Skip to main content

Advertisement

Log in

Abstract

Using a structural model of time allocation between household production and market work, I estimate the shadow wage of household production. I combine data from the American Time Use Survey and the Current Population Survey’s Food Security Supplement to estimate the time cost as well as the total cost of food preparation at home. Using information on whether a household received food stamp benefits, I compare actual food stamp recipients’ time cost to those who do not receive food stamp benefits. Previous literature that also uses American Time Use Survey data suggests that “typical” food stamp recipients incur a higher time cost of preparing food at home. I cannot confirm this finding using actual data on food stamp recipients and non-recipients. In fact, the lower shadow wage of household production of food stamp recipients more than offsets the larger amount of time spent preparing food at home, generating lower time cost of food preparation for them.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Notes

  1. The Nutritional Assistance Program of the Commonwealth of Puerto Rico is also administered by the United States Department of Agriculture. This program requires 75% of a household’s benefit to be used directly for food purchases, while up to 25% of the benefit may be withdrawn as cash. See Fraker et al. (1986) for a discussion of the economic effects of ’cashing out’ on food stamps. The data and discussion in this paper relate to the United States only, excluding the Commonwealth of Puerto Rico. We also assume that food stamps are used for ingredients for everyday meals. While it may be possible to purchase food items such as wedding cakes with SNAP benefits, due to the nature of the data we cannot make distinctions between the different types of food items that SNAP participant purchase.

  2. Of course, food does not only have to be prepared, but also eaten. For example, Hamermesh (2010) has studied the frequency and duration of actual food consumption and their relationship to heath outcomes. He found that eating occurs often as a secondary activity; that is, people eat while actually being engaged in another activity. Since many foods can be taken ’to go’ once they have been prepared, the focus of this research is the time cost associated with preparing the food, not actually eating it.

  3. Davis and You (2009) used single non-white, non-professional females, with some high school education from metropolitan areas as the typical Food Stamp participant demographic profile.

  4. I aggregated Race into three groups because the original data provide more detail than is useful. The Race category ’Other’ contains all 15 possible races other than white-only or black-only, which are indicated by themselves accordingly. Note that Race and Ethnicity are treated as two separate variables in the data, meaning that it is possible to be white-hispanic, black-hispanic, etc.

  5. The maximization was carried out using the BHHH algorithm for numerical optimization. Other (Quasi-Newton) methods all provided very similar results. Analytic gradients used in the optimization were tested for accuracy against their numerical counterparts. Starting values were chosen based on the Heckman Selection model where the marginal product equation from the present model served as the selection equation in the Heckman procedure. Results are highly robust to different choices of starting values (e.g. random starting values).

  6. When attempting to test for the overall significance of the model there are two things to consider: First, the hypothesis that all the regression coefficients except the intercepts (and variances) are zero is meaningless here when they include γ; in fact, I require γ < 0 for identification. A technical problem is that the Jacobian of the transformation term in the log-likelihood is infinite whenever γ = 0, but this is merely a symptom of the fact that the first order condition of the model only has a unique solution when γ ≠ 0.

  7. A Smearing Estimate along the lines of Duan (1983) is often the preferred method of computing the expected response after fitting linear regression models with transformed scales. However, since the left hand side of the specification is not directly observed (though the shadow wage follows a linear specification in the parameters, the model is not a linear regression per se), these methods do not apply here.

  8. All differences in this paper are tested using a Welch’s t-test in order to account for the possibility of unequal variances of the two samples.

  9. One potentially strong assumption has to be made here: The marginal productivites of different types of household work is assumed to be identical. Ideally, each activity such as vacuuming, taking out the garbage, or cooking would be modeled separately and have different marginal productivities associated with them. Due to the multitude of different possible activities and the design of the model used in this study this is not possible.

  10. The dollar amount of food stamps received by the household is available for the month of November preceding the December CPS interview. Since that amount is reported as a monthly total, I divide by 30 in order to obtain a per diem value.

