"Developing Country Expansion and Relative Wages in Industrial Countries," by Robert McDougall and Rod Tyers, 
Chapter 7 of "Global Trade Analysis: Modeling and Applications", Thomas W. Hertel(editor). 
Based on the README file prepared by Betina V. Dimaranan.
****************************************************************************
I. REGIONS AND COMMODITIES

Regions:
AUS	Australasia
NAM	North America
EC	European Union
NRG	Rapidly Developing Economies (RDEs)
ORG	Japan
ROW	Slowly Developing

Sectors:
ali	crops
gaki	other agriculture, forestry and fishin
min	mining
mkh	highly capital-intensive manufacturing
mkm	moderately capital-intensive maufacturing
mlm	moderately labor-intensive manufacturing
mlh	highly labor-intensive manufacturing
pcp	petroleum and coal products
services	labor-intensive services
ski	capital-intensive services

II. EXPERIMENT FILES: SHOCKS, CLOSURES AND SOLUTION METHOD- :

To replicate the experiments with RunGTAP, go to "SOLVE" in RunGTAP, click on "Load Experiment" and choose one of the experiments EXP1 or EXP2. This will automatically load the shocks, closure, parameter file and solution method for the experiment of interest.

EXP1 Simulates the impact of slower growth and less openness in the rapidly developing Asian economies (RDEs) on the older industrial economies (OIEs). Slower growth in the RDEs (this is the same region as NRG in the command files, more on this below) is accomplished by applying shocks to population, labor, capital, investment (cgds) to bring these growth rates down to average OIE levels. Less openness is achieved by providing a uniform shock on import taxes (tms).

EXP2 Simulates the impact on the OIEs of slower growth of the RDEs. As in the first experiment, this involves bringing down population, labor, capital and investment to average OIE levels.
 
CLOSURE
The closure chosen for all scenarios fixes factor use and renders unit factor rewards endogenous.

NOTE: Due to the use of cgdslack, walraslack in the model is not computed correctly.

SOLUTION METHOD 
 
The solution method used by the authors was:
	method = euler;
	steps = 16 32 64 ;
This method is very time consuming because of the number of iterations it uses. In the experiment files for this version with RunGTAP the method Euler 4 8 16 was used, so differences may appear in the final results. To replicate the exact results in Chapter 7 change the solution method to Euler 16 32 64.

REPLICATION OF THE EXPERIMENTS
The results reported in the chapter can not be obtained directly from the solution variables in the above two experiments. In the chapter, the results reported under the heading "open" are  calculated as:
 - (results from expt. 1) + (results from expt. 2) 
 
 The results reported under the heading "open with growth" are calculated as:
 - (results from expt 1) .

The authors give the following explanation for the approach that they have taken in their simulations and in reporting their results:

" Over the period 1970-1990, both the OIEs and the RDEs have grown and become more open, but the changes in the RDEs have been much faster than in the OIEs. To estimate the effects on the OIEs of these rapid changes in the RDEs, we consider three different scenarios for economic development over the period. Each scenario starts from the actual state of the world in 1970. In each scenario the OIEs grow and become more open at the rates historically observed but different assumptions are made about the RDEs:
 
 Scenario 1 (counterfactual): the RDEs grow and become more open at 
 the same rate as the OIEs;
 
 Scenario 2 (counterfactual): the RDEs grow at the same rate as the 
 OIEs, but become more open at their historically observed 
 rates; and 
 
 Scenario 3 (actual): the RDEs grow and become more open at their 
 historically observed rates.
 
 These three growth scenarios lead to three different states of the
 world in 1990, which we number 1 to 3 in the same way as the 
 scenarios. States 1 and 2 are counterfactual and 3 actual. "
 
The two experiments start at Scenario 3, the actual period (1990), and brings the RDE growth rates down to OIE growth rates. The development achieved by the RDEs over the 1970-1990 period may be divided into two parts: the first being of the same rate as the OIEs and the second being that in excess of OIE growth rates. Shocks are imposed to bring the RDE growth rates down to average OIE growth rates. For example, average growth rates of labor use over the period is 14% for the OIEs and 41% for the RDEs (see Table 8). To bring the RDE's growth rate for labor use down from 41% to 14%, a shock of -19% is needed. This is calculated as:
 [(1+.14) - (1+.41)] / (1+.41) = -.19 * 100 = -19%.
 
The results obtained from experiment 1 (E1) represent changes going from state 3 to state 1. The results from experiment 2 (E2) represent changes going from state 3 to state 2. In each case the changes are measured as percentages of state 3.
 
In the chapter, the results reported under the heading "open" represent changes going from state 1 to state 2, measured as percentages of the actual state (state 3). The results reported under the heading "open with growth" represent changes going from state 1 to state 3, again as percentages of state 3.
Therefore, "open" = - E1 + E2, and "open with growth" = - E1.
 
The results which can be replicated, noting the above procedure, are limited to those reported in Tables 9, 10, and 11. The results reported in Table 13 were calculated by the authors based on employment changes by sector for the 3 OIE regions resulting from the model.

Of the results reported in Table 2, only the shares of direct industry expenditures on domestic factors can be obtained using the standard GTAP code. Total industry expenditures on domestic factors include both direct and indirect factor requirements, the latter being the factor content of intermediate inputs into the industry and of intermediate inputs into intermediate inputs, etc. The calculation of indirect factor requirements is not an available feature of the standard GTAP code.

Note: ALTERTAX is not available.