6
Care of the Elderly: Division of Labor Among the Family, Market, and State

Beth J. Soldo and Vicki A. Freedman

INTRODUCTION

Although disability is not the inevitable consequence of human aging, the risks for both chronic disease and disability are highly age related (Manton and Stallard, this volume). Because of this, the aging of the U.S. population as a whole and the implosive growth of the very oldest segments within it are associated with an increased demand for all types of health care, including nonmedical personal care (Suzman et al., 1992). As shown in Table 6-1, the current demand for such care is substantial. In 1985, nearly one in five (19.2%) persons aged 65 and over was disabled but living in the community, while another 5 percent were disabled and institutionalized, mostly in nursing homes. The vast majority of these disabled elderly call on informal sources of support—family and friends—to assist them. At any point in time, however, slightly more than 5 percent of the elderly are disabled and rely in whole or in part on paid helpers in the community. In all, nearly 7 million older persons, or about 24 percent of all elderly, are disabled in some way.

Personal care is assistance that compensates for chronic limitations in basic self-care activities, such as shopping, meal preparation, eating, bathing, or dressing (Soldo et al., 1989). Because it is inherently a nontechnical service bundle, no one social institution claims unique competency in its

Support for this research was provided by National Institute of Aging Grant No. R01-AG08651.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



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

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

OCR for page 195
Demography of Aging 6 Care of the Elderly: Division of Labor Among the Family, Market, and State Beth J. Soldo and Vicki A. Freedman INTRODUCTION Although disability is not the inevitable consequence of human aging, the risks for both chronic disease and disability are highly age related (Manton and Stallard, this volume). Because of this, the aging of the U.S. population as a whole and the implosive growth of the very oldest segments within it are associated with an increased demand for all types of health care, including nonmedical personal care (Suzman et al., 1992). As shown in Table 6-1, the current demand for such care is substantial. In 1985, nearly one in five (19.2%) persons aged 65 and over was disabled but living in the community, while another 5 percent were disabled and institutionalized, mostly in nursing homes. The vast majority of these disabled elderly call on informal sources of support—family and friends—to assist them. At any point in time, however, slightly more than 5 percent of the elderly are disabled and rely in whole or in part on paid helpers in the community. In all, nearly 7 million older persons, or about 24 percent of all elderly, are disabled in some way. Personal care is assistance that compensates for chronic limitations in basic self-care activities, such as shopping, meal preparation, eating, bathing, or dressing (Soldo et al., 1989). Because it is inherently a nontechnical service bundle, no one social institution claims unique competency in its Support for this research was provided by National Institute of Aging Grant No. R01-AG08651.

OCR for page 195
Demography of Aging TABLE 6-1 Disability and Care Arrangements of the Elderly Population, 1985   Percentage N (thousands) Community     Nondisabled 76.3 21,761 Disabled 19.2 5,465 Informal only 13.5 3,853 Formal only 1.2 344 Informal + formal 4.5 1,273 Institutionalized 4.6 1,310 Total 100.0 28,536   SOURCES: Manton (1989); Soldo et al. (1989). provision. As a result there is no natural division of labor among family care givers, private pay helpers recruited from the marketplace, and state subsidized providers in meeting the nonmedical care needs of the elderly. In this chapter we explore the alliance of the three major sources of personal care in the U.S.: the family, the marketplace, and the state. By family we mean the network of relations and obligations defined by blood, marriage, or adoption and not simply the family of coresidence. We confine our attention to parent-adult child relations 1 and acknowledge that both generations may be deemed elderly in terms of chronological age.2 The 1   Although most elderly have at least one sibling (Cicirelli, 1985), siblings are far more likely to provide emotional support and companionship than hands-on personal care. Children are considerably more likely to be the major source of assistance to the frail elderly largely because same-aged kin (spouses as well as siblings) are exposed to similar risks of morbidity and mortality (Scott, 1983). 2   Recent sociodemographic trends have converged in a way that blurs the distinction between age and generational status. Among the more important of these trends are secular declines in fertility, unanticipated improvements in old-age life expectancy (Manton and Soldo, 1985), increased incidence of divorce, and disruptions in the ''normal" order of some life-cycle events (Rindfuss et al., 1987). In combination, these trends have (a) "verticalized" the contemporary kin network (Bengtson and Dannefer, 1988) by increasing the number of coexisting generations even while the size of the nuclear family within a generation declines; (b) extended long-term intergenerational bonds, allowing for the possibility of "life overlaps" (Hagestad, 1986) where a single life-cycle stage encompasses two generations; and (c) increased the complexity and diversity of kin networks through the dissolution and reformation of families within a generation, thereby creating a matrix of latent kin relationships (Riley, 1983).

OCR for page 195
Demography of Aging market refers to the private sector in which consumption is determined largely by price. Our notion of the state is broad and references a variety of structured intergenerational transfer programs under the aegis of federal, state, and local governments. For the most part, the state provides personal care indirectly by subsidizing market transactions. Of particular interest in this chapter are how the volume and type of care and support received by older persons is distributed across these sectors, and the range of factors that affect this distribution. Demographic interest in the division of labor between the family and other sectors of society is fairly recent and largely confined to research on the child care arrangements of working mothers (see, for example, Liebowitz et al., 1988). But population aging requires a society to focus on the dependencies of both the very young and the very old. The needs of the young and the old are not equivalent; societal strategies for accommodating dependency are thus transformed in the wake of population aging. At least in general principle, the division of labor among the various sectors of society in providing for the needs of the young is well established. Parents bear primary responsibility for socializing children and providing them with basic care and amenities. Other child care arrangements are viewed as alternatives for which parents still bear primary responsibility. Through various structured intergenerational transfers (e.g., property and income tax) the state provides or regulates formal education. The public sector also subsumes or shares aspects of parental responsibilities in some instances of social or economic incapacity. In contrast, the lines partitioning responsibility for the daily care and well-being of the elderly are not as cleanly drawn. Understanding how social institutions adapt to population aging can be approached from either a macro- and a micro-level perspective. The chapter by Lee in this volume provides a macro account of intergenerational transfers that bear on society's adaptation to an older population. This chapter largely considers the micro-level features of these emerging accommodations in the United States. SUBSTITUTIONS Because we consider the division of labor between the family and the state in the context of a market economy, issues of substitution emerge. Financial transfers from either the family or the state may be used to purchase care services in the marketplace. Inter vivos financial transfers to the frail elderly and their claims on structured programs augment their own privately held economic resources, including those that result from past

