Trade causing net loss

[Antti-L]

A TIMES user approached me with an interesting question (shortened by me):

We have modelled power system of two countries in TIMES. We have two stand-alone models, and then we have one integrated TIMES model for analysing trade. In the integrated model, we have an inter-connection technology that transmits power between two countries. When we run the trade case, the second country makes much higher expenses in its power system (to produce export quantity), than it earns from its export revenues, causing a net loss.

It would be difficult to communicate the result to the policy/decision makers. No country would be interested to export if there is an economic loss. The exporting country needs to make some gains from its export revenue, if export needs to happen in reality. It would be good to show trade benefits for both countries; while one benefits from cheap electricity, the other gains from export earnings. However, it is not that case here.

TIMES is used for multi-regional modelling in several occasions. Is there any option available in the model to deal with a trade case to get a win-win situation for both countries?

I responded to the user by saying that such a loss from trade should not be possible. According to my understanding, under a competitive equilibrium (such as in TIMES), any voluntary trades (i.e. not forced ones) should be guaranteed to be Pareto efficient. But as I am not a very proficient economist, I decided to present the question here and ask for others' opinions.  So, if you have any insights about this issue, please comment.  

[Pernille.S]

Here are my comments to the question:

• As I understand, the model decisions in TIMES are made to minimise the system cost for all the model regions. This implies that the model decisions in a model with two regions are not necessarily the optimal model decisions for one of the regions separately. If both regions should optimise their decisions in accordance to what is optimal for their region, you need to use a complementary model using game theory.
  


• Personally, I think this issue can cause a dilemma with the regional coverage of TIMES models. Using one region provides a set of policy recommendations that are optimal for this region, but does not necessarily capture the interaction with surrounding regions properly. On the other hand, using several model regions, endogenously capturing the interaction with neighbouring regions, do not provide optimal solutions for one specific model region.   
  

Pernille

[Antti-L]

Thanks, Pernille, for your comment, which is a very useful reminder of the fact that there are many other different equilibrium conditions that, in general, cannot be directly analysed with TIMES, such as Cournot equilibrium, Nash equilibrium, Stackelberg equilibrium etc.

I agree with all the points you made. However, I think your comment does not actually address the original question about trade causing a net loss in TIMES (or my confusion about it).

Both the original question and my reformulation of it could be summarized as follows:  What should a TIMES user do (if anything) in order to ensure that a trade between two regions will show a net benefit for both regions in the TIMES solution, when compared to the non-trading case?

Economic textbooks state that under so-called competitive equilibrium all possible mutually beneficial gains from trade are exhausted, which in my understanding also means that the trades are Pareto efficient. And by definition, a Pareto efficient economic transaction leads to a net gain, without anyone being made worse off.  

Therefore, would it be possible to conclude that if the TIMES equilibrium can be qualified as a competitive equilibrium, we could be assured that there is no net loss from trade to either region?  Any further insight to this question would reduce my confusion...