National Academies Press: OpenBook
« Previous: 3 Key Research Issues
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 61
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 62
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 63
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 64
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 65
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 66
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 67
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 68
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 69
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 70
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 71
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 72
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 73
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 74
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 75
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 76
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 77
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 78
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 79
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 80
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 81
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 82
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 83
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 84
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 85
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 86
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 87
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 88
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 89
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 90
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 91
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 92
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 93
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 94
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 95
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 96
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 97
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 98
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 99
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 100
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 101
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 102
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 103
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 104
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 105
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 106
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 107
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 108
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 109
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 110
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 111
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 112
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 113
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 114
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 115
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 116
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 117
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 118
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 119
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 120
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 121
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 122
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 123
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 124
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 125
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 126
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 127
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 128
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 129
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 130
Suggested Citation:"4 Data Needs." National Research Council. 1997. Assessing Policies for Retirement Income: Needs for Data, Research, and Models. Washington, DC: The National Academies Press. doi: 10.17226/5420.
×
Page 131

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 Data Needs Within the next few years, policy debates about the retirement income security of current and future generations of Americans are likely to require a range of modeling capabilities with which to evaluate and project the likely effects of alternative policy proposals. However, as is clear from the preceding chapter, there are important gaps and uncertainties in what is known about the behaviors and processes that affect retirement income security. These gaps stem from deficiencies in available data, which hamper or preclude the development of robust analytical models and parameter estimates from them. In some cases, notably for employers, there are insufficient data with which to describe the distribution of relevant employer and employee characteristics, much less to support analysis of behavioral change over time. These deficiencies need to be remedied and the knowledge base further developed before it will be possible to construct reasonably adequate projection models with broad capabilities. Moreover, existing retirement-income-related projection models and the associated databases have many limitations and do not generally provide an adequate platform on which to develop improved models once new data and research knowledge become available (see Chapter 5 and Appendix D; see also Hollenbeck, 1995~. Thus, there is a great deal of work to do to prepare for the policy debate. With very tight budget constraints, the question is one of priorities. We conclude that agencies should devote the bulk of their limited resources over the next few years to data collection and analysis rather than making significant investments in large-scale projection models. This conclusion is based on our assessment that some of the gaps in needed data and basic research are so critical that projection models, no matter how elaborate or elegant, cannot compensate for them. An 61

62 ASSESSING POLICIES FOR RETIREMENT INCOME example is the failure of existing research to adequately explain observed savings patterns in the population. Moreover, past experience suggests that it takes more time to collect new data and analyze them than it does to build a projection model to use data and research in estimating the likely consequences of policy changes. There are more than a few instances in the history of policy analysis when models were built in a span of weeks or months. As an example, the prototype of the Carter admin- istration's welfare reform projection model, KGB, was completed in a few weeks (see Citro and Hanushek, 1991: 107-114~. It is very rare that needed new data can be obtained and analyzed sufficiently in so short a time, particularly if the data set is rich enough to be useful. A small-scale, quick-response survey of employers' health care costs was completed for use in the recent health care reform debate within 10 months from initial design to final output (Ponikowski, Scheible, and Wiatrowski, 1994), but its scope was very limited. More detailed information on employers' health care plans and costs that would have been useful, from a large survey for which the design work had begun in spring 1993, was still not avail- able by the end of 1996 (Hing et al., 1995~. THE LESSON FROM HEALTH CARE REFORM The experiences and reflections of policy analysts who provided estimates for the 1993-1994 health care reform debate underscore the panel's conclusion about giving priority to investments in data and research. Box 4-1 describes the major players in health care reform estimation and the models and databases they used.1 More lead time and prior investment would have facilitated the development of usable projection models for estimating the likely effects of alternative health care reform plans. Indeed, some timely investments that were made in model building were helpful (e.g., the extension of the TRIM2 model to simulate em- ployer-provided health care benefits). Conversely, inexperience with building health care projection models, particularly with a database not previously used for this purpose, was a handicap. That was the case, for example, for the Agency for Health Care Policy and Research (AHCPR), which based its new AHSIM model on the 1987 National Medical Expenditure Survey (NMES). However, the model builders themselves pointed to major difficulties that stemmed from the absence of critical data and research; see Box 4 2.2 Existing data were so inadequate that it was difficult to develop an agreed-upon "baseline" scenario that is, a representation of the current distribution of health insurance coverage, utilization of services, costs, and other characteristics of consumers, Information for this discussion and Box 4-1 comes from Bandeian and Lewin (1994), Bilheimer and Reischauer (1996), Citro and Hanushek (1991, esp. Chap. 5), Nichols (1996), Office of Technol- ogy Assessment (1993, 1994), Shells (1996), and interviews with analysts. 2See footnote 1.

DATA NEEDS 63 providers, and insurers let alone simulate the likely effects of alternative re- forms relative to the baseline. Bilheimer and Reischauer (1996:149), speaking from the Congressional Budget Office (CBO) experience, flatly concluded: "To construct a comprehensive picture of the health care system is impossible with today's databases. What is known must be pieced together from several inad- equate or dated surveys and sources." Also lacking was up-to-date research with which to estimate behavioral responses to changes in the health care system. Bilheimer and Reischauer (1996:152) noted that "such studies can credibly illuminate only the effects of marginal changes in the current environment. The effects of large, systemic changes that major health care reform proposals would generate are far outside the boundaries of knowledge that can be gleaned from existing economic re- search or even from social experiments." Nonetheless, they identified several areas in which better data about the current system would have made it possible to develop more credible estimates of the effects of reform proposals (see Box 4- 2; see also Bandeian and Lewin, 1994~. In the absence of key data and research, rough estimates based on very inadequate information or simply guesses were used for values of behavioral parameters, and no projection model, however complex or elegant, could com- pensate for the lacking information. Different models incorporated widely differ- ent assumptions in key areas, and consequently, there were significant differ- ences in estimates of the likely effects for the same reform plan (see Office of Technology Assessment, 1993, 1994~. Differences in databases for example, between the March 1994 Current Population Survey (CPS) and the 1987 NMES aged to 1993-1994 also contributed to differences in estimates. Moreover, in the heat of debate, it proved difficult, if not impossible, to develop new sources of needed information on a timely basis. Subsequently, and anticipating future health care policy debates, AHCPR and the National Center for Health Statistics (NCHS) are working to implement a major reorganization and expansion of health-related surveys that could meet many of the information requirements identified by participants in the 1993-1994 effort (Hunter and Arnett, 1996~. The picture is much the same for retirement-income-related policy analysis, namely, that key descriptive and analytical data with which to develop credible projections of the likely effects of current and alternative policies are missing or incomplete. As with health care reform, even the best data and analysis are unlikely to resolve the uncertainty associated with major policy changes, such as privatization of Social Security (which would resemble a system of universally mandated Individual Retirement Accounts), because there is no historical experi- ence on which to base any models.3 For example, an important issue about 3However, research on the experience of other countries with privatization schemes may help develop projections for a u.s. system.

64 ASSESSING POLICIES FOR RETIREMENT INCOME ............................................................................................................................. ^,,u,,m,~, ~ ~ -ion An, l, -up u ~ Or -m,e,,nl or ne,a,,,~ ,,a,,nu num,,a,,n privatization is whether it will increase or decrease personal saving. One can argue that privatization will educate people about saving and what it can do for them and thereby lead millions of people who now save little or nothing to save much more, in addition to their mandatory privatized accounts. But one can also plausibly argue that people will be more confident of actually obtaining payments from their dedicated personal accounts than they are of receiving Social Security benefits and thus will curtail other forms of saving (see Mitchell and Zeldes, 1996). Nonetheless, as with health care reform, filling key gaps in data and research knowledge can go a long way to make it possible to develop credible projections of the likely effects of many retirement-income-related policy alternatives. We urge that priority be given to strengthening the base of data and research for

DATA NEEDS 65 - ............................................................................................................................ ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: retirement income modeling through improvement of existing data sets when- ever possible and through new data collection when necessary, and including appropriate levels of funding for analytical research and validation. The remainder of this chapter addresses: the dimensions of databases that should be considered in designing and evaluating cost-effective retirement- income-related data collection systems, whether new or modified; issues in con- tinuing existing panel surveys of middle-aged and older people in order to pro- vide sufficient longitudinal observations for analysis of consumption, savings, and retirement behavior of individuals; issues in developing new and improved cross-sectional and panel data for employers and their workers in order to under- stand labor demand and employer decisions about pensions and other benefits; issues in linking administrative and survey data, which can be a cost-effective

66 ASSESSING POLICIES FOR RETIREMENT INCOME ............................................................................................................................. MA l D A e D e 1 A De ~-v-~---v^--l--^---~-^r-~ r--v--~ ~--~-~-~ ~ ~ ~-m-~ N ..... . . of nouseno at surly pm-w a ED a E-e so off ne u-~ meow ..... ... """"""""' ' ' ' """"h'' " " ' 'h" 'I'd""""' '' ' 'b' ''--' ''""""h'' ' '1' h"""' at' '' """'' ~ 'h'' ' Ith~'' ' ' """'' ti'l'i' ' ti'' ' ~ 't'h'' """"""""""""""""""" ........ .. ..... . . ..... ..... . . . . ......... . . . ... .. .................... .... . . ............... - tit i i 1 l 1 i I r :':':':':':':':':':':':':':':~:I':'E:V O=:':':':'E:':I'V :~:':':':LV' ':':':':'V :I':':':':W:I':'E:~.:L I :'t='I:::: WV :I':'n='E:'O':':':'~::I::~:I:::I:::::::~:I:::I:~:I: V : 1 - :~:~:I:::::::~:V::V ::~:1::~:~::::::: - V:~:I:~::::::::I:::I:~:V::~.:::::::::::: stat ES a' d ................. ........... .......... ..... ............... . ......... ................. .......... ...... ... ............................................................................................................................. E j E EKE i 1l · .- r f i -~- ~ ~- ~~- -.~. -~-~--~.- -~I-~-Ily~ l~l ~ l~lLl l w~l~ l~lvl~ ll w ........................................................................................................................................ ~\IO.E .~ .~ne. ~l-.-~.-t-. 1~1 EtO EaU ~ ':':':':':':':':':':':':':':'' :':':':':':':':':':':.': ':':':':':':':':':':'':':':':':,':':':':':':':':':':':':':':':r:':':':':':':':':':':':':':':':':' :':':':':':':':':':':':':':':':':':':':':':':':':':':'A':':':':':':':':':':':':':':':':':':':':': Ta-m-l-- y---slal' -s~ oT~ e-m-p- oye-es~ ^n~ e-mp- -oye-r---su-' ey---conc -u-cTeo~ ~y---T 1-e~ -e-a ~ 1~ ::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::: ::::::::: :::::::::::::::::::::::::::::::::::::::: .................................................. ~/ ........ ....... .. ~............... ... . f '~ '#-~-~ -'&~-6-~ ~ '--~-~-'&~-6-~ ~-'&##-l'#-~&~--'-'VI~ l'~l-~l'-'-'-~-'W'-'--'V-'&~ '~l'-~-'-' E-'~ /~ l~ e ~-eal~ E ~aE ~lnaE cing ^~mlE !sirmlon p~pam$ aE E'U'al"""'eSl!""""""""" ~ ' ''~''' ' ' '' ' ' '~' ~ 0'i~ ' E~ ' 6'~ '1~ ~t' t' -~ E~ t' ~ ~ ' 1 ~$ 1 ............................................................................................................................ ''''''''''''''" ' "'''''''' " ' ' ' ' ",''''' ' d' ''''i''' "' ' ''i' ' '''"'""""''' """" ' ' i'l'' b'l'' """t' ''""'' t' ' ' """" ''"""' ob' ' t

DATA NEEDS 67 - ............................................................................................................................ ............................................................................................................................. ................................................................................................................................................................................. :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ................................................................................................................................................................................ :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ............................................................................................ :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ........................................................................................................................................................................... ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

68 ASSESSING POLICIES FOR RETIREMENT INCOME ............................................................................................................................. ............................................................................................................................. means of obtaining high-quality measures of key variables with minimal added expenditure; and issues of data validation, internally and in comparison with other sources. DIMENSIONS OF DATABASES Databases differ on a number of dimensions, including source, reporting unit and universe, type and frequency, scope, size, data collection methodist, accuracy or validity, uncertainty, ease of use, level of aggregation, and cost; see Box 4-3. There are trade-offs to consider among these dimensions. For example, the level of uncertainty in survey responses can be reduced by increasing the sample size; however, such a decision will increase costs. Similarly, an expansion of the number and detail of survey questions will make a survey more useful for a wider

DATA NEEDS 69 range of purposes, but increase its cost and the burden it places on respondents. Such an expansion may also make the data more difficult for analysts to use. There are also trade-offs with regard to the use of administrative records instead of survey data. Administrative records are usually thought to be more accurate than survey responses. They are also relatively inexpensive to use for analysis because the costs of data collection and processing have already been incurred for administrative purposes. However, such records usually lack de- tailed content, and their content may change from year to year to reflect changes in program data requirements. Also, administrative records are not without errors (e.g., Social Security and Medicare records may have inaccurate information on whether individuals are still alive). Indeed, when comparing information for the same variable in an administrative records data set with a survey estimate, it is important to take account of likely errors in both sources and of differences in definitions and other features that could affect the comparison. Finally, adminis

70 trative records are often inaccessible to researchers because of concerns about maintaining confidentiality for individuals and other reporting units. Data sources are rarely satisfactory for both analytical and projection model- ing purposes on every dimension. In fact, analytical and projection models often use different types of data. For example, analytical models of individual behav- ior generally require rich longitudinal data from panel surveys, but models that project individual outcomes rarely use panel surveys as their primary database because of small sample sizes and restricted universes.4 Yet if the projection model database does not contain a similarly rich set of variables as were used to estimate key behavioral relationships in an analytical model, it will not be pos- sible to take advantage of the most advanced behavioral models. Instead, the behavioral relationships will have to be reestimated with a reduced variable set, or such procedures as statistical matching (see Cohen, l991b) will have to be used to impute needed variables to the primary database. (We discuss this issue further in Chapter 5.) ASSESSING POLICIES FOR RETIREMENT INCOME il PANEL DATA ON INDIVIDUALS An underlying theme throughout our report and the papers we commissioned (Hanushek and Maritato, 1996) is the need to understand how people reach their retirement years. What enters into decisions about working as people age? What are the implications of different employment paths for pension plan participation and the level of benefits received in retirement? How do government and em- ployer policies affect personal savings behavior and the ultimate wealth accumu- lations that influence both retirement decisions and well-being in retirement? Questions such as these emphasize two key issues that have implications for data collection and analysis. First, many of the antecedents of retirement out- comes are present long before any actual retirement decisions. Second, behavior that is related to policy often has a long time horizon, with individuals looking many years into the future as they make decisions. To obtain suitable data for analysis, it is essential to follow individuals over many years in order to under- stand their retirement behavior and outcomes. This central fact leads us to emphasize the development of panel surveys that obtain longitudinal data by interviewing the same individuals over time. Panel surveys, which have become increasingly common to study individual behavior, permit investigation of behavior that evolves and that has implications over long periods. Moreover, panel surveys provide a variety of ways for dealing with the heterogeneity across individuals that can complicate analyses based 4An exception is a recently developed public assistance model STEWARD (Simulation of Trends in Employment, Welfare, and Related Dynamics) which directly uses data from the Na- tional Longitudinal Survey of Youth (NLSY) to simulate the effects of welfare reform proposals on program participation (Jacobson and Czajka, 1994).

DATA NEEDS 71 solely on a cross-section of individuals. Finally, panel surveys can often permit corrections for measurement and observational errors because consistency checks for individuals over time can aid in separating errors from true changes for individuals. Of course, the need to follow the same individuals over long periods implies that a panel survey is likely to be expensive certainly more expensive than a one-time cross-sectional survey of equivalent size and perhaps more expensive than a repeated cross-sectional survey.5 Also, for cost reasons, it may be difficult in a panel survey to refresh the sample frequently enough to address such ques- tions as whether patterns of behavior remain the same for newer cohorts or to maintain representation of a changing population (e.g., to represent immigrants). The trade-offs often suggest the need for cross-sectional data collection. For example, we argue below for collecting data to understand employer behavior that is relevant for retirement income security, but we believe that the first step is to improve cross-sectional data. Although a panel may later be appropriate, the initial efforts which include learning about what data to collect and how and what the sampling frame should be would most appropriately be thought of as a cross-sectional effort. Also, there is a need for regularly updated descriptive information on trends in the characteristics of employers, work forces, and ben- efits that more efficiently comes from repeated cross-sections than from panels.6 Similarly, repeated, nationally representative cross-sectional surveys are needed to provide important data on trends in the population that are relevant to tracking and understanding retirement outcomes (e.g., trends in ages at retire- ment). Nonetheless, the central longitudinal data with which to analyze indi- vidual behavior and individual decisions should almost certainly be gathered through panels of individuals. Although cross-sectional surveys can use retro- spective questions to collect longitudinal information, such as employment and earnings histories (and in some cases this is done), the quality of retrospective information is much less, compared with panel surveys, because of recall and other errors, which may be large (see, e.g., Kennickell and Starr-McCluer, 1995~. Also, cross-sectional surveys are limited in the amount of retrospective informa- tion that they can collect due to considerations of respondent burden. Panel surveys, in contrast, can obtain a wealth of information with which to understand different life courses and retirement outcomes. 5Whether a panel survey is more or less expensive than a repeated cross-sectional survey with the same number of sample members is affected by many factors, such as frequency of interviews, costs of obtaining an interview (a panel survey may have higher costs to locate sample members but lower costs to obtain an interview once the sample member is located), and others. 6Pane1 surveys will provide consistent time series for a population as well if a new panel is introduced on a frequent basis, such as every year; however, costs will be prohibitive unless the size or length (or both) of each panel is reduced, which will, in turn, reduce the usefulness of each panel for longitudinal analysis.

72 ASSESSING POLICIES FOR RETIREMENT INCOME Features of Long-Term Panel Surveys Several completed and ongoing panel surveys sponsored by government agencies have made possible a wide range of retirement-income-related analyses. Table 4- 1 presents the basic features of major retirement-income-related panel surveys, which have followed or are intended to follow samples of individuals over long periods; see also Figure 4-1. (Appendix B provides more detailed descriptions for these and other relevant panel surveys of individuals.) In particular, the Retirement History Survey (RHS), sponsored by the Social Security Administration (SSA), and the National Longitudinal Surveys of Labor Market Experience (NLS), sponsored by the Bureau of Labor Statistics (BLS), have supported extensive research on labor supply behavior and the retirement decision. As of 1988, over 200 articles and reports were identified that drew on the 1969-1979 RHS (Smith, 1988~. An annotated bibliography of research con- ducted with the various NLS cohort panels over the period 1968-1995 fills a 2- inch volume and numbers 2,540 entries (Fahy, 1995~. However, surveys like the RHS that were initiated several decades ago do not contain the richness of data needed for research on retirement and savings decisions. Thus, the RHS, which followed a cohort of single men and women and married men (aged 58-63 in 1969) and their spouses over 10 years, lacks questions on respondents' expectations about such factors as their own life span. The RHS also lacks detailed information on pension plan coverage and health status and health care benefit coverage. It contains a fair amount of information about assets and expenditures, but no direct measure of total consumption. The NLS cohorts that began in the late 1960s (mature women, older men, young men, and young women) and the cohort of youth (young men and women) that began in 1979 (NLSY) also lack the full set of information that researchers would like. Boxes 4-4, 4-5, 4-6, and 4-7 show, respectively, the information in major retire- ment-income-related panel surveys about expectations, pension coverage, health status and health insurance coverage, and assets and expenditures. The older panel surveys also vary in the length of time with which they followed, or are continuing to follow, sample members. For example, the Na- tional Longitudinal Survey of Young Women has followed for over 25 years a sample of women who were aged 14-24 when first interviewed in 1968; they are now in their 40s. However, the companion National Longitudinal Survey of Young Men stopped following a sample of men who were aged 14-24 when first interviewed in 1966 after only 15 years. Hence, there are gaps in the extent of longitudinal data that are available across cohorts of men and women; see Table 4-1 and Figure 4-1 (below). Spe- cifically, there are no NLS surveys that cover men who are currently in their 40s, 50s, or 60s or that cover women who are currently in their 50s or 70s. A new Text continued on page 88.