OCR for page 195
Demography of Aging TABLE 6-2 Potential Substitutions of Time and Financial Transfers Within and Across the Family, Market, and State: Personal Care for Disabled Elderly   Sectors     Currency of Transfer Family Market State Provision of services Informal care Formal care Paid providers Local area programs Financing of services Subsidized market transactions through direct payment to providers Long-term care insurance; home equity conversion Subsidized market transaction through Medicare and/or Medicaid employment, consumption, and savings, as well as private pensions, housing equity, and consumer durables.3 Table 6-2 lists potential substitutions involving the family, market, and state in two of the currencies of inter vivos transfers: time and money. Potential substitutions can be identified by reading this table vertically or horizontally. We focus on only two types of substitution: (1) substitution involving transfers in the same currency across sectors (e.g., the substitution of services purchased in the marketplace for family services) and (2) substitutions in different transfer currencies within a sector (e.g., substitutions of inter vivos family financial transfers for the family's time). The former type of substitution results from the family sharing its traditional responsibilities for the care of dependents with the marketplace and state whereas substitution of the latter type arises from the opportunities afforded by the marketplace to convert money into (purchased) time. Reading across the columns of Table 6-2 suggests the scope of potential substitutions across the three sectors. A frail older person can recruit family members to provide the necessary personal care; purchase private care 3   Financial transfers from family and state benefits are not truly independent alternatives for the elderly whose own resources are inadequate to purchase care services in the marketplace. Although receipt of some benefits, such as those from the Medicare and Social Security programs, is conditional only on age and/or past employment history, other benefits, such as those from Medicaid and the Supplementary Security Income programs, are means tested, requiring depletion of resources. Income flows from family are figured into the income base by these latter programs. Welfare rules in some states also take account of coresidence in determining eligibility or benefit levels (Hill, 1990).

OCR for page 195
Demography of Aging or assistive technologies (LaPlante et al., 1992); or avail him or herself of the limited number of state-funded programs, such as those providing geriatric adult day care.4 Reading down the rows highlights the nature of substitution from the perspective of the provider. A family can respond to a parent's need for personal care, for example, by providing hands-on care directly or by financing the purchase of such care, either at home or in the nursing home. The marketplace also offers both service and financing options. In increasing numbers, private home health agencies provide a range of fee-for-service options to the frail elderly as do independently employed home health aides (Estes and Swan, 1992). Insurance for long-term care, a relatively new product of that industry, underwrites some of the costs of community care and nursing home care for a limited number of subscribers (Meiners and Greenberg, 1989; Cohen et al., 1992). The state's response to the personal care needs of its older dependents is confined largely to subsidizing the costs of such care through the reimbursement of services under a capped schedule of fees.5 By focusing only on personal care we mask an important aspect of inter vivos transfers: transfers responsive to one area of need often yield secondary benefits. By coresiding with an adult child, a frail parent, for example, receives implicit service and financial benefits (Morgan, 1983; Cox and Raines, 1985; Burch and Matthews, 1987; Jackson et al., 1991). Direct financial transfers from one or more children can supplement the parent's disposable income with which market services or other amenities can be purchased. In theory, the family also can provide an older parent with an annuity that serves the same function as regular out-of-pocket transfers to the parent.6 Similarly, market substitutes for income maintenance (private pensions, trusts, and other assets) can be directed to the purchase of needed services or other expenditures. Considering the ways in which transfers 4   There are very few publicly financed nursing homes operating in the United States. Veterans Administration nursing homes are a notable exception. Many communities run programs, which typically offer a range of services (such as home-delivered meals, transportation assistance, chore services, and case management) and benefit from public funding through Title XX of the Social Security Act and the through the Older Americans Act. 5   The major public programs paying for care services for the elderly have large gaps in coverage of chronic, personal care services. Medicare covers only those long-term care services associated with acute episodes. Skilled nursing facilities are covered up to 100 days, along with home health for recovery purposes. Medicaid covers unlimited nursing home and home health care visits; however individuals are required to spend down assets to become eligible for coverage. Other federal programs, provided under the Older Americans Act and Title XX of the Social Security Act, provide care services directly on a limited basis in the form of nutritional and social services (Davis and Rowland, 1986). 6   Altonji et al. (1992), however, reject the hypothesis of family risk-sharing behavior.

OCR for page 195
Demography of Aging convey side benefits responsive to areas of need other than personal care is beyond the scope of this chapter. In the following sections we consider two types of substitution that have received considerable research attention from diverse disciplines: the substitution of paid providers for family care and the substitution of family financial transfers for their time. For each type of substitution we identify unresolved conceptual and methodological issues. Substitution of Paid Providers for Family Care Research on the substitution of formal care for informal care has been motivated by two related concerns. First, demographic trends portend a reduction in the supply of informal helpers, particularly the adult daughters of widowed or divorced older persons, as the family size of the elderly declines in the next century (Treas, 1977; Doty, 1986; Himes, 1992). Sustained labor-force participation by adult daughters of the frail elderly may further deplete the potential supply of care providers. Second, and more distinctively a policy concern, is apprehension of a possible "woodworking" effect if state subsidized home care benefits were to encourage family helpers to withdraw or reduce their efforts in response to a decline in the price of purchased care. For the most part, the gerontological literature has concerned itself with the substitution of market (or "formal") services for family (or "informal") care (Moscovice et al., 1988). As commonly used in this literature, market services refer to the assistance provided by both paid helpers with various levels of training and volunteers from community programs.7 Thus this literature blurs the distinction between private pay market services and state subsidized transactions in the marketplace. Numerous studies based on national samples using various definitions of disability show that the bulk of personal care received by the frail elderly in the community is provided without compulsion or compensation by family (Branch and Jette, 1983; Stone et al., 1987; Soldo et al., 1989). Family helpers also are the providers preferred by the elderly themselves (McAuley and Blieszner, 1985). Kin are recruited to care giving much along the lines postulated by Cantor's (1979) theory of "hierarchial compensation" with 7   Defining "formal care providers" solely in terms of nonprimary group relations results in a grouping that is both ambiguous and heterogeneous. To overcome this imprecision, various researchers have introduced other considerations. Litwak (1986), for example, equates formal helpers with those recruited from structured bureaucracies whose organization reflects an ideology of rationality and efficiency. The term "paid helper" (which includes compensated but unskilled home health aides as well as skilled providers delivering in-home medical care) has largely replaced the less precise term "formal helper" in policy research (Liu et al., 1985).

