Identification and estimation of a class of household production models

Abstract

We consider a class of household production models characterized by a dichotomy property. In these models the amount of time spent on household production does not depend on the household utility function, conditional on household members having a paid job. We analyse the (non-parametric) identifiability of the production function and the so-called jointness function (a function describing which part of household production time is counted as pure leisure). It is shown that the models are identified in the two-adult case, but not in the single-adult case. We present an empirical application to Swedish time-allocation data. The estimates satisfy regularity conditions that were violated in previous studies and pass various specification tests. For this data set we find that male and female home production time are q-substitutes. Contract/grant sponsor: Netherlands Organization for Scientific Research (NWO). 1 The value of the output of home production can be estimated by the total value of the production factors used (input approach). Normally, however, we do not have information about all of these inputs (e.g. auxiliary goods). But even if we could measure all the input quantities, this approach would systematically ignore any returns to scale and therefore the value added in home production. Fitzgerald et al. (1996) evaluated the value of home production by measuring output in physical units and attributing market prices to them (output approach.) Since households choose not to purchase these market alternatives, this approach only provides an upper bound to the value of home production. 2 In a recent theoretical analysis Apps and Rees (1997) and Chiappori (1997) stress the importance of household production in analysing intrahousehold behaviour.

Spillovers from High-Skill Consumption to Low-Skill Labor Markets

“…These shortcomings are clear in the case of childcare. See Graham and Green (1984) and Kerkhofs and Kooreman (2003) for attempts to address them. These problems do not affect our basic results, however, as long as there is a nontrivial number of activities for which the assumptions are plausible.…”

The Role of Preferences and Opportunity Costs in Determining the Time Allocated to Housework

“…Kerkhofs and Kooreman (2003) and Gørtz (2011) estimate household production models that allow for the possibility of such 'process benefits' using Swedish and Danish time use data respectively. Both papers include time spent on a wide array of activities in their measure of housework -including the mundane tasks of cleaning and laundry as well as cooking, gardening, and do-it-yourself work that often constitute hobbies.…”

“…Both papers include time spent on a wide array of activities in their measure of housework -including the mundane tasks of cleaning and laundry as well as cooking, gardening, and do-it-yourself work that often constitute hobbies. Kerkhofs and Kooreman (2003) even include childcare which is well known to include a leisure component.…”

Nonmarket Household Time and the Cost of Children

“…This shortcoming seems particularly disturbing in the case of childcare." Graham and Green (1984) and Kerkhofs and Kooreman (2003) suggest some solutions to the restrictions pointed out by Pollak and Wachter (1975), that allow for the identification of household production functions, but they impose other restrictions. Such identification problems pertain both the estimation of econometric demand systems and the calibration approach of macroeconomists, as Gronau (2006) explains in detail.…”