When the Work is Never Done: Time Allocation in US Family Farm Households

Abstract

In this paper we use a Nash-cooperative bargaining framework to examine how members of US family farm households allocate their time between work and leisure. Time allocation categories for parents include farm, off-farm, and household work, as well as leisure time; for children, the categories are farm work and leisure time. The analysis includes 227 Wisconsin dairy farm households. Most notably, the results confirm that US women and children make significant labor contributions and that both women and men are decision-makers regarding their own and their children's time allocation. The results also show that intra-household time allocation on US farms is gender specific, and that the father's economic status has the largest impact on the time allocation of household members. The findings also confirm that children's labor makes an important economic contribution to the operation of their family farm.

Keywords:

• Time allocation
• household bargaining
• household labor supply
• child labor
• farm household

ACKNOWLEDGMENTS

This research is a part of the first author's doctoral dissertation. The research was supported by a Hatch grant from the College of Agricultural and Life Sciences at the University of Wisconsin-Madison. The authors would like to thank Paula Kantor, Mark Purschwitz, and Karen Holden for their valuable comments. Gratitude is also expressed by the second author to the Helen Riaboff Whiteley Center and the staff of the University of Washington Friday Harbor Marine Lab for facilitating the completion of this article. This article is jointly authored; the authors are solely responsible for any errors.

Notes

JEL Codes: C7, J2, D1

Household production is particularly high in farm households because they have more members, on average, than nonfarm households; also, farm households lack market substitutes for some goods due to remoteness, and household members spend more hours at home.

In this study children's time is assumed to be allocated to two activities: farm work and leisure. Unfortunately our data did not include household work or school. However, given that the data are from one state in which school standards are fairly uniform, it can be assumed that the time allocated to school is fixed. What may differ is the time allocated to homework and school activities, which generally compete with time allocated to farm work.

The evidence that children's consumption of goods depends on their parents' preference structures supports this argument. Children's well-being is significantly improved when women have control of income and expenditure in developing countries (Duncan Thomas 1990). Similarly, in US family farm households, women are more likely to spend off-farm income on children and household expenditure than men are (Zepeda et al. 1997).

This aggregation does not permit recognition of individual children's leisure consumption.

Many children also frequently work on household chores. This study, however, focuses on children's farm work because data for time allocation for household work by children were not available.

The equations differ from those in the unitary model because they include personal nonlabor income. When the individual nonlabor incomes are summed to a household nonlabor income, the reduced form of the equations is the same as its counterpart in the unitary model. Therefore, the unitary model is nested within this cooperative bargaining model.

Exclusion of demand for goods is based on a two-stage budgeting between goods and leisure (Angus Deaton and John Muellbauer 1980). While there is some debate, many studies support the assumption of separability between goods and leisure (Singh, Squire, and Strauss 1986).

Flexible functional forms such as the translog and almost-ideal demand system are not commonly used for time-allocation systems (Solberg and Wong 1992).

The advantage of the two-step model is that it is more general than the Tobit model. For example, Shonkwiler and Yen (1999) showed that the Tobit model is a nested version of this two-step specification.

The three counties have a higher proportion of intensive rotational grazing (IRG) technology adoption permitting statistical comparison between IRG and conventional technologies. Farm technology choice may influence children's labor provision and the actual amount of time devoted because IRG is a labor-saving technology.

Observations with no farm revenue or no time on farm work are omitted for this study.

Leisure here includes personal care and sleep.

This is consistent with the results from a Wisconsin survey in 1993 by ATFFI (1996).

Father's farm work is not included in the first step because 100 percent of fathers work on-farm.

Since this study focuses on aggregate farm labor by children, rather than the farm labor of individual children, the shadow wage with respect to the age categories is not considered.

Using the endogenously determined wages may generate smaller standard errors. However, given the complexity of this error correction process, we assume that endogenously determined wage rates do not make a large difference in standard errors. The errors are corrected for the two-stage process, and we found that the corrected standard errors were not very different from the uncorrected ones and did not affect the implications of the hypothesis tests.

The coefficients for leisure equations can be recovered from the adding-up restriction.

Additional information

Notes on contributors

Jongsoog Kim

JEL Codes: C7, J2, D1