Are there trade-offs in the representation of women and workers? I analyze Italian archival

data to examine how an exogenous increase in the number of women in elected office affects

legislature class composition. For this, I leverage an natural experiment, in which some

municipalities temporarily introduced gender quotas on party lists in the mid-1990s. Findings

show that having a singular election with a gender quota has positive effects on the descriptive

representation of women who belong to the lower middle and working classes. The political

opportunities of upper-middle class women were unaffected by the quota, as are those of

working-class men. Instead, it is middle-class men who lose out as a result of quota adoption.

Results highlight that—in contexts characterized by low levels of party gate-keeping—gender

parity rules can work to bolster the prevalence of social groups that are the most under-represented

in politics. When this occurs, these improvements come mainly at the cost of groups

that are traditionally over-represented.