73 0 x ca o c' A 0 3 R ~ ~ .~ ~ ~ ~ ~ ¢ ~ ~ ~O .00 ca =4 =4 , ~,= ~ ~ ~, a, c 3 ~R ~ R R ~ . ~. ~ . R O O R ~ ~ R Rat ·^ ·^ ~4 JO ~ 0 ;^ ~ ~ ~ ~ ~ ~ ~ O Cal ~ ~ ° ~ O ~ ca ~ ~ ~ O ~ O ~ ~ Vet ca ca E ~ ~ ~ ~ ~ a To ~ ~ Y a ° Y ~ I o D e ° R ~ ~ ~ ~ R ~ - ° ~e ~ C ~ ~ ~ u O Cal a' ~R ~ Y o ~ ~ ~ ~C E~ v ~¢ ~ ~ ¢ o ~ ~ ~¢ Z Z ~ ~ ~ Z

74 a' .= o ¢ EM ca sit o C) sit ~ o 3 · R . ¢ o ;^ ca C) V, ·0 .~ ~ .0 V, Ct I ~ ~ ·0 lo Us ~ ~ O ~ Cd .O ca ¢ ;^ sit V, O O O O Z Z Z Z . ~ ~ . ·^ ~ V, ~4= ca ~ ~O o ~3 i= A' 3 4 ~ ~ ~ . ~ ~ ~ o ~3 ~ ~ o ;^ ~ ~ ~ ~ ;^ ~ ;^ ~ ~ ~ 0 O0 0 0 ~o ;~. ~O ~ R O S~ , ~O S~ , ~O S~ , ~O S~ ,~ ~ ca ° ^ O ~Om o ~O ~ Ct O O ca ~ ca ·^ ~ ~.^ ', ~ ~ ~ ~O ~ ~ ~ ~ ~O 0 c 0 ° e ~ u a C.a ~j~ 8 o ~ c ~ ° ~ o 8 ^ ~ ^ ~ ~ >` ~ O c ~ ~ ~ ~o ~ ~ R ~.0 ~ ~ ~ R ~ ~ Z ~ V) O Z Z ~ V) ~ Z Z ~ ~ ~ Z Z ~ ~ ~ ~ ~ Z

75 ca o ~ · ~ ca ~,= ca ·= R ~ -c, R R ~r ~ R ·= Vim ;^ ~ Hi ~ ·0 ;^ ~ ~ ·0 O ;^ ~;~6 ;~ ~ ·^ ·4 ~ ·4= ~ ~4 I= ~ ~ U:) at,_ Emory O a ~ 8 ~ -~ ~ All,'== ~ ~ O ~ ~ ~ ~ ~ ~ O ~ O ~ ~ .0 ~ ~ ~ ~ 0 ~ ~ =0 =0 ~ ~ ~ '~ ~ so ~ 8 ;^ ~ O ;^ ca R ~ ~V) ~ R Ct .^ m x ¢ ~ O.4 ca O t,.4 ~ t,.4 ~ .. ~ .5 ;^ =4 ca ~ O =4 4= ~ o u, O ·~ R ~ t004 ~,4 .0 so HO sit · I= X 4= 4= O ~ w '~ ~ z ~ · ~ u:) ·= ~o w ca ·~ R s-. ¢ t004 .~ ~ c) ca V, ~ ~ W W O ~ ~ O ¢^ t004 .. ~ C) E~ Z O ca t.4 o w ~ .= W W w ~ W O ca ~W ~ ~ ;~. o ~n _' W O ~W 4= ca an O .= O C) ¢ ·= O C) ¢ t.4 V, ~ 1 ~ oo W . s~ W ._4 ~ W ~ . O C=) ~ W ~ .0 ¢ V, ~ O ca W o C)

76 85 r an 75 7a 6C <~, 55 5C 45 cog </' 4C a 35 30 an 15 ASSESSING POLICIES FOR RETIREMENT INCOME a. Women 10 5 1 966 1 1 1 1 1 1 1 1 1 1968 1972 1976 1980 1984 1988 1992 1996 2000 Year of Data FIGURE 4-1 Ages of participants in retirement-income-related panel surveys. NOTES: AHEAD, Asset and Health Dynamics Among the Oldest Old; HRS, Health and Retirement Survey; NLS-MW, National Longitudinal Survey of Mature Women; NLS- OM, National Longitudinal Survey of Older Men; NLS-YM, National Longitudinal Sur- vey of Young Men; NLS-YW, National Longitudinal Survey of Young Women; NLSY, National Longitudinal Survey of Youth (Young Men and Women); RHS, Retirement History Survey. See Table 4-1 for frequency of data collection for each survey; inter- views after 1996 are subject to provision of funding.

DATA NEEDS 85 So-b. Men 75- 70 65 60 55- Q ~ 50 77 45 40 35 tan ~................................ ~........................................................................................ .,.,., ~""""'~"""""""""""""""""""""""""""""""""""""""""""""'~ ....... ~ ................................................................................... ............................................... ~ . ~ ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ~^~\ ~ \~5 V ..................................... ....................... , - - ~e\~`e~\e hem ~- ~ A........ ......................... ~5 ~ . \~....................................................................... ............................................................................................................ .................................................. ,,,,, o`5 gem 10 _ 5 1 966 , ~ ~ 1968 1972 1976 1980 1984 ~~~ ~~~ Year of Data 1988 1992 1996 2000 aSingle women and spouses of married men in age cohort. bNo upper age limit; 4.5 percent of female respondents in the initial AHEAD interview in 1993-1994 were aged 90 or older (Hurd et al., 1994:Table 5~. CNo upper age limit; 2.6 percent of male respondents in the initial AHEAD interview in 1993-1994 were aged 90 or older (Hurd et al., 1994:Table 5~.

78 ASSESSING POLICIES FOR RETIREMENT INCOME ............................................................................................................................. Wh h Id t h s ~--1- - t ff~ .................... ................ ............... - . ............ ...... ............. ....... ................ ..... .... ..... - 1 i h Ia d b ti ti e .................................................... ................................................. .......................... ''''W."""'h""e't2h"'e-2-"'''' i'' ' 'i'' ' i' '' i'''' ''''' ' it''' '' ' ~ ~ ill i $i i 1n $ -- -- -- -it-::--:- ::::~:~:~:::: :.'.'.:':':-A- it' ' '- '' ~''':':':.'~IV'.':'::~ --:: :- :.:~:.:::.::::::.:.:::.::::::.:::.: ~::::::::.:::V:~ :V-: - : -: :-:-:-::::::::::::: :::::::::::::: :: :: :::::::::::::::::::::::::::::::::-::::::-::::::::::::::::::::::.::::::::::: :::::::::.:::: ::: :::::::::::::-:, w-- eon. exp-ect.-.-to~ e.ce.l e . ............................................................................................................................. W h.e.ther w 1 j - i ~i 1 . ~ Ta-m-i--ly~ in-co-m-e or cut lye a l-n-l--ro-,~ war, :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

DATA NEEDS 79 - ............................................................................................................................ ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

80 ASSESSING POLICIES FOR RETIREMENT INCOME ............................................................................................................................. ......... .... .... ... .... .... ......... ... .... .... ............. .... ......... .... ......... ... .... ........ .. S ..... .... .. .. .. . . . . . ... ....... .. .... . .. .... .. , .... . ... ,._ .. ........................... :::cLr -~-c-rw-~---l---c-u~ rcl~Ol~l~ Aim -~-~A~-E---U-~---~-U~-~--I~ JUO-- -- - --- - -- - - ~i- -i-la -tin f ~i- --j- -by if to kin Dot flu eac ~ Jo a~ In Jo- ~ .............. ................. .... ....... ..... ..... Bernie -bust ante -owl m-uc ~ ~-e-n-eIl~ wo-u- -us ~-e~ rec -u-ce-u~ w e-- her p -any Eon Wh'eth-2e-'r2---"'v2'e"'''' ' d""" '' 'd"""h" ' """ " '' ' I""'' .................... .................. ......... . ......... . ........ ~......... ... ::: Y: ...... ....... .... M :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

DATA NEEDS 8 - ............................................................................................................................ ................................................................................................................................................................................ ................................................................................................................................................................................ ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................. ................................................................................................................................................................................. ................................................................................................................................................................................ ................................... ................................................................................................................................................................................. :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................................................................ :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

82 ASSESSING POLICIES FOR RETIREMENT INCOME ............................................................................................................................. Di i .......................................... .~ .. A ~ . .. ~.. . .~ _ .. .. .... hi by .. All .. . Hi. _ . .. .... - .~ .... :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ................................................................................................................................. :::::::::::::: :::::1 ::::::::::::::::::: :::::::1:::1 ::::::::::1 1::::::::: ::::: ::::::::::::::::::::::::::::::::::::::1:::::::::: ,,, Mel T reponeo yearn ~,,,a,,,,lus now and compared wiln a year ago . ~ i ,. : : , : ~ ~..ITT.Ic-U-Ity Wltn actlVities Otis a-a-i-l-y~ i-iv-i-n-g~ i-nc-l-u-a-l .. .. ............... - - - - - - - -.- .-. - - - - . ............................................................................................................................. . ~, , i Inl~n clean Bar ^~^t^~ Enter ^~^n`^ '' Ink nine '''''''''''''''''''''''''t!''1'l - !''l''~'1~'~'''~!'~' - = "'l'~'''' - !'~'4~'~'1''~'1'1~'1''''' - l''l!''~l'!'l~'' I "!''1 - ''~'1~'1''~ 4'1'~ - At'''''''''''''' E EmE^Tl~na l nrmE l~m~ ~ "~ rlTl~ E T: EE Er nrE ~| Em~ E r^E "n ~^n E~ r =~n ~ E · :! ~:~:! ~:!: E ~ E E t ~ ~ · ~ E :!: ~ ~::: :~ E ~! E E t ~ ~ ~ j ~:~:!: E ~ E ~ · ~ ~ E ~ E t :! ~, E ~:! :~:~:!: E ~ E IT E- ~ E ~ E ~ E T ~ m E ~ I n ~ ,1 ~ I E~ "Y~ rE 1 ~= ~ ~-oa-n-lt-lo-n~ oade-~ a-n-a~ mo-oa~ assess-E -e-nt~ an E~ cl-l-n-l-c-al~ oe-~ress-l-on~ oaue-~ ............................. ~ i . E ~ ~eir reponea woE K alsann IIes a a e Eployer acco E EOa~lOE S -- -- - -- - j-- N-at-lo-nal~ Lon-a-lt E-d-l-n-a-l~ ~-u-~ev~ ot~ Mat E-~ ~-m-e-n~ (--N-L~- -M-~-~ . . · · . .~-ITT-I.C. '.l-ly.~ .pe-~O-' el EO Vat10 eS JOD ....................................... .................................... ~.................. JE ~ | ~---- -- -- :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: -~:r:~-:-~-~-l:-l:-:~-~-:l-:~-^n~-:-:-l:-rl-~-:l-:-:\A-/~:-~-l:E :~:-:-:r ~- - ~ - -n r~r~ l-"m - -r~= l- -l o=- -^ l- - -r:l=:~:-l:l-:r 1 ~ ~ ~ tAl~ ~' n~ l n V~ ~EE E~Cl~E E~ E' !~C ~ El~ ~ ~ ~! ~!~·! l ~V~ - E~ V! ! '~I~! I \~ y~l ~l !` :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: .-:-:-:-:-:-:-:-:-:-:-: A t ~ E i f ~ ~ ~` n~tn-=r --n P E ---~-~rl^~nt --tnP t':':'O'TT=:~t=:E ---n-=P I-tn ~ ~:~:~!':'~:~'l::::~:~ E ~~V~ ~C bl !~C ~ ~ ~V~- E ! l~l ~l ! Wh th E h Ith is b ~E V E sa E a d t I E y N- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - . - E- - - - - - - - - - - - - - ~ - - -~- - - - - - - - - - - - - - - - -~-. - - - - - - - -~- E- - - - - - - - - ~ - - -~- - - - - - - - - - - - - - - - - - - - - - - - - - ~E~l ~ ~ ~E L ~ ~ "E ~ l T l i ~ l E 9 1 ~ I ~ ~, ~T ~ ~ E I E ~ ~ ~ E ~' ~ ~ E ~ E E ~ ~ ~ _ E E E~' ~ ~ ~ ~ ~ ~ ~E E E ~ ~ ~ E ~ ~ ~ E E ~ ~ ~E ~ E / ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: MealCal proolems nave or nave er nac [anoul 4U con lIlons ln all ln ::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ~ E ~ E VlUal g~a$Crl'E -a~ ~Ee-a'~l~ ~- V'U' E-l-e~ r-~--l-~-r-~ O-~ oW ~lovu pre$s '-re~ a-l-l-e- a-l-e a vi te tiE I CE ebl de e si E adh i ci$ a al si

DATA NEEDS 83 """"""""""""""'' ' ' ''""'' '' ' ' """' '' 1" ""'' ''' """' '' " " ' ' 'i""" ' '' """"" '' '1'' ""'' ""'' '' ' """' " " 'i'i'' ""''" ' '' '' '' " ' i" '' """"""""""""""""""""""""""""" . . ................... ....... . ........... . . d ....... . - ~VV- He 1-e-r~ Lena- 1~ 1-S~ De ear wars ~ vv- pelt -err ~we~a~wo-r (~u-!Sa~! lTy COnUIT!OnS ~ em 0 <-l-n-g~ c -u-rl-n-~ p-~-gnan-cy~ c ~-rl-n-~ -nag c rug us-e ae-p-res$-lo-n~ l-na-l-c-alo-~ .... .... .. .... .... ~ ~ ~ ~Oun^~wc me v cnM=c All.... an.. am. E- reman. -me I r`I^`, . ~ _ . ............................................................................................................................ . - I - -. - - -. ........ t- 1 h th 1 ~11 t 1 ............................................................................................................................ ...................................... -. i - - . ~ . n-etn-e-~ ~l--lao-l ............................................................................................................................ ............................................................................................................................ ~ hche i s t h $a ea$ th 1 e h h s e ................... .................... ~' ....... ............. ..... :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::: ::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::. ::::::::::::::::::::::::::::::::::~::::::::::::::~::::::::~::::::::.:. :::::::::::::::::: ~ can~ ~-e~ cove- eo~ an-a~ wn-e~-n-e-E~ etl-- e-es~ pav~ ~-n-e~ sa-m-e~ To-~---$po-u-s-e~ cove-~ :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::: :it:::::::.: l::::::::::::::::::::::::::::: ::::::::::::::::::i::::::. ::::::::::::::::i:::::::::::::::::::::::::: :::::::::::::: ::::::: :::::::: :::::: ::::: ::::::::: ::::::::: ::::::: ::: ::::: ::: ::::::::::::::: ::::: :::::::: 1:::: ::::: :::::: ::::: ::: ::::: ::::: ::::::::::::: :::::1::: ::::: ::::::::: :::::1:: :~::: .. I l=~=Vl E 1 I~L ~"I I ~I I LI V ~VV-=I == Ll I~ I IVVY l~t l~ .............. lA{~t^.~r t t ~ 1 ~ E+^ I 1 1 .. r A ~r *1 ~Y -v -~-1~ ~1 1~1 ~A ~ ~ ~ ~ ~- I I~ V ~ =t I I ~ I ~ V- ~ I I I =~ I LI I I I I ~ ~ I C I -~- - ~V v ~ l ~ E ~ ~ l t~ ~l ~ ~ l l ~. N. ~. . . - - .. . ............................................................................. ............................................... ............................................................................................................................ ~ i h E Id E E tE pl f E i I d d bill ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::: ~ :::::::::::::::::::::::::::~:::::::~:~:::::::::::::::::~::::::::::::~:::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::~:~:~:~:t^:~::::::::::\A:Ir`:^T:~:~:r::::::r-~:l:"S:~S:::::::~S:~:~:::::::~:::::::i~~:~:~::10TI^:I:~::::::::::\ A 1^ ST smr ~ rs^m lT S:::::::~:~:l:^T^:-:-:-::::::::::::::::::::::::::::: ~ w~-!--e~~y-~w ~ VV-I-!-wL!--ew-e~ ~l-!~ l-!-..~..----~---~w~-~.- ~~!-w~ V-Vi-I~-~-!-~'!~ '~'~ F'-'!-~-.-~'!-~'-'!~ I-'!-'-~.. :::::::::::::::::::::::::::::::::: :,.:::::::: :,.,.: ::::::::::::::::::.:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::..:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ~ '' ' e'' -' 'E""'' ' ' ' Ee E"""' ' ~'e' -' 1'' ' e'~' 'e'E'c'l"""' ''""' ' 'i'' 'E"""' a"e"""i E' ' "' E ' e"""""t' pes""'' '"""""""""""""' :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::..:::::::::::::::.:::::::::..:::::::::::::::: .......... ~i' ' ' ' ' ' ' -' 'd~ '' b'' tE " ''""' '-'-'d'' d-' -' ti'b-l'' ~ '' E'' t'E " ''""' ' t' ' -'-'d-' E-- '' t'i" ' '-'-f~ .'''W.'''h.''e.'t''h..' " " :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:..-:-:-:-:-:-:-:-:-:-:-:-:-:-:-. :-:-:-:-:-:-:-:-:-.. :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: T' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ':':':':':':':':':':':':':':':':':'~':':'·:':':':-:-:-:-:-:-:-:-: t-:~-:1-:-:-:-:-:-~:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:- - - - - - - - - - yp-e. oT. n-e-al-l- `. !- ~s I-ra- Ece. cove-rage. al - '""WE o cove-re E---E E"""E -ou-s-e EO' E. ~

