A recent paper by researchers at the St. Louis Federal Reserve looks at variations is workforce participation by women across cities. The authors suggest that commuting time is correlated with women's participation in the work force. As commute times increase in metro areas, women's work force participation slows. The metro areas with the highest commute times have the lowest rates for women workers. So why is this?
The first explanation probably starts in the home. As the authors were looking at married women's workforce participation, there are likely children involved. Unlike single mothers, married mothers have a higher likelihood of being able to stay at home. Since commute times are a direct cost (though through time, not out of pocket) of working, higher travel times reduces the incentive to work relative to the cost of caring for a child (or children).
The second reason builds on women's obligations at home. There are two potential explanations for longer commute times. Either distances are greater or speeds are lower. Either of these possibilities leads to the idea that as times increase the predictability of the journey decreases. Unpredictable travel times make it hard to deal with day care, after school activities and other home obligations. This reason also increases the costs of working for women.
There are more possible explanations that explain the differences. I do wonder what the intraregional differences are. Controlling for communities within a metro area, will the results hold? There are variations in travel times in all urban areas. This points to the limitations of the data used (PUMS). It would be interesting to see this analysis done on a micro scale.