Theoretical Justification for Using the Gravity Model
Classical gravity models generally use cross-sectional data to estimate trade effects and
trade relationships for a particular time period. However, cross-sectional data observed over
several time periods (panel data methodology) result in more useful information than cross-
sectional data observed in say a year. The advantages of this method are that panels can capture
the relevant rela0tionships among variables over time, and panels can monitor unobservable
trading -partner pairs’ individual effects (Rahman, 2003).
Deriving its origins from Newton’s law of gravity in mechanics, gravity models
analogously determine trade flows between two or more countries as a function of their
respective economic masses, the distance between the economies and a variety of other factors.
Therefore, the Newtonian physics notion provides the first justification for using the gravity
model. A second justification derives application from the partial equilibrium model of export
supply and import demand presented by Lineman (1966). Based on some simplifying
assumptions discussed in the model derivation section, the gravity model is derived as a reduced
form equation with characteristics similar to the original Newtonian model.
The gravity model has been applied to evaluate bilateral trade flows of aggregate
commodities between pairs of countries and across regions (Oguledo and Macphee, 1994).
Gravity modeling was originally developed by Tinbergen (1962), but with little in the way of
theoretical justification. Recently, it has found empirical application in determining trade flows
and policy analysis (Koo and Karemera, 1991), boarder effects inhibiting trade (McCallum,
1995; Helliwell, 1996 and 1998), and impacts of currency arrangements on bilateral trade (Rose,
2000; Frankel and Rose, 2002; Glick and Rose, 2002).
Koo and Karemera (1991) state that research in this area (such as Takayama and Judge,
4