State tax incentives for person giving informal care of the elderly.

Programs for informal caregivers of frail elderly can be adopted by States to address some of the problems associated with an expanding and costly long-term care system. In this article, highlights are given from a 3-year study of Idaho and Arizona tax incentive programs. Characteristics of informal caregivers and elderly participants are described, and elderly participants are compared with elderly nonparticipants and with the general elderly population. Tax incentives were positively related to the level of service and financial support provided by informal caregivers. Data were inadequate to determine whether the induced informal help substituted for public expenditures.

Introduction

During the last decade, there has been a sharp and dramatic increase in the Nation's elderly population, especially in the oldest and most impaired segment. Associated with this aging of America has come mounting pressure on publicly subsidized long-term care and support service programs.

Medicaid, a State-operated program jointly funded by the States and the Federal Government, is the principal source of public funding for long-term care. Because the Federal Government contributes to Medicaid programs through a matching percentage rate program, both State and Federal officials are concerned with rapidly increasing spending requirements. Consequently, both levels of government seek less costly alternatives as well as incentives by which a larger portion of these costs might be supported by private sources.

Several States are currently interested in addressing long-term care problems with legislation that will promote the purchase of private long-term care insurance. Tax incentives may be such a mechanism. However, it is clear that those who are frail now or likely to become frail in the near future may not benefit from such a program, given the long-term planning such programs would require. Also, long-term care insurance will not by itself stimulate an increase in the supply of informal caregiver support services to keep pace with rising demand.

The vast majority of community-based frail elderly, nearly 75 percent, rely solely on family and friends for informal care; most of the remainder depend on a combination of family care and paid help (Soldo, 1983; Liu, Manton, and Liu, 1986). Therefore, one approach that has been attractive to States is tax incentive programs that might support the informal caregiver system. Consequently, the tax incentive programs described here need to be understood as investments that a State makes in order to support the informal caregiver system, expecting to reduce either current or anticipated demands on the publicly subsidized long-term care system.

Idaho and Arizona were among the first States to create tax incentive programs. Although each State chose different eligibility criteria, the focus in both States was on the frail elderly. Iowa, Oregon, and North Carolina have adopted similar tax vehicles, and the Federal dependent care tax credit is applicable to all States. These programs, however, were not fully implemented at the time of this study.

Methods

Findings discussed throughout this article are based on a 3-year study conducted by the Center for Health and Social Services Research and its subcontractor, the Hebrew Rehabilitation Center for the Aged. An exhaustive description of the study methodology and limitations can be found in the final report (Hendrickson et al., 1988). The study methods are summarized here.

Primary data sources

Information about the configuration of the tax incentive programs was gathered using structured in-person interviews with State officials, including legislative and Governor's office staff and key agency personnel from health, social services, aging, finance, and tax revenue departments. Hearing transcripts, reports, and various manuals were reviewed as well. In addition, discharge planners, social workers, and tax preparers were surveyed by mail to determine their level of knowledge about the respective programs and to help determine the level of information diffusion in a State.

Telephone interviews were conducted with a sample of elderly persons claimed on tax returns, their caregivers, and a control group of elderly-caregiver pairs not participating in the tax incentive program. Two waves of person-level data collection were conducted at 1-year intervals. During 1983, data were gathered simultaneously for 1981, the baseline year prior to the tax program implementation, and for 1982, the first program year; 1 year later, data were gathered for 1983. Final sample sizes were 398 pairs of users and 203 pairs of nonusers in Idaho and 78 pairs of users and 358 pairs of nonusers in Arizona.

Secondary data sources

Other data sources used included a study of elderly in Massachusetts and the 1982 National Long-Term Care Survey, or LTCS (Hanley, 1984). In addition, over the 3-year study period, the 50 States were monitored through contacts with the Intergovernmental Health Policy Project and the Council of State Planning Agencies, as well as through direct contact with selected States to identify proposed State tax incentive legislation.

