F. Thomas Juster

An Overview of the Health and Retirement Study

This paper examines the scientific, public policy, and organizational background out of which the Health and Retirement Study emerged. It describes the evolution of the major parameters of the survey and the unique planning structure designed to ensure that the substantive insights of the research community were fully reflected in the content of the database, highlights key survey innovations contained in the HRS, and provides a preliminary assessment of the quality of the data as reflected by sample size, sample composition, response rate, and survey content. The paper also describes the several types of administrative data that are expected to be added to the HRS data: earnings and benefits from Social Security files, and health insurance and pension data from the employers of survey respondents.

The measurement and structure of household wealth

This paper deals with methodological issues that arise in measuring household wealth. Two prominent American household surveys--the PSID and SCF--rely on different methodological approaches to the measurement of household wealth. In particular, SCF oversamples high-income households and has a far more extensive set of questions. In the top one percent of the wealth distribution, better measures of wealth are related to over- sampling of very wealthy households and the number of questions that are asked. However, one can characterize total household wealth holdings for the overwhelming majority of households with a relatively moderate number of questions. When successive waves of wealth modules are used to compute savings, the verdict on quality is more cautious, in part due to the inherently lartger role measurement error plays in any first difference formulation.

Improving the Quality of Economic Data: Lessons from the HRS and AHEAD

Missing data are an increasingly important problem in economic surveys, especially when trying to measure household wealth. However, some relatively simple new survey methods such as follow-up brackets appear to appreciably improve the quality of household economic data. Brackets represent partial responses to asset questions and apparently significantly reduce item nonresponse. Brackets also provide a remedy to deal with nonignorable nonresponse bias, a critical problem with economic survey data.

Well–Being: Concepts and Measures

Well–being, or quality of life, is a continuing goal for individuals and a major criterion for the evaluation of governments and societies. As a research concept, however, it has been marked by persisting problems of definition and measurement and by uncertainties about its changing pattern over the life course. In this article, these issues are discussed; the concept of well–being is “unpacked” and a model of stocks and flows is described as applicable to the analysis of well–being. Finally, the concept of resilience is proposed as important for research on well–being in relation to age.

The Decline in Household Saving and the Wealth Effect

Using a unique set of household level panel data, we estimate the effect of capital gains on saving by asset type, controlling for observable and unobservable household specific fixed effects. The results suggest that the decline in the personal saving rate since 1984 is largely due to the significant capital gains in corporate equities experienced over this period. Over five-year periods, the effect of capital gains in corporate equities on saving is substantially larger than the effect of capital gains in housing or other assets. Failure to differentiate wealth affects across asset types results in a significant understatement or overstatement of the size of their impact, depending on the asset.

An Assessment of Alternative Measures of Time Use

Although time use has received much attention by social scientists as an index of resource allocation and social relations across groups, only a few studies have carefully assessed the relative strengths and weaknesses of the existing methods of measuring time use: time diary (TD), stylized (S) respondent report, and experiential sampling method (ESM). We note the varying degree of biases that arise in part from the extent of detail in the information collected by the three methods. Using findings from our analysis of the structure of these methods, we hypothesize that there are empirical exceptions to previously reported common findings that TD provides less biased information on time use than does S—namely (a) when labor market workers report their time spent on labor market work, and (b) when the historical trend in time, rather than the absolute level, is studied. Empirical results confirm our prediction and show that, among individuals who work regularly, TD and S estimates of labor market work hours reported by the same persons correspond closely to one another. In addition, when assessing historical trends, TD and S values correspond closely to one another, although TDs yield some inexplicable deviations from the trend even when the sample and the codes are carefully standardized. We also provide notes on a strategy of standardization for diary codes that are distinct across historical or national contexts.

Rethinking utility theory

Economists have had a long-standing concern with understanding and modeling the behavior of consumers, both in the narrow sense of their behavior as buyers of goods and services in product markets and as suppliers of services in labor markets, and in the broader sense of their collective purchases of public goods-schools and roads, air and water quality, national security, etc. Analysis of these issue has been cast as a constrained optimization problem: Consumers are perceived as choosing a particular bundle of goods and services, or a combination of work and leisure hours, or a set of taxes and public goods, that represents the optimum mix subject to the constraints of income and prices. Optimum simply means that no change can improve matters, and what is being optimized is utility or satisfaction. This paper provides a brief history of the development of utility theory, and suggests a reconceptualization of the basic sources of utility. The new formulation extends the range of interesting and important phenomena encompassed by the theory, reexamines the role of goods and services in producing utility, and simplifies the conceptual structure-at the cost of complicating the measurement problem as well as the analytic properties of the system. Some recent data reflecting the new conceptual structure are examined, and we note some implications of the theory and data for both scientific and policy issues.

Economic Status Measures in the Health and Retirement Study

Variables measuring economic status represent a straightforward but crucial part of Round 1 of the Health and Retirement Study (HRS). In addition to studies focusing directly on the economic well-being of this cohort of the population, the economic status variables are likely to be inputs into other analyses that focus on retirement and savings behavior, on variations in health status, on intrafamily transfers and support, and on poverty status. As a first look at these data, this paper has several modest goals: to offer a flavor for the major economic status variables on the HRS, to indicate some preliminary analysis of the quality of the data, and to take a preliminary look at the interrelationships among economic status measures like income and wealth and other important variables such as health status, pension rights, and health insurance coverage.

Enhancing the Quality of Data on Income Recent Innovations from the HRS

This paper evaluates two survey innovations introduced in the HRS that aimed to improve income measurement. The innovations are (1) the integration of questions for income and wealth and (2) matching the periodicity over which income questions are asked to the typical way such income is received. Both innovations had significant impacts in improving the quality of income reports. For example, the integration of income questions into the asset module produced in HRS an across-wave 63 percent increase in the amount of income derived from financial assets, real estate investments and farm and business equity. Similarly, asking respondents to answer using a time interval consistent with how income is received substantially improved the quality of reports on social security income. Fortunately, we also suggest ways that these innovations can be introduced into other major social science surveys.

The role of panel studies in research on economic behavior

The analytic and monetary costs and benefits of panel surveys are assessed in light of experiences from the Panel Study of Income Dynamics, an l&year panel swey on the economic status and behavior of the U.S. population. The analytic benefits of panel are for- midable, ranging from description of gross change to various analytic advantages of continuous and discrete time modelling. Analytic costs such as the conditioning of responses in subsequent participation or nonresponse bias are possible in panel surveys, but their effects can be minimized with proper data collection procedures and analytic adjustments. Surprisingly, the monetary costs of panel surveys are less than the costs of comparable repeated cross-sectional surveys.