Data and Methodology
This study uses data from the Health and Retirement Study (HRS), a nationally representative survey of older Americans. In addition, we match the survey data to the restricted Social Security Administration Administrative Summary Earnings data. Starting in 2001, a random sample of 5,000 households was selected to participate in the Consumption and Activities Mail Survey, which asks households to report the amounts they spent on various categories of consumption. The CAMS has been administered biannually from 2001 to 2007, which allows us to track households born before 1947 and directly measure any consumption changes after children leave the household. Ideally, we would examine the spending patterns for younger households, at ages when the majority of people experience their children leaving home, but panel consumption data do not exist. However, the older households that comprise our sample are the ones for whom retirement savings are especially salient at the time when their children leave home. We divide respondents into three categories:
• those who either never had children or whose children moved out prior to 2001;
• those who had resident children throughout the period 2001 to 2007; and
• those whose children moved out between 2001 and 2007.
The first two categories are the control groups, since the household composition remains constant during the time period. 5 The third category contains the households of interest. If the household consumption smoothing assumption holds, then the change in total household consumption should be similar among all three groups, but the growth in percapita consumption would be significantly higher for the third group after children leave. If the per-capita consumption smoothing assumption holds, then the growth rate of total household consumption of the third group should be lower than that of the first two groups, while the growth in per-capita consumption of all three groups would be the same.
Results
To test the alternative theories, the analysis examines two separate types of consumption: housing and nondurable goods.
Housing
We start with the illustrative example of housing. Housing expenses include insurance, real estate taxes, maintenance, and standard expenses such as mortgage or rent, electricity, water, heat, and phone. Spending on housing is basically fixed, since the costs associated with owning the family home vary little with the number of bedrooms that are actually being used. This fact would suggest that, as long as families do not move, the three household types should exhibit similar rates of growth of household-level spending. But per-person spending of households whose kids move out should increase simply because the fixed expenses are shared among fewer individuals.
Non-durable Consumption
Non-durable consumption includes purchases of housekeeping supplies, personal care products, apparel, leisure and hobby items, vacations, vehicle insurance, food purchases (including dining out), and gasoline. These types of consumption involve fewer economies of scale and are therefore potentially the most responsive to changes in household size. Thus, this category of spending allows us to test whether households are smoothing total household or percapita consumption.
This figure [which we have not shown] shows that, between 2001 and 2007, households without children living at home were spending an average of $8,800 to $10,300 per person a year on non-durable goods. Note that their consumption was much higher than that of households with children still present, who spent only $4,700 to $5,800 per person. The cross-section pattern is the first indication that per-capita consumption may increase after children leave. The consumption of both of these types of households declined during this period. Households whose children left between 2002 and 2004, however, showed a marked increase in per-person non-durable consumption. In 2001, they consumed $5,100 per person, which is very similar to the amount spent by other households with children.
By 2007, after their children have left, these households spent $6,500 per person, almost $2,000 more than the average per person spending of those who have children. This descriptive analysis supports the assumption of household-level consumption smoothing, and provides evidence that households increase per-capita consumption when children fly the coop.
This brief is related to broader analysis from Coe and Webb (2010).
(For the complete brief - with diagrams - go to http://crr.bc.edu/images/stories/Briefs/IB_10-18.pdf)
Pages: 1 · 2
More Articles
- Facing Financial Ruin as Costs Soar for Elder Care
- Congressional Budget Office: Federal Budget Deficit Totals $1.4 Trillion in 2023; Annual Deficits Average $2.0 Trillion Over the 2024–2033 Period
- Supreme Court Surprises The Public in LGBTQ Ruling: What is Sex Discrimination?
- Federal Reserve Chair Jerome Powell Addresses Current Economic Issues: For Some, a Reversal of Economic Fortune
- “That’s Not the Government Calling: Protecting Seniors from the Social Security Impersonation Scam”; Isolating Their Victims by Instructing Them Not to Tell Anybody What is Going On
- GAO: A Comprehensive Re-evaluation Needed to Better Promote Future Retirement Security
- Another Turbulent Day on the Market: Three Reasons to Do Nothing
- Listen to What the Concerned Scientists Union States: Hurricane Michael Threatened Gulf Coast Homes and Military Bases: Update: Thomson Reuters Foundation Film: Home Beyond the Water
- Gender News: City of New York Agrees to Pay $20.8 Million to Settle Federal Discrimination Charges Made by Registered Nurses
- Stateline: Why Most States Are Struggling to Regulate Airbnb