Problem with non-zero profits in a CGE model

GAMS
1

Dear forum,

I have a problem with the formulation of a CGE model. The model represents a simple economic system, with 3 economic sectors, maximizing the individual utility.

In the baseline scenario, the profits of each sector are close to 0, as it should be given that it should find the equilibrium.
However, as soon as I modify the baseline scenario, the solution indicates that profits are positive or negative, and not close to 0.
In particular, the way I modify the baseline scenario is simply by adding an additional cost.

Example:

  • in baseline scenario COSTS are = rK + wL + pw*W
  • in scenario 1 costs become = rK + wL + pwW + phiEMIS.

I couldn’t identify the cause of this problem, but it seems like it doesn’t optimize the solution anymore.
Attached the baseline scenario and scenario 1.

Moreover, I am concerned with the way I formulated the model, as It seems that GAMS doesn’t accept some formulations.
For instance:

  • It doesn’t work (meaning it doesn’t solve with Profits → 0) if I impose WD = e = AWW + AWD as a equation, or in the Loop
  • It doesn’t work if I add the equations C(‘3’)=Y(‘3’)-sum(j,AWD(j)). However, It does if I add this as a statement in in the loop.

Could someone help me out?
I can provide additional information if needed!
Thank you so much for your help and support!

Best,
SM
210104_scenario 1.gms (11.2 KB)
Sofia
210104_scenario 0.gms (11.2 KB)

2

Dear GFA,

I received your PM, but unfortunately I’m unauthorized to respond for some reason. Thank you for your reply!
I indeed have quite some doubts about the fit of this model…
And you are right! I am indeed at WUR-ENR. May I ask how you got to know?

As to the model, I wonder if negative profits as a optimal solution isn’t eventually the correct answer. I tried to change and rewrite the model in all ways, but always obtaining the same result (negative profits) as soon as I change the smallest thing in the baseline scenario.

In this regard, I have another question (perhaps someone can help).
In the model, shadow prices are initially defined as parameters - and therefore defined with a initial value. However, later on in the model, prices are defined inside the LOOP statement as marginal values of equations. Therefore, I would assume that the initial definition should have no effect on the optimal solution.
However, I find the opposite, which is that the initial values given to the price parameter do affect[/u] (and substantially!) the optimal solution.

How can that be explained? And how do I know how to correctly define the initial value?

Best,
SM