84 ASSESSING POLICIES FOR RETIREMENT INCOME a :::: ::::: hi: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ; :: ::~:::::~ _ _:::: ::: l -:a:-: : -: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ~ ~ :A :::~::: J:::: ~it- : -:a: : : ~ .~. N - :::::::::::::::::::::::::::::::::::::::::::::::::::: aT : : a:- ::: :-:-:~:~-a :aTi:a:~:~:^ a- ~:a:-a-~-~a or HAT -~e ~ - ~ ~ -- ::::::::::::::::::::::::::::::::::::::::::::::::::::: . . :a ~ : -ha:- :: --:~: ~ a:a: ~ a:~:~:~:- ~:~: a-i-:-:~-~:~ :~:a:~ ha:-:--: . :~:: :~ :~ -I: :~:: ::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::: - :::::::::::::::::::::::::: T:-:-:-:-:-:-:-:-:-:-:-:-:-:- :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-~-:-:- 1:-:-:-:-:-:-:-:-:-:-:-:~-:.-:1-:-:-:-:-:-~:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: . ----yp-e~ oat n-e-al-l-n~ !-n-s-u-ra-nce~ cove-rage an-a wn-o~ con ................... .... ................................................................................................... ....... 1::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: 1- - - - - - - - - - - - - - _- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -. - - - - - - - - - - - - - - - - - - - -. - - - - - - - - - - - - - - - - - - - - -. - - - - - - - -'-.- - - - - - - - - - - - - - - - -'- -. '':':':':':':':':':':':':':':':': :':':~:':~':'- - - :',:':':'- - - -: :':':: :':': :'- - - - -:-:-:-::::':':'E.. --:::-:-.:.:.::-:::-:::-:::-:::'.:.:.:.':::':::':::'J':':':':::':::::':::':::'::-:-:-'::-::::':'::::-:-:-::':::':':'::-:-:-::'-:-':::': l-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: ':-:-:-:-:-:-:-:-:-:-:-:-:-:- _:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:r -:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-~:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-~:-:-:-:-:-:-:-:-:-:-~-:-:-:-:-:-:-:-:-:-~:-:-:-:-:-:-:-~:-: E:':':':':':':':':':':':':':':':'l:':'t't^'=O':':'^T':':~:^'m'/l:~:~:':':~:l=:r~ E:':':':':\A:ln^:':':'l:n:':':'M'^:l':'l¢'=:n:^l^:':':~:t^'r~ E:':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':': [':':':':':':':':':':':':':':':':!':':~':~:~:':':~'I:':': ~-:!: :V :! ~ ~-::-~ :V :~'1:':~:~' ~ :V-Y : I :'~':':':!'I:':!':':':!':'~:v'-':':!' ~ ~':':'~:Y:'w':!':~':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':' l-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: . .-:-: .-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-.-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::a::a:_:A: :::_:a::a::: :: :a :_:::i:a _:::: ::a:a ::: : ~_ :::: ::::::::::::::: : ::::::::::::::::::: :::::::: :::::::::::::::::::~: ::::::: :::::::::::::::::::::::::::::::::.:::::::::::::: ::::::::::::::::::::::::::: ::::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:-:~- T^:~:-t^:-:-: 1:~:-:-:-~-~-~-~-1 T~-l:-:-:-~-~-:-:-~-l:-: I-:~:l-~:-:-:-~-~-~:~:-:-:-I:~:^T -:-:-~-:-.-:~-:-:-:~:~-^ T - -~-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: :':':':':':':':':':':':':':':~: L"'V:~:':': I :I':'E:':':'E:':l'v.'~'l:'L=l:':':W'l':':':'E:':l'U':l':~:l'l':l E:':':'E:':l'vl:':l':l=:':':'E: LO:L':':':'I':~:':':E':'E:'I:w'E:':I'`I:':I'O':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':' I:::::::::::::: :::::::::::::::::::::::-:::::::::::-:::::::-:::::::::::::::::::::::-::::::::::::::::::::::::::::::::::::::::::::::::::-::::: [':':':':':':':':':':':':':':'t':':l:~l:'l:':l=':':':l':'E:~.:C 'E: Ll:'l':':' - =l:':~.:':':'l:C A: L':':':':E':'~:':':I':'E:''~'E:':E':~:I':'E:~:':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':' l-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:- l:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: E:':':':':':':':':':':':':':':~:' '-:-:-::'-:-:-:-:-:-''::::::::'-:-:-::':' ':':'-:-:-:-:-:-:::-:-:-::'' '.::-:-:-:-:-:-:-':1':-:-::-:-:---:-:-:--::-:-:-:-:-::-:-:-:- ::-:-:-::- - -:::-:::-:- :- -~::-:-:-:::-:-:-:::-:-:-::-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: l:-:-:-:-:-:-:-:-:-:-:-:-:-:-:~:-I:~-m-~7:^f -I- - -a An --am- l- - -ma- .- - -- ~ ~-l - l-r ---I:-l:^n:~:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: E ~I-'~'l-'t'-l't~'-'-'t'-l'-'~-~-~'~-l'-'-' V~-l'-~'-'-'~-~'Y~'L'l-V-l'-'l-~ t-:-:-:-:-:-:-:-:-:-:-:-:-:-:_-:-:-:-:--::-:-:-:-:-:~-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:,:::::::::::: ::-::::::: ::::::::::::::: [ -""""""""""'!""vta'!""""a' E'a""""v' Et' OT pOC et .-ea-!-c-a-E~ -are exp-e-. -a-ltu-- es ~-as-Ke-a~ E-' ~ specs- .-a~ l . ~ - - - l~ w-a e--} e- n-a-.d-ed~ to se- Ella ::: :t:,:::: . :. : ~ : : : : : :.:.:.: t:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::: :::::::::::::::::i ::::::::: ::::::~:::::::: ::::::::::: ::::::::::::: :: :::::::::::::: :~::::::::::::::::::::.::::: -no Esu--Etea~ aocto-r~ Tori otn-e-~---t-n-an~ E'E-n-O'i '-'-E-E'E- Ee$S~ 'En 'E-aSt~ ~ Sea S""""""" :::::::::::::::::::::::::.:::::::::::::::::::::::::::::::::::::::::~::::::::: :~:::::::::::: ::::::::::::::::::::::::::::::::::: .l-...Esu~nce ::::::::::::::: :A:::::::::::::::::::::::::::::::::::::~:::::::::::::::::::::::::::::::::::::::::::::::.::::::::::,::.::::::::::::::::::::::::::::j:::::::::::::::::::::::: :,::::::: air:::::::::: on-- EN al p-~-scrl-pil-o- E art ~ costed \sa- Ee InTorm~lon as acclor DIIIS} ~ i ~ i. , ~ AnE al Cv$:IS to. EmeOval Se~lCeS al E S pp!!eS t$aEme Inivrma Iv E as ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: E. ~ : : E:. :~: :-':-l':':~'tHE:~:':'~':':~:l':'l:~:':':'-:~'! :':l'~ ':'::-::::: l: :l :~: ~ ~ ::::::::::::::: :::::::::::,::::::::::::::::::::::::::::::::::::::::::::::: ::,::::::::::::,::':~:::::.::::::::::::::::::::::::::: t-mp-l-oye-E ~ s-E --E y~ oT~ n-e-al-t-n~ l-nsu--. ' -ce~ pEan~ p-' v-lslo-n-s~ a E0""'COS S"""tTO'E""'e'E " """""""""""""" ::::::::::::::::::::::::: ::::::::: ::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: l::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ~ 2-2~ EOTE'S"""""A E-EAD""' ''' ' j- E' ' "" E' '' ""''i'' il t th f ERS Q ti t ~ . l ~i - i * E~ essarl-ly---as-E ec~ l- '~ eve-' ~ l-n-~~l-ew---o.T---a---s '-~.e; .~ ~ee~ uox---~--4---Tor--q- Eestl-o-n-s~ on~ expe~ .. .. ............... - - i - i - i i - - - ~i tation$ ~t t t E Ith t t $ ~ ~Et E r ~.

DATA NEEDS 85 ........................................................................................................................... D ~--A-' ~-A~ ! - . ~ Hi-- ^:-~-:/: :- : a ^''''A':' t: ':':':': - r in:' ':~ ~ 1:- :' n -- ':': :' A': A' E:':':':':':':':':':':':': :::::::::::::::::::::::::::::::::::::::::::::: .-. A :':': ::A:::: ! : : ::::~A::--:: : : ::_:~::::A : :::: . a. :::: : :: : : :: : O~nQr T E9' - :':':'. a - ~:l:^n:::9:a E:':': Ea :a: E E':':'a 'a a a a ram, ~ .~9'|'~' E:a':':':':':':':':':':':':':':':':':':':':':':':':':':':':' :':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':': -a F -- a -if- ~ -hi- - - a OF: ~ a ~ ~ l: a::: '-if -:~ _ -- By:: :: a :~: :F: My:: :~: ~ ~ A ~ E :: :':':':':':':' ............. "E.=T:.:-`f:~:|-.E-:|~:-:-:^T:: :m:E~-ET~ r: : n E~ m =22~2r222~-:~^ E.-=2_E7.E~ n ~ :1: : t/ ~ n l r l ~ E E ~ ~V :~l: E ~ ! l E' !'~ E :~:!::: :I: l:~-: E | l ~ E | ~ V l - ~ ~ E ~ |' '| ~'| ' ' 'Y' - :'|:':|'|:V:|'w':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':': :':':':':':':':':':':':':':':'::-:-':':':':':':':':':':':':':':':':':':':':':::':':':':':':'-::'::::::-:-:::::::-:::-::::-::: -:- -:- -:-:-:-:-: : -: ---: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ':':':':':':':':':':':':':':' :'1":':':" :':':::':'::::':':":1':::'::::::':':":': :':':'::~: :':':':' ::':':'::':':' 1: ::::':':'::::: :':':':::::: :1:::: :1:: :~:::: : : ::::::: -:::::::::::::: E~I ^:I:: :~:I E:I I: I ^:: :^ I::: ~ I E^~::::: m I: I :I :I: I:~: E::: E: I: :I n r E~ ............. ~Eo-' -ey~ l-n~ ce-E !T-Ica~es oT acposll to.E s)~ gove-' n-me-' l~ s.a vl-n-gs~ ~-on- ES""""""""""""""""""""""""""" :':':':':':':':':':':':':':' ' :':' :':: :':' ':':: :': :'::: :'1:: :': :'::: :':':':':: :':' ':':'::: :' 1:::::::::::::::::::::::: i::::::::::::: 1 ::::::::::::::::::: I :::::::::::::: :::::1_ ::::::: :::::::::::::::::::::::::::::::1_::::::::::::::::::::_1::: :::::::::::::::::::::1 :::::::::::::::::: E E e rilan ces w e n aE ~T E O E W~ O E reve lvou wo ~aT I' e ~l . e ~ :-:-:-:-:-:-:-:-:-:-:-:-:-:-:t:-:-:-.:~-:-:-:-:-:-:-:.-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-- :-- :- :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:~-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: -:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: E ITe l E$ ~nCe--SeTI!-eme ES E en E cel e- E O~-n---aT--TE e--t-m-e~ EO Lar~-e~ u--n-ex-~-ecte ~ expe-n-ses~ o e-E~ I-ast~ 20~ vears~ tn-~ m-aGe~ lt~ mlc Elt to ~1 ~::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ~-ap-lla-l~ oal- Es~ co-m-po-' -e-n~ oT~ assel~ v.-al- Ee~ l-E -c-E eas-es~ ane-~---TE-rs~ l-E le-~-l-ew ~ j N-at-lo-nal~ ion-a-ltu-d-l-n-al~ ~-u-~ev~ ot~ Matu-~ ~-m-e-n~ (--N- i~- -M-~-~ - - . , , j - - E Eo -ey. l-n. c-n-ec- l--n-~. savl-n-~-s. ~'E -s. moE ey EaE ' el TUE as c eall unions ~ :::::::::~|: :1:1:;: :: ~:::E t E E ::::~: : : EE :::: : :: :|:I+I I I t; l ^::: :E :E :::::::::::::::::::::::::::::::::::::::::::::::::~ V--~I- E~ ~l E~--E ~ 6---~V-I:-:!-~:,~ :~-I:~ :!-:-I:-E ELU~:!:::!:U:!:I: E::~l::!~:!:~-::::::::::::::::::::::::::::::::::::::::::::::

86 ASSESSING POLICIES FOR RETIREMENT INCOME - - . 't ~' ~ . ~ ^~^ ~ / ~ ~ ~ . ~ ~ ^~` .^ ~^ ^. ~ ~ ~ . .^ ~ if ^^ 1~ ~ ~ ! ~ ~ ~ r.-^ ~ 1` ~ ~ ~ 1 t~ . - ~ -~= V=.~ --- - -- - - -- == ~ L== 1 . 1 - -= -- ~ - ~=1 11 -~--- I VI-~-I---l-=- ~ I Vent - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -. - - - - - - - - N .. , , ,, , _ Fit ^^. E ~ ~^ fit ff - - l ~ ~ ~ Off Vet ------Jimmy - l me ~ Vito ..... I ~ ·-~~-~~~-~i---~--i--~--A--- B--~-y~-~-~---~-~-~---------_ -B---~-e-l - ::::::::::::::: :::::1::::::::::::::: :: :::::::::::::::::::::::::::::::::::::::: ::::::: :::::1:::::1::: :::::::::::::::::::::::: :.:.:.:.:.:.:.:.:.:.:.:.:.:.:.~:~:.E:I.:.I:P.:.:.~T:.:.T=:.rm:.:.:.:.:.~.~:l.:.l:.nr:.:.nT.:.:~.PnT.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:':':': ~ An' _ ~ An.- 1 ! An' ' ! ', An' ' 'A ~ ' '' An.- ' ~ ~ am.- ' - - - - - - - - - - - - - -.- - -. - - - - -, - - - - - - - - - - - - - - - - - - - - - - --- - - - - - - - - i- - - - - - - - - - -- - - - - - - - - - - - - - - i- - - - - - - - - - - - - - - - - - - - - - - - - - - - --- - - - - - - - - -, - - -- ~ / E I i t t ~ ~ n ~ l l i l l E t l l E :::::::::::::::V:~:I - ~:::Ve:::~L - ~: ---~Vl !~ - vl ! l tyL - ~! ! - ! l - - :::::::::::::: :r::::: ::::::::::::::: ::::::::::: :::::::::::::::I::::.l:::::::::::::::::::::::::::::::::::: ::::::::r-:::::l:::: ~laEl Ea nT aE lTnm^mEEa=~ OmnE EnT ~r EamT ::::::::::::::::~:~:~Y:::~.:~:::~:~:~.:~:~::~..~:~:~.:~: :::~::~:~:~:~:~: ~ ·:::~:~:::~:~::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::: - :::::::::::::::::::::::::::::::::::::::::,::::::::::::::::::::::::::::::::::::- ::::::::::::::::::::::- ::::::- :::::::::::::::::::::::: -:::: :.::::::::::::: -:::: :~:::::: 't ~I I~r r~l`/~^ ~r^~ ~. rm - ~rl,^ mff~^ ~l~. rr~ r.-^l~l`~^ .............. ~I I.IV.U.~.IL I.~=I.V=. ~I.I wI.I.1 ~.~.~.1 11 ll l~l l\"l l~=l EllLO - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -. - - - - - - - - N---- ~ -------- ----I E ---- -------- ~ -- - E~ -------- l ~ - ------- - a~l-on-al---Lo-n~-l-~u-u-l-nal---a-urveV Ol ~OU-~'l---u~-~-~--E~ ----------------------------------------------------- - - - - --- - - . ~- - - - - - - --- - -- - -~---- - --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - --- - - - - - - - - - - - - - - - - - - - - - - - - T .... ......... ........ ................................................................... ................................................ ~TOI rt~n'= En~lllmlnr' mn"~= na^~ TaY.~ ann n~nT~ n~t=~ ~ - l lil . ~.~! ~ - j ll !~l~l! ly t! l~'~y~y~ :':':'~I:~.':':':~''':':':~!':'I:~':':'~:~:~,,:'h~':':~':V:~'I:':':':~:~:':':':':':':':':':':':':':':':':':':':':':':':':':': .................................................................................. ... ........................... ... ................ :::::::::::::: : :::: :1::::::::::::::: :::::: ::::::: :1::::::::::: ::::::::: 1 1 W ~ -E I- -I a- - -^T - - OT^r ~£~ ~^ n ~-£~ ^ r - -~ 1- ITI- -I ~ -E - - Tl - l-n- E - - ~ n ~ -r- P ~ ~ - - - - - - - - - - - - - - _ - - - - - - - - - - - - - - - - - - - - - - i- - - - - - - - - - - - - - - -- - - i-. -- - - - - - - - i- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -. --- .- E t l l! 1 I l r 1~1 .~. . I-= Y ! I ! ! ~ I ! I ~ ~ !- - . .~. ~ l-. ~ - ~ I ! I !-! LI w - - u-eot---to~ sto-d s- m=~ca' cade~ o v~oers oa E~S otE ~ ieE ~=~

DATA NEEDS 87 ................................................................................................. - -'''''''''''Fii'' '' ' ' i" !"""" ' $'i' ' ' '' ' """'~"""$5'00"""' '' """' ' '' ' """i"' """" ' ' ' """'1"'2"""'' '' " ' h" """' ' """' 'h'i'!'d"' ' '' """'' " """"""""""""""I , . . . --loo- per wee <~ or loon ~ ~ -l-n-c- -u ............................ v -e-a so e-a En out t-no ~ co-u-n~-l-n-a~ air wolf <~ of sc solo . N atio-na on ItUC Ina MU O Tutu ---wom-en--- N ~ VOW ~ N - - - - - - - - - - - .- - _ .............................................................. -ate N - - - - -. -. - - - . -. - - - - - ::::::::::::::::::::::::::::::::::::::::::::::::::::: : Hi: : 1:::~:::~: ~:~.i:i:~:~: Ql:::~:: :f::~:::V : .~:::I:~I:~::~:V:~:::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::: ":~IVl: I":~:::"VI :~:" 1:~1 1 :~:::~:~::Y:. V :: V l::::i::~:~:~:l ::~:1 ~:i::i:::::::::::: - - - - - - - - - - - - - - _-~r l- - -"Yn-^n,- l l- l-l- -l r"~- - -l-~-~-l - - -\A-I=~-~- - -t-~-AII I ,-~- - -~-m~ - - f -I-~-l-~- l-l - - -^n - - l `/n=- l- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - :::::::::::::: :I::: V:V:~:: :~A -~-~.-:I-:-l:~:!:c:~:!:~:: t~ ~:: :~Y:~':~:: \:~: I:~I: ·: :~': :! :I~.-:: :~ ~:!':: :~t :I:: :~, :~ ~ :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-. :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-. :-:-:-:-:-:-:-:-:-:-: - ::: :1:::::::::::::::: :::::: :::::: ::::::: ::::: :~::: ::: :: :: :: . , :-:-:-:-:-:-:-:-:-:-:-:-:-:-:~:~-T:~:~-1:~-~-~- 1-:-:-~-:~:~:-:-:-~:~-01~:1-:-:-:~:~:1- 7-~-T:I-~-~-:-:-:~-I :-:1-~-~-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: :::::::::::::::1::::1:V:I=:~OI:wI::I"~::~:I:E:~:::~I=I:::wI:~=I:I:I~:~:IVl::l::: - U=~::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: M:::::::: ::::::: ::::: :1::::::::::::::::::::::::::: ::::::::::::::: ::::::::::: :::::: A:::: ::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:-: ^-A-~:-~:-:-:~:~:-r^:-:-:~:~-~-~-~:-:-:-~-:~:-:-:-~-~-:-:-, 1:-'-'-'~-:1 :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: :.:.:.:.:.:.:.:.:.:.:.:.:.:.: ~ ~ t-~ ~:: :~:~ l: ~:: :~.1:.:!.~.:.:.\.~:~:.:.:.~. ~.A :.:.~.~ V.:I:.:.:.:.:.:.:.:.:.:: ~ j Ii t f t """"""""""""""''''''i'l ~ j~ j~ j~ f ~............. ,.,., ................................................... -:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:. -:-:-:~-1 :-:-:-:-:-:-:. -!:-:-:-:-:-:-:-:-:-:-:-:-:-:~-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:. -:-:-:-:-:-:-:-:-:-:-:-:-:-~:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-~:-:-:. -: on~ =nmoull-on$~ lo~ pe-n-s-lon~ pl-an$~ ana~ n-ealln~ l-n$u-ra-n-~ p-rem-l-~-m-s~ re$pecll v.el-y. ~ :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::.:::::::::::::::::::::::::.:::::::::::::::::::::::::::

