When women are induced to enter the labour force, what activities do they sacrifice? Do they spend less time enjoying leisure? Do they give up time with their children? Do they cut back on household chores?
Over the past thirty years, US policymakers tried to increase participation of single mothers in the labour force by expanding the Earned Income Tax Credit and reforming the welfare system. One key motivation for reform was the perception that some single mothers were choosing to be idle and instead ought to contribute more productively to society by working (Moffitt 2006). But did the policy reforms induce single mothers to shift from one productive activity – work at home – to another – work in the market?
In a new paper (Gelber and Mitchell 2009), we address this issue by examining how single women spend their time when faced by a tax policy that shifts them between working and staying at home. If tax rates are lower, single women are more likely to participate in the labour force. When single women do participate in the labour force as a result of lower tax rates, we study changes to the rest of their time. We find that time spent participating in the labour force comes largely at the expense of time spent on housework.
New insight on the increase in working hours
Employed individuals take substantially less leisure time than the unemployed (Burda and Hamermesh 2009), and we confirm in our data that the employed do only a bit less housework than the non-employed. It is perhaps surprising, then, that when we examine women’s reaction to tax policy, the majority of the increase in market work is accounted for by decreases in housework, such as cooking and cleaning, and not a decrease in leisure time.
Striking patterns in data from the Panel Study of Income Dynamics from 1975-2004 suggest that tax policy had a very important effect on the labour supply and housework decisions of single women over this period. As shown in Figure 1, from the mid-1980s to the mid-to-late-1990s, the incentive to participate in the labour force greatly increased for single women with children relative to those without children. This was largely due to major expansions of the Earned Income Tax Credit. We capture this incentive using the “net-of-tax share,” the percentage of earnings that a woman would keep if she participated in the labour force.
Figure 1. Changes in tax rates over time: Mean imputed average net-of-tax share by year for single women with and without children
Note: Average tax rates are calculated using the Taxsim model of the National Bureau of Economic Research, by calculating a woman’s tax liability if she works and if she does not work, and then calculating the fraction of her earnings that would be taken away in taxes if she works. The average net-of-tax share for women with children is greater than one primarily because the EITC transfers a substantial amount of money to a low-income woman if she works, often implying that the effective tax rate is negative.
Figure 2 shows that over the same period there was a large increase in hours of market work for single women with children relative to those without children. Over the periods before and after the major policy changes, by contrast, there was little relative change. This suggests that the changes in policy were responsible for the large changes in market work over the same period.
Figure 2. Mean usual hours of market work and housework of single women with and without children, 1975-2004
Note: The figure shows mean usual hours worked and usual hours of housework for single female heads of household aged 25-55, excluding cohabitators, with and without children, in the Panel Study of Income Dynamics.
The pattern for housework looks almost like a mirror image. Hours of housework fell substantially for women with children relative to those without children over the period of the primary policy changes, with little relative change outside of this period. The relative fall in housework accounts for over half of the relative increase in market work, suggesting that most of the change in market work came out of housework.
Other comparisons also point to the same conclusion. As shown in Figure 3, the net-of-tax share rose much more for lower-income women with children than for higher-income women with children, again primarily because of expansions in the Earned Income Tax Credit, which largely affects those toward the lower end of the income distribution. The figure shows that, correspondingly, the change in market work was substantially more positive, and the change in housework substantially more negative, for lower-income women with children than for higher-income women with children.
Figure 3. Mean change in usual weekly hours of market work and housework (y-axis) plotted against mean change in net-of-tax share (x-axis), among women with children
Note: “High income” refers to individuals with imputed income above the median, and “low income” refers to all others. The “change” in market work, housework, and the net-of-tax share is computed by calculating the change in the mean of the variables from the 1975-1986 period to the 1987-2004 period.
In line with this evidence, our statistical analysis shows that, among single women, labour force participation rises significantly when the fraction of their earnings taken away in taxes falls, as in previous literature (for example, Eissa and Liebman 1996 and Meyer and Rosenbaum 2001). We find that the resulting decrease in time spent on housework is significantly larger than the decrease in other non-housework or non-market work time. For every additional hour that a single woman spends working in the market in response to a change in tax policy, she spends about 40 minutes less time working at home.