Data analysis

Extensive descriptive analyses of the tax systems, level of program dissemination, and claimant and dependent characteristics were conducted. In addition, because the quasi-experimental study design included both experimental and control groups, it was possible to control for baseline differences between the users and nonusers using regression techniques. To this end, hierarchical regression analyses were conducted using formal and informal services received as the dependent variable while controlling for demographic characteristics, other background variables, and baseline services.

Structure of tax incentive programs

In Idaho, the program allows either a deduction or a credit for in-home care of persons 65 years of age or over. A $1,000 deduction from taxable income can be claimed for each elderly person for whom care is provided. A refundable tax credit of $100 is available for taxpayers with insufficient income to take advantage of the tax deduction. Informal caregivers may claim a maximum of three persons per tax year. These incentives are available irrespective of the nature of the expenditures made and can be combined with a dependent deduction and/or a personal care deduction. The care must be provided in the home of the taxpayer, but the elderly person does not need to meet any disability requirements or income qualifications.

According to the Idaho State Tax Commission, which analyzes Idaho tax returns, the program costs little in administrative overhead or revenue loss. When the program was initiated in 1982, the State had no projections on rates of participation or on the expected amount of revenue loss. The State's experience for the 1982, 1983, and 1984 tax years is described in Table 1. No audits of tax returns have been provoked and no additional staff have been added as a result of this program.

Table 1

Idaho tax incentive program statistics: 1982-84

Year	Number of claims filed	Refundable credits	Deductions	Revenue loss	Cost per claim
1982	701	$60,000	$5,018	$65,018	$92.75
1983	692	61,000	3,500	64,500	93.21
1984	769	69,000	4,000	73,000	94.93

NOTE: All dollar amounts reported are estimates.

SOURCE: Department of Revenue and Taxation, Idaho State Tax Commission, Boise, Idaho.

In Arizona, the tax incentive program was an amendment to a nursing home deregulation bill. In the 1981 session, Arizona legislators were concerned with reducing nursing home costs, including regulatory costs, and stimulating the private sector to both provide and finance long-term care. Currently, long-term care in Arizona is coordinated and financed at the county level, with the State assuming principally a regulatory function. The Arizona Health Care Cost Containment System, a project designed to test the effectiveness of establishing a Medicaid program based on competitive principles, was still an experimental program at the onset of the tax study.

The tax incentive measure allows a $1,000 personal exemption for every taxpayer who paid 25 percent or more of the total yearly costs of an elderly person's nursing care in an institution, supervisory care facility, or foster care home; 25 percent or more of home health costs; or at least $800 toward other medical costs. The program also allows for a deduction of payments of any amount made on behalf of an elderly person, whether or not the person is a blood relative. Subsequent amendments have changed the benefit from a personal to a dependent exemption and lowered the exemption to $600, the level for all other dependents.

The State Department of Revenue originally estimated that more than 8,813 persons would be eligible for the tax incentive program and that total revenue loss might exceed $500,000 per year. Because the item(s) related to the tax incentive program are not computer coded on the Arizona tax return, it was impossible to determine the actual number of claimants or the amount of revenue loss. However, in the tax incentive study, an attempt was made to identify Arizona tax incentive claimants from a randomly drawn list of 10,000 Medicare recipients, and only 78 elderly were identified as participating in the program.

Recipients of care

In Idaho, an overwhelming percentage of the elderly persons claimed were female (78 percent), widowed (88 percent), and poor (82 percent with incomes of less than $5,000 per year). However, although many would presumably be eligible for both Medicaid and the Supplemental Security Income (SSI) program, only a small proportion (14 percent) were on Medicaid and even fewer (not even 5 percent) claimed SSI benefits. Most of the elderly (93 percent) were receiving social security payments. A large proportion of the elderly persons claimed were among the old-old: 34 percent were 85 years of age or over, compared with 9 percent of all elderly in the Idaho population. This is a much larger percentage of the old-old than the 20 percent reported in the 1982 LTCS. The majority (58 percent) of those 85 years or over were cared for by the young-old, claimants who were themselves 60 years of age or over.

