
-301,989,247,884.5 0.0 0.0 0.0 573,242,036,028.5
-202,183,040,391.3 0.0 0.0 0.0 633,496,178,627.1
-331,389,235,701.4 0.0 0.0 0.0 438,834,307,743.9
-45,986,907,899.1 0.0 0.0 0.0 86,910,647,702.6
-1,803,930,819,874.8 0.0 0.0 0.0 3,222,364,156,856.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
, note: scenario 2 adjustd for NTBs=unacollected taxes.
Downstream/Final Consumer Effects Total Welfare Employment
K L M N = F+G+H+I+J+K O P
Change in
Protected base protected
consumer prices, % total consumption, % employment, employment,
change in consumer surplus change change Net welfare effects thousands thousands
-400,481,307,895.8 13.50 -17.30 -129,228,519,751.8 100 -26
-439,011,813,034.9 13.50 -17.30 -7,698,674,799.2 100 -26
-318,002,201,315.2 13.50 -17.30 -210,557,129,272.7 100 -26
-61,086,273,121.5 13.50 -17.30 -20,162,533,318.0 100 -26
-2,331,208,173,238.5 13.50 -17.30 -912,774,836,257.3 100 -26
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
This version 25 April 2007
Instructions.
Overview: This spreadsheet implements a multi-region non-linear model of Armington-based international trade. It
supports up to 35 regions. Policy options include the implementation of changes in the bilateral tariff structure (as would
follow for example from imposition of the injury margin in antidumping cases) as well as the implementation of price
undertakings (defined as an agreed increase in price relative to benchmark prices) and domestic or export subsidies or taxes.
Data requirements: The following data are required to calibrate the model.
o The bilateral matrix of trade flows in value terms. These should be expressed c.i.f.
o The value of domestic shipments (entered in the trade matrix as own-trade).
o The vector of initial output taxes or subsidies, and also the matrix of bilateral trade taxes (import and export).
o Elasticities of supply, composite demand, and substitution. Composite demand means, for example, “all wine” where
the model then examines varieties of wine differentiated by country of origin, and where the elasticity of substitution
measures the degree of substitution between national varieties.
o Base employment for the protected and downstream industries.
o Employment share of value added for protected and downstream industries.
If no baseline policy data are available, one can approximate the marginal impact of new duties by setting all initial taxes to
zero.
If baseline employment is unknown, the value of 100 can be entered as an index. Value added shares are needed to calculate
employment effects.
Note that taxes are entered as T=1+t where t is the tax rate.
Experiment definition: The experiment is defined by supplying a set of counterfactual policy matrices. This includes the
option to specify price increases through a price undertaking, defined as the ratio of the agreed price to the benchmark price.
Spreadsheet Structure:
Tab 1: RESULTS summary: This worksheet summarizes the results of the model simulation experiment.
Tab 2: INPUT1 data and parameters: This is where the user specifies region names, elasticities, and the initial bilateral trade
matrix (including own shipments).
Tab 3: INPUT2 experiment definition: This is where the initial and counterfactual set of policy matrices are specified.
Tab 4: CALIBRATION: This is where CES share terms, supply and demand curve coefficients, benchmark prices and
quantities, and derivations of these values are calculated. This page should be left alone.
Tab 5: MODEL SOLUTION: After all data have been input, the user should invoke the Excel solver from this page and
solve the model. Note that the model may violate the default time limit or iteration limit on the solver a few times before
solving.
Underlying Theory: This model implements a non-linear, non-nested Armington model with up to 35 regions engaged in
bilateral trade. Supply is based on assumed constant-elasticity national industry supply curves, while demand is derived
from national constant elasticity composite demand schedules, combined with a CES-based aggregation technology for
composite goods. The details are spelled out in
J. Francois (2007), "An Extended Global Simulation Model: Analysis of Tariffs & Anti-Dumping Policy Impacts on Prices,
Output, Incomes, and Employment ," Tinbergen Institute mimeo, 2007.
© J. Francois 2004-2007
,XXXX
SUMMARY OF GLOBAL RESULTS
Produce Prices, Quantities, and Revenues -- change
A B C D E
% change in % change in % change in
% change in PRODUCER FOB (or ex- producer value of change in producer
OUTPUT PRICE factory) PRICE revenues revenues
EU -17.30 -12.69 -12.69 -27.80 -724,153,743,983.1
US -17.30 -12.69 -12.69 -27.80 -484,823,902,489.6
China -17.30 -12.69 -12.69 -27.80 -794,653,311,103.0
Singapore -17.30 -12.69 -12.69 -27.80 -110,274,096,719.1
ROW -17.30 -12.69 -12.69 -27.80 -4,325,727,708,023.0
… 0.00 0.00 0.00 0.00 0.0
… 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
0 0.00 0.00 0.00 0.00 0.0
, Producer Surplus Tax Revenue Effects
F G H I J
part 2, rents from change in output change in export
part 1, change in producer surplus under-takings taxes taxes change in import taxes
-301,989,247,884.5 0.0 0.0 0.0 573,242,036,028.5
-202,183,040,391.3 0.0 0.0 0.0 633,496,178,627.1
-331,389,235,701.4 0.0 0.0 0.0 438,834,307,743.9
-45,986,907,899.1 0.0 0.0 0.0 86,910,647,702.6
-1,803,930,819,874.8 0.0 0.0 0.0 3,222,364,156,856.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
, note: scenario 2 adjustd for NTBs=unacollected taxes.
Downstream/Final Consumer Effects Total Welfare Employment
K L M N = F+G+H+I+J+K O P
Change in
Protected base protected
consumer prices, % total consumption, % employment, employment,
change in consumer surplus change change Net welfare effects thousands thousands
-400,481,307,895.8 13.50 -17.30 -129,228,519,751.8 100 -26
-439,011,813,034.9 13.50 -17.30 -7,698,674,799.2 100 -26
-318,002,201,315.2 13.50 -17.30 -210,557,129,272.7 100 -26
-61,086,273,121.5 13.50 -17.30 -20,162,533,318.0 100 -26
-2,331,208,173,238.5 13.50 -17.30 -912,774,836,257.3 100 -26
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
0.0 0.00 -100.00 0.0 100 0