88 ASSESSING POLICIES FOR RETIREMENT INCOME NLS cohort will begin in 1997, but it will be limited to young people aged 12-17. These kinds of gaps are important, given the economic changes that are affecting middle-aged workers, such as layoffs and employer incentives for early retire- ment. Analysis is needed of whether and to what extent middle-aged people respond differently to economic and policy changes than either older or younger people. For almost 30 years the Panel Study of Income Dynamics (PSID), sponsored by the National Science Foundation and other agencies, has followed a sample of the entire population that includes all age groups (Table 4-1~. However, the sample for middle-aged and older men and women is not as large as in the NLS panels or in the Health and Retirement Survey (HRS) or the Asset and Health Dynamics Among the Oldest Old (AHEAD) survey, discussed below. Also, the focus of the PSID is on income and program participation. Supplemental mod- ules have covered such retirement-income-related topics as savings behavior, net worth, extended family ties, financial situation and health of parents, disability and illness, and retirement plans and experiences, but they are not regular parts of the survey. HRS and AHEAD, sponsored by the National Institute on Aging, are specifi- cally designed to provide the kinds of comprehensive, high-quality, multivariate data that are needed to develop improved analytical models of retirement- income-related behaviors of individuals. (Features of these surveys are shown in Table 4-1 and Boxes 4-4 to 4-7.) HRS covers men and women who were aged 51-61 in 1992, and AHEAD includes people who were aged 70 and older in 1993. The two surveys include married as well as single women and men in their samples. In contrast, the RHS included only single women in the sample, along with married and single men. (Married men's spouses were also interviewed, but they were not necessarily the same age cohort.) HRS and AHEAD also obtain extensive information about spouses of sample members. HRS and AHEAD obtain a wide range of information for sample members on employment, income, savings, assets, family status and kinship networks, health status and health care arrangements, and retirement-related expectations. From analysis of the initial waves, the surveys appear to be making important improvements in data quality; for example, they obtain more complete reporting of wealth than most other household surveys. Further enriching the HRS and AHEAD databases are descriptions of pen- sion and health care plans for sample members obtained at the first interview from their current and former employers. Information about health care plan provisions is obtained directly from the employer; information about pension plan provisions is abstracted from Summary Plan Descriptions provided by the employer (which employers are required to have on file). The NLS-Mature Women survey also obtained Summary Plan Descriptions from the current or prior employer of sample members in 1989, when they were aged 52-66. Also planned for HRS and AHEAD is the addition of data from Social Security and

DATA NEEDS 89 Medicare administrative records. Having a rich array of data in the same survey is important for at least two reasons: researchers can explore behavioral models of joint decision making (i.e., joint labor supply and savings decisions), and comparisons among different researchers' analyses of the data are facilitated by the use of common concepts and variables. Features of Other Surveys We stress the importance of long-term panel surveys of individuals and their families, but there are also repeated cross-sectional and short-term panel surveys of the entire population that provide needed data for retirement analysis. The Survey of Consumer Finances (SCF), sponsored by the Federal Reserve Board, has been conducted every 3 years since 1983 under its current design (predecessor surveys were conducted as early as 1963~. The survey is a source of detailed information about household wealth in the United States that is fre- quently used as a benchmark with which to evaluate the quality of wealth report- ing in other surveys. The SCF obtains information on pension wealth by obtain- ing descriptions of pension plan provisions from sample members' employers. Features of the survey that contribute to the high quality of the wealth data include its sample design and detailed editing and imputation procedures. The sample design combines an area-frame household sample with a list-frame sample of high-income households drawn from Internal Revenue Service (IRS) records.7 The SCF originally included a panel component: the 1986 SCF was a reinterview of the 1983 sample, and the 1989 SCF included a subset of the 1983 sample in addition to a new cross-sectional sample. However, the 1986 survey data were not used because of quality problems, and, for cost reasons, the 1992 and 1995 surveys have not repeated the panel feature of the 1989 survey. Other limitations of the SCF are its small sample size (about 3,000-4,000 households) and limited or no information on some topics, such as health status and health care use. Nonetheless, the SCF is important for its periodic reporting of trends in household wealth. The Survey of Income and Program Participation (SIPP), conducted by the Census Bureau, is a repeated, short-term panel survey of individuals that began in 1983. From 1983 to 1993, a new panel was introduced every year that followed members of about 12,000 to 24,000 households for 2-1/2 years, with interviews every 4 months. The survey collects detailed monthly information on employ- ment, income, and program participation. Periodically, it asks about a wide range of other topics, including assets, health status and health care utilization, pension plan participation, and retrospective information on employment and family his- tory. Beginning in 1996, the SIPP design will have larger panels (37,000 house 7Households selected from the IRS list frame must give written permission in advance to be included in the survey.

9o ASSESSING POLICIES FOR RETIREMENT INCOME holds for the 1996 panel), whose members will be interviewed every 4 months for 4 years, with a new panel introduced every 4 years.8 SIPP has been used extensively for analysis of trends and transitions in income receipt, program participation, health insurance coverage, and other top- ics. The survey supports cross-sectional analysis by combining panels. Its use- fulness for retirement-income-related behavioral analysis is limited by the short panel lengthy SIPP may gain added relevance for analysis of retirement-income related trends in the future; for example, it may replace supplements to the Cur- rent Population Survey, not only as the source of official poverty statistics, but also as the source of information on pension plan participation. The Current Population Survey (CPS) is a monthly survey of about 50,000 households (previously 60,000), which is sponsored by BLS and has been con- ducted by the Census Bureau since the 1940s. It collects information on labor force participation each month to use in calculating the monthly unemployment rate; the March income supplement collects information on sources and amounts of income for the preceding calendar year; a supplement conducted every 5 years has provided information on employer-provided pension and health care plan coverage. The CPS has been used to follow employment and earnings trends for many years. The CPS supports limited kinds of longitudinal analysis: there is overlap in the sample across pairs of months and years;l° but the overlap is limited, and the sample households may not contain the same individuals (the interviewers return to the same address and do not follow household members who move). The Consumer Expenditure Survey (CEX), which is sponsored by BLS, has been conducted on a continuing basis since 1980 (predecessor surveys were conducted as early as 1901~. Its major uses are to provide the market basket for the Consumer Price Index and to provide data for analysis of expenditures in relation to demographic and other characteristics. The CEX includes two compo- nents: the Interview Survey and the Diary Survey. The Interview Survey collects detailed information for most expenditures from about 5,000 consumer units that are interviewed every quarter for 5 quarters. (In the CEX rotation group design, each month one-fifth of the sample is new and one-fifth is completing its fifth and 8The design may change if the survey becomes the source of official annual poverty statistics as recommended by a National Research Council panel and if additional funding can be obtained. Under consideration is a design of 3-year panels, with a new panel introduced each year. One of the panels in the field each year would have a sample size of about 37,000 households; the other two panels would include about 11,500 households each. 9The 1992 and 1993 panels will be extended with annual interviews from 1997 through 2002 to collect data for tracking the effects of recently legislated changes in social welfare programs. The focus of this Survey of Program Dynamics will be on families with children. 10In order to reduce the variability of estimates of month-to-month change in unemployment, household addresses are retained in the sample for 4 months, dropped for 8 months, and brought back into the sample for another 4 months.

DATA NEEDS 91 final interview.) The Diary Survey collects records of daily expenditures for 2- week periods from about 6,000 consumer units; interviews are spread over the year. BLS makes use of data from both surveys to develop a total picture of consumer expenditures. Each of these surveys serves as a benchmark in its area and provides infor- mation on trends that are important to monitor for purposes of retirement- income-related analysis and projections: trends in wealth (SCF), income (SIPP), labor force participation (CPS), and expenditures (CEX). Each is also well established; we assume that they will and certainly support that they should- continue. Our recommendations pertain to long-term panel surveys that are critical for retirement-income-related behavioral research and whose benefits may be less apparent than the benefits of these other surveys. We discuss below the advantages of coordinating aspects of the question- naire design and content of long-term panel surveys with the SCF and SIPP, both of which provide (or have provided) relevant data for longitudinal analysis over short periods in addition to key cross-sectional estimates. In that regard, it could be useful for future rounds of the SCF to include a panel component to help measure change in savings. We discuss the contributions that such surveys as the SCF and SIPP can make to data validation and (in Chapter 5) the possible role for SIPP as a microsimulation model database. Directions for the Future HRS and AHEAD HRS and AHEAD promise to make possible important new and refined analyses that can add materially to understanding savings and retirement decisions and other topics that are relevant to Americans' retirement income security. How- ever, both surveys are very new HRS began in 1992 and has completed three waves of interviews; AHEAD began in 1993 and has completed two interviews. To achieve their full potential, the original HRS sample must be continued until its members are well past retirement. The original AHEAD sample must also be continued for a significant length of time in order to understand patterns of saving and dissaving, health care utilization, and mortality among the very old. In addition, new cohorts need to be introduced periodically in HRS to make it possible to analyze differences in behavior across cohorts in response to socio- economic and policy changes. At present, funding is in place for HRS to introduce in 1998 a new cohort of people aged 51-56 and to continue the original cohort for another interview (for a total of four interviews). Funding is also in place to conduct two more interviews with the AHEAD sample and to introduce a cohort that fills the age gap between HRS and AHEAD. The sample size of the new and filler cohorts will be smaller

92 ASSESSING POLICIES FOR RETIREMENT INCOME than the current cohorts in order to contain overall costs (see Table 4-1 and Figure 4-1, above). When the new and filler cohorts are added in 1998-1999, HRS and AHEAD will be fully integrated and will use the same questionnaire with appropriate skip patterns (e.g., younger sample members will be asked fewer questions than older sample members about health care use, while older sample members will be asked fewer questions than younger sample members about employment). The intent for HAS/AHEAD is to continue to introduce new cohorts of people aged 51-56 every 5 years, so that the combined survey is both continuously represen- tative of the population 51 years or older and follows each new cohort until its sample size is no longer useful because of mortality. We support the implemen- tation of this plan, which will likely require a modest increase in future funding until the system reaches a steady state and then continuation of funding levels in real terms. Other Long-Term Panel Surveys HAS/AHEAD is not the only useful panel survey of individuals. As indicated above, several cohorts of the NLS are continuing to provide valuable data, as is the PSID. We support the continuation of these surveys at their current funding levels. In particular, the NLSY, if the questionnaire is appropriately modified (see below), can play an important role in explaining the substantial heterogene- ity in savings and net worth levels that is already evident by the time people reach age 50. (NLSY covers people who are now aged 31-38.) Similarly, the NLS- Young Women cohort (women who are now aged 42-52) and NLS-Mature Women cohort (women who are now aged 59-73) can provide useful data for cohorts of women who are not yet covered by HAS/AHEAD. Unfortunately, there are no equivalent surveys for male cohorts in these age ranges. In the future, HAS/AHEAD may obviate the need to continue NLS-Young Women or NLS-Mature Women. Content Enhancements Some content additions to existing panel surveys would be very desirable for analysis purposes, although the marginal costs of each proposed enhancement need to be carefully evaluated against the expected benefits. In the case of the younger NLS cohorts, it may be that, as they age, some existing modules that are more appropriate for younger people could be cut back to make room for new modules that are more relevant for retirement-income-related behavior. For analysis of consumption and savings behavior, it would be very useful to have direct measures of total consumption in all of the major panel surveys. It would also be useful to have direct measures of housing and medical care expen- ditures, which are two important elements for assessing retirement income secu

DATA NEEDS 93 rity. Although it is possible to estimate total consumption as a residual by subtracting change in net worth from income, direct measures are useful for at least two reasons: first, changes in asset values will reflect not only realized capital income, but also unrealized gains and losses; second, there is a consider- able amount of "noise" in the measurement of asset values and change in net worth over a time period. Typically, in surveys, most people are willing to indicate the types of assets (and liabilities) they hold, but they are not always willing or able to specify the value of each asset, or they may underestimate the value of their assets. Because of substantial imputation for nonresponse (which does not usually capture all of the missing information) and underreporting, most surveys underestimate the value of families' asset holdings. Curtin, Juster, and Morgan (1989) found that estimates of wealth from the 1984 SIPP panel and the 1984 round of the PSID were only 61 percent and 79 percent, respectively, of the estimates from the 1983 SCF. Juster and Kuester (1991) similarly found that estimates of wealth from the 1979 round of the RHS and the 1981 round of the NLS-Older Men survey fell short of those from the 1983 SCF.ll In HAS/AHEAD, range cards permit holders of an asset to select a category (e.g., $1,000 to $5,000) if they do not know or do not wish to state the exact amount. "Unfolding" or "bracketing" techniques are also used to increase re- sponse. In the bracketing technique, holders of an asset who don't know or refuse to provide an exact value or a range are asked if the value is above a certain amount; if yes, whether it is above another (higher) amount, and so on.l2 HRS/ AHEAD obtains high rates of response to the asset value questions by this method; however, the response categories are very broad for example, less than $1,000, $1,000 to $10,000, $10,000 to $50,000, $50,000 or more. The resulting wealth estimates appear reasonably robust; Smith (1993:13) estimates that HRS obtains 88 percent of the wealth aggregates in the SCF. However, it is difficult to obtain a precise estimate of consumption by subtracting change in net worth from in- come, given the broad categories for so many of the asset value responses. HRS/ AHEAD plans to separate out the capital gains and losses component of change in net worth from the component of savings per se. This step will be helpful for estimating consumption, but the calculation will still be hampered by the inherent imprecision in estimating the two components. The NLS surveys do not include questions on expenditures, except for child care and commuting costs (see Box 4-7~. The PSID obtains information about 1lThe SCF is taken as the standard for asset valuation in household surveys. Its wealth aggregates compare very favorably with aggregate figures on household balance sheets from the Federal Re- serve flow of funds accounts, when proper adjustments are made to achieve conceptual compatibility (Antoniewicz, 1994). 12The SCF pioneered the use of range cards and also uses bracketing, which was initially devel- oped for the PSID; see Kennickell (1996).

94 ASSESSING POLICIES FOR RETIREMENT INCOME housing and food expenditures but not about other expenditures. HAS/AHEAD has questions similar to those on the PSID, as well as questions on medical care expenses that use the bracketing approach; Wave 3 of HAS/AHEAD includes a one-question assessment of total expenditures. There are obvious problems in attempting to measure consumption directly in surveys that cannot afford to include the detailed questions used in the CEX: people may not be able to provide accurate information on their aggregate expen- ditures, and the lack of detail can cause problems for such issues as how to treat durable goods expenditures in deriving conceptually appropriate estimates of total consumption. Despite the problems, we encourage work on measuring consumption in such surveys as HAS/AHEAD and NLS, including careful evalu- ation of the information obtained from the new HAS/AHEAD questions and further experimentation with question detail and wording. (HAS/AHEAD has a subsample explicitly for experimenting with new content and alternative question wording.) Another desirable content enhancement would be to add retirement-income- related questions to the NLS surveys for younger cohorts NLSY and NLS- Young Women whose sample members are in their 30s and 40s. As noted above, by the time people are in their 50s, significant disparities in income and wealth are evident. To understand how these disparities arise, it would be very useful for the younger NLS cohorts to include modules for pension coverage, savings and net worth, health status, and retirement- and savings-related expecta- tions, similar to the modules in HAS/AHEAD (see Boxes 4-4 to 4-7~. NLSY and NLS-Young Women include members of the baby boom cohort, and it would be particularly useful to have information about their savings behavior as they ap- proach their pre-retirement years. (The baby boom generation will ultimately be picked up in successive new cohorts of HAS/AHEAD, but only on reaching age 51.) Retirement-related modules could be included in NLSY and NLS-Young Women without necessarily increasing questionnaire length, given that modules related to education and training may be less relevant as these cohorts age and likely could be curtailed without loss of important analytical information. Finally, it could be useful to expand the content in HRS (and in the retire- ment-related modules that are added to the NLS) that is intended to help explain savings behavior, in particular, why so many people appear to be saving so little. HRS already includes a large number of relevant questions (see Box 4-4), but it could consider adding a few more. For example, in addition to the question on how much personal savings respondents expect to have accumulated by the time they retire, there could be a question on how much they think they will need (in today's dollars) to finance their retirement. Also useful would be questions on respondents' awareness of and participation in employer financial education pro- grams, in order to analyze the effectiveness of such programs in increasing em- ployee pension contributions and personal savings.

DATA NEEDS Cross-Survey Reviews 95 We urge collaborative efforts among NLS, PSID, and HAS/AHEAD to review their questionnaires and data collection procedures on a regular basis to deter- mine ways to improve the quality, utility, and comparability of the data. It would also be useful to include the SCF and SIPP in these reviews so that analysis and validation of key questions can be carried out across surveys in a comparable manner both cross-sectionally and longitudinally. Each of these panel surveys has its own orientation and user community, and each tends to consider content or procedural changes from its own perspective. However, the return to the investment in all of the surveys would be increased to the extent that each includes some common modules on key topics that permit comparable analyses across surveys or with data pooled from two or more sur- veys to increase sample size. Furthermore, with common modules on key retire- ment-income-related topics, it would be possible to combine the samples from several panel surveys into a rich and broadly representative database for projec- tion models (see discussion in Chapter 5~. In this regard, plans for the newest NLS cohort are to include such questions as expectations that HAS/AHEAD has pioneered (e.g., of one's own life expect- ancy). This development is very welcome, although the newest NLS cohort, which includes people in their teens, will not be relevant for retirement-income- related research for several decades to come. Finally, when one panel survey pioneers techniques that demonstrably im- prove data quality such as the bracketing or unfolding technique in HRS/ AHEAD, which is used for medical and total expenditures as well as assets it seems highly desirable for other surveys to adopt them. Such techniques should also be adopted in other household surveys, such as SIPP, that are a resource for tracking assets and other retirement-relevant variables in the general popula- tion.~3 Mechanisms to provide for regular cross-cutting reviews could include joint meetings of major panel surveys' advisory groups and principal investigators and commissioning papers on specific areas of possible interchange and congruence in questions and data collection techniques. We recommend that the National Institute on Aging organize an interagency working group of survey sponsors and investigators and facilitate means to bring together and learn from the experi- ences of the major retirement-income-related panel surveys of individuals. (See discussion in Chapter 6 of issues of interagency coordination to support improved data and models for analysis of retirement income security.) i3The 1996 SIPP panel uses bracketing for measurement of asset values and interest income.