Although no screening for frailty takes place in the Idaho tax incentive program, a large majority of elderly participants (83 percent) had one or more chronic illness that limited their activities. Indeed, 94 percent had one or more functional limitation, and almost 76 percent were vulnerable (i.e., deemed unable to survive at an acceptable level in the community without services, formal or informal).

In Arizona, an overwhelming proportion of the elderly participants (95 percent) were 75 years of age or over, with 64 percent being age 85 or over. The average age of the Arizona participants was 86 years, compared with 81 years in Idaho. A larger percentage of the Arizona sample were still married (14 percent, compared with 4 percent in Idaho), and a smaller proportion were widowed (81 percent, compared with 88 percent in Idaho). The sex distributions in the two States were essentially the same: 80 percent female in Arizona, 78 percent in Idaho. Arizona elderly participants were better off financially than Idaho elderly; nonetheless, 66 percent had incomes of less than $5,000. SSI benefits were claimed by 10 percent of Arizona participants, compared with only 5 percent of Idaho participants.

The participants in both Arizona and Idaho programs were considerably more impaired in activities of daily living (ADL) and instrumental activities of daily living (IADL) than the national population of elderly persons, as reported in the 1982 LTCS. Three ADL measures were compared. About 1 percent of the national sample had problems with eating, compared with 25 percent in Arizona and 22 percent in Idaho. Significant differences between the national sample and the two study States also were found with respect to dressing. A more severe problem of incontinence existed among Idaho and Arizona study elderly than among the LTCS population. Grocery shopping and handling money were the tasks most likely to cause problems for all three groups; however, the problem was more severe among Idaho and Arizona study elderly than among the national population. Problems with meal preparation were significantly greater for recipients of the two tax incentive programs than for the national population. Table 2 contains a summary of ADL and IADL comparisons.

Table 2

Percent of elderly persons with selected impairments: United States, 1982; Idaho, 1981-83; Arizona, 1981-83

Impairment	United States	Idaho	Arizona
ADL1 impairment	Percent
Dressing	3.9	15.5	26.6
Eating	1.4	22.4	25.0
Continence	5.9	26.3	32.8
IADL2 impairment
Preparing meals	7.0	54.9	82.8
Taking medicine	4.1	32.3	60.9
Grocery shopping3	13.5	—	—
Managing money3	6.2	—	—
Grocery shopping and managing money	—	75.4	95.3
Light housework3	6.1	—	—
Doing laundry3	9.7	—	—
Light housework and laundry	—	57.3	90.6

Activities of daily living.

Instrumental activities of daily living.

These IADL questions were asked as one question in the tax incentive study.

NOTES: National data are for community-based elderly persons surveyed in the Long-Term Care Survey. Idaho and Arizona data are for elderly persons cared for by participants in State tax incentive programs. Sample sizes used for this table were 393 in Idaho and 64 in Arizona.

SOURCE: (Hendrickson et at, 1988).

Caregivers

In Idaho, the vast majority of the tax claimants (more than 94 percent) were children or stepchildren, and the remainder were relatives other than children. Seventy-two percent of claimants were married, compared with 64.7 percent of taxpayers in the general Idaho population. A sizable minority of caregivers, 45 percent, were 55 years of age or over, compared with the general Idaho population, of which 18 percent were 55 years or over. Among caregivers interviewed, 18 percent themselves received social security benefits.

This older group of Idaho caregivers is particularly interesting because it represents an increasing phenomenon of the young-old, who are becoming disabled themselves and are often on fixed incomes, nonetheless accepting primary responsibility for the care of the old-old. About 19 percent of claimants in the age group 60 years of age or over indicated that they were experiencing health problems that limited their daily activities. A majority (58 percent) of these older claimants were in the income category less than $20,000, compared with 44 percent of the claimants under age 60.

