Preferential Trade Agreements and Trade Liberalization Efforts in the OIC Member States

With Special Emphasis on the TPS-OIC

interest and tries to look for the effect of this known event into the variables of interest. The

main aspect of the ex-post analysis is trying to disentangle or control for all the other elements

that might have had an effect on the variables of interest, leaving the pure effect of the

experiment under study. Econometric analysis typically using the gravity models are the main

technique applied.

Many methods have been developed to perform ex-ante analysis. Although they differ with

respect to their technical aspects, they share the normative nature in their conception. This

means that these methods require the definition of some assumptions on how the variables of

interest are expected to react as a result of the policy exercise. This contrasts with the

econometric approach where, although implicit, there are no behavioural assumptions in the

formulations. Simulation techniques, such as general equilibrium or partial equilibrium

models, use mathematical equations to represent the relationship between different variables

of the system. In this type of ex-ante analysis, a parameter (a tariff, for example) is changed in

order to evaluate the effect that this might have on a given variables (imports, for example).

The equations are part of a consistent system of equations that must be in overall equilibrium.

These might take the form of requirements of equilibrium in each market in isolation (partial

equilibrium) or that all markets must clear simultaneously (Walras Law) in general

equilibrium models.

Data requirements are particularly important in this type of simulation analysis. In addition to

consistent trade and tariff data, it is necessary to know in advance the value of some

behavioural parameters such as demand and supply elasticities that provide information on

the responsiveness of a variable with respect to another variable. For example as tariffs go

down, typically domestic prices would go down which in turn would increase demand. The

modeller has to make some (mathematical) assumption as to by how much demand will rise as

price go down. In addition to the problem of obtaining these parameters, all the set of data

must be mutually consistent in the sense that in the initial solution they must constitute

equilibrium of the system. Therefore, a calibration stage is necessary before the use of the

model to assure that data put in the model constitute equilibrium of the system. Given these

technicalities, this type of tool is, in general, outside the scope of the standard trade analyst

and specialists are needed in general to operate them. Moreover, the interpretation of the

results tends to be less intuitive and, frequently, interpretation of the results can be

challenging.

Econometric techniques, on the other hand, such as gravity models try to capture the effects of

a certain variable on another variable of interest. This implies empirically isolating, among all

the different events and policies, the single effect of the experiment under study. For example,

the effect that an FTA has on the observed trade values controlling for other structural or

short-term effects. The aim is to identify the different factors that impact on a given outcome,

such as bilateral trade between countries. Although data requirements tend to be less

stringent than needed for the simulation techniques, and in principle, these methods do not

require the formal use of assumptions about the relationship between the variables; the