OCR for page 195
Demography of Aging the spouse as the preferred provider, followed by children in the absence of a spouse, and distant relatives and friends as a last alternative for the unmarried, childless elderly. There also are distinct racial and ethnic variations in helping patterns (Mutran, 1985; Jackson, et al., 1992). As shown in Table 6-1, only about one in four disabled elderly in the community depend in whole or in part on paid helpers (Liu et al., 1985). Approximately 3.3 million disabled elderly were estimated to have had at least some episodes of paid care in 1988, which generated $3.1 billion in out-of-pocket costs for the frail elderly and their families (Wiener and Hanley, 1992). Like other forms of health care, the consumption of formal community care services is highly skewed. Using data from the National Long-term Care Demonstration (Applebaum, 1988), Kemper (1992) estimates that the top 10 percent of formal care users consume 63 percent of all paid care.8 An ongoing debate over what constitutes in-home or personal care is reflected in alternative definitions of care found in prior studies of substitution in the gerontologic literature. Kemper's (1992) definition of personal care is broad and includes in-home assistance with both medical treatments and services supportive of daily functioning. Tennstedt and McKinlay (1986) argue, however, that in-home medical care is inconsistent with the notion of personal care, and in subsequent analyses they exclude health care professionals from their inventory of formal helpers (Tennstedt and McKinlay, 1989; Tennstedt et al., 1992). Neither has consensus emerged in the aging research literature on the exact meaning of substitution. In an economic sense, substitution refers to an increased demand for good x1consequent to an increase in the price for good x2 (Varian, 1990).9 In extending this framework to the care arrangements of the disabled elderly, family care and market care are assumed to be two distinct goods. The demand for these goods may be measured in a variety of ways, such as hours of help per day, days of help per week (Christianson, 1988; Edelman and Hughes, 1990; Hanley et al., 1991; Tennstedt et al., 1992), or number of tasks with which assistance is provided (Greene, 1983). The price of family care may be proxied by the value of family members' time in the 8   The National Long-term Care Demonstration is a 10-site project in which publicly subsidized care services were made available to older persons deemed to be at risk for nursing home admission. 9   In the simplest case of perfect substitution, an individual chooses not to buy any of x1 if its price (p1) exceeds the price of the second good (p2); if p1 < p2, the individual will buy m/p1 units of the first good, where m is the budget constraint. More realistic is the notion of imperfect substitution, which implies that the demand for a good (x1) will increase if the price of an alternative good increases. Formally, the first good is a substitute for the second if Δx1/Δp2 > 0.

OCR for page 195
Demography of Aging marketplace (opportunity costs), while the price of formal care often is equated with out-of-pocket costs. In an economic sense, formal care would be seen as a substitute for informal care if changes in the price of formal care are positively associated with the demand for informal care. Whereas economics has a precise definition of substitution embedded in a theoretical formulation of marginal utility, the concept of substitution as typically used in the gerontological literature refers to a discrete trade-off or simple replacement of family care by market care. Analogues to the notion of price elasticity are implicit in most prior research. Although seldom developed in any detail, the tacit hypothesis is that as care needs increase, the "price" of informal care (measured in terms of caregiver burden or foregone wages) increases and the demand for market care alternatives increases as well. Most often, the demand for such care is equated with the actual use of these services.10 In practice, studies of substitution of formal for informal care have not explicitly modeled the price of alternative goods, or the budget constraints on the disabled individual or on his or her family. Instead, the demand for formal care—usually proxied simply by utilization—often is used to predict the level of family care (see, for example, Tennstedt and McKinlay, 1989). Although there also are numerous instances in which family care has been used to explain the demand for paid care (Soldo, 1985; Wan, 1987; Coughlin et al., 1990), the major policy focus has been on how formal care affects levels of family care. Because the price and the demand for a good are typically related indirectly, substitution between formal and family care is inferred if an increased use of paid care is inversely associated with the volume of family care consumed. Neglecting for a moment the potential substitution of nursing home care for any type of community care, it is likely that disabled elderly and their families decide how much total care they prefer to consume and the preferred mix of providers, subject to a variety of constraints. Simple cross-sectional regression models predicting the receipt or level of family care can not render this simultaneity. More sophisticated approaches for cross-sectional data have been described by Greene (1983) and Hanley and colleagues (1991) who jointly model the use of formal and family care with 10   Weissert and colleagues (1988), for example, identified some 53 experimental studies of the effect of formal community care on the level of informal care. Their conclusions are expressed in terms of the simple use of services (i.e., "informal social support tended to decline with home and community care use"). More recently, Weiner and Hanley (1992) have challenged these findings on the grounds that there were not significant reductions in the total volume of informal care across all areas of need.

OCR for page 195
Demography of Aging multistage estimation procedures. Such studies find limited evidence for the substitution of paid care for family care. Many of the deficiencies found in the existing literature on the substitution of market services for family care reflect nascent theoretical accounts of the demand for personal care in general. Although analysts inevitably model the use of care services as a function of need (usually indexed by physical or cognitive limitations in basic self-care activities) and the individual's economic resources11 (usually measured only in terms of annual income), treatments of price and supply factors differ markedly. Consider first approaches for indexing the potential supply of care providers, both family helpers and paid helpers. Although early models of family caregiving commonly omitted any measure of the potential supply of family helpers, more recent analyses include at least crude measures of family structure (such as the "number of daughters") as proxy measures of supply. Even with summary measures of family structure, most prior studies estimate what might be deemed "demand" models of care. A distinctive, albeit implicit, feature of such models is viewing the beneficiary of the transfer as the sole decision maker (i.e., the use of any kind of care is determined only by the beneficiary's need and budget constraints). By ignoring constraints on potential donors of care, demand-based models inevitably treat kin as interchangeable resources for the frail elderly in spite of accumulating evidence to the contrary (Soldo et al., 1990).12 With the advent of data that link attributes of adult children living both inside and outside a parent's household to the specific care behaviors of each child, more realistic models of the full range of transfers are possible (see, for example, Wolf and Soldo, 1990). Yet in cross section, such data pose formidable statistical problems. At a point in time the attributes representing the constraints on potential providers, such as employment, child care responsibilities, and proximity, are endogenous with the elder care outcomes of interest (Wolf and Soldo, in press). Nonetheless, such data bases provide the opportunity to model the use of care services as the outcome of a joint decision-making process involving both the frail elderly and the potential kin donors of care. They also allow for models in which the care decisions of one potential provider are conditional on the behaviors of other possible providers of care, including paid helpers. Such models can be extended to test theories of joint family decision making by taking 11   In-home care by paid helpers is assumed to be a normal good (i.e., other things being equal, individuals will consume more paid care as income rises). 12   Coward and Dwyer (1990) also have examined the extent to which adult children are interchangeable resources for parent care, although their model assumes independence of the behavior of each child. In combination with Soldo et al. (1990) these studies indicate, however, that the characteristics of others within the pool of potentially available helpers largely determine which child has primary responsibility for the personal care of a parent.