In both the Idaho study and the 1982 LTCS, it was found that a majority of caregivers who provide services to the elderly are female. The proportion of female caregivers, 71.5 percent in the nationwide study, was 88 percent in the Idaho program. More than one-half of the Idaho claimants were employed at least part time. A comparable proportion (50 percent) had at least some college education. About 11 percent of Idaho caregivers were college graduates, compared with 16 percent in the general U.S. population.

Arizona tax claimants represent a significantly different mix of informal caregivers from those in the Idaho program. Although a majority of Arizona claimants were adult children (80 percent, compared with 94 percent in Idaho), 13 percent were spouses of the care recipient. Of the remaining number, 5 percent were siblings or grandchildren and 3 percent were nonrelatives.

Almost 72 percent of Arizona claimants were 55 years of age or over, and an even larger percentage of spouses (78 percent) were in this age category. Almost 36 percent of the claimants were 65 years of age or over, compared with only 12 percent in the general Arizona population. Nearly 55 percent of the tax claimants in Arizona already received social security benefits.

In Arizona, as in Idaho, a sizable percentage (25 percent) of those in the age group 60 years or over indicated that their health conditions limited their function as an informal caregiver for an elderly person. Of these older Arizona claimants, 41 percent had incomes of less than $20,000, compared with 37 percent in the age category under 60 years.

Female claimants represented the majority of those providing services to elderly persons (82 percent), yet nearly 38 percent of the female claimants were employed at least part time. About 56 percent of claimants had at least some college education, and 15 percent were college graduates. As a group, the Arizona claimants were significantly more affluent than those in Idaho. This is not surprising, given the structure of the Arizona program. More than 61 percent of Arizona claimants reported incomes of $20,000 or more, compared with 49 percent of Idaho claimants.

Tax claimants were asked their opinions about the importance of the tax incentive programs. Although 43 percent of claimants in Idaho believed that the tax incentive made a difference, more than three-quarters of the Arizona claimants indicated that the incentive was of real importance. As might be suspected, in Idaho, the percentage who perceived the program as important was higher among those with incomes of less than $20,000 because they could claim a tax credit without listing expenses. The reverse was true in Arizona, with a higher proportion of more affluent claimants perceiving the program as important, perhaps because Arizona claimants were required to itemize all expenses on the tax form.

Assistance level and service use

Idaho tax claimants demonstrated a significantly higher level of commitment to caregiving responsibilities than caregivers who were nonclaimants did. Nearly one-half of those interviewed spent 2 to 3 or more hours per day helping their elderly person with activities of daily living, even though 55 percent of all claimants were employed outside of the home at least on a part-time basis.

Elderly Idaho participants also required fewer formal services provided by community agencies than nonparticipants did. Only 15 percent of elderly persons claimed received any assistance with activities of daily living from community agencies. Study results also suggest that elderly participants required somewhat fewer nursing home days than did nonparticipants, although there were no significant differences in hospital or other institutional days. Given the average age of 81 for elderly participants, it was surprising to discover that only 1 percent had been institutionalized by the end of the first year of the program.

Although Idahoans generally did not contribute to the payment of elderly family members' medical bills, claimants provided more financial support for both medical and daily living expenses than nonclaimants did. More than 69 percent of claimants spent $1,000 or more on the older person's living expenses, with 12 percent contributing $500 or more for out-of-pocket medical expenses.

Arizona claimants were more likely to contribute financial support for medical and ancillary expenses (83 percent contributing support) than either Idaho claimants or Arizona nonclaimant caregivers were. Among Arizona claimants, 37 percent spent $1,000 or more on medical services, with 18 percent contributing $2,000 or more. In terms of ancillary expenses, 50 percent of claimants expended $2,000 or more.

