Despite shrinking families and dramatic increases in women’s time spent in the workforce, the amount of time married parents spent with children rose from 1975 to 2000. ... Linking twelve time use surveys between 1965 and 2007, we show that the increase in time spent with children occurred starting in the mid-1990s. We also show that the increase in childcare has been twice as great for college-educated parents. ...
The first leading explanation argues that because parenthood is mostly voluntary now, individuals can choose whether and when to be parents. This selection effect suggests that parents are now more likely to be individuals who enjoy interacting with children, and hence to spend more time on childcare. Our findings continue to hold, however, even when we consider time spent in childcare averaged over all adults. Thus, compositional changes in the population of parents cannot explain our results.
Second, time spent in childcare may be positively related to income. If this is true, one could explain both the rise in overall childcare as well as the increase in the college differential. We test this explanation and find that once we control for education, there is no quantitatively significant relationship between income and time spent in childcare. Thus, the effects of rising income cannot explain the trends.
As a third possibility, heightened safety concerns induce parents to accompany their children in their activities and to substitute structured activities for the free, unaccompanied play on neighborhood streets that was the norm in earlier times. According to this explanation, there has been a form of “negative technological progress” in raising children. But trends in crime rates and parents’ perceptions of safety correlate negatively with trends in childcare, which is at odds with this explanation.
A fourth possible explanation may be that childcare is best classified as leisure because it generates high enjoyment. We show that surveys of enjoyment of basic childcare in 1985 and in the mid-2000s do not support this notion. We also show that our trends are robust to the exclusion of the high enjoyment childcare activities involving playing with children.
A fifth explanation is that parents’ work schedules are more flexible, so that they can spend more time with their children. However, we find that the greatest increases in childcare time occurred among nonworking mothers.
The inability of existing explanations to account for the evidence leads us to offer a new explanation for the upward trends. We argue that the increase in time spent in childcare, particularly among the college educated, may be a response to an increase in the perceived return to attending a good college, coupled with an increase in competition in college admissions. Importantly, the size of college-bound cohorts rose dramatically beginning in the early 1990s, coincident with the increase in time spent on childcare. ...
We provide support for this explanation by comparing childcare trends in the U.S. to those in Canada. The U.S. and Canada are similar in many respects, but differ in two ways that are key to our explanation. First, Canadians do not experience the intense rivalry to gain admission into higher rated colleges. Second, the returns to a college degree have increased much less in Canada. ... Employing time-use data from Canada’s General Social Survey, we show that time spent in childcare by educated parents in Canada changed very little over this period, corroborating our theory.
--"The Rug Rat Race," by Garey Ramey and Valerie A. Ramey. HT: Freakonomics blog