proper unit of observation to compare against the consumer expenditure survey

Hi. We would like to link the average spend per consumer unit from the BLS Consumer Expenditure Survey (CE) with ‘household’ counts from the American Community Survey.

From our understanding, the consumer unit is more granular than a typical “household”, as a single household can contain multiple families. This would lead us to consider counting the number of families in the ACS rather than households.

However, we noticed that in the CE published tables, the weighted count of CUs is around 120-130 million. This is much closer to the number of households in the US than to the number of families measured from the household-weighted tabulations from the ACS-- which our calculations suggest about 270 million.

Do you have any suggestions on the most comparable unit of observation (i.e. family or household) to the CE consumer unit? On a related note, would the household or person weights be more appropriate to achieve similar counts as the CE?

Thank you!


I’m not sure we (IPUMS) are the best people ask about this comparison. From what I can tell, from the BLS FAQs, a consumer unit is defined by one of three definitions. (1) all members of a particular household who are related by blood, marriage, adoption, or other legal arrangements; (2) a person living alone or sharing a household with others or living as a roomer in a private home or lodging house or in permanent living quarters in a hotel or motel, but who is financially independent; or (3) two or more persons living together who use their incomes to make joint expenditure decisions.

The key detail when defining a consumer unit seem to be financial independence, which is determined by spending behavior with regard to the three major expense categories: housing, food, and other living expenses. Therefore, to be considered financially independent, the respondent must provide at least two of the three major expenditure categories, either entirely or in part. This seems to be similar, but slightly different, to the way the Census Bureau defines families within households.

I’d encourage you to ask the BLS about their recommendations for this comparison.