How not to model transition economies

In the nineties, Eastern European economies went through a profound and painful transformation. There were hit by two major shocks: A change in the price system from administered prices to market price, and an opening of the economies to competition from abroad. The first impact of these shocks were very significant drops in the wages and in all manufacturing indicators. As mentioned, these changes were painful, and one could ask whether something could have been done to help ease this transition, for example with foreign aid. It is difficult to imagine where this question could matter in the future (North Korea?), but anyway.

Mohsen Fardmanesh and Li Tan ask this question. They build a two-sector three-factor small open economy model with some particular features: capital never depreciates, trade is fully open, and firms maximize profits. To make model resemble in some way the pre-transition situation, they add a constant to the market clearing condition for non-traded goods to represent shortages. Removing this constant is supposed to represent the change that happened. Then the economy is a new state, and they analyze the impact foreign aid would have had.

This is completely silly. The object of study here is the transition from a steady state to another one. To do this properly, you need to model properly the dynamics. In this case this means: 1) model properly the command economy, showing the allocation of factors across sectors, and in particular modeling technology vintages; 2) model properly the market economy the model is converging to; 3) using the initial factor allocation, show how the economy evolves on the path to the new steady state (it is not instantaneous...) as factor need to be reallocated and in particular become obsolete as the economy is exposed to foreign competition; 4) it helps to model a labor market in order to say anything about wages; 5) introduce households so that anything could be said about the welfare impact of foreign aid, which is the research question after all.