OCR for page 195
Demography of Aging TABLE 6-3 Predicted Probabilities of Caregiving by Various Types of Helpers, by Level of Need: Widowed Mothers with Two Children (an unmarried son and an unmarried daughter)   Level of Mother's Need       Predicted Probabilities IADL Only 1-2 ADL 3-4 ADL 5-6 ADL Primary caregivera         P (child) .504 .424 .375 .346 P (paid) .140 .197 .287 .354 P (other) .298 .309 .304 .295 Care involvementb         P (child) .701 .642 .652 .664 P (paid) .222 .305 .446 .575 P (other) .500 .549 .614 .624 NOTE: ADL = Activity of daily living; IADL = Instrumental activity of daily living. a Primary caregiver defined by respondent report of frequency and level of help from each provider. b Probability that a provider of a specified type will appear anywhere in the care network. SOURCES: Predicted probabilities estimated under the model described by Wolf and Soldo (1990). Data are from the 1982 National Long-Term Care Survey (Macken, 1986). into account not only the constraints on each potential helper but also purchased care alternatives. Table 6-3 illustrates the potential of models that exploit linked child and caregiver rosters by simultaneously taking into account: (1) the needs and resources of an older parent and his or her array of potential providers (the option set); (2) the underlying process by which the care involvement of one child is conditional on the behaviors of other adult children; and (3) the potential substitution of paid helpers for informal care. The predicted probabilities shown in Table 6-3 were estimated under a five-tiered model of the care networks of elderly, widowed mothers living in the community in 1982 (Wolf and Soldo, 1990).13 For purposes of this illustration, the 13   The model estimated uses a multinomial specification of care behaviors and assumes a hierarchical rather than a simultaneous decision-making process. Effects were estimated for the first five levels of care graded by the intensity of the care given. Caregivers at the first level of the hierarchy are termed primary caregivers. The estimated model includes indicators of the mother's needs (functional limitations and cognitive impairments), her resources (income and housing equity), and basic demographic traits of each child available for parent care at each level of the model. A child assigned the primary caregiver role at the first level of the model is considered ineligible to provide lower levels of care. The sole indicator of the price of paid care is the mother's current Medicaid eligibility.

OCR for page 195
Demography of Aging pool of potential kin helpers is limited to two children—an unmarried daughter and an unmarried son. The mother's need is indexed in terms of the number of activities of daily living (ADL) for which care is needed. In this hypothetical two-child family, a child is likely to provide primary care at all but the most extreme level of disability. At that level of need, a paid helper is more likely to be the primary caregiver. Other unpaid helpers, such as more distant relatives and friends, have a relatively small probability of assuming such care responsibilities. Turning attention to the full array of helpers, however, it is clear that a diverse network is likely to be assembled by drawing on the resources of both the family and the marketplace. Because the data for this analysis are cross sectional (the 1982 National Long-Term Care Survey) and wage rates could not be estimated for all potential family helpers, this specification is inadequate for testing complete theories of substitution. Nonetheless, the results suggest that future research will benefit from a network perspective rather than the more commonly used dyadic approach. It also highlights the importance of allowing for market substitutes at varying levels of intensities within the care network. The supply of paid care alternatives has received scant research attention. Lacking direct estimates of the supply or market prices of paid care, analysts typically rely on proxies. Among those more commonly used are urban-rural or city size distinctions (Soldo, 1985; Hanley and Wiener, 1991; Kemper, 1992) and Medicaid eligibility of the care receiver (Soldo, 1985; Kemper, 1992). More importantly, models of community care arrangements typically do not condition outcomes on the supply of nursing home alternatives. Although there are numerous examples of equations predicting nursing home admission and others predicting type of community care arrangements, there are only a few instances in which even crude indicators of the supply of other care alternatives are incorporated (Sloan et al., 1992). Variations in the supply and price of nursing home care are a source of unmeasured heterogeneity in models of home care; variability in the supply of informal caregivers is confounded with other factors (such as marital status) in models predicting entry to a nursing home. Even the most sophisticated cross-sectional models do not accommodate the diachronic nature of disability or compensating care. From the older person's perspective, the propensity to use different types of care arrangements is likely to change over the course of a disability or during the final years of life. During that time, changes also may occur in the level and type of care needed, as well as in the kin available to provide such care. Furthermore, the implicit price of care may change over the course of an illness. The out-of-pocket price of institutional care, for example, drops markedly if an older person spends down to Medicaid eligibility. A longitudinal model of substitution has been proposed by Tennstedt