Although Arizona tax claimants contributed greater financial support, their elderly also used more nursing home days than nonparticipants did. Arizona tax claimants contributed less informal services than Arizona nonclaimants or Idaho claimants did. However, the differences between Idaho and Arizona are understandable in light of program differences. Arizona's tax incentives are focused on stimulating informal caregivers to pay a larger amount of the costs for formal services, whereas the Idaho program is focused on encouraging provision of informal services available to the frail elderly. Arizona credits medical and ancillary expenses incurred and excludes informal services, whereas Idaho provides a refundable tax credit for informal services alone. Impact results from this study are summarized in Table 3.

Table 3

Summary regression results comparison of tax incentive participants and nonparticipants, by dependent variable: Idaho and Arizona, 1981-83

Dependent variable	Idaho participants vs. nonparticipants	Arizona participants vs. nonparticipants
Informal support
Total using informal services:
First year of program	+	−
Second year of program	+	*
Percent using informal services:
First year of program	+	*
Second year of program	+	*
Helper contributions
Medical care dollars:
First year of program	+	+
Second year of program	+	+
Daily expense dollars:
First year of program	+	+
Second year of program	+	+
Institutional placement
Nursing home days:
In past 12 months	*	+
In past 24 months	*	+
Hospital days:
In past 12 months	*	*
In past 24 months	*	*
Noncommunity days:1
In past 12 months	*	+
In past 24 months	*	*

Days in group quarters—domiciliary care facilities, mental institutions, etc.

NOTES: + = participant greater than nonparticipant; significant at the p = .05 level. * = no significant difference. - = participant less than nonparticipant; significant at the p = .05 level.

SOURCE: (Hendrickson et al., 1988).

Summary and conclusions

This article contains a description of tax incentive programs in the States of Idaho and Arizona and characteristics of informal caregivers and elderly participants. Participants are compared with both nonparticipants and a national sample from the 1982 Long-Term Care Survey. Although limitations exist with study sample sizes in both Idaho and Arizona, making comparisons somewhat problematic, the results are consistent with those from other national and regional studies: Informal caregivers provide a great deal of support to the frail elderly. In such studies, it has been estimated that nearly three-quarters of the disabled elderly who live in the community rely solely on their informal caregivers for support (Horowitz and Shindelman, 1983; Lammers and Klingman, 1984; Soldo and Sharma, 1980; Stone, Cafferata, and Sangl, 1987).

According to results from the study discussed here, tax incentives for informal caregivers are worth considering. Such programs might decrease the amount of formal services used by increasing the amount of informal services available to frail elderly. Moreover, they might serve to shift some of the costs for long-term care from public to private sector sources by encouraging informal caregivers to assume such costs for their elderly. Several study findings lend support to these premises.

First, elderly participants in these programs are considerably older and more impaired than the general elderly population in these States or the national sample from the 1982 LTCS. It appears that the Idaho and Arizona tax incentive programs, without explicitly screening for frailty, are attracting caregivers of the frail elderly. Ninety-six percent of the elderly participants in Idaho and 100 percent of those in Arizona were impaired.

A second finding relates to the types of informal caregiver behavior the tax programs have served to strengthen. In Idaho, tax incentives are principally used to stimulate family caregivers to expand the informal services available to frail elderly. Credit is based on living arrangement rather than documentation for actual expenses incurred. The success of the program is indicated by the significantly greater amount of informal services provided by claimants than by nonclaimants as measured by both total hours of informal services provided and money contributed for medical care and daily living needs of the elderly. As previously noted, elderly participants in Idaho receive a smaller percentage of their help from community agencies and somewhat fewer nursing home days than nonparticipants do.

These findings indicate the effectiveness of Idaho's tax incentive program in stimulating informal services and financial support for daily living expenses of community-based elderly. It achieves this despite the rather modest $100 credit given. These positive findings assume added meaning in light of the fact that almost all of the elderly participating in the program had levels of impairment similar to those of elderly persons who are in nursing homes or are candidates for nursing home admission within a relatively short period of time.

