Click for next page ( 238

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 237
APPENDIX B Data Needs and Research Agenda DATA NEEDS Adequate data and sound research are indispensable foundations for pol- icy making. Throughout this report we mainly used data collected by gov- ernment agencies, supplemented by surveys done by nonprofit institutions and, to a much lesser extent, surveys by businesses and business associa- tions. It is to be hoped that, despite tight budgets, government data collec- tion will not only be continued, but will also be expanded to include more data on individuals, families, and firms, which are now all too scarce. Such efforts need to be on a substantial scale in order to avoid problems of selection bias. Nonprofit institutions also need adequate funds to pioneer new ideas and innovative methods. To the extent that rules of confiden- tiality permit, data collected from employers and unions should be made available to the public in a timely fashion. It would also be useful if private firms could find ways to share the information they gather with each other and make that information available to researchers. Because most of our suggestions are for additions to work that has al- ready proven to be fruitful, the payoff can be expected to be large in rela- tion to relatively modest costs. Thus, the data collection efforts we propose should be attractive even in this era of budget cutting. Skimping on data collection is a false economy, because the costs of shaping policies without adequate information are likely to greatly exceed any short-term savings. Data on Individuals and Families As discussed in this report, understanding of work and family relation- ships has been substantially enhanced by the availability of the relatively 237

OCR for page 237
238 APPENDIX B recent longitudinal data sets, which are particularly well suited to studying the impact of changes. They include the widely used Michigan Panel Study of Income Dynamics, the Ohio State University National Longitudinal Surveys, the Census Bureau's Survey of Income and Program Participa- tion, and the University of Wisconsin's new National Survey of Families and Households. Important contributions have also been made by one-time surveys that focused on particular issues, such as the Long- Term Care Survey, and particular populations, such as the National Survey of Black Americans and the National Survey of Americans of Mexican Descent. Because the federal government is best equipped to carry out large-scale, long-term data collection, it should consider institu- tionalizing some of these projects initiated by research organizations. All of these surveys cover basic income data and some household characteristics. In addition, new initiatives are needed in the areas des- cribed below. One subject on which there is very little information is people's use of time. Far less is known about how people spend their time than about how they spend their money, although awareness is growing that in Amer- ican society there are many individuals and families who are "time poor" (Vickery, 1977~. Some researchers, notably Walker and Gauger (1973), Walker and Woods (1976), Robinson (1977a, 1977b), and Juster and Staf- ford (1985) have done pioneering work in this area. aids in evaluating the usefulness of various approaches to helping people manage the competing claims of work and family. Large-scale samples would provide information about the differences in use of time among individuals by gender, race, ethnicity, and class, as well as work and family status. Better data on time use would also help to develop more reliable esti- mates of the value of nonmarket time and, more particularly, nonmarket work. Such data are crucial for the development of a useful measure of real income, a far better proxy for well-being than money income alone. Ignoring nonmarket time and work leads to such unrealistic conclusions as considering as equal two families with the same number of persons and the same earnings, whether or not there is a full-time homemaker. Just as a great deal of data is collected on expenditures of money but not on expenditures of time, so are large amounts of information gathered about earnings but not about the rewards workers receive in the form of benefits. Asking questions about benefits, and perhaps also about working conditions, in addition to the usual ones about wages, will give a fuller picture of the total compensation employees receive. reveal whether differences in earnings overstate or understate differences in compensation. This is particularly important for a better understanding of the situation of low-wage workers, and hence of the situation of minorities Such information Such data would

OCR for page 237
DATA NEEDS AND RESEARCH AGENDA 239 and women, who are heavily represented among them. It would be particu- larly helpful for accurate estimates and fruitful analysis to have matched data from employers and employees. One issue involved in using money income as an indicator of value received that has been generally recognized is the importance of adjust- ments for the cost of living. Adjustments are not entirely satisfactory, however, when a single index is used, because it does not take into account that expenditure patterns vary considerably among individuals and groups. Thus, for instance, using the same index for the elderly as for the rest of the population does not reflect the fact that costs for health care and prescription drugs a much larger proportion of expenditures for older people than for the general population have been increasing much more rapidly than prices of most other goods and services. Use of separate cost- of-living indices for some population segments would significantly im- prove understanding of real cost-of-living trends. Finally, most data are collected only for individuals, not families, yet for many purposes the former are not adequate. A good illustration of this lack is when there is information on the number of persons who earn the minimum wage, but not on the total income of the families in which they live. Similarly, there are data on how many workers are covered by their own health insurance, but little about how many of them are or could also be covered as family members of another worker. Establishment-Level Data As detailed in Chapters 5 and 6, a fair amount of information has been collected on the extent to which public employees and workers in large- and medium-sized firms have access to various types of benefits. However, data for smaller firms, which employ a large and growing part of the labor force, are generally inadequate. For all types of establishments, little is known about utilization (as opposed to availability) of programs. Government and research organ- izations are most likely to play a major role in collecting the needed data, but businesses need to cooperate, at least, and might even take the initiative. In some cases, employers are already collecting such data for their own use. The value of the knowledge gained should more than out- weigh the costs for businesses so long as proper care is exercised with regard to issues of confidentiality. Such data would make it possible to relate the availability and use of programs to characteristics of both employers and employees, as well as to the laws governing that estab- lishment. Surveys of employers, such as that by Trzcinski (1989), are a promising beginning. In-depth case studies of individual companies also can be a rich source of useful information for initiating and eval

