councilors that are women is strongly explained by whether or not
therewerewomen councilors in the past. From themarginal effects
column we see that, holding all other variables at theirmean, forest
associations that had women councilors in the past are expected to
have 1.67 additional councilors than those that have not hadwomen
councilors in the past (significant at the 0.01 level). Forest
associations where there is high wealth inequality are predicted
to have 0.58 fewer women compared to associations with low
wealth inequality (holding all other variables at their mean this is
significant at the 0.05 level). When there is competition for council
seats women are predicted to hold 0.47 fewer seats in the council
than when there is not competition (holding all other variables at
their mean this is significant at the 0.01 level).
The marginal effect of a change in the gender wage differential
increases the number of seats held by about 0.016. Although this
result is statistically significant it is substantively very small. This
is more easily seen by examining the effects of a discrete change in
wage differential. Even going from 80% wage differential (women’s
daily wage is 80 percent less than men’s) to 0% wage differential
(equal wages), which are the minimum and maximum values in
the data, only 0.09 additional seats are expected to be taken by
women on the council (this calculation is not reported here but is
available from the author upon request).
The remaining independent variables, including the percent of
women that labor outside the home, the number of children per
adult women (both measures of norms), and the percent of
households headed by women (a measure of personal endowments/
attributes) are insignificant.