OCR for page 195
Demography of Aging and colleagues (1992), who define substitution as an increased number of days of formal community care and a decreased number of days of informal care at successive interview dates. These researchers also add a new dimension to studies of substitution by disaggregating help into six common care activities and exploring the possibility of substitution occurring only within specific tasks. As with most cross-sectional models, the price of formal and informal care, and the family's budget constraints are not measured. The major weakness of this approach, however, is the loss of information that ensues regarding the extent of substitution in terms of the total volume of care provided by different sectors. More elaborate models of dynamic substitution have not been proposed, most likely due to existing data limitations. To model substitution over time requires information on the types of care received on a regular basis over the course of a chronic condition or illness. Such data are not available on a national basis. Finally it is important to note that the supply of potential providers of care is itself dynamic. Exogenous change (e.g., death or migration of the primary caregiver) can initiate change in the size or structure of the care network. For the most part, prior research using panel data has allowed only for changes in the underlying structure of needs (Tennstedt et al., 1992). To date no published research has simultaneously modeled changes in the pool of potential providers and in the price or supply of paid providers along with changes in need. Limitations in currently available data preclude such an analysis (for reasons discussed below), but at the present time we also lack sufficiently developed theoretical frameworks to guide such analyses. Gerontologists, for example, have speculated about a ''tolerance threshold" for informal care, but the nature of such an underlying process has not been given serious consideration in the context of shifts in the real and opportunity costs of parental care. A full account of the substitution of market resources for family resources awaits not only panel data merged with information on the supply and price of care alternatives but also considerably more developed theoretical formulations. Substitution of Financial for Time Transfers The framework shown in Table 6-2 also highlights the substitution of financial transfers for time transfers.14 As discussed above, this type of substitution organizes the state's response to the care of frail elderly. Yet 14   In this section we consider only those financial transfers from a family member to another relative. Structured intergenerational transfers, such as retirement benefits from Social Security, are not considered. Kingston et al. (1986) provide a macro-level perspective on structured intergenerational transfers.

OCR for page 195
Demography of Aging considerably less is known about the extent to which the family commonly substitutes financial transfers for its collective time. In part, this situation results from limitations in data collection activities. Many national surveys fail to distinguish between services paid for out-of-pocket by the older consumer and those for which the family (regardless of residential status) pays (LaPlante et al., 1992). In other instances, high rates of item nonresponse have resulted when the frail elderly were asked to specify who pays for the care they receive (Liu et al., 1985). High-quality information on both direct (e.g., payment to the older parent) and indirect (e.g., payment to the provider on behalf of the parent) financial transfers from each potential provider needs to be combined with data on the care behaviors of all family members to complete an analysis of the ways in which the family as a whole substitutes financial transfers for its collective time. More finely grained analyses of substitutions within the family (in which one child provides time, for example, while another provides financial support) would require data on the transfer behaviors and relevant attributes of all potential donors within the family network. Nonetheless, existing data suggest that about one-fifth of the elderly may receive either regular or episodic financial transfers from adult children (Soldo and Hill, in press; Stephen et al., 1992). Other data indicate that most of the costs of in-home personal care by paid helpers is borne out-of-pocket either by the elderly themselves or in combination with their families (Liu et al., 1985; Wiener and Hanley, 1992). The elderly and/or their families also absorb most of the costs of assistive technologies and housing modifications (LaPlante et al., 1992). In cross-section the income and asset profile of the most disabled elderly is markedly lower than that of healthier older persons (Hurd, 1991) suggesting that the frail elderly with large out-of-pocket health care costs may be drawing on the financial reserves of their extended families. Although there has been little empirical analyses of the extent to which families substitute financial transfers for their time, there is considerable literature, primarily economic, concerned with the motivation for intergenerational financial transfers. Just as public policy concerns have motivated much of the research on the substitution of market care for family care, concern for the potential "crowding out" of private financial transfers by public benefits compels much of the economic research on the issue (Cox and Rank, 1992). Ultimately the redistributive outcomes of public transfers are influenced by the motivation of private donors for financial transfers. With only a few notable exceptions discussed below, bequests rather than inter vivos transfers have been the focus of most empirical research on financial help from a parent to a child or from a child to a parent (Soldo and Hill, in press). The two primary economic models of transfer behavior are distinguished by the motive hypothesized. Theories of altruism postulate

OCR for page 195
Demography of Aging that children make financial transfers to their parents because they care about their well-being. No matter how the recipient disposes of the transfer, the donor's utility will be increased (Becker, 1981). But if state transfers are increased, altruism models also imply that there will be no net gain for the older parent because the state subsidies will simply replace dollar-for-dollar the transfers previously made by the family. Alternatively, exchange may motivate a child to make an inter vivos transfer. Services or financial transfers might be given, for example, in expectation of a bequest or a larger inheritance. Exchange-based accounts of inter vivos transfers predict that increased state subsidies will improve the well-being of the elderly because family transfers would not be reduced. These theories, and variations on them, are useful for identifying the factors important to an adult in deciding whether to make transfers to his or her parent and, if so, its magnitude. In the selfish exchange model, for example, there are various qualities of the frail parent that a child wants to obtain in return for transfers given. The individual may want the parent to remain in a separate dwelling (and thereby protect the privacy of the child), to maintain independent functioning, or to avoid depleting the entire estate (spending down) in order to qualify for Medicaid. In the tied transfers model the issues are similar, except that the transfers from the individual are designated for particular purposes. Service transfers or transfers of goods would allow the adult child more control over the beneficiary's behavior than would direct transfers of money. Purchasing services on behalf of the older parent is an ultimate form of restricted exchange. Alternatively, reciprocal exchanges could be a major motivating force in the decision to provide any transfers and the level of services to provide. The type of transfers from the parent to the child in the past would then affect the type of transfers the child chooses to give to the parent. The empirical evidence concerning the various economic models is mixed and inconclusive.15 To date there has been no empirical test of the substitution of financial transfers for time. Nonetheless, standard economic theory yields predictions of the sort that the higher the value of a child's time the more likely he or she is to make a financial rather than a time transfer. But various theories also point to other important considerations. Once a child 15   The empirical evidence generally does not support the "wealth" implications of the altruism model; parents tend to equalize bequests rather than wealth across children (Behrman et al., 1990; Wilhelm, 1991). Support of both the altruism and the exchange models has been evidenced by monetary transfers received being positively related to donor's socioeconomic status (Cox, 1987; MacDonald, 1990); but Rosenzweig and Wolpin (1990) also describe a negative relationship.