These patterns of market work and housework are not limited to annual survey data. Over five decades, several studies have asked survey respondents to fill out diaries containing complete lists of activities and their durations from the previous day (see Aguiar and Hurst 2007). This allows us to disaggregate time into detailed categories.
We investigate a variety of definitions of “home production” and “leisure” and find evidence that the increase in market work corresponds to substantial and significant decreases in home production. We find that time spent eating and preparing food decreases and that time spent sleeping – a hallmark of “idleness” – changes insignificantly. Importantly, we find no evidence that time spent with children decreases significantly.
How does this affect consumption patterns?
Analysis of the Consumer Expenditure Survey allows us to investigate how these changes in market and home work interact with what people buy. We find evidence that expenditures on food away from home, such as takeout and restaurant meals, increase in response to an increase in the incentive to participate in the labour force. Expenditure on home-cooked meals, meanwhile, decreases significantly. This makes sense. Women are busier when they enter the labour force and make up some of the time by purchasing food prepared by others instead of themselves. We also find some evidence that overall food expenditures rise significantly.
Interpretation and implications
Our results have implications for several areas of economic inquiry.
- First, in line with the Becker model (1965), when single women have an incentive to work more, they use market goods, such as purchases of food, to substitute for household production, such as time spent cooking.
- Second, our findings support the literature in macroeconomics that explains the magnitude of business cycle fluctuations in part through substitutability between market and home goods (see Benhabib, Rogerson, and Wright 1991). Our estimates imply that the elasticity of substitution of market and home consumption is 2.43. This means that households are very willing to substitute market goods for home goods, and that a rise in the net-of-tax wage will cause a large increase in work in the market relative to work at home.
- Third, our finding that the increase in market work corresponds largely to a decrease in housework suggests that public policies affecting labour force incentives may largely shift people from one productive activity to another. In a world of ideal data, we would also be able to observe the intensity with which they perform home and market work, and how much they value what they produce. Nonetheless, since the policy reforms we examine were motivated in part by decreasing the “unproductive” activity of “idle” single mothers, it is notable that the policies seem to have shifted individuals from work at home to work in the market.
Many papers have looked at how incentives and policies affect hours worked in the market. Our paper complements this work by examining how income taxes affect time allocation in the other two-thirds of the day, as well as what people buy. This promises to give a richer understanding of how taxation affects the lives that people live.
Aguiar, Mark and Erik Hurst (2007), “Measuring Trends in Leisure: The Allocation of Time over Five Decades”, Quarterly Journal of Economics, 122: 969-1006.
Becker, Gary (1965), “A Theory of the Allocation of Time.” Economic Journal, 125: 493-517.
Benhabib, Jess, Richard Rogerson, and Randall Wright (1991), “Homework in Macroeconomics: Household Production and Aggregate Fluctuations”, Journal of Political Economy, 99: 1166-1187.
Burda, Michael and Daniel Hamermesh (2009), “Unemployment, Market Work, and Household Production.” NBER Working Paper 14676.
Eissa, Nada and Jeffrey B. Liebman (1996), “Labour Supply Response to the Earned Income Tax Credit”, Quarterly Journal of Economics, 111: 605-637.
Gelber, Alexander and Joshua Mitchell (2009), “Taxes and Time Allocation: Evidence from Single Women”, NBER Working Paper 15583.
Meyer, Bruce and Dan Rosenbaum (2001), “Welfare, the Earned Income Tax Credit, and the Labour Supply of Single Mothers”, Quarterly Journal of Economics, 116: 1063-1114.
Meyer, Bruce, and James X Sullivan (2008), “Changes in the Consumption, Income, and Well-Being of Single Mother Headed Families”, American Economic Review, 98: 221-241.
Moffitt, Robert (2006), “Welfare Work Requirements with Paternalistic Government Preferences”, Economic Journal, 116: 441-458.