96 ASSESSING POLICIES FOR RETIREMENT INCOME Recommendations 1. Existing panel surveys of middle-aged and older people should re- ceive continued government support. Longitudinal data from these surveys are essential to analyze retirement and savings decisions and determine be- havioral responses to changes in public and private sector policies. Such analyses in turn are essential to develop better models for projecting the likely effects of alternative policy proposals on retirement income security. In particular, the HRS and AHEAD surveys should receive continued sup- port. These surveys should be refreshed periodically with new cohorts in order to offer insight into how behavior changes over time. 2. Panel surveys of middle-aged and older people should experiment with methods to develop measures of families' total expenditures and expen- ditures on housing and medical care. Such consumption measures are im- portant for projections of economic well-being in retirement. 3. Panel surveys of younger people, such as the National Longitudinal Survey of Youth (NLSY), should include detailed questionnaire modules on pension coverage, wealth, health status, and retirement- and savings-related expectations. Such information is needed to understand more fully life-cycle behavior and to track the disparities in income and wealth that are evident by middle age. 4. Agencies and researchers involved in retirement-income-related panel surveys of individuals, and other surveys as appropriate (such as the SCF and SIPP), should collaborate regularly in reviewing questionnaire content and data collection practices to identify ways to improve data quality and utility. For example, the bracketing technique used in HRS and AHEAD that has been demonstrated to reduce nonresponse to important items should be adopted in other surveys. Also, such surveys might include a common core of questions on specific topics. The National Institute on Aging should facilitate such collaborative efforts. DATA ON EMPLOYERS The discussion in Parsons (1996) and in Chapter 3 makes it clear that there are major gaps in understanding employer demand for older workers and what causes employers to adopt, modify, or drop pension and health care benefit coverage for current and retired employees. There has also been little systematic study of the consequences for workers and retirees of changes in the mix of employers with different hiring and compensation preferences. These gaps in understanding cripple the ability to develop useful projection

DATA NEEDS 97 models of retirement income security, given the prominent role of employers in generating earnings and benefits for so much of the population. The gaps in research knowledge about employers in turn stem from deficiencies in available data: relevant cross-sectional data are not adequate, and relevant longitudinal data are almost entirely lacking. We would have discussed the needs for improved employer data before the needs for improved panel data on individuals except for the cost implications. The first priority, we believe, is to continue and improve existing panel surveys of individuals for which budgets are already largely in place. To improve em- ployer data to the extent that is needed is likely to entail costs for new data collection as well as modification of existing data systems. New data collection is not undertaken lightly in a time of constrained budgets. There must be careful planning to evaluate existing databases and what can be done relatively inexpen- sively to improve them and then to evaluate and choose among alternative strat- egies for obtaining needed new data in the most cost-effective manner. Government Sources Federal agencies collect substantial amounts of information about business enter- prises and other employers in the United States (e.g., payrolls, sales), but rela- tively little information is directly relevant for retirement-income-related de- scriptive and behavioral analysis. Moreover, the relevant data sources all have important deficiencies. Table 4-2 summarizes features of these databases (see also Appendix B). Employee Benefits Survey (EBS) The EBS of the Department of Labor's Bureau of Labor Statistics (BLS) began in 1979 as part of an effort to develop information on the comparability of wages and benefits between the federal government and other employers. The focus of the survey design and published tabulations is on benefits available to workers in several broad occupational categories for example, professional, technical and related; clerical and sales. From surveys of large private employers in one year and smaller private employ- ers and state and local governments in alternating years, the EBS obtains exten- sive information on features of pensions, health insurance, and other benefit plans. However, the data are limited for analytical purposes for several reasons, some more important than others: · The sample of about 6,000 establishments excludes agricultural enter- prises, private household employers, and the federal government, so that it covers most but not all of the employer universe. · The sample design, which is based on unemployment insurance records, is in terms of establishments, not employers (or enterprises). Linkages to the

98 4= o v so a' to o Cq a' Cq a' a' o Cq a' a' Cal ¢ EM , , 3 , & ~ , ~ ~ 9- ~ A, ~ 3 ~ Fly ~ ~ ~ R ~ O. ~ ~ ,,, ~ ~ . ~O =~ mu_ a ~ in ,, a , em ~= 2 A. . At 3 ~ O ~ 3 it ~ , ~. ~ ~ ~ ~ e 3 =, ~ ~ ca ~ ~ ~ O ca ca ~ ~ ca ~ ~ ~ ~ ~ ~O ~ ~ 4= o c' o v, ca 4= e A ~ ·, e S e A I ~a ~ So ~ ~ ~ R · - ~ R ~0- ~ ~ e ~0 ~ e Ill ,^ A s ~ e == ~ ~ ~,,, as

99 e i. ' ~ ° ~ ~ ~ ~ ~ e ~ ~ ~ ° ~ 2 ~ ca ca ~4= ~S. ~ O .e ;~. ~ ~.~ ~ O F.~ ~ ~ z ~ ~ ~ ~is ~ ~ ~ ~C ~ Hi ~C - O ~.~ ~ ~ ~0 . So + C ~ C ~ ~ ~ ~ 3 ~ ~ c ~ ~ ~ c m.~ ~ ~ ~ e ,, ~ O u°~ O ,, ~= e ~ ~ e O ~ ~ ~¢ ~ ~ ~ ~ ca ~ ~ ~ ~ ~ ~ .^ ~ ca ~So ~ ~ ~e ~ = ~ ~ - ~ ~ , . ~ ~ ~ ~ ~v ~ ~ ~ ~ ~ O v => ~ ¢ ~ ~ e ~ ~ ' ~ ~ e ~ ~ ~ ~' =' == ~ ~ o ~ R ca =4 ._4 .g ¢ ca V, ~ . ~ m Z O ~ 0 . ~ ·0 ca ¢ ~ m 0 ~ ·^ 0 0 4= ~ ~ 0 V, ·~ ~ Z O ~ ~ Z . ~ ca . ~ .= ~ ca ~ ·~ ~ V~ ~Q S~ ~ m 0 X . 0 ¢ ~ m V ~V ~ ~ p~ . ~ ~V~ ~ ~- -> O° 4.;, V,z V,

100 ASSESSING POLICIES FOR RETIREMENT INCOME unemployment insurance records system would be necessary to identify and appropriately characterize establishments that are part of larger enterprises. · The survey publishes estimates of how many employees in various cat- egories are covered by certain kinds of benefits, but not estimates of how many employers are providing these benefits; the latter estimates are needed to help understand employer benefit behavior. . The survey microdata could be a source of employer-based estimates, but very few characteristics of employers are available for cross-classification pur- poses (these include whether public or small or large private employer, standard industrial classification, and geographic region), and sample sizes are small for analysis of specific employer types. . The survey obtains detailed information on benefit plan characteristics (e.g., vesting period and type of pension plan) but not on employer (or employee) costs. (The EBS sample was queried on a one-time basis in 1993 about employer health care plan costs to provide information for the health care reform debate; a related BLS survey on Employment Cost Trends (ECT) obtains information about wages and benefit costs in a form that pertains to types of jobs and not employ- ers.) · Microdata files that can be obtained for analysis purposes only upon request cannot be linked with other data sources, such as the Form 5500 database (described below). Form 5500 Database Private employers have been required since 1975 to file information annually with the IRS about pension, welfare, and fringe benefit plans. More detailed information is filed for large pension or welfare plans (those with 100 or more participants) than for smaller plans; more detailed information is also filed for defined benefit plans than for other pension plans. The IRS transmits the Form 5500 data to the Pension and Welfare Benefits Administration (PWBA) in the U.S. Department of Labor (DOL), which edits the information and makes available a research database of all large pension plans and a 5 percent sample of smaller pension plans. PWBA also publishes summary information from the database twice a year (see, e.g., Pension and Welfare Ben- efits Administration, 1995a). The information for each plan includes various financial characteristics, number and type of plan participants, and total employ- ees and those excluded from the plan for various reasons. Researchers have used employer identification number (KIN) and plan number to link the Form 5500 database records for longitudinal studies; in many of these same studies, they have also linked the Form 5500 data to financial information for publicly owned, for-profit employers abstracted in the Standard and Poor Compustat database from annual reports filed with the Securities and Exchange Commission (see, e.g., Bajtelsmit, 1996; Ghilarducci, 1996; Kruse, 1991~. While an increasingly important resource for researchers (see, e.g., Clark and McDermed, 1990; Employee Benefit Research Institute, 1996b; Papke, 1995;

DATA NEEDS 101 studies cited in the preceding paragraph), the PWBAIDOL Form 5500 database is limited in many respects: · The database pertains to pension plans of private employers, but informa- tion that is filed for health insurance and other welfare benefit plans (e.g., life insurance, disability insurance) is not included because of inconsistencies in reporting, and there are no reports for public employers. · Within the universe of private employer pension plans, the database ex- cludes some plan types: for example, employers are not required to file reports for "model simplified pension plans" or for plans that do not qualify for tax- exempt status, and reports filed by self-employed people whose plan covers the person and his or her spouse (Form 5500EZ) are not picked up. Also, special "window" offers are not documented in the database. · The database has limited information about employers per se, although researchers have obtained some added information for publicly owned, for-profit employers by linkage with the Compustat database. Because the reporting unit is the plan, the linkage to the employer is not always accurate (e.g., when subsidiar- ies do not indicate that they are part of a holding company).l4 The database has limited information about an employer's work force. Beginning in 1992, defined benefit pension plans but not defined contribution plans are asked to report the number of participants and average compensation by 5-year age and service groupings (average compensation is not to be reported for any grouping that contains fewer than 20 participants). However, these data are not being keyed or provided to PWBA. · Summary Plan Descriptions that provide additional information about features of both pension and welfare benefit plans are available as paper copies only and are often out of date: the filing requirement is once every 5 years or every 10 years if there has been no change. (For defined benefit pension plans, there is an added requirement to file annually a brief description of features that are used in calculating funding requirements.) . As part of an effort to reduce regulatory burden, beginning in 1996 IRS and DOL curtailed some of the information required of smaller employers on the Form 5500. Census Bureau Longitudinal Research Database (LRD) The Census Bu- reau maintains a longitudinal database on U.S. manufacturing establishments that derives from the quinquennial Census of Manufactures (with data back to 1963) 14Longitudinal matching of employees and plans is hampered by similar problems; achieving a satisfactory match rate over several years requires the use of other information besides KIN and plan number.

102 ASSESSING POLICIES FOR RETIREMENT INCOME and the Annual Survey of Manufactures (with data back to 1972~. The database has been used for a growing number of innovative studies about such topics as the relationship of business investment and technology use to productivity (see Cen- ter for Economic Studies, 1995~. LRD data on establishments were recently matched to data from the 1990 census long-form sample on their workers, includ- ing demographic characteristics, occupation, and income (see Troske, 1995~. The LRD is limited to manufacturing establishments. The Census Bureau is working to develop a Longitudinal Business Database (LBD) that will include comparable data on establishments in other industrial sectors (e.g., services). If both the LRD and LED were linked to census data on workers, they could be a resource for studies of employer demand for older workers, although neither database has or will have information on employer benefits. To protect confiden- tiality, research with these databases is restricted to on-site use, although the Census Bureau recently set up secure offices for research use at two locations outside the Washington, D.C., area (see discussion below in "Expanded Use of Administrative Datable. National Employer Health Insurance Survey (NEHIS) The NEHIS was first conducted by the National Center for Health Statistics in 1994 of a large sample of public and private employers, including self-employed people, in order to provide needed data for the health care reform policy debate (although data were still not available as of the end of 1996~.15 As part of the plan of the U.S. Department of Health and Human Services (HHS) to redesign and integrate the department's health surveys, the NEHIS will become an annual survey, begin- ning in 1997, with a sample size of about 25,000 employers and a design that permits state estimates.l6 The survey covers the universe of employers and has extensive information on health care benefit plan provisions and costs, but it lacks detailed employee characteristics or any information on pension plans and other nonmedical benefits. A longitudinal component, as yet unspecified, will likely be built into the NEHIS sample design. U.S. Establishment and Enterprise Microdata (USEEMJ File The Small Business Administration (SBA) maintains a file of basic information about pri- vate for-profit businesses that derives from the Dun and Bradstreet Market Iden- tifier File (which NEHIS also used in drawing the largest part of its sample). The 15Prior to the NEHIS, the Health Care Financing Agency (HCFA) in HHS conducted a Survey of Private Health Insurance Plans (SHIP) in 1989, which covered about 4,000 private employers, 1,000 state and local governments, 700 labor unions, and 800 membership associations (see Garfinkel, 1995). 16Administration of the NEHIS will be integrated with the health insurance provider survey com- ponent of the Medical Expenditure Panel Survey (MEPS). This component obtains health insurance plan information from employers of household respondents included in MEPS.

DATA NEEDS 103 SBA USEEM File (see description in Jack Faucett Associates, 1990) has occa- sionally been used for one-time special surveys of employer benefits and work force characteristics. Scott, Berger, and Garen (1995), with financial support from SBA and the National Science Foundation, conducted such a survey in 1991 of about 2,250 employers. Results from the survey indicate that employers with health care plans are less likely to hire people aged 55-64 than are other employ- ers, but that whether the employer offers a defined benefit or defined contribution plan has no effect on hiring of older workers. However, the response rate for the survey was very low (29%), so that the resulting analysis sample is small and very likely biased. Private Sources Private benefit consulting and actuarial companies and professional and trade associations collect significant amounts of information about employer benefits; see Table 4-3 for features of selected private sources. Some of these sources have the potential to fill important gaps in government databases: for example, the Public Pension Coordinating Council provides information on features of public employer pension plans that is not otherwise available from federal sources (which include surveys of governments by the Census Bureau). On the whole, however, private sources tend to be even more limited in focus than federal sources of employer data: many cover only defined benefit pension plans, or health care plans, or large employers. Information in these sources on employer and employee characteristics (other than plan features) tends to be very limited. Also, many private data sources rely on convenience samples of their clients, which may not be representative of all employers or even large employers (see Mitchell, 1991, for reviews of several private data sources). In addition to the databases developed by benefit consulting companies and associations, individual researchers and research groups have carried out original data collection. Some researchers have conducted special-purpose, one-time surveys of employers; others have obtained detailed information from one or a handful of companies to use in case studies of employer hiring and compensation policies and their effects on workers; see Box 4-8 for examples. These data sets have supported some very innovative research, including the few attempts to measure and relate worker productivity by age to compensation patterns (Kotlikoff and Gokhale, 1992; see also Medoff and Abraham, 1981~. However, the results of special surveys and, even more so, case studies of one or a few employers are inherently limited in their generalizability, although if they are focused on certain groups, such as large employers, they may yield some gener- ally-applicable results for those groups.

104 4= o v Cq a' o o Cq a' Cq a' .> a' C) a' a' VO o Cq a' a' V, ca so .~ ;~6 ~ C) o V, ¢ ~ EM ~ R O c' ·0 ~ ~ ca ca - ~ ~ ~ S ~ ~ ~ ~ - ~ ' - ¢ ~ ca ~ O ~V ~ ~ ~ ~V ^ ca . 3 ~ ' ~He ~= ~ ~ 3 in ce ~R ~ v, ~ ,,, ,~ ~ :r, C ~ C ~ ~ ~ O ~C ~ ~ ~ ~R ~ R ~ o ~ ~ ~ ~ O _ ~ R . . .. ~ O ~ ·0 E~ ^ ~ca ~ 0 0 =, ,, ~ ~ ~ a ~ ~ O ~ v' c m ~ ~ ~ 0 m m ~ z ~ ¢ ~z ~ ~

105 of x ~o ~ ~ r ~ ~ ~ , ~ ~ ~ ,~ r - ~ " ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ O ~ .0 ~ ~ 4= :O 4= SO o 4= · · C) ~ 2 ~ ~ '^ ~ ~ ~ . ~ em ', 'a ~ · ~ m ~ ~ ~ ~0 ca Omit ca ;^ 0 ~ ~ ~ to 0 0 i ~ ~ ;^ . ~ ca . ~. ~ 4= ;^ Ct ;^ ;~. C) ~ 0 ~ O Vim ~Vim Vim C) ca .^ ~by ~ ~ 0 ~0 ~ _ ,, c,, ~ ~ o ~3 ~O _ ,~ O _ A ~~0 ~ ^ sit ~3 c, c, ~ ~ ~ ~ ~vim ~vim ~

106 a' .= o ¢ EM R o V so ~ ~ V, C) ;~. ~ C) O ~ ca 4= ~ .^ ~ ~;^ _, a ~ ~ ~ 2 me, ~ [ ca Ct 4= ~ 3 0 R ~ ~ - 0 V ~ To ~ ~ ~ ~ ~. - to ~ ~ . - . V ~ O O ~ ~ ~ ~ ~ O ~ a ~ -em a ',e,~!,,~e ~ R ~ ~ ~ O ~ ~ ~ so o ~ at R t ~d ~ ~ R ~ ~ ~ tD .~ R R ~ V, ~V, .0 ~ ~ ~ ~ C) .0 ·- ~ ~ ~ ~ ; o c) o v o · - ~ 4= z ca s~ 4= s~ 4= ca .~ .e ¢ 4= s~ 4= 4= v, o o · - ~ 4= c) ca ¢ o ~ 4= s~ ) ~ cd o E~ · - ~ ·~) o o ~ ca c) ¢ ~ u, o s~ v c) . o .0o ~ z 4= o 4= s~ s~ R ~V ca 4= ;^ V, 4= ;^ o C) · _4

DATA NEEDS 107 Problems with Employer Surveys Academic and private and public sector analysts have made imaginative use of a variety of data sources about employers. However, the available data are frag- mented and greatly deficient for comprehensive analysis of trends and behavior with respect to employer demand for older workers and benefit offerings and their effects on workers. A reason for the lack of much analytically useful data for employers and their workers, in contrast to the relative richness of data for households and people, is the greater difficulty of surveying employers (see Cox et al., 1995; Federal Committee on Statistical Methodology, 1988~. The prob- lems involve every step of the process. Developing and Maintaining a Comprehensive List of Employers as a Sampling Frame Business organization in the private sector is complex and dynamic: mergers, splits, liquidations, relocations, and, especially, start-ups make it difficult to keep a list complete and up to date and to identify which employers are the same over time. The same phenomena also make it hard to characterize employers by size, number and location of establishments, and industry, vari- ables that are important stratifiers for cost-effective sample designs. In the public sector, incorporations, annexations, and other jurisdictional changes complicate maintenance of an accurate employer list. A growing trend toward self-employ- ment requires a method of sampling these people (e.g., selecting them from a household survey, as was done for the 1994 NEHIS). Identifying Appropriate Sampling and Reporting Units Enterprises with multiple establishments (i.e., separate business locations) complicate decisions about survey design whether to have an establishment-based sample (as is com- mon) or an enterprise-based sample, each of which has advantages and disadvan- tages. There are also issues of the appropriate reporting unit and how to mini- mize respondent burden. For example, information about benefit plans may be maintained at the establishment level, or it may be maintained at headquarters for all establishments, even when the benefit plans differ among establishments for such reasons as state regulation (see discussion in Chapman, Moriarity, and Sommers, 1996~. Simply characterizing the enterprise-establishment structure can be difficult in many situations (e.g., holding companies, conglomerates, mul- tinational companies). Nonetheless, it is vital to understand the enterprise-estab- lishment distinction in comparing survey results; distributions of workers by size of employer, for example, will likely differ between the two kinds of surveys (see discussion in Zarkin et al., 1995~. Identifying Appropriate Respondents Finding the right respondents in an enterprise or establishment can be difficult: for example, to understand the

108 ASSESSING POLICIES FOR RETIREMENT INCOME ............................................................................................................................. reasons for changes in benefit plans, the chief financial officer may be a more relevant respondent than the chief personnel officer. Obtaining Information It can be difficult to obtain comparable information across employers: for example, definitions of fiscal year differ, as do definitions for other characteristics. More important, it is often difficult to obtain any re- sponse to surveys (or administrative records systems) that have many or very detailed data items. One reason is that the more complex the data request, the more burden it likely imposes on employers to consult more than one set of records in multiple locations and departments. Another reason is employers' concerns about revealing information that might affect their competitive position.

