The Panel Study of Income Dynamics (PSID) is justly considered a cornerstone of empirically-based social science research. With thirty-plus years of data on American families, it is one of the most widely used social science data sets in the world.
By the end of 2003, the PSID will have surveyed the economic and social well-being of over 65,000 individuals nationwide, spanning as much as 36 years of their lives. This abundance of knowledge has compelled both researchers and policy-makers to confront and learn from the dynamic economic and social processes affecting our lives.
The PSID was created in 1968 by research scientists Frank P. Stafford, director of the Institute for Social Research (ISR) at the University of Michigan, and James Morgan, former director of ISR's Survey Research Center (SRC).
The study has been the recipient of major NSF support via a continuing grant since the mid-1980's. The long-running study has produced a wealth of valuable knowledge, including both the intuitive and the surprising.
It Runs in the Family
The PSID is the only longitudinal database that includes data on wealth across
all age groups. In one study using PSID wealth statistics, Stafford and ISR
researcher Ngina S.
Chiteji found that young adults raised in families that owned bank accounts
or stocks are more likely to have these kinds of financial assets themselves
than those adults whose families did not have such resources.
"Our study results suggest the presence of a relationship that is quite intuitive,
but one that has not been previously analyzed," Dr. Stafford says. "We
find substantial intergenerational effects in that children's portfolio choices
are influenced a great deal by the holdings of their parents."1
Time is Relative
Time diary studies using PSID data show that children in two-parent families spend more than twice as much time directly interacting with biological or adoptive parents as children in one-parent families, and that in the two parent situation, half of this time is spent with both parents simultaneously.2 See Chart at right
Rich or Poor?
In perhaps one of its most notable achievements, PSID data has transformed research on poverty from a static view of "poor" and "rich" to
a dynamic one in which many families experience episodes of relative poverty or affluence.
The PSID data have influenced national welfare policy formulation.
Home is Where the Mortgage Is
PSID research results indicate that the huge differences between minority and non-minority wealth accumulation are driven in part by home ownership, or lack thereof. In their study on the likelihood of home ownership, researchers Kerwin Kofi Charles and Erik Hurst examine racial differences with regard to mortgage application and approval. Using PSID data, the study tracked African American and white renters over a period of time and showed that not only are African Americans less likely to apply for a mortgage, they are less likely to be approved for a mortgage if they apply.3
Not the Stigma It Used To Be
PSID data enabled researchers to show that bankruptcy has increased as a result of two things: 1) more families are overcommitting to debt and thereby erasing a larger debt burden through bankruptcy, and 2) over time, there has been reduced social disapproval of, and reduced stigma to, declaring bankruptcy. The results imply that private debt limits the effectiveness of monetary policy: in a recession, people respond to lower interest rates by borrowing on their homes, but they could experience repayment difficulties and bankruptcy if the recession is protracted. 4
Results forthcoming from current research based on the PSID reveal:
- Children's well-being is more strongly shaped by persistent poverty and persistently high income than by a single year's income. This relationship is to be expected, but required data from the PSID to demonstrate. The results suggest that public policies should be targeted to those children in persistently difficult conditions.5
- Charitable giving is strongly influenced by expected variables such as income and wealth, but data from the PSID show the permanent or multiyear average income is more important than income from a single year. Beyond that, education has a very strong influence on giving both time and money.6
- Employer-based health insurance coverage is less likely to be available to men working in industries and occupations with high labor turnover (e.g. construction or retail). But the unavailability of health insurance is overcome by marriage, since marriage is a signal to providers that an individual is a lower risk. Moreover, since employer-based health insurance covers the family, the men in high turnover industries but married to wives with good jobs (earnings of $25,000 per year or more) have no particular problem in getting health coverage.7
By periodically reinventing itself, the PSID retains its value as a viable and robust tool. More than 10,000 customized data sets are delivered each year to researchers via its Internet Data Center. Since 1968, over 2,000 journal articles, books and chapters, dissertations and other works have drawn on PSID data, as have undergraduate courses at a number of universities and publicly-available online learning resources. With its on-going commitment to high-quality data, the PSID is well poised to maintain its position as the premier research tool of social science scholars.
For more information, please see:
Panel Study of Income Dynamics (PSID)
1 "Asset Ownership Across Generations", Ngina S. Chiteji and Frank P. Stafford, manuscript presented at the meetings of the American Economic Association, January, 1999.
2 "The Distribution of Children's Developmental Resources," Frank P. Stafford and W. Jean Yeung, Research manuscript presented at the International Research Consortium in the Economics of Time Use in the Household in St. Gerlach, The Netherlands, May 26-27, 2003.
3 "The Transition To Home Ownership And The Black-White Wealth Gap", Kerwin Kofi Charles, Erik Hurst. The Review of Economics & Statistics, May 2002, 84(2), 281-297.
4 "The Household Bankruptcy Decision," Scott Fay, Erik Hurst and Michelle J.White, American Economic Review, vol. 92:3, June 2002, pp. 706-718
5 "Evolution and Change in Family Income, Wealth and Health: The Panel Study of Income Dynamics, 1968-2000 and Beyond," Greg J. Duncan, Sandra L. Hofferth, and Frank P. Stafford, Chapter 6 from forthcoming publication from the Institute for Social Research (ISR).
6 First-round results from national study on philanthropy by the Center on Philanthropy at Indiana University and the University of Michigan's Institute for Social Research (ISR).
7 "Distribution of health insurance coverage by occupational characteristic: Results from the Panel Study of Income Dynamics," Stafford, Frank P. Working paper, University of Michigan, May 2003.
This research is supported by the Economics Program with contributions from the Sociology Program, the Methodology, Measurement and Statistics Program and other federal agencies.