The comments are relatively interesting. I would add that it's worth looking at areas in the U.S. with a traditionally Hispanic dominant population, such as parts of the upper and lower Rio Grande Valley as a test of the institutionalist explanations. They tend to be much richer than Mexico, but much poorer than the rest of the U.S., thus showing the institutionalist theory's glass is part full and part empty.
I would also add that a lot of Mexico isn't terribly poor anymore. Overall, the current life expectancy in Mexico is 97.5% of the life expectancy in the U.S.