DATA NEEDS 109 - ............................................................................................................................ ............................................................................................................................. ............................................................................................................................. ................................................................................................................................................................................ More generally, many employers have "gatekeepers" to steer away inquiries, particularly those directed to management. As a consequence, even when limited information is requested, response rates to voluntary business surveys tend to be lower than response rates to house- hold surveys. Rates are particularly low for surveys conducted by mail and by private organizations and researchers, compared with those conducted by govern- ment agencies (see Christianson and Tortora, 1995; Paxson, Dillman, and Tarnai, 1995~. With intensive efforts, the BLS Employee Benefits Survey obtains quite high response rates for medium and large private establishments (85% in 1989) and for governments (94% in 1990), but its rate for small private establishments is significantly lower (72% in 1990~. Also, 21 percent of the original sample of

110 ASSESSING POLICIES FOR RETIREMENT INCOME small private establishments was determined to be out-of-business or out-of- scope (see Bureau of Labor Statistics, 1992a). In comparison, the household response rate to the BLS-sponsored Current Population Survey averages about 95 percent each month. Providing Anonymous Data Many employers are concerned that they could be identified in survey data releases. Because of the relatively small numbers of establishments and enterprises and their skewed distribution by size, it is in fact difficult to provide analytically rich microdata for employers that protects the confidentiality of individual replies. Confidentiality concerns also affect the maintenance of a government-wide employer list: the Census Bureau Standard Statistical Establishment List is not shared with other agencies. Agencies like BLS maintain their own lists, and there is evidence of discrepancies in complete- ness of coverage and industry categorization among lists (see, e.g., Federal Com- mittee on Statistical Methodology, 1990~. Directions for the Future Clearly, the development of analytically useful databases on employers is a ne- cessity for retirement-income-related research and projections, and we believe it should be an important government goal. Below, we identify priority areas and strategies to move toward that goal. However, recommendations like ours that are developed primarily from a data user perspective need careful consideration in a broad context. The gaps in available data on employers are sufficiently large and the problems of employer data collection sufficiently formidable that a mea- sured and comprehensive approach is needed to determine feasible and cost- effective goals for improvement, both in the near term and the longer run. To this end, we urge the Department of Labor to establish an interagency task force on retirement-income-related employer data. The charge to the task force should be to develop and implement an integrated, multiyear plan for pro- viding the data needed to understand and project employer demand for older workers and the causes and consequences for workers of changes in benefit offerings. We suggest that BLS and PWBA jointly lead the task force: PWBA has responsibility for oversight of pension and other benefit plans and hence has the most policy interest in improving employer data; BLS brings extensive statistical experience and expertise. The task force should also involve other agencies with relevant analytical interests and expertise, such as the Pension Benefit Guaranty Corporation, relevant HHS agencies, and the Census Bureau. In addition, the task force should involve people in the research and modeling communities- including the principal investigators of panel surveys of individuals, such as HAS/AHEAD, that collect employer data along with private benefit consult

DATA NEEDS 111 ants and representatives of public and private sector employers. Involving data users is essential to focus the work of the task force on priority research and projection modeling issues. Involving employers and benefit consultants, who are often both users and providers, is essential to develop cost-effective methods for collecting and providing analytically useful information. The task force needs to get information about useful and feasible data collection from large and small employers in various sectors of business and with different benefit struc- tures: sponsors of private single-employer pension plans, private multi-employer plans, and public plans are subject to different regulatory regimes with possibly differing data implications. Benefits: Descriptive Information A critically important area for the task force to consider is how to improve data on employer benefits. At present, there is no comprehensive series with which to characterize and track benefit offerings for the entire universe of employers. The Employee Benefits Survey comes closest, but it is designed to describe benefits available to categories of employees and not benefits offered by types of employ- ers; also, at present, it lacks important information, such as benefit costs, and has other limitations. The recommended DOL task force needs to consider how best to develop a baseline picture for public and private employers and the self-employed and update it on a regular basis with information on benefit features and costs (includ- ing pension, health care, disability, and life insurance plans), employer character- istics (e.g., number of employees, financial characteristics, wage structure), and work force characteristics (e.g., age structure). Frequent updating is critical because benefit plans are dynamic, changing frequently. For reasons of cost and feasibility, the task force should give priority to modifying and coordinating existing data systems as far as possible toward the goal of a comprehensive database. Involvement of public and private sector employer representatives is critical in determining modifications to enhance data quality and utility that are feasible to implement. Given the overlap in existing surveys, it may be possible to reduce reporting burdens or, at least, to identify some information that can be dropped or curtailed to make room for needed new items. Data systems to consider for modification include the Employee Benefits Survey, the Form 5500 database, and the National Employer Health Insurance Survey. In addition, data collected by the Public Pension Coordinating Council, if coverage were improved, could serve as an important resource with which to augment federal agency sources on state and local government employer benefits at little or no added cost. Although the federal government does not regulate state and local government benefit plans, retirement-income-related research and policy analysis needs to consider the full range of employers.

2 ASSESSING POLICIES FOR RETIREMENT INCOME Employee Benefits Survey Improvements to the EBS and the companion Employment Cost Trends survey (which includes the data published as the Em- ployment Cost Index) were recently discussed in a meeting of researchers con- vened by BLS (MacDonald, 19951. Suggestions included adding questions on health care and other benefits for retirees, providing more data for establish- ments, facilitating the construction of estimates for establishments, and providing microdata for research purposes that could be linked with other sources of em- ployer data. Currently, BLS is working to integrate the EBS more closely with the Employment Cost Trends survey and with local-area occupational compensa- tion surveys. Such an integrated system could potentially bring together informa- tion on benefit plan provisions, employee participation in benefit plans, and employer costs. The task force should review these and other suggestions and goals for the EBS in the context of developing an overall plan for comprehensive employer benefit information. The Form 5500 PWBA Database This database has many limitations for research use, but it offers information not otherwise available on pension plan costs and coverage for all large employers and a large sample of small employers. Efforts to reduce government regulation may curtail the reporting requirements for the system. We urge the recommended task force to determine the usefulness of the system, both as it exists now and if it were improved, in the context of an overall plan for obtaining comprehensive benefits data. The use of the system for this purpose should be a factor in decisions about reporting requirements. Relatively low-cost improvements for this database include: . linking Form 5500 records over time to provide panel data; · merging Form 5500 records with employer financial characteristics such as those in the Compustat database for publicly owned, for-profit companies; · making the data more timely and accessible, perhaps by providing the data files on the Internet. With regard to linking and merging, researchers have carried out linkages on their own, but it would be useful for PWBA to develop standard linking procedures that handle such problems as business mergers and splits. It would also be useful for PWBA to perform linkages on a regular basis and to determine if there are ways to fill gaps in Compustat and other information.~7 If staff resources are a constraint, consideration could be given to contracting for these activities. We also suggest exploring ways to improve the consistency of reporting of welfare benefit plans, particularly health care and disability plans. Form 5500 lain the future, the EDGAR database of the Securities and Exchange Commission may provide more complete financial coverage than Compustat.

DATA NEEDS 113 data for these plans are not currently provided in the PWBA database. Similarly, consideration could be given to including in the database the newly required information on demographic characteristics of participants in defined benefit pension plans and to obtaining such information from defined contribution pen- sion plans and other benefit programs. A more difficult question is how to obtain useful information at reasonable cost from the narrative Summary Plan Descriptions that are filed with the Form 5500, which does not itself describe plan features. An information-rich but costly procedure would be to require more frequent filing of the descriptions (which are often out of date) and to regularly abstract and code analytically useful informa- tion from them. Many private benefit consulting firms categorize features of a large variety of plans, and their categories could be the basis for a coding scheme, as could the categories used in the EBS. A less costly and less information-rich procedure would be to develop a limited set of categories of plan features to add to the Form 5500 itself and to drop the requirement to file Summary Plan De- scriptions. If this option were adopted, it could still be possible to obtain detailed plan feature information and the cost and coverage data in the Form 5500 data- base for a sample of employers by providing the applicable Form 5500 records to BLS to merge with the EBS data and prepare a file that could be tabulated at researchers' request. Finally, to reduce regulatory burden and costs but provide information for monitoring and research purposes, consideration could be given to conducting the Form 5500 data collection program on a sample basis. If it is necessary to have complete reporting in order to monitor compliance with pension law, then consideration could be given to asking a sample of employers information that would be useful for research purposes but is not necessarily needed for compli- ance. The sample design should consider the need for information from employ- ers of different types and sizes. The design should also build in a longitudinal component, so that changes in benefit provisions can be analyzed behaviorally as well as on a time-series basis. Finally, the design should consider the degree of integration that is feasible with the sample design for the Employee Benefits Survey; without some degree of integration, it will not be possible to relate the data from the two systems. National Employer Health Insurance Survey The U.S. Department of Health and Human Services plans to conduct NEHIS annually to obtain detailed information on employer health care benefit plan features and costs. NEHIS will play an important role in an integrated system of health status and health care surveys. In this context, it is probably not practical to turn NEHIS into a full- scale employer benefit plan survey, including pensions and other benefits. How- ever, it might be useful to broaden the survey on a periodic basis. If this were done, there would be less need to find ways to improve reporting of health care and disability plans in the Form 5500 series.

4 ASSESSING POLICIES FOR RETIREMENT INCOME Determining the set of possible options for improved data on employer ben- efits and then determining which options appear more cost-effective is a chal- lenging task. Is it preferable, for example, to enhance the Form 5500 system and rely less on such surveys as EBS or NEHIS? Alternatively, is it preferable to take advantage of the large sample size of NEHIS by periodically expanding the NEHIS questionnaire to cover pension as well as health care benefits? The recommended task force will need to address such questions in order to find cost- effective ways to provide needed information while minimizing duplication among data collection systems. Labor Demand: Case Studies Another important but difficult area for the task force to consider is how to obtain data for analyzing and projecting employer demand for older workers. For this purpose, detailed information is needed for employers with which to relate work force composition by age with compensation, benefit costs, and worker produc- tivity by age. Obtaining such information presents formidable problems, includ- ing not only employer reluctance to furnish the level of detail required, but also the fact that some of the information may not be readily available. For example, employers may not know the details of their health care benefit costs by employ- ees' (and dependents') ages. Estimating productivity differences by age presents particularly difficult con- ceptual and measurement problems. Pioneering work by Kotlikoff (1988) and Kotlikoff and Gokhale (1992) determined age-productivity differences among employees of a large company by comparing the expected values of total com- pensation, including wages and pension benefits, across age cohorts. This method is indirect and involves some strong assumptions (for commentary, see Lazear, 1988~. Medoff and Abraham (1981) related performance evaluation and salary data for white collar employees of a large company, not to age per se, but to length of service. However, the suitability of performance evaluations (where they exist) for measuring age-productivity differentials has not been established. While case studies, such as those conducted by Kotlikoff and Gokhale and Medoff and Abraham (see Box 4-8 for other examples), are not readily generaliz- able, they offer a feasible and cost-effective way to begin to address the data collection problems in the area of employer demand for older workers. We suggest that the recommended DOL task force, with the cooperation of private and public employer representatives, arrange to sponsor a series of employer case studies. The studies should be chosen to include a variety of employers confront- ing a variety of situations. They should focus not only on substantive issues of the factors that affect labor demand, but also on ways of obtaining needed data that could be feasible for more structured, representative surveys. They can do so by exploring a wide range of potentially useful information for example, atti- tudes and perceptions of operating, financial, and personnel officers and of em

DATA NEEDS 115 ployees and by investigating cost-effective methods and sources of data collec- tion that could be implemented on a larger scale. Medoff and Abraham (1981) support such an approach. They comment (p. 215) on the need for richer data with which to empirically test theories of experi- ence-earnings differentials: It is our belief that major steps [to obtain needed data] can be taken through interaction with those who formulate company compensation policies and with those affected by these policies. In particular, interviews with the members of top management who are responsible for the outlines of a company's pay prac- tices should be conducted; discussions with supervisors about how they deter- mine the proper salaries for their subordinates should be initiated; and the atti- tudes of employees toward different compensation schemes should be assessed. . Moreover, we should seek data that would permit analysis of the impact of changes in the nature of firms' compensation practices on things such as pro- ductivity, quits, discharges, ability to attract new hires, absenteeism, and job satisfaction. Panel Data for Behavioral Analysis improvements in such data systems as the EBS, Form 5500 series, and NEHIS can provide valuable information for tracking trends in employer benefit offer- ings and analyzing some of the factors involved. However, such systems are not likely to provide the richness of detail that is needed for in-depth behavioral analysis of employer compensation policies and their effects on workers. Case studies will also not suffice to establish the underlying mechanisms in employer decisions about recruitment and retention of older workers. What is needed, ideally, is an employer-employee panel survey that provides detailed information on characteristics of a sample of employers linked with characteristics of a sample of their workers. Alternatively, given that confiden- tiality concerns could prove a barrier to implementing such a survey, an employer panel survey that provides aggregate information about the work force character- istics of sampled employers would be useful. In either case, the employer sample would need to be refreshed periodically to include new businesses. The survey should include data on the financial and other characteristics of employers and skills, earnings, benefit plan participation, and demographic and family charac- teristics of employees. Similarly to HAS/AHEAD for individuals, the survey ideally should also include data on attitudes and perceptions of corporate officers, supervisors, and workers, although confidentiality concerns may make it difficult to obtain such information. The recommended employer data task force should carefully consider the i8See Gustman and Mitchell (1992) and Parsons (1996) on the need for such a survey.

116 ASSESSING POLICIES FOR RETIREMENT INCOME feasibility of an employer-employee (or employer) panel survey and its costs and benefits. Such a survey would undoubtedly be expensive. It could also encoun- ter more than the usual problems of employer cooperation and response (given the level of detail and possible sensitivity of the information required), but the data provided would be very valuable for needed research on factors in employer behavior and its consequences for workers. There is little experience with this type of survey in the United States. In the 1970s, BLS sponsored several Quality of Employment Surveys of small samples of workers that included a panel component. Information was collected in these surveys on many aspects of the employment situation (e.g., work-related prob- lems, work attitudes and behaviors, job/task characteristics, earnings, fringe ben- efits, and noneconomic forms of compensation), but no questions were asked of the sample members' employers or about benefit costs. An advisory group recommended expanding the program to include a survey of employees that also obtained limited information from their employers, together with a survey of employers that also interviewed samples of their employees (Kalleberg, 1986~; however, no further surveys were conducted. The Census Bureau recently matched 1990 census data on workers with employer data from its Longitudinal Research Database. Future matches may be possible if plans are approved for a very large continuing household survey with census-type content (the American Community Survey). However, the LRD is limited to manufacturing establishments, which include only 20 percent of the work force, and census data for workers are very limited in scope. (A longitudi- nal database for nonmanufacturing establishments is under development.) Of possibly more relevance, Statistics Canada (the national government agency) is launching an employer-employee panel survey that may offer guid- ance for a U.S. effort (Statistics Canada, 1995~. A pilot test of the Workplace and Employee Survey was conducted in early 1996 of 1,000 establishments and 6,000 of their employees. The production version of the survey will begin in 1998 and continue yearly; the sample will include 5,000 establishments and 30,000 em- ployees. Establishments will be tracked over time with periodic replenishment of the sample; employees will be tracked as long as they remain with the same company and for one period after they leave. The survey content covers a range of topics, including: work force characteristics, compensation, training, business strategy, financial performance, and technology use for employers; and job char- acteristics, pay and benefits, training, participation in decision making, educa- tion, recent work experience, and family situation for employees. A way to begin providing panel information on U.S. employers for behav- ioral analysis and to explore feasible methods of data collection would be to build a continuing employer survey into HAS/AHEAD. In the first interview wave, HRS asked information with which to locate sample members' employers and obtain pension and health care plan descriptions from them. The survey also

DATA NEEDS 117 asked employed sample members about benefit plan features and other aspects of their job and work place (e.g., perceptions of employer attitudes toward older workers); see Box 4-9. At a minimum, it appears desirable and feasible for subsequent waves of HAS/AHEAD to collect employer benefit data. It is important that the collection cover all employed sample members, including those with the same job, because of the likelihood that some employers will change their benefit provisions. If reporting in the Form 5500 series is improved, it would be possible to obtain plan descriptions from this source for a large fraction of HAS/AHEAD respondents without having to contact their employers. In addition to using the employer benefit plan descriptions for such pur- poses as estimating future benefit entitlements, the information can be compared to self-reports in HAS/AHEAD. One important question such data would ad- dress is whether employees' knowledge of benefit plan features and the implica- tions for how best to prepare for retirement is improving over time. Research has shown that workers often do not know or give incorrect answers to questions about important pension plan provisions, such as whether and at what age early retirement is possible: see, for example, Mitchell (1988), who compared worker responses in the 1983 Survey of Consumer Finances with information provided by employers about the workers' pension plans. For this purpose, it would be useful to ask both the sample members and their employers about employer programs for educating workers about pension and other benefits. The Depart- ment of Labor is making such employer education programs a priority initiative. The Social Security Administration is also attempting to provide covered work- ers with more information about benefit entitlements under current lawful Going further, HAS/AHEAD could conduct a full-fledged survey of sample members' employers, with information on such characteristics as return on in- vestment, benefit plan costs, wage structure, and age distribution of employees. Some of this information could be obtained from publicly available sources, such as the Compustat database, thereby reducing respondent burden. There are several advantages of incorporating an employer survey into HRS/ AHEAD. The employer sample is readily generated from HRS sample members' responses and would have a built-in longitudinal component. It would provide a vastly richer set of employer characteristics than ever before available with which to analyze individual savings and retirement behavior. However, for analysis of employer behavior as such (e.g., factors in benefit plan decisions), a piggybacked survey is not an efficient sample design. Important categories of employers, such i9See Employee Benefit Research Institute (1995, 1996b, 1996c) for research, based on case studies and special surveys of workers and participant-directed defined contribution pension plan sponsors, on employer financial education programs and their effects on workers' pension plan contributions.

