« Back to Results

Intra-household Allocations: Advances in Identification and Recent Applications

Paper Session

Sunday, Jan. 6, 2019 8:00 AM - 10:00 AM

Atlanta Marriott Marquis, L506
Hosted By: Econometric Society
  • Chair: Denni Tommasi, Monash University

Identification of Random Resource Shares in Collective Households Without Preference Similarity Restrictions

Geoffrey Dunbar
Bank of Canada
Arthur Lewbel
Boston College
Krishna Pendakur
Simon Fraser University


Resource shares, defined as the fraction of total household spending going to each person in a household, are important for assessing individual material well-being, inequality and poverty. They are difficult to identify because consumption is measured typically at the household level, and many goods are jointly consumed, so that individual level consumption in multi-person households is not directly observed. We consider random resource shares, which vary across observationally identical households. We provide theorems that identify the distribution of random resource shares across households, including children's shares. We also provide a new method of identifying the level of fixed or random resource shares that does not require previously needed preference similarity restrictions or marriage market assumptions. Our results can be applied to data with or without price variation. We apply our results to households in Malawi, estimating the distributions of child and of female poverty across households.

Sharing the Pie: Undernutrition, Intra-household Allocation, and Poverty

Caitlin Brown
Central European University
Rossella Calvi
Rice University
Jacob Penglase
Boston College and University of Bordeaux


Anti-poverty policies often assume that targeting poor households is effective in reaching poor individuals. However, intra-household inequality may mean many poor individuals reside in non-poor households. Using Bangladeshi data, we show that undernourished individuals are spread across the household per-capita expenditure distribution. We then quantify the extent of food and total consumption inequality within families. Based on a collective model, we develop a new methodology to compute individual-level poverty rates that account for intra-household inequality. We show that women, children, and the elderly face significant probabilities of living in poverty even in households with per-capita expenditure above the poverty threshold.

The Engel Curves of Noncooperative Households

Pierre-André Chiappori
Columbia University
Jesse Naidoo
University of Pretoria


We provide a set of necessary and sufficient conditions for a system of Engel curves to have been generated by a noncooperative model of family behavior. These conditions fully characterize the local behavior of household-level consumption in the cross-section, i.e. as a function of total income and distribution factors. In this setting, any demand system compatible with a noncooperative model is also compatible with a collective model, but the converse is not true. We describe how these nested conditions may be tested using standard instrumental-variables strategies.

Validating the Collective Model of Household Consumption Using Direct Evidence on Sharing

Olivier B. Bargain
Bordeaux University
Guy Lacroix
University Laval
Luca Tiberti
University Laval and PEP


Collective models have been designed to perform welfare analysis at the individual level but have rarely been operationalized for that purpose. A recent literature suggests ways to do so using assignable goods and assuming the stability of individual preferences across household types. We suggest an original validation of this approach. We exploit a unique dataset from Bangladesh, in which the detailed expenditure on private goods by each family member is collected. Thus, we can derive `observed' resource sharing within families and compare it with the resource allocation predicted by collective models of consumption. Sharing between parents and children is well predicted on average (and for demographic subgroups), and tends to validate the Rothbarth approach. Sharing among adults is less accurate, which reflects the more fragile identification in this case. Results depend on the identifying good: clothing provides the best fit while other nonfood goods or protein/dairy consumption perform poorly. Models
identified on clothing provide very accurate measures of inequality and poverty based on individual resources. These results are particularly important in the context of poor countries where (i) within-household inequality is potentially large, (ii) child poverty is possibly
underestimated and (iii) pro-boy discrimination prevails.
George-Levi Gayle
Washington University-St. Louis
Ethan Ligon
University of California-Berkeley
Alexandros Theloudis
Luxembourg Institute of Socio-Economic Research (LISER) and University College London
Denni Tommasi
Monash University
JEL Classifications
  • D1 - Household Behavior and Family Economics
  • I3 - Welfare, Well-Being, and Poverty