IRTSPROF: Increasing returns to scale calibrated to NON-zero profits 

Accompanies the paper titled: "Economic integration and market structure: An analysis of 
the Morocco-EU free trade area", by Aziz Elbehri and Thomas Hertel
GTAP Working Paper, Center for Global Trade Analysis,Purdue University

This research employs a modified model incorporating Cournot, imperfect competition and 
increasing returns to scale in selected sectors. In doing so, it builds on GTAP
Technical Paper #14 by Joseph Francois. The latter has been adapted to include entry/exit
of firms.

I.	SUMMARY

        This version of GTAP uses a 28-good,3-region aggregation 
	(Morocco, European Union,and Rest-of-World).
        Manufacturing sectors are imperfectly competitive with oligopolistic
        Cournot behavior and increasing returns to scale. 


II.	SHOCKS

	FTA: Eliminate Morocco's tariffs on EU imports of manufactures, with exemptions in food
        manufacturing based on agricultural content.

	Multilateral Liberalization: Cut global tariffs by 30%

III. 	CLOSURES

	There are a wide range of alternative closures that combine the following choices:

	FTA = FT or Multilateral Liberalization = ML
	No entry = NE or Entry = EN
	Full Employment = FE or Unemployment = UN
	Tax Replacement = TR or No Tax Replacement = NR

	So, FTNEFENR = FTA with no entry, full employment and no tax replacement.

	It DOES NOT MAKE SENSE to implement the unemployment closure without tax replacement.
	If you do so, you will get an illusory rise in employment as real wages automatically fall.


IV.	CREATING THIS VERSION                      

	1. To create this version of the data base, begin with the data obtained from GTAPagg.
	This is available in the version: MorV53, for example.
	2. This must be modified to reflect existing agreements between the EU and Morocco, as 
	well as additional information on Morocco's protection rates that is not in the 
	standard GTAP data base. These shocks have been computed for the version 5 data base
	and are reported in TMSADJ.DAT. 
	3. Running ALTERTAX with these shocks generates an 
	4. To this must be added the data arrays pertaining to market power (CV_RATIO) and 
	scale economies (SCALE). We compute these by inferring CV_RATIO from the Herfindahl
	Index, combined with the assumption of symmetric, Cournot firms. In order to be 
	compatible with zero profits, we must set SCALE = [1/(1 - CV_RATIO/DELAST)] - 1.
	This combination of CV_RATIO and SCALE can then be added without contradicting the 
	implicit assumption of zero profits.
	5. If the user wishes to assume a different value for SCALE, thereby giving rise to 
	excess profits, this must be inserted into the data base via a simulation. SCALINS.EXP
	does this for the SCALE values used in the above-mentioned paper. The resulting data 
	base is saved in version: IRTSPROF. 
	THIS IS THE PRESENT VERSION
	6. Note that for the reader interested in comparing results with the CRTS/Perfect 
	Competition model, the data base may be modified by setting SCALE = CV_RATIO = 0.
	This re-produces the (non-nested Armington) perfectly competitive, GTAP model. The
	latter may be found under the version: CRTSZERO. It can only be run in "no entry" mode.
	updated standard GTAP data base saved in version MorUPD. 