Over the past 35 years, the proportion of women in cabinets has increased substantially. The upward global trend line, however, masks significant variation across states and within states over time. Some states experience a “ratchet effect”: the proportion of women in their cabinets steadily increases as the years go by. In other states, representational gains are ephemeral: upward ticks in representation are followed by declines that offset initial gains.
In this paper, we conceptualize and analyze within-country temporal variation in women’s cabinet representation. We introduce a concept called gender volatility. Gender volatility occurs when levels of women’s representation rise and fall over successive temporal observations. We are particularly interested in explaining why some countries that experience representational gains backslide to lower levels of women’s representation while other states do not. Taking advantage of a global dataset spanning from 1979-2010, we show that cabinets are significantly more prone to backsliding than legislatures. We also show that levels of backsliding in the cabinet differ across states and test hypotheses about the domestic and international factors affecting this variation.
While most studies of cabinet dynamics focus on domestic variables, we argue that the most important brake on states’ likelihood of backsliding is international. Specifically, we argue that inter-governmental organizational (IGO) memberships matter. States that associate, in IGOs, with partners that experience steady, upward growth in women’s representation are more likely to experience this kind of growth themselves. Conversely, states that associate with partners that backslide are more likely to experience declines in the proportion of women in the cabinet that erase earlier gains.
By investigating the conditions under which states tend to lock in representational gains, we take up Darhour and Dahlerup’s (2013) call to determine factors that promote the “sustainable representation of women in politics.”
