ABSTRACT
This thesis analyses how International Trade –and more particularly how trade intensity - influences upon the wage gap between genders and differs between tradable and non tradable sectors.  Following Gary Becker’s model, the hypothesis that was tested is that in a country were women discrimination has a tendency to exist one could expect that it should  be less in tradable sectors and even it should not exist. 
The data used to empirically test  the hypothesis was the Household Survey Bolivia in 2002, which form part of MECOVI (Improving the Living Conditions Survey and the Population), which is the only database in the country that classifies individuals according to their main activity and in accordance to International Standard Industrial Classification.
Bolivia was chosen for being one of the poorest and unequal countries of the Americas. 
In addition, Bolivia is one of the pioneers of political liberalization which was widespread in most of the continent during the Nineties by the Washington Consensus, with consequences that were not fully studied.
Using econometric estimates corrected for selection bias in the level of wages, the Oaxaca-Blinder decomposition was estimated, since this methodology is the most used to discuss issues of discrimination.
The results obtained shows that in Bolivia there is a general tendency to discriminate against women in the labor market. The outcome of the decomposition,  separated in  non-tradable  and tradable sectors, shows that for non tradable sectors there is no  statistical evidence of discrimination, while the opposite is true for non-tradable sectors, although the statistical tests suggest that the difference between the two is not statistically significant, demonstrating -Curiously- the strong hypothesis but not weak.
Finally, there is a new variation of the theoretical model relaxing partially the assumption of perfect market and treating the agriculture as a sector that internationally is an imperfect market and locally a perfect market. The results show that, in the case of agriculture, the propensity to discriminate in the imperfect market is stronger than in the perfect market.
Furthermore, if the agricultural sector is excluded, the new estimates proved, the strong hypothesis and the weak hypothesis, since the evidence found allowed to conclude that, in some perfect markets, both local and international trade intensity decreases the propensity to discriminate based on gender.