OCR for page 237
240 APPENDIX B uating policies and programs. Finally, more information on the presence and role of labor unions would add to our understanding of establishment- level data. A RESEARCH AGENDA Data, of course, are useful only to the extent that they are used in relevant research. Our suggestions for topics on which additional invest- igations would be especially useful cover three broad categories: the im- pact of employment on families, the impact of families on work perfor- mance, and the costs of employer benefit programs. Impact of Employment on Families The great increase in the proportion of women who are employed- particularly, the influx of mothers of young children into the labor force- as well as the growth in the number of female-headed families have led to a great deal of work examining the effects of these trends on women, men, and, especially, children. Nonetheless, there is much still to be learned. For the most part, only short-term outcomes have been examined, although, particularly for children, long-term outcomes are of great interest. Little work has been done on issues relating to the care of elderly and handi- capped family members. Relatively scant attention has been given to the changing role of men in the family and the effects of this development. Much of the research on the well-being of families has relied on subjec- tive perceptions, rather than on objective measures that might be expected to be more reliable. Another question that has been neglected is the pref- erences of workers and their families when given choices of different types and different amounts of benefits, as well as the effect of their choices on levels of stress and satisfaction. Among the main lessons learned from existing research are that well- being is not unidimensional, that outcomes are therefore often mixed, and that outcomes are generally not the same for all family members. These insights make clear that it is far too simplistic to ask, for example, if the effect of mothers' working is good or bad for children, let alone whether it is good or bad for families. In view of these complexities, it is not surpris- ing that much remains to be learned about policy interventions that are most likely to be successful in mitigating the problems of the growing number of women and men with dual roles. At the employer level, further research on both barriers to the adoption of family-related policies and ways to facilitate their adoption and imple- mentation is also strongly recommended. Such research would require the use of carefully selected samples of the labor force in private- and public

OCR for page 237
DATA NEEDS AND RESEARCH AGENDA 241 sector workplaces, both unionized and nonunionized, with information from management as well as employees. Such work would help to understand why some employers adopt new policies and others do not and would pro- vide insights on the dynamics of their successful implementation. For ex- ample, additional investigation of the effectiveness of employee assistance programs in addressing family issues would be very helpful. Among the most promising ways of pursuing such further research is systematic evaluation of programs that have already been implemented, whether voluntarily by individual enterprises, as a result of negotia- tions with unions, or because of government requirements in particular states or other countries. Studies of individual companies in the United States, such as that by Burden and Googins (1987), and in other countries, such as that by Galinsky (1989a), begin to provide this kind of information, although problems of confidentiality have been a barrier to more such work. In addition, demonstration projects, or social experiments, would be very useful in some instances. A thorough discussion of their advantages and limitations is found in Ferber and Hirsch (1982~. Such projects are likely to be very expensive; even so, labor unions, business associations, and governments may find them worthwhile because instituting full-fledged programs without adequate knowledge may be far more costly. Impact of Families on Work Performance Although much remains to be learned about the effects of work on families, even more research is needed about the effects of the family on work. To the extent that research has been done on this subject, it has focused on the relation of various characteristics of the worker to particu- lar types of behavior, including tardiness, absenteeism, work interruptions, and turnover and how such behavior can be influenced by various policies. While these variables have some interest in themselves, they are often used as proxies for the "bottom-line" issue of productivity, or how much a worker contributes to output and hence to revenue. As discussed in Chap- ter 6, this question has so far proven to be rather intractable, although there is a body of research concerned with productivity primarily among production workers. Work on this important subject would neces- sarily involve studies of individual enterprises with a focus on the effects of implementing innovations. Just as demonstration projects can be ex- pected to be useful for learning about the effects of workplace condi- tions and policies on families, so can they be useful for learning about the impact of family situations and family-oriented policies on work performance.

OCR for page 237
242 APPENDIX B Costs of Employer Programs As discussed in Chapter 1, estimates of direct costs to employers do not show who really bears the costs, which may be shifted, to a greater or lesser extent, to workers, consumers, or the public. How much of this can be done will vary from one business to another and is likely to vary over time even for the same enterprise. Unfortunately, there is little work to date that offers promise on this topic. Observations from existing programs, let alone demonstration projects, are not very helpful because so much depends on the context of the economy and the society. The record of public finance specialists in demonstrating how much shifting of tax bur- dens takes place has not been convincing; nevertheless, the topic is im- portant and should be studied. At the very least, experts can make it clear that one cannot simply assume that no shifting takes place. SUMMARY Our proposals for data and research are based on the assumption that it is possible to make significant progress in gaining more knowledge about, as well as understanding of, family-workplace interrelations at an expense that would be justified by the possibilities for improved policy making. Learning from programs that have been adopted by some companies, by individual states, or by other countries is often particularly expeditious. Even so, it is impossible to learn everything about the consequences of proposed programs. It is to be hoped that recognition of the limits of possible knowledge will act as a safeguard against large-scale implementa- tion when small-scale experiments and model programs are feasible. At the same time, businesses and governments should not fail to implement use- ful programs while waiting for perfect solutions. It is as important to count the price of inaction in the face of serious challenges as it is to avoid costly mistakes.