  11. I thank a referee for this insight.

References

  • Aguiar, M., & Hurst, E. (2007). Measuring trends in leisure: The allocation of time over five decades. Quarterly Journal of Economics, 122(3), 969–1006.

    Article  Google Scholar 

  • Becker, G. S., (1965). A theory of the allocation of time. The Economic Journal, 75(299), 493–517.

    Article  Google Scholar 

  • Bollinger, C., & David, M. (1997). Modeling discrete choice with response error: Food stamp participation. Journal of the American Statistical Association, 92(439), 827–835.

    Google Scholar 

  • Bollinger, C., & David, M. (2001). Estimation with response error and nonresponse. Journal of Business and Economic Statistics, 19(2), 129–141.

    Article  Google Scholar 

  • Davis, G. C., & You, W. (2009). The time cost of food at home: General and food stamp participant profiles. Applied Economics, pages 1–16: iFirst.

  • Davis, G. C., & You, W. (2010). The thrifty food plan is not thrifty when labor cost is considered. Journal of Nutrition, 140(4), jn.109.119594.

  • Duan, N. (1983). Smearing estimate: A nonparametric retransformation method. Journal of the American Statistical Association, 78(383), 605–610.

    Google Scholar 

  • Fraker, T., Devaney, B., & Cavin, E. (1986). An evaluation of the effect of cashing out food stamps on food expenditures. The American Economic Review, 76(2), 230–234.

    Google Scholar 

  • Gronau, R. (1977). Leisure, home production, and work—the theory of the allocation of time revisited. The Journal of Political Economy, 85(6), 1099–1123.

    Article  Google Scholar 

  • Hamermesh, D. S. (2010). Incentives, time use and BMI: The roles of eating, grazing and goods. Economics & Human Biology, 8(1), 2–15.

    Article  Google Scholar 

  • Heckman, J. (1974). Shadow prices, market wages, and labor supply. Econometrica, 42(4), 679–694.

    Article  Google Scholar 

  • Kiker, B. F., & de Oliveira, M. M. (1990). Estimation and valuation of non-leisure time. Oxford Bulletin of Economics and Statistics, 52(2), 115–41.

    Article  Google Scholar 

  • Meyer, B., & Goerge, R. (2010). Errors in survey reporting and imputation and their effects on estimates of food stamp program participation. Working paper, University of Chicago.

  • Meyer, B. D., & Sullivan, J. X. (2007). Reporting bias in studies of the food stamp program. Working Paper 08.01, Harris School of Public Policy, University of Chicago.

  • Meyer, B. D., & Sullivan, J. X. (2008). Changes in the consumption, income, and Well-Being of single mother headed families. American Economic Review, 98(5), 2221–2241.

    Article  Google Scholar 

  • Moffitt, R. (1983). An economic model of welfare stigma. The American Economic Review, 73(5), 1023–1035.

    Google Scholar 

  • Moffitt, R. (1992). Incentive effects of the US welfare system: A review. Journal of Economic Literature, 30(1), 1–61.

    Google Scholar 

  • Rose, D. (2007). Food stamps, the thrifty food plan, and meal preparation: The importance of the time dimension for US nutrition policy. Journal of Nutrition Education and Behavior, 39(4), 226–232.

    Article  Google Scholar 

  • United States Department of Agriculture. (2010). Supplemental nutrition assistance program. http://www.fns.usda.gov/pd/15SNAPpartPP.htm.

  • Vickery, C. (1977). The Time-Poor: A new look at poverty. The Journal of Human Resources, 12(1), 27.

    Article  Google Scholar 

Download references

Acknowledgments

I would like to thank Naci Mocan, Kaj Gittings, seminar participants at Louisiana State University, two anonymous referees, as well as the editor for their valuable comments. Any remaining errors are my own.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Christian Raschke.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Raschke, C. Food stamps and the time cost of food preparation. Rev Econ Household 10, 259–275 (2012). https://doi.org/10.1007/s11150-011-9128-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11150-011-9128-3

Keywords

JEL Classification

Navigation