OCR for page 195
Demography of Aging decides to help a parent in some way, the form of the transfer would depend not only on the needs and wealth of the parent but also on the costs of inputs or preferences, the qualities or behaviors the child wants to encourage in the parent, the cost of market substitutes and the child's budget constraint, and finally, the type and volume of previous flows between the child and the parent. The various hypotheses about motivation for transfers suggest a need for information to sort out whether past exchanges are important to future exchanges because they build expectations for reciprocity or because they reflect differences in orientation toward giving/receiving (heterogeneity that persists through the life course). Failure to control for unobserved permanent differences across households can lead to serious biases (Rosenzweig and Wolpin, 1990). The life-cycle timing of transfers also may be important; transfers given early in life, for example, may subsequently reduce assets (Kurz, 1984). The chronicity of need also may be important because individuals may resist receiving help if needs are transitory (Kurz, 1984). CLOSING THE GAPS: FUTURE RESEARCH ON SUBSTITUTIONS It has become commonplace for investigators of intergenerational transfers to argue that there are multiple currencies of exchange and alternative donors in any transfer currency. Nonetheless, a number of issues remain unaddressed. The simple conceptual framework of substitution discussed above highlights these gaps. In part, data limitations preclude analyses of theories of substitution in currencies (e.g., time for financial transfers) or in providers (e.g., paid providers for family helpers). In other cases, the formidable methodological challenges of specifying a model of substitution have given rise to simplistic, and oftentimes misspecified, models. Both cross-sectional and projection models suffer from such oversimplifications. As a result, various gaps exist in our knowledge of current patterns, no less the future demand for alternative care arrangements. Until recently, virtually no data sets provided information sufficient to analyze transfers from which the elderly benefited. Nationally representative surveys that focused on the elderly and their health care needs rarely obtained information on the family structure of the respondent. The Supplement on Aging to the 1984 Health Interview Survey and its follow-up (the Longitudinal Survey of Aged), for example, provide detailed information on household members, but only count the number of sons and daughters living outside the household. Care providers are identified only generically (e.g., "a relative"), and information is not available on financial flows to the older respondent. An early exception to this approach is the 1982-1984-1989 National

OCR for page 195
Demography of Aging Long-Term Care Surveys. In this panel study of disabled elderly, demographic information (including marital status, education level, and employment status) is recorded for each child regardless of household residence. This data array enables researchers to specify the supply of potential informal caregivers in a micro-level analysis (see, for example, Wolf and Soldo, 1990). But because this survey does not provide the data needed for estimating the wage rates of all children, employment-related substitutions (i.e., time for money) cannot be explored fully. In addition, data are lacking regarding which children make regular financial transfers to parents. Such detailed information on both time and financial transfers is necessary to model the full family network of helpers and substitutions within it. Data are currently being collected that will allow care and substitution issues to be addressed in more detail. The University of Michigan recently released the first wave of the Health and Retirement Survey (HRS) to the research community. This survey targets persons born between 1931 and 1941 (i.e., those aged 51 to 61 in 1992). More than half of these middle-aged respondents have one or more surviving parents. To provide the data necessary to analyze the extent to which parent care obligations effect the labor supply of those nearing retirement, the HRS collects limited data on the care needs of parents. For respondents with living parents, information also is collected on the siblings of the HRS respondent. Each sibling is identified in terms of demographic variables, as well as crude measures of wealth and attachment to the labor force.16 Transfers, whether of time or money, to the parent of an HRS respondent are identified in terms of which adult child provides help and the intensity of the help provided. Parent and sibling data are collected for both spouses in a household, providing the opportunity to explore the extent to which two sets of family obligations compete for the resources of the middle-aged couple. An even more detailed listing of potential helpers and their transfer behaviors is incorporated into the new study of Asset and Health Dynamics (AHEAD), a national panel survey of persons aged 70 and older, also being fielded by the University of Michigan. Both HRS and AHEAD are supported by the National Institute on Aging. Although a number of surveys will support the analyses of a broad range of substitution issues, all of them suffer from a common limitation— the absence of data on the price and supply of market substitutes for care. As noted above, only crude proxies are generally available. Confidentiality issues may be raised when survey data are linked with detailed area resource and pricing information, including Medicare, Medicaid, and private 16   Complete information is available on all siblings only if the total number of siblings is less than five. Otherwise, such data are available only for a random sample (N = 4) of the sibship.

OCR for page 195
Demography of Aging health insurance subsidies. Creative masking techniques are needed to close this final loop in the data needed to examine various aspects of substitution. Understanding substitution of currencies or of providers depends, in part, on understanding the supply of resources available for accommodating disability in the older population. In this volume, Wolf discusses the demography of the family, including variations in family size, structure, and proximity. The supply of family resources is not adequately indexed in terms of the sheer number of different types of kin alive at various ages. Factors that reasonably constrain the time available for parent care, such as work and obligations to other kin, have proven difficult to measure in surveys in which the sole respondent is the potential beneficiary of the transfer. Even when crude measures of competing constraints are available they are difficult to model in cross section because these constraints and transfer behaviors are likely to be jointly determined. Thus, the tradeoff between parent care and work is poorly understood (Wolf and Soldo, in press). Similarly, we know very little of how families allocate their resources of both time and money across the extended family, including parents, in-laws, adult children, and grandchildren. As four-generation families become increasingly common, (and, at the same time, more fragmented through divorce and remarriage), an important goal of research is understanding how such demographic changes will reconfigure the supply of potential helpers. Additional research also is needed on how racial and ethnic differences affect both the supply of family resources and transfer behaviors. Fertility differentials suggest that the elderly in these subgroups have an abundance of potential helpers. Further research is needed, however, to determine the extent to which these seemingly large helper networks respond to the declining health of older parents. In large segments of minority populations, severe resource constraints operate, which would seem to favor family care over paid care in the community and time transfers over financial transfers. Alternatively, variations in kin bonds may discourage going outside the family network to secure personal care help, independent of the family's resource constraints. Research also is lacking that evaluates the effects of family living arrangements on transfer and substitution behaviors. Coresidence anchors one end of the proximity spectrum and presumably those who live together are more inclined to pool their resources of both time and money so as to maximize the utility of all household members. But because coresidence also allows for economies of scale, the substitution of paid care for family care may be more feasible. At the same time the coresidence of a disabled parent with one child may encourage other children to provide financial support as well as less intensive care. Considerable effort is needed to understand how state transfers that