118 ASSESSING POLICIES FOR RETIREMENT INCOME ............................................................................................................................. D _ . . . - - _ _ _ - _ _ _ _ . _ . ::::::::::::::::::::::::::::::::::::::::::::::: :::A::: :::A::::::::: : ~ ':::: ::: : n ~ v ~A A: :A:: :T: :: A:: : ~:~: :: _~:::: :: ~ I::::::: . :: A: A:::::::::::::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:-~-:-~-:-~t:-:-:-:-:-:-:-.:- 1:-:-:-:-:-:-:-:-:-:-:-:-:1 -:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-.:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-~:~-:-:-:-:-:-:-:-:-~:-:-:-~-~:~-:.-:-:-:-:-:-:-:~-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: ~nln=~n-=r--l^-n-~---T^rm--- El-~-~-nt-~-~--~nel leaned or. :::::::::::::::~:~:~:I~.-:~!:I:~l::::l:V:!:l:~ :~:1:~:!:1:::~1:~. :I:11:tY:::l:!:l:~:~:l:~·-:I~.~ :~. 1 v-v---: ::::::::::::::: ::::::::::-:::::::::::::::::::::: :: :::::::::::: ::-: :::::.::::::::::::::::::::::::::::::::::::::::::::::::::::: ~-emU--mS ~ nealtn lE SE Pranced cove- ane Tor sent lsee Box ~ 0 ::::::::::::::: :::::::::1:::1:::::::::::::::::::::::::::::1:::::::: ::: ::::::::::::::::::::::::::::::1::::::::::::::::::::::::: E n ~ r ' n r= ~ n ^ n n a n r e n ~ r P a ~ ~n ~ l l rO n n E m E E . _ . ~ _ :-:-:--m ~ ~^~- - -~--^^ ~- - -^T- - -~ -m ^-l-~r - - A ^~-^m- m^~-~-l~-~ :::::::::::::::::::::::::::::::::::::::::::::::: :~:li~:~:~.:::~::~:~: ~::~:~::::_:~:~:. ~:~:~:~::::A~:~:~:~:.':~:~:~:~:~::::::::::::::: -:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-..-:-:-:-:-:-:-:-:-:-:.. :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-..-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: ,.,.,.,.,.,.,.,.,.,.,.,.,.,.,o,,,,pe,,,c,,,l,,l.,l.c,,,.,.,KlnEs ol a'Iu ................................................................................................................................ :::::::::::::: :: :::::::::::1 :::::::::::::: ::::::::::::::::: :::::::::::::::::::::::::: ::: 1 :'::::::::::::::: 1 1:*:::: :::: E EUm~e~ OT ~mpEvypes al lE |S =~a~lES E E=nl (EoCalloE E~ -:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-.-:-:-:-:-:-:-:-:-:-.-.-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: ::::::::::::::: :i ::::: 1::::::::::::::::::::::::::::::::::::::1::1::::::::::::::::::-::::::::::::::::::::::: ::::::::::: :::::::: - - - - - - - - - - - - - - .,'. ~-T^-^ r - -~-~-~-r^~- - -~-/ - - l- -e ^-l^~- - -~-~-'r.~- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - :::::::::::::::~:V:1:~`I::!~:!:::~:V:~:~::~:Y-:::~:!: l:lV:!:l::: - VI::!:~:!:~:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: E---- .:.:.-m-~-ln~/~---~-r^^nTInn-~--nT---~-~- ~^IaT-^ ~^r.-nnn~l ~r E~l'^~-':':':':':':':':':':':':':':':':':':':':':':':':':':':':' ~ .................................................................................................................................. - t Eolover wv Ela let vluer WVE ~e-rs~ m-vve---tv~ a~ tess~ ~-e Eanaina JOo WltE ~ l-ess~ ~ E - - - - _ -. - - :::::::::::::::::::::::::::::::::::::::::::::::: --: : 1 : ::~:~ - : : ~ : 57 t. : :: ~f ::::. :~::: ~:~:ie: : i :. ::::::::::::::::::::::::::::::::: :,, ,- - - i,:~:,~-~:~ V-:V -~-:-:~:~: B-- -:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-~-:-:-:-:-:-:-:-:-:-~-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: :::::::::::::::: :::::::-:: :::::::::::: ::::: ::::::::::::::::::::,::::::::::::::::: ::::::::::*:::::::: ::::::: ::::::::-:::::: v-v-netn-e-r--Jv-~ rea-u-l--res~ p-n-vslca-l~ eno-~ av-oo~ eves-l-a-nt~ ~-oo-~ m-e-m-o-~ l-ots---vI~ ~ :::::::::::::::::::::::::::::::::::::::::::: ::::: ::::::: :::: :::::: ::::! ::::::: ::: ::: ::::::: ::::::::: : ::: ::: ~: ::::::::::: ::::: ::::::::::::: ::::::: ::: ::: ::::::: :::: ::

DATA NEEDS 119 as companies with predominantly younger workers, may have high sampling variability and hence large standard errors of estimates. The alternative, as discussed above, is to design a continuing employer- employee (or employer) panel survey as a wholly separate data collection effort. Thorough analysis will be required to determine the costs and benefits of alterna- tive approaches to obtaining needed longitudinal employer data for behavioral research. At the least, HAS/AHEAD could offer a vehicle to experiment with methods of employer data collection to improve measurement and data quality. Recommendations The development of analytically useful databases on employers for retirement- income-related policy analysis and projection purposes should be an important government goal. At present, important information on the characteristics and determinants of employer benefit plan offerings and employer demand for older workers is lacking, incomplete, or not provided in a usable manner. Given the central role of employers in providing retirement income and health care benefits, the lack of an adequate database is a major handicap to evaluation of alternative policy proposals in these areas. 5. The U.S. Department of Labor should establish an interagency task force on employer data to specify an integrated plan for collecting retire- ment-income-related information. The plan should specify short-term and long-term goals that consider user needs, resource constraints, and the prob- lems of obtaining information from employers due to such factors as low response rates, locating the appropriate respondents, and confidentiality concerns. The task force should involve researchers, private benefit consult- ants, and representatives of public and private employers in its work. 6. The employer data collection plan should include short-term and long-term goals for obtaining improved information on the distribution across employers of all benefit plan offerings (including pensions, health insurance, disability insurance, retiree health insurance, life insurance). Comprehensive baseline information is a priority need, along with a plan for regular updating. Needed data elements include benefit plan characteristics and costs, employer characteristics (e.g., number of employees, financial characteristics, wage structure), and work force characteristics (e.g., age structure) for public and private employers and the self-employed. 7. The employer data collection task force should give priority to rede- signing and enhancing existing data collection systems on employer benefit offerings and related topics. Such systems include the Employee Benefits Survey, which currently provides information for broad categories of em

120 ASSESSING POLICIES FOR RETIREMENT INCOME ployees but not for employers, and the Form 5500 data series, which serves regulatory purposes and currently has limited research use. Consideration should be given to improvements to the Form 5500 series, including: · making the data more timely and accessible (e.g., on-line); · linking records over time to provide panel data; · merging the Form 5500 benefit plan information with the kind of employer financial characteristics found in the Compustat database; · working to standardize the reporting for health care and disability plans, so that they can be added to the Form 5500 database; and · finding ways to add information about benefit plan features to the database, perhaps by abstracting analytically useful information from the narrative plan descriptions that are filed with the Form 5500. 8. The employer data collection plan should include short-term and long-term goals for obtaining information on labor demand for older work- ers and the factors that may affect that demand. Needed data elements include employment patterns of older workers, compensation and benefit costs by age, and worker productivity by age. Very little information on these topics is currently available, and some raise difficult measurement issues. A reasonable short-term goal is to sponsor case studies of employers that can help identify important variables and feasible means of collecting them on a larger scale. 9. The employer data collection task force should consider the feasibil- ity and cost-effectiveness of a panel survey, which is periodically refreshed, that collects detailed information on employers and their workers. Such a survey should cover the full universe, including private for-profit, nonprofit, and government employers, and the self-employed. Longitudinal data from an employer-based survey are needed to analyze the factors that affect em- ployer decisions about recruitment and retention of older workers and ben- efit plan offerings and how these decisions, in turn, affect workers. 10. HRS and AHEAD should develop and implement a plan for obtain- ing information on a continuing basis on the pension and health insurance offerings of the employers of the HAS/AHEAD sample members. EXPANDED USE OF ADMINISTRATIVE DATA A recurrent theme in retirement-income-related research is the need for data from administrative records, either as stand-alone databases for analysis or linked with survey data. Such records can often provide important variables for individuals

DATA NEEDS 121 and employers at very low marginal cost. The major difficulty concerns how to provide access to such data for research and modeling purposes when their use raises concerns about maintaining the confidentiality of respondent information.20 Records on Individuals Greater access to Social Security Administration earnings and benefits records could advance many important areas of retirement-income-related analysis and modeling. As a stand-alone database, SSA records have the potential to improve U.S. data on mortality at older ages and to study the relationship of socioeco- nomic status (as measured by earnings levels) to mortality.21 Such studies could be carried out by SSA staff or by researchers who are sworn in as SSA employees to prevent disclosure of confidential data (as has been done for some Census Bureau studies). Given the importance of mortality projections for projecting retirement income security, we urge that priority be given to mortality research with SSA records. More problematic from the perspective of confidentiality protection are pro- posals to link SSA records with survey responses. Some studies have been done, but they have been limited. Exact-match files of SSA records with the March 1973 and 1978 CPS, developed by the Census Bureau, were made publicly avail- able (the 1973 file included an exact match with IRS records), as were exact- match files of SSA records with the Retirement History Survey. However, no exact-match files of SSA records with CPS data for years later than 1979 have been developed for public use. The Census Bureau has developed exact-match files of Social Security records with the 1984, 1990, and 1991 panels of the Survey of Income and Program Participation (SIPP), but these files are made available only to SSA analysts with strict restrictions on use. The Census Bureau recently released a public-use, exact-match file of the March 1991 CPS with selected data from IRS administrative tax records. In this file, techniques of data- switching and the addition of noise were used to mask the data so that no sensi- tive information that could identify specific individuals was released. More extensive matches of IRS data with CPS and SIPP files have been used to evalu- ate the quality of income reporting in the March CPS and SIPP and for research on improved weighting schemes to reduce the variance of SIPP estimates, but these files are only available internally to Census Bureau staff. The Department of Labor sponsored a 1977-1978 Survey of Private Pension Benefit Amounts that linked private employer pension plan records on beneficia 20See Duncan, Jabine, and deWolf (1993) for a review of confidentiality and access issues for federal statistical data and promising avenues for addressing the difficulties. 21If SSA tracked marital status of all beneficiaries, then SSA records could also support needed analysis of the relationship of marital status to mortality.

22 ASSESSING POLICIES FOR RETIREMENT INCOME ries with SSA earnings and benefits records (Office of Pension and Welfare Benefit Programs, 1985~. This survey used the Form 5500 database to sample private pension plans and obtain information from plan administrators on ben- efits paid to individual plan participants. The matched records of pension and Social Security benefits and earnings were used to analyze the contribution of employer pensions to retirement income security (e.g., to calculate earnings re- placement rates). The response rate from plan administrators was low (about 50%), and large defined contribution plans were underrepresented. However, the matched data were viewed as more accurate than household survey estimates of pension retirement benefits, which are typically underreported. No public-use files were made available from the survey, and it would presumably be difficult to do so if it were to be repeated. Legislative restrictions are one reason that publicly available exact-match files of SSA and survey data have not been developed in recent years. Another reason is that statistical agencies have become more concerned with questions of privacy and confidentiality of data and the potentially adverse effects on survey response rates if people believe that their replies are not held in strict confidence. Nevertheless, there is a strong need for exact-match files. Calculations of expected Social Security benefits require either complete histories of covered earnings or summary variables, such as average indexed monthly covered earn- ings over a worker's span of employment, that in turn derive from earnings histories. Such histories are difficult to obtain retrospectively in surveys and would require decades of data collection to obtain prospectively. Earnings histo- ries, including earnings above the payroll tax ceiling (available in SSA records beginning in 1979), are also helpful in calculating expected benefits from the types of employer pension plans that calculate benefits on the basis of several years of highest earnings with the employer or that specify employer contribu- tions as a percentage of earnings. Finally, benefit histories are useful to evaluate and augment survey responses of Social Security income. Plans are now being implemented to make available on a restricted basis exact-match files of HAS/AHEAD and SSA records that will provide very valu- able information for analysis purposes. (Links will also be made with HCFA Medicare data and possibly with state Medicaid data.) A three-pronged strategy will be followed to protect confidentiality. First, linked data files with complete earnings and benefits histories will be made available on a limited access basis only to researchers who sign nondisclosure agreements that include penalties for violation. Second, public-use files will include only summary variables derived from the earnings histories. Third, estimated Social Security entitlements that have been computed under a variety of assumptions will be made available to HRS users under restricted conditions (Mitchell, Steinmeier, and Olson, 1996~. We support the preparation of exact-match files that link SSA and other administrative records with HAS/AHEAD and urge that arrangements be made to perform these linkages on a regular basis. We also encourage the Census Bureau and SSA to consider the development of SIPP-SSA exact-match files that can be

DATA NEEDS 123 made publicly available by following the strategy of HRS and AHEAD, namely, to provide summary variables derived from the earnings histories that facilitate the calculation of expected Social Security benefits. (lams and Sandell [1996i, SSA researchers who are using matched SIPP-SSA files for Social Security ben- efit modeling, make a similar recommendation.) There are plans to include SSA information on Social Security benefit type, and whether the respondent has died, in publicly available SIPP files. We support these efforts and also urge consider- ation of developing SIPP files for public release that include derived variables from SSA earnings records. Records on Employers Administrative records for employers, such as financial statements that are ab- stracted in Compustat and the Form 5500 data series, provide useful information for analytic purposes. These particular data sets, unlike SSA records, are derived from public documents, but problems can arise when they are merged with other data for which confidentiality protection is promised (e.g., BLS or Census Bu- reau surveys). Employers are sensitive about the release of data that could be useful to competitors, and it can be very difficult to mask such variables as employer size sufficiently to prevent disclosure and at the same time maintain the analytical value of the data. Indeed, microdata from employer surveys, let alone matched survey and administrative records data, are often not made publicly available at all. Sometimes agencies are willing to retabulate confidential data at the request of researchers. For example, BLS has linked Form 5500 data with the EBS and run analyses for outside researchers. However, the researchers were not them- selves given access to the microrecords, and they found this mode of data access very limiting (MacDonald, 19951. One possible strategy to provide greater access to matched employer data is to adopt the strategy proposed for exact-match files of SSA earnings histories with HAS/AHEAD. Under this strategy, researchers could gain access to the complete data sets under very strict conditions of use. At the same time, public- use files could be developed in which key administrative records variables are summarized in a manner that is most relevant for research needs and other steps are taken (e.g., limited geographic identification) to prevent disclosure. If this approach is adopted for matched employer data, it would be important for agen- cies to consult with researchers to determine the appropriate summarized vari- ables. The Census Bureau is pursuing another very promising approach for re- search access to its employer data files, including the LRD, which have not been available for use except at the Bureau's headquarters. This approach may pro- vide a model for other agencies. Several years ago, the Census Bureau, in

24 ASSESSING POLICIES FOR RETIREMENT INCOME collaboration with the National Bureau of Economic Research, a private organi- zation, established a secure Research Data Center at its Boston regional office. Researchers may come to the center, be sworn in as special Census Bureau agents, and use the data sets on site. Census Bureau employees must review any output that researchers take with them to ensure that it does not identify specific respondents. Although more limiting than use of microdata at one's own institu- tion, this arrangement is far preferable for researchers in the Boston area than having to come to the headquarters in the Washington, D.C., area. The success of the Boston data center has led the Census Bureau to set up a second center at Carnegie Mellon University in Pittsburgh, and the agency is exploring research- ers' interest in having similar centers in other major cities around the country. Recommendations 11. Matched files of panel survey responses and key administrative records should be regularly produced for retirement-income-related policy analysis and projection purposes. Examples include exact matches of survey records with Social Security earnings histories and benefit records, Medi- care and Medicaid records, and the National Death Index. The added infor- mation in matched files is obtainable at low marginal cost and is essential for analysis of retirement and savings decisions and the effect of medical care use and expenditures on retirement security. 12. Agencies should collaborate on the development and oversight of matched data sets for individuals and employers, with input from research- ers on content. They should also vigorously explore creative solutions for providing research access to exact-match files that safeguard the confidenti- ality of individual responses. Possible solutions include: (1) developing public-use files that contain summary variables derived from the adminis- trative records portion of the matched file; (2) requiring researchers to sign nondisclosure agreements with significant penalties for violations; and (3) providing researchers with access to matched files on site at secure data centers. DATA VALIDATION Validation of databases that are used in behavioral and projection models is as important as validation of the models themselves. Sampling errors in data inputs are one source of uncertainty of model estimates; more important, nonsampling errors can introduce both uncertainty and bias into model estimates. Data valida- tion is essential to identify the types and magnitudes of such errors. It is also essential for survey methodological research, which should be part of every data

DATA NEEDS 125 collection program to determine procedures for improving data quality at the outset by improving questionnaire design and data collection procedures. There are many sources of nonsampling errors in both surveys and adminis- trative records. One source is unit nonresponse, that is, failure by a reporting unit to provide any information at all. Panel surveys are subject to cumulative unit nonresponse over time, or attrition, as people become tired of cooperating with the survey or move and cannot be traced. Other sources of error are nonresponse to specific items, overreporting (e.g., a false positive report of pension coverage), underreporting (e.g., reporting an amount less than actually received for an in- come source), and misclassification (e.g., reporting a defined benefit pension plan as a defined contribution plan or vice versa). Yet another source of error in surveys is undercoverage of the population because the sampling frame does not include all people or employers in the universe or other reasons. For example, household surveys of the general population almost always have low coverage rates of such groups as young minority men.22 Surveys and administrative records systems use several methods to try to compensate for nonsampling errors, such as adjustment of survey weights for population undercoverage and attrition, imputation for item nonresponse, and editing for misclassification or inconsistency in reporting. However, these proce- dures are not likely to maintain all of the underlying relationships and may themselves be a source of bias. Validation Methods Validation involves estimating overall error rates and the contribution of indi- vidual sources of error to them, including the contribution of weighting, imputa- tion, and editing procedures. The problem is to determine appropriate bench- marks for comparison. There are several approaches to validation; see Box 4-10 for examples of their use. Reinterviews Asking a sample of respondents the same question in a reinterview cannot establish which answer is correct, but it can indicate whether the responses are robust in the sense that there is a high level of consistency between the answers given originally and in reinterviews. Use of Alternative Question Wording Experimentation with different ques- tion wording, or other aspects of questionnaire design (such as the order in which questions are asked) may determine that the responses are sensitive to such 22Coverage rates are developed by comparing survey population estimates by age, race, and sex to census population estimates updated by births, deaths, and estimated net immigration; Medicare records are used for the elderly.