In Arizona, no credit is given for the amount of informal services provided to the elderly. Consequently, it is not surprising to find the following.

A significantly larger proportion of Arizona than Idaho elderly participants were institutionalized residents of either skilled nursing homes or personal care institutions.

Arizona elderly claimed were sicker and older than participants in Idaho or elderly persons in the national LTCS sample.

The vast majority of Arizona claimants made cash outlays for medical and ancillary expenses for their elderly.

A larger proportion of Arizona than Idaho elderly participants received help in major activities of daily living from community agencies.

The goal of the Arizona tax incentive program, to stimulate informal caregivers to assume greater financial responsibilities for their frail elderly, especially for such services as medical care and formal support services, appears to have been achieved. As was found in Idaho, the Arizona program reached informal caregivers of an extremely vulnerable group of community-based elderly.

In conclusion, a number of policy implications result from this study. First, appropriately structured and implemented tax incentive programs appear to stimulate and strengthen the informal caregiving system for frail vulnerable community-based elderly. Tax incentives appear to serve as an inducement for families and friends to provide a greater amount of the informal care and support in daily living services required by the frail elderly, especially those who are at risk of being institutionalized. It appears that such programs contribute to a slowdown in the rate of migration of elderly from informal to formal services, at least until a critical level of need is reached.

In particular, findings on the Idaho program, with its focus on encouraging intergenerational living arrangements, support this conclusion. The tax program contributed to the amount of support services available to disabled elderly in the community. Also, the use of refundable tax credits does not create a bias toward either formal or informal services, nor does it eliminate low- or moderate-income caregivers from participating in the program. Even those whose income is below the level required in order to pay State income taxes are stimulated to provide informal services.

Second, tax incentive programs can be structured to stimulate informal caregivers to assume some of the costs for formal and ancillary services that might otherwise require public support. The tax incentive program in Arizona was structured to shift costs for medical care, custodial care, and other formal services to the private sector. A potential danger in the current design of the Arizona program is that it will accelerate institutionalization among frail elderly because of the bias of the program toward formal services. However, to date, the Arizona program is benefiting elderly who are older and sicker than those in the Idaho or nationwide samples.

The Arizona experience supports the conclusion that tax incentives can be used to promote the private sector to share costs for formal health care services for frail elderly. The extent to which this will result in a significant savings for the State is unknown at this time. Arizona, unfortunately, does not have ready access to information on the amount of revenue lost as a result of the tax incentive program.

The results of this study suggest that tax incentive programs can be used to stimulate a number of informal caregiver behaviors, ranging from increasing informal caregiver services to paying for formal health care and ancillary services. However, legislators should clearly define which behaviors of informal caregivers are to be affected and should structure incentives and eligibility criteria to match these specific objectives.

Tax incentives represent a major change in policy development. To date, most publicly subsized programs directed at the frail elderly are in the form of entitlement programs, primarily Medicaid and Medicare, which compensate for services delivered by formal caregivers. These programs require huge bureaucracies to administer them at Federal, State, and local levels. Further, because entitlement programs currently reimburse formal caregivers directly for their services, there is a natural tendency for providers to encourage frail elderly to use their services even when less costly alternatives are available. Study data were inadequate for determination of whether the induced informal help substituted for public expenditures. Tax incentive programs are feasible because both Federal and State Governments currently have mechanisms to audit tax forms on an annual basis. However, the incentives may redistribute a portion of the subsidies for long-term care from the welfare or low-income population to others.

Family and informal support services should be utilized to the greatest possible extent, and appropriate recognition and incentives should be given for such services. The tax benefits of caregiving will never equal the costs incurred by informal caregivers of frail elderly. However, such programs may serve as a signaling device to the elderly, informal caregivers, and members of the long-term care system, indicating that informal caregivers are an essential and valued component of the State's approach to long-term care.