OCR for page 195
Demography of Aging subsidize market care not only substitute for family care but also affect the price and supply of paid care services. The elderly's use of medical services clearly responds to shifts in consumer prices or provider subsidies (Morrisey, 1993). There is little evidence to date, however, on how the supply or price of home care services changes as the population ages and demand increases. Finally, research is needed to understand not only the factors that promote various kinds of substitutions but also the consequences of these substitutions. How family and state transfers interact in terms of various outcomes has been largely neglected in previous research. Those studies that have examined such issues typically focus only on the risk of institutionalization (usually over a relatively short period of time) as the sole criterion of efficiency or efficacy of substitution. Yet other outcomes are reasonable to consider. Substitutions in either provider or currency may improve the overall well-being of the older beneficiary. The net effect of substituting paid care for some aspects of family care may be to redirect time transfers from the family to previously unattended areas of care. To the extent that paid care encourages the sustained work activity of potential providers, substitution also may protect contributions to structured intergenerational transfer programs financed through tax and payroll deductions (e.g., Social Security and Medicare trust funds). Demographic research on substitution draws on many of the themes common to more traditional demographic research areas—family, resource allocation, labor economics, and dependency. Although some of the key issues have been explored with respect to child care dependencies (see, for example, Liebowitz et al., 1988), research on substitutions involving old age beneficiaries lags far behind. REFERENCES Altonji, J., H. Fumio, and L.J. Kotlikoff 1992 Is the extended family altruistically linked? New tests with microdata. American Economic Review 82:1177-1198. Applebaum, R.A. 1988 Recruitments and characteristics of channeling clients. Health Services Research 23:51-66. Becker, G.S. 1981 A Treatise on the Family. Cambridge, Mass.: Harvard University Press. Behrman, J., R. Pollack, and P. Taubman 1990 The Wealth Model: Efficiency in Education and Distribution in the Family. Institute for Economic Research Working Paper, University of Washington, Seattle. Bengtson, V.L., and D. Dannefer 1988 Families, work, and aging: Implications of disordered cohort flow for the twenty-first century. In R.A. Ward and S.S. Tobin, eds., Health in Aging: Sociological Issues and Policy Directions. New York: Springer.

OCR for page 195
Demography of Aging Branch, L.G., and A.M. Jette 1983 Elders' use of informal long-term care assistance. Gerontologist 23:51-56. Burch, T.K., and B.J. Matthews 1987 Household formation in developed societies. Population and Development Review 13:495-511. Cantor, M. 1979 Neighbors and friends: An overlooked resource in the informal support system. Research on Aging 1:434-463. Christianson, J.B. 1988 The effect of channeling on informal caregiving. Health Services Research 3:99-117. Cicirelli, V.G. 1985 The role of siblings as family caregivers. Pp. 93-107 in W.J. Sauer and R.T. Cowards, eds., Social Support Networks and the Care of the Elderly. New York: Springer. Cohen, M.A., N. Kumar, and S.S. Wallach 1992 Who buys long-term care insurance? Health Affairs 11:208-223. Coughlin, T.A., T.D. McBride, M. Perozek, and K. Liu 1990 Home Care for the Disabled Elderly: Predictors and Expected Costs. Working Paper No. 3972-01. Washington, D.C.: The Urban Institute. Coward, R.T., and J.W. Dwyer 1990 The association of gender, sibling network composition, and patterns of parent care by adult children. Research on Aging 12:158-181. Cox, D. 1987 Motives for private income transfers. Journal of Political Economy 95:508-546. Cox, D., and F. Raines 1985 Interfamily transfers and income redistribution. In D. Martin and T. Smeeding, eds., Horizontal Equity, Uncertainty and Economic Well-Being. Chicago: University of Chicago Press. Cox, D., and M.R. Rank 1992 Inter vivos transfers and intergenerational exchange. Review of Economics and Statistics 74:305-314. Davis, K., and D. Rowland 1986 Medicare Policy: New Directions for Health and Long Term Care. Baltimore: Johns Hopkins University Press. Doty, P. 1986 Family care of the elderly: The role of public policy. Milbank Quarterly 64:34-75. Edelman, P., and S. Hughes 1990 The impact of community care on provision of informal care to homebound elderly persons. Journal of Gerontology 45:S74-S84. Estes, C.L., and J.H. Swan 1992 The Long Term Care Crisis: Elders Trapped in the No-Care Zone. Newbury Park, Calif.: Sage Publications. Greene, V.L. 1983 Substitution between formally and informally provided care for the impaired elderly in the community. Medical Care 6:609-619. Hagestad, G.O. 1986 The aging society as a context for family life. Daedalus 115:77-118. Hanley, R.J., and J.M. Wiener 1991 Use of paid home care by the chronically disabled elderly. Research on Aging 13:310-332.

OCR for page 195
Demography of Aging Hanley, R.J., J.M. Wiener, and K.M. Harris 1991 Will paid home care erode informal support? Journal of Health Politics, Policy and Law 16:507-521. Hill, M.S. 1990 Shared Housing as a Form of Economic Support for Young Unmarried Mothers. Paper presented at the annual meeting of the Population Association of America. Himes, C.L. 1992 Future caregivers: Projected family structures of older persons. Journal of Gerontology 47:S17-S26. Hurd, M.D. 1991 The Income and Savings of the Elderly: Final Report to the AARP Andrus Foundation. Unpublished manuscript, Department of Economics, State University of New York at Stony Brook. Jackson, D.J., C.F. Longino, Jr., R.S. Zimmerman, and J.C. Bradsher 1991 Environmental adjustments to declining functional ability: Residential mobility and living arrangements. Research on Aging 13:289-309. Jackson, J.S., L.M. Chatters, and R.J. Taylor 1992 Aging in Black America. Newbury Park, Calif.: Sage Publications. Kemper, P. 1992 The use of formal and informal home care by the disabled elderly. Health Services Research 27:421-451. Kingston, E.R., B.A. Hirshorn, and J.M. Cornman 1986 Ties That Bind. Washington, D.C.: Seven Locks Press. Kurz, M. 1984 Capital accumulation and the characteristics of private intergenerational transfers. Economica 51:1-22. LaPlante, M.P., G.E. Hendershot, and A.J. Moss 1992 Assistive Technology Devises and Home Accessibility Features: Prevalence, Payment, Need, and Trends. Advance Data Number 217. Washington, D.C.: National Center for Health Statistics. Liebowitz, A., L.J. Waite, and C. Witsberger 1988 Child care for preschoolers: Differences by child's age. Demography 25:205-220. Litwak, E. 1986 Helping the Elderly: The Complimentary Roles of Informal Networks and Formal Systems. New York: Guildford Press. Liu, K., K.G. Manton, and B.M. Liu 1985 Home care expenses for the disabled elderly. Health Care Financing Review 7:51-58. MacDonald, M. 1990 Family Background, the Life-Cycle, and Inter-Household Transfers. Center for Demography and Ecology Working Paper, University of Wisconsin, Madison. Macken, C. 1986 A profile of the functionally impaired elderly persons living in the community. Health Care Financing Review 7:33-49. Manton, K.G. 1989 Epidemiological, demographic and social correlates of disability among the elderly. Milbank Quarterly 67(2 part 1): 13-58. Manton, K.G., and B.J. Soldo 1985 Dynamics of health changes in the oldest old: New perspectives and evidence. Milbank Quarterly 63:206-285.