126 ASSESSING POLICIES FOR RETIREMENT INCOME ............................................................................................................................. :::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::: ::::::::::: ::::::: :::::::::::: ::::::::: ::::::::::: ::::::::::::::::::::::::: ::::::::::: :::::::: :::::::::::: ::::::::: S::::::::::::::::::::::::::::::::::::::::::::::::: .: of: _' :_::: ::: :: ::: ' ,,.A :': :':'A:':':':':~:':'A _' . ' ,,.A :': :: : ::, ::::~:: :: : ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ~ Aaa~aalO HepO=lna toners in ASSel virtues DO Lyme gremlin - ... a.. a... .... ~ ................. - - ~ ~-l.-.Y.Y-~-:--~--~- -a-- ! advises ~ co-m-pa em ag-g-reg-a Ed asleep val-u-e . ~.. ......................................................................................................................... ... ~ ............................................................................................. _ ............................................................................................................................. .......................... - -. - - '''''''''''''''~!''L1 1 't~'''1 - ''1 '';' 1~!''1''!'1'!'''V `!''I A.. l - - l v ~ Y - '.''''''1''1'!'w!'w''~'1'w'' - !'~w'1'~ ~1'tw!'~=''1!''1''1'w~V l'`;t'l'!'~''''''' - _ :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::: . ~ A~te--Re on~ng--~ ~ In on--l.n ome--Amo .n~ C d ~ .............................. . ....................... .................... ............................................ ............................................................................................................................ ~ acne Coon-- -~og-ers~ ~---~--~-~-:-~---~--- ~ I a--e ~ co-m-p-a-reu~ l-n-c-ome~ ~-po-~l-n-g~ lint .... .. ~ . . . . . . . - ~trl l~t^^ ~^m tn~ ~l~tl^~l ln~d m^---~r ^---v~-~-~-l-l- .~.t:':':'~:~^d I lr tO---l-~t-l-~-~-~ l--n~l-r~ '''''''''''''''~'t'i'" .:L.~'''!'!'~'i'!'!''''t'l'!' - .-'''i'~'~ L!'~t'l~!''''I'I'!~'I'l'!'~''~I'!'~'''I''''I'~L'''~'Y'I'!'t~'''t'I'~!'I'''~`'~'''''''!''I'! ~ ~ ~| h I Ih Tj i T ~ |~ l I I P j T ~ i T :::::::::::::::.:~..b~j:::: :..:.:.:~:.:.:~ :~:~-~..:~:~...:.y :~.::.:::~.:.:.: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::....::::::::::::::::::::::::: '~ 't~'d t~ n-ot~ p- -ol-l-c)- e -p-love-d ~ p-e-nslo-d S--~-md tne ~l~A ano maKl' -a~ oin-e-r~ aD~ ::::::::::::::: ::::::::::::::::::::::::: ::::::::::::::::::: ::::::::::::: :::::::::J:::::::::: :::::::::::::::::::::::::::: :::::::::::::: :::::::::: :::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::: : :::::d::::::::::::::::::: :: ::::1:::::::::::::::::::::::::::':::::':':::':':::':'':':':':':::' :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:~- T^:T.-~-:-:-:d :~:~-:-:-~-~-~-I:-:-:-~-~-~-J ^-:r~-~-~-: - T -:-:-~-~-~-1:-~-~-~-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:~-~-:-:-:-:-:-:-~:-:-:~-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: .................................. ~ L".L= "I I ~ tW~=I ~ V V ~ I 1 ~ l ~ l ~ l l ~ ~t l~l ~ l l ~ I :2:2:2:2:2:2:2:2:2:2:2:2:2:2: 4:: 1:l: t: ::: :1:1: i T .::: : ::: : T:: d 1: :I T: t ' '~''T r' ' ''' '' ;' ''' i' ''' ' ' ' ;'' ' '''' '' '' ''I' t i 1 1 U 1 1 1 = ~ U 1 1 1 1 =~' ~ LO 1 =. 1 I V ~ ~ U ~ L 1 "~ L~ ~ 1 ~ VI 1 1 ~1 I V "~. ~ =t 1~1~1 1 "1 1 1~ U 1 1 L ~11 1 I I A ~ ~X A ~ ^ 1 ' ' ' ' ' ' ' ' ' ' ' ' ' ' `1 't~' ' ' !' '~'1'i' ' ' ~ ' ' ' ' `1 't~'1 ' !' ' ' 'i'1 ' !' - .-' ' ' V l 'l ' ' ' 'l ' ' ' ' ' '~ !' 'l ~' ' 'i'V l:~l:'w'l:':!':':':~'i':':':'V:: :~:~:L! l: :! :l~::: ~:! :~:: :V:!: l:l:~:: :~ ~. . ~' ' ' :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: . . . ...... ~ ...... .......... ......... ............................................................................................................................. ............ ~ t'entS""'OT""'!"d 'd 'd ' $ d-'d '-'-'d enSl'On''g'!$lr!0ul!-o~ S""are""'t ep-'t l'nq""'l'n'em"'aS'''d -q- d-'l-ar~ l'd '' '~

DATA NEEDS 127 :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: - lmoulallon turrets on onus enor 1naor aarnmas Lo em ~ ~ ........................................ A ....................... ................................. ... ... ......... ....... ............ ................. .............. .... By. ....... ........... ::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ................ - . - ~ ....... ......... co-m-pa en ea-m-l-n-as~ e-oo-r so off m-a--e-c ~ coup -es lint t 1e calm 1 l ad -hi alas 1~ ~ ~ -~ waling exact ye malc-n-ea~ l-~-~ reco-ras~ ~-e~ To-u-no~ Doing .............................................................................................................................. - ~ u--nc-e-r epoding anD ~rrepoding pmolems Me Tuhner Determines tnat ::::::::::::::: It:::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::: :::::::::::: ::::::::::::::::::::::::::::::.:::::::::::.:: :::::::::::::::::::::::::::::::::::::::::::::::::::::: ~ knee -ensues Flu-- earl so l-m-p-ulal-l-on~ p-roc-ea-u- es Tor missing reports Inc ease-a Q . . . . . . .... . ~ . u e.-s.~.ion~.vr.om.ing--- ~-~.Is---on---c m picy.men-~---- Janus expense v. e---q-u-es~ .... ............... , - - . . i .. ~ t Io-n-nal--re~ tem-l-n-~ l-n-c-l- ulna :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: - , :::::::::::::::::::::::::~:::::::::::::::::::::::::::::: ............................................................................................................................. ......... I !~^ To r=~Imn To" ~ - ` ^' l==TI^n RIO! A.-= TO r.-=TI= ~one nOmO ~ n =mO ':':':':':':':':':':':':':':'w'w'w'~:':':':~ .~.-.:':':':'I:'~:~:~:~:!'~:I':'l:':':':~!':'l:~':':':'~:!':':':':~:':':':~:~:~'~'l:~:!':'l:'l:':l~l:'l:':~:':':':~.-.:':':':'I:':~:I':I'~:~:':':'w'I:':I'~:I':'I:y:~:':':':!':l':'l:':':':~: I :'1:':1'~'1 ~ ' e'' to a e''' -so d'-'-ta' i-l-itate~ ' es'' ' -' ' em O-' We'd e '-'-' t~ ' ' ti'' ~ '' h' ' Be' ""e'l*' ''I ........................................... ........................................................................................................................................ .................. ............................................................................................................................... . , , ~ -e-a more e-p-o-ns~ -room women wo-rK'ng ................. -. - - - - - - -. ~ lOem as nol ln lHe la~or Iome ~se oT me new quesTIonnalre anu uaTa .............. ~.~.''''' t'r' ' '''' '' ' ' ^'i''''' ' ''' t~' ' i' ^''' '' ' ' '''~' ''''I' ^' ''''t' 'r.' ' ''' +' 't.'' t'r' ' '''1' ' ' '''~' 'lr''' ''''''''''''''' ......... ''''''~'l 1~ ~ I'~t' 1 ''~'E'V~ U I '=O'''" I' l'=. ~ L=. "'' MV'I'I'I'~''~= V'''t=~V I''''t'V I'~.''= L"'L'I ~ ~ I'~O''l'~='~''l''''V l'I'V ~ '''''''''''''' - ~r see - ~ ~-.-Ae~Qr~.~ _rr.~r~ ~l er.-~^l ~ ~ t~^ ~ --~^r .^ l l ~ t ~l~ l l-- .-~.~-~-'&lV~-~'-'-'~V'-~,'-~-'~y=-'-'_~-'l-'Vl'--'~ '~'U'-I'-='~U'~ V'!~ ~2) e s ea e te i the Ma h 1994 CPS b a e e a d e ....................... ............. ............................................................... ............................................................................................... :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::: - I::::::::::::::::::::::::::~::::::::::::::::::::::::::::~::::::::::::::::::::~:::::.:.:::::::::::::::::::::::::::: ~ l~ ne~ wo-rSl~ cove-ma~ po-p-u--lmlo-n~ g-~-u-ps~ a-~ yo-u-n-g~ an-a~ m-l-a-a-l-e---age-a~ ~-l-ac-K~ ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::: ::::::::::::::::::::::::::: ::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ~ me-n--:~ co *e-~ge~ ra-les~ are~ ~ l~ p-e-rc-en-l~ I--~ pe-me-n-~ ~--f~ p-e-rc-en-l-l~ a-n-u~ ~ ...................................... - - - - - - -. - - ' ' ' ' ' ' ' ' ' ' ' ' ' ' '~= I '~='I' I L - ' ' ' ' l' '=O'~ Ll'V'= i'V '- ' ' ' ' 'I ~'I' ' ' ' '~ I=~ ~' ' ' 'I'l' l'= I 't' ' ' ' "~'=~ ' ' ' '~ V.-~ - '- ' ' ' ' = - ' ='= -' ' ' ' '~V '~ - - ' ' ' 'M'1 'l'~' ' ' ' ' ' ' ' ' ' ' ~ a a a i il I I .... ::::::::::::::::::::::::::::::::::::: :::::::::::.::::::: ::::::::: ::: ::::::::::: :::::::::::::::: : ::: ::: : : :::::::::::::: ::: ..... 2''''''''~ ^^'''~O t^'O'''l ^''tn^'''N'n'~'r.-^n'''1''a"~''' ''---'w'`-'''~'M'M ''` I' v v V'''''''B'' n ^''~'t ^^'rt'''' ''M'^ n ^'r~'l'l' t'' r O' r^''''''''''''''' ::::::::::::::~:~:::I:~:L~:::I I::!::Lt:l~:::!:V I:~I:~I::!::::I::~:V~:: *:!::::V::~!:I:~::~l:t:::t:::):.::::::1::1::!~:::~:!~:~:1:1: y:::~:~l::l - .:l :~l:l:y:::l: ~ b'ed'''"""' ' ' ' "a' '' """' ' s'""""""94"""' ' '' ' " '""""'96"""' e" ' '' '""""'93""" ' '' e" '""" .................. ...................... ~ ................. ............ ~- j ~......... .... . ~ ce-n~ res-n-ecilv.-elV ~orolac~men olac~women nonolac~men ano non ~ 1 ~ :::::::::::::::::::: ~: t:::::::::::::::::::: ::::::::::::: ::: ::::: ::: ::: ::: ::::::: ::: 1 ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::-:::.::::::::::::::::::::::::::::::::::::-'::::::::::::::::::::::::::::::::::::::::::::::::::::::-::'-'::::::::::::::::::::::-:::: ~ $ tl ~ h-- e ti- -t-d ~eae t ~ f~hi- h ~ t-han~ ................................................................................................................................ . - . - *::::::::::::::::::::::::::::: ~ pe-me-nt---a-u-e---to~ ag-e~ repo-~-n-g~ e-~-ro-~-*~ ~ee~ Te~---To-~---a-eT-I-n-l-l-l-on~ a-n-a~ m-eas-u-re~ . ... .. .... variations. In the past 10 years, federal statistical agencies have made increasing use of techniques from cognitive psychology to study in greater depth the ways in which respondents react to and interpret specific question wording. The results of such methods, which include one-on-one sessions in which a researcher probes the respondent after each question to ask what he or she had in mind when answering, have often shown startling differences in perceptions between respon- dents and survey personnel (see Jabine et al., 1984~. Aggregate Comparisons of Two or More Surveys Comparing aggregate estimates from one survey with aggregate estimates from another survey that is believed to be superior can provide an overall measure of data quality. For

28 ASSESSING POLICIES FOR RETIREMENT INCOME example, as discussed above, estimates of household wealth from such surveys as HRS or SIPP have been compared with estimates from the SCF. Another ex- ample is comparing estimates of retiree pension and health care benefits from the March CPS income supplement with estimates from the detailed supplements that have been conducted occasionally on these income sources (most recently in September 1994; see Pension and Welfare Benefits Administration, 1995b). However, aggregate comparisons do not generally shed light on the sources of error in survey estimates. Also, they need to be carefully made to ensure that definitions of the reporting universe and data items are comparable between the surveys being compared. Aggregate Comparisons of Surveys with Administrative Records Data Sur- vey and administrative records comparisons are often viewed as a preferred method of measuring overall data quality, on the assumption that the administra- tive records estimates represent "truth." For example, validation studies of the quality of income data in such surveys as the March CPS and SIPP have used estimates from IRS tax records, food stamps and other program records, and the National Income and Product Accounts (NIPA) as benchmarks. However, such comparisons often require extensive adjustments of the ad- ministrative sources, which cannot always be completely made, for consistency of coverage and definitions with the survey data. Thus, comparing NIPA and survey income estimates requires adjusting the NIPA estimates to exclude in- come of institutionalized people, Armed Forces members overseas, and others who are not covered in household surveys (including nonprofit institutions in some cases). In another example, comparing the percentage of private wage and salary workers who participate in employer pension plans between the Form 5500 data series and the periodic supplements to the CPS on pensions requires several adjustments (see Belier and Lawrence, 1990~. The two series do not include exactly the same types of pensions; also, the Form 5500 series includes nonvested participants who left their jobs less than 1 year previously, and it double counts workers with more than one job in which they are covered. Finally, administrative sources are not always error free. For example, there is evidence that earnings are underreported to assistance program caseworkers, which suggests that household surveys are not necessarily inaccurate when they find higher proportions of public assistance recipients with earnings than shown in case records. Also, Medicare records are not an entirely accurate representa- tion of the older population, given the problem of phantom enrollees (records for people who have already died). Microlevel Comparisons of Survey and Administrative Records Exact- match files make it possible to carry out detailed validation studies that decom- pose overall error levels into specific sources of error, including overreporting, underreporting, misclassification, erroneous imputation for nonresponse. Again,

DATA NEEDS 129 care needs to be taken to assure comparability of universes and data items: for example, not everyone is required to file a tax return. Because of confidentiality restrictions, the opportunity for microlevel error analyses has generally been limited to federal statistical agency staff. One analy- sis by outside researchers is Herzog and Rubin (1983), who studied the quality of March CPS Social Security benefit imputations with the publicly available 1973 CPS-SSA-IRS exact-match file. David et al. (1986) carried out a similar study of earnings imputations with a 1981 CPS-IRS exact-match file that they used while working at the Census Bureau as special sworn agents. Validation Needs To improve the capability for accurate modeling and analysis of retirement- income-related policies and behaviors, validation studies of key data sources should be carried out on a regular basis. Such studies can provide important feedback to data collection agencies to improve data quality at the source. They are also needed to enable researchers and policy analysts to determine appropri- ate strategies to compensate for data problems in their models. For these pur- poses, it can be useful to develop data quality profiles that are regularly updated as new information becomes available. Quality profiles bring together the results of validation studies for a particular survey or administrative records system into a comprehensive document that describes sources of error and their magnitudes, where known, and that identifies areas for which more validation work is needed (see, e.g., Jabine, King, and Petroni, 1990, which is a quality profile for SIPP.) Several kinds of data validation studies could be useful for retirement- income-related databases. Comparing CPS, SIPP, and HRS Reports of Pension Participation SSA recently completed a comparison of the May 1993 CPS pension supplement with 1993 data from the 1992 SIPP panel, finding that participation (coverage) esti- mates in the two surveys are almost identical (lams, 1995~. A similar analysis should be performed for all three surveys for the HRS age cohort. Comparing Household Survey Reports of Pension Participation with Esti- mates from Employer Administrative Records Aggregate comparisons, such as the study by Belier and Lawrence (1990) of the CPS pension supplements and the Form 5500 data series, should be carried out on a regular basis. More work is needed to improve the validity of such comparisons to account, for example, for worker participation in more than one plan and in plans of more than one em- ployer. Microlevel comparisons of household survey reports of pension plan provi- sions with employer records are possible and should be carried out for sample members of HRS, although the quality of the analysis may be affected by the

130 ASSESSING POLICIES FOR RETIREMENT INCOME relatively low rate of employer response. About 25 percent of sample members' employers did not respond to the request for Summary Plan Descriptions, and another 10 percent provided inadequate information with which to code relevant pension plan features. This level of employer response is typical of the experi- ence of other surveys that have requested the descriptions, such as the 1989 SCF and 1989 NLS-Mature Women (see Juster and Suzman, 1995:44-45~. Comparing Household Survey Data on Income and Assets Across Surveys and with Administrative Records Comparisons should be regularly performed of household survey reports with other surveys (e.g., comparing wealth estimates from HAS/AHEAD or SIPP with the SCF) and with NIPA and other administra- tive records sources (e.g., income tax records). Such comparisons, particularly with administrative records, require considerable care. With regard to pension income, a major issue is the treatment of the rapidly growing phenomenon of lump-sum pension distributions, which are treated dif- ferently in different surveys and records. Lump-sum distributions are included in the NIPA accounts and in income tax returns; according to the income concept of the March CPS, lump sums are not to be reported; SIPP has a separate category to report lump sums of all types; and HAS/AHEAD has questions on several types of lump sums, including pension distributions. Comparisons of March CPS, SIPP, IRS, and NIPA data suggest that some CPS and SIPP respondents may be reporting lump-sum pension amounts as regular income, but the extent to which this happens is not clear (Coder and Scoon-Rogers, 1994:21-24; see also Schieber, 1995~. Careful analysis of pension income reporting in the March CPS and SIPP in comparison with HAS/AHEAD for the HAS/AHEAD age range could be helpful, as could cognitive research with respondents to determine their knowl- edge of types of pension income and, in particular, whether they distinguish lump sums from pension distributions that are spread out over time. To make house- hold surveys more useful for retirement-income-related analysis, it would clearly be desirable to obtain as complete reporting as possible of both regular and lump- sum pension amounts. To the extent that these and other validation studies identify serious data quality problems, behavioral and projection models will need to be adjusted or their results qualified in an appropriate manner. For example, some microsimu- lation projection models have a provision to adjust March CPS income data for comparability with NIPA estimates. Such adjustment procedures must be care- fully worked out, not only to be sure that the NIPA estimates are in fact compa- rable with CPS income concepts, but also to preserve key relationships among income amounts and other variables.

DATA NEEDS 131 Recommendation 13. Budgets for retirement-income-related surveys should include suffi- cient resources for regular evaluation of data quality. Evaluation methods include reinterviewing subsamples of respondents to measure consistency of reporting; experimentation with alternative question wording to identify possible reporting problems; and comparing survey estimates with adminis- trative records to determine the completeness and accuracy of survey re- porting, taking care to adjust for differences in definitions and other aspects of the two sources.

Next: 5 Development of Projection Models »
Assessing Policies for Retirement Income: Needs for Data, Research, and Models Get This Book
×
Buy Paperback | $55.00 Buy Ebook | $43.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

The retirement income security of older Americans and the cost of providing that security are increasingly the subject of major debate. This volume assesses what we know and recommends what we need to know to estimate the short- and long-term effects of policy alternatives. It details gaps in data and research and evaluates possible models to estimate the impact of policy changes that could affect retirement income from Social Security, pensions, personal savings, and other sources.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!