and this income is spent on purchasing goods and services. Household budget constraints now imply that
and demands for goods and services by each household are derived from maximization of Cobb-Douglas household utility functions subject to the household budget constraints (19).
Taxes charged by national governments on S2 purchased at home can be incorporated into this framework, and distort the relative prices of both services and goods, and among services, reducing purchases of externality generating international services.

Table 1 presents parameterizations we have used in a numerical example showing the impacts of network related services liberalization in this two services type model. Country 2 has double the resources of country 1 at 1000 rather than 500. We consider 3 households in country 1 and 5 in country 2, with identical preferences within countries. We consider liberalization to be a removal of a pre existing 50 percent tax on international services in both countries.
As the results in Table 1 show, we can construct cases, such as that in the first specification where, as before, the absolute value of the gains from liberalization in network related services are similar across the two countries. It is also possible with a change in specification to construct a case where the gains to the two countries go mainly to the large country (opposite to trade in goods), as in the second specification.
In this more realistic but more complex world, then, the theme is the same as above; namely that with network externalities and cross country service liberalization, there is no presumption (as with goods) that the benefits of liberalization will accrue disproportionately to the small country. Per capita effects and country size seemingly once again operate in opposite directions.