OCR for page 195
Demography of Aging McAuley, W.J., and R. Blieszner 1985 Selection of long-term care arrangements by older community residents. Gerontologist 25:188-193. Meiners, M.R., and J.N. Greenberg 1989 Improving the role of private markets in financing long-term care services. Pp. 87-115 in K. McLennan and J.A. Meyer, eds., Care and Costs: Current Issues in Health Policy. Boulder, Colo.: Westview Press. Morgan, J.N. 1983 The redistribution of income by families and institutions and emergency help patterns. In G.J. Duncan and J.N. Morgan, eds., Five Thousand American Families. Patterns of Economic Progress, Vol. 10. Ann Arbor, Mich.: Institute for Social Research. Morrisey, M.A. 1993 Retiree health benefits. Annual Review of Public Health 14:271-292. Moscovice, I., G. Davidson, and D. McCaffrey 1988 Substitution of formal and informal care for community-based elderly. Medical Care 10:971-981. Mutran, E. 1985 Intergenerational family support among blacks and whites: Response to culture or to socioeconomic differences. Journal of Gerontology 40:382-389. Riley, M. 1983 The family in an aging society: A matrix of latent relationships. Journal of Family Issues 4:439-454. Rindfuss, R.R., C.G. Swicegood, and R.A. Rosenfeld 1987 Disorder in the life course. American Sociological Review 52:785-801. Rosenzweig, M., and K. Wolpin 1990 Intergenerational Support and the Life-Cycle Incomes of Parents and Children: Coresidence and Intergenerational Financial Transfers. Paper presented at the annual meeting of the Population Association of America, Toronto. Scott, J.P. 1983 Siblings and other kin. Pp. 47-62 in T.H. Brubaker, ed., Family Relations in Later Life. Beverly Hills, Calif.: Sage Publications. Sloan, F.A., T.J. Hoerger, and G. Picone 1992 Effects of Public Subsidies and Strategic Behavior on Families' Savings and Long Term Care Decisions. Paper presented at the Workshop on the Economics of Aging, National Bureau of Economic Research, Cambridge, Mass. Soldo, B.J. 1985 In home services for the dependent elderly: Determinants of current use and the implications for future demand. Research on Aging 7:281-304. Soldo, B.J., and M.S. Hill In press Intergenerational transfers: Economic, demographic, and social perspective. Annual Review of Gerontology and Geriatrics, Vol. 13, Aging, Kinship, and Social Change. New York: Springer. Soldo, B.J., E.M. Agree, and D.A. Wolf 1989 The balance between formal and informal care providers. In M.G. Ory and K. Bond, eds. Aging and Health Care: Social Science and Policy Perspectives. London: Routledge. Soldo, B.J., D.A. Wolf, and E.M. Agree 1990 Family, households, and care arrangements of frail older women: A structural analysis. Journal of Gerontology 45:S238-S249.

OCR for page 195
Demography of Aging Stephen, E.H., V.A. Freedman, D.A. Wolf, and B.J. Soldo 1992 Intergenerational Flows of Support: Transfers Between the Elderly and Their Adult Children. Department of Demography Working Paper, Georgetown University, Washington, D.C. Stone, R., G.L. Cafferata, and J. Sangl 1987 Caregivers of the frail elderly: A national profile. Gerontologist 27:616-626. Suzman, R.M., K.G. Manton, and D.P. Willis 1992 Introducing the oldest old. In R.M. Suzman, D.P. Willis, and K.G. Manton, eds., The Oldest Old. New York: Oxford University Press. Tennstedt, S.L., and J.B. McKinlay 1986 An Examination of the Interface of Informal Care and Formal Services for Frail Elders. Paper presented at the 39th annual meeting of the Gerontological Society of America, Chicago, Ill. 1989 Informal care for frail older persons. In M. Ory and K. Bond, eds., Aging and Health Care. London: Routledge. Tennstedt, S.L., S. Crawford, and J.B. McKinlay 1992 Is Family Care on the Decline? A Longitudinal Investigation of the Substitution of Formal Long Term Care Services for Informal Care. Paper presented at the annual meeting of the Gerontological Society of America . Treas, J. 1977 Family support systems for the aged: Some social and demographic considerations. Gerontologist 17:486-491. Varian, H.R. 1990 Intermediate Microeconomics. New York: W.W. Norton. Wan, T.T.H. 1987 Functionally disabled elderly: Health status, social support, and use of health services. Research on Aging 1:61-78. Wiener, J.M., and R.J. Hanley 1992 Caring for the disabled elderly: There's no place like home. In S.M. Shortell and U. Reinhardt, eds., Improving Health Policy and Management: Nine Critical Issues for the 1990's. Ann Arbor, Mich.: Health Administration Press. Weissert, W.G., C.M. Cready, and J.E. Pawelak 1988 The past and future of home- and community-based long term care. Milbank Quarterly 66:309-388. Wilhelm, M.O. 1991 Bequest Behavior and the Effect of Heir's Earnings: Testing the Altruistic Model of Bequests. Paper presented at the annual meeting of the Population Association of America, Washington, D.C. Wolf, D.A., and B.J. Soldo 1990 Family Structure and Caregiving Portfolios. Paper presented at the annual meeting of the Gerontological Society of America, Boston, Mass., Nov. in press The trade off of time between parental care and paid employment. Journal of Human Resources.