What Affects the Demand for Long-Term Care Insurance? A Study of Middle-Aged and Older Adults in Hong Kong

Abstract

Despite the potential of private insurance in financing long-term care (LTC), its coverage remains rather limited. This study is built on a comprehensive framework to examine the demand for LTC insurance in Hong Kong, a rapidly aging Asian society. A telephone survey was conducted in 2016 to collect data that formed a sample of 1,474 middle-aged and older adults. Multivariate analysis reveals more nuanced characteristics of Hong Kong middle-aged and older adults who tend to show a demand for LTC insurance, including: (a) being younger, better educated, relatively high status, and financially literate; (b) living with children but reluctant to be a burden on the family; (c) being in a better financial situation and able to afford premiums; and (d) anticipation of LTC needs and dependence, and a preference for formal care. This study offers preliminary evidence to understand the demand structure of the LTC insurance market in an Asian society.

Keywords

long-term care insurance aging financing disability Hong Kong

Introduction

Rapidly aging populations, coupled with the increasing prevalence of elderly chronic disability, have created enormous demand for long-term care (LTC). In the coming decades, how to finance LTC will become a policy issue of strategic importance for many governments, as population aging augurs a sharp increase in the demand for LTC services while sustainable funding mechanisms have not been put in place. A lack of awareness of LTC costs leads to poor financial preparedness among the elderly, who often find the expenses too large to absorb, resulting in financial distress. Paying LTC costs out of pocket represents a catastrophic financial burden for many elderly people and their families. Conventional wisdom would suggest that LTC due to dependency is a high-cost low-probability event that is theoretically insurable; the random and costly nature of LTC expenditures makes them, ideally, the type of risk for which risk-averse individuals would find insurance valuable (Costa-Font, 2010; Gleckman, 2007). Contrary to theoretical predictions, the paradoxical reality is that the private LTC insurance market is remarkably underdeveloped and makes a low contribution to financing (Feder, Komisar, & Friedland, 2007).

Despite the potential of voluntary private insurance, some intrinsic weaknesses have hampered its expansion into LTC. First of all, private LTC insurance is often too expensive (Feder et al., 2007). Second, most policies have major limitations in terms of the benefits covered, inflation adjustments, and benefits in case of lapse (Wiener, 2010). Third, and more fundamentally, Barr (2010) argues that private insurance is ultimately incapable of addressing the nature of risk associated with LTC needs, largely due to severe asymmetric information. Therefore, it is unsurprising that a private insurance market for LTC has thus far failed to thrive even in free-economy aging societies. In most Organisation for Economic Co-Operation and Development (OECD) countries, for example, less than 2% of total LTC expenditure is financed through private LTC insurance (Colombo & Mercier, 2012).

A number of studies have sought to further explain the low demand for private LTC insurance, some attributing it to supply-side imperfections and others shedding light on demand-side issues. Yet, most studies are based on the Western context while little is known about LTC insurance demand in societies where there is neither a private LTC insurance market nor social LTC insurance system. As the demand for LTC insurance may be shaped by cultural and sociological factors that are inherently contextual (McGarry, Temkin-Greener, & Li, 2014), further studies are vital to broaden the received knowledge by examining the situation in other societies. This article attempts to further scholarly understanding on factors associated with the demand for private LTC insurance with an empirical study of both middle-aged and older adults in Hong Kong, a rapidly aging Asian society, where government-funded formal LTC services are available but unable to meet vast demands, resulting in significant private responsibility for financing.

Literature Review

Insurance is considered an effective tool to finance LTC because its intermittent payment reduces people’s risk of catastrophic out-of-pocket expenses by paying a predictable affordable premium. The increasing burden of both public LTC expenditures and out-of-pocket payment could be mitigated by engaging the private sector. A few theoretical frameworks have been proposed to explain the low purchase rates of private LTC insurance. Some attribute the underdevelopment of the LTC insurance market to supply-side factors such as transaction cost, asymmetric information, and issues related to long-term contracting (Sloan & Norton, 1997). Other studies, however, have found that even when these issues are addressed, the demand-side imperfections, such as a lack of awareness regarding future LTC risk, intra-family moral hazard, and adverse selection, would still hamper the growth of the market for private LTC insurance (Brown & Finkelstein, 2009; Brown, Goda, & McGarry, 2012).

Theories in the field of behavioral economics provide a useful conceptual framework for understanding demand-side factors. Psychological factors such as loss aversion and mental accounting may affect purchase decisions regarding LTC insurance (Chen, 2003). Over the last decade, researchers have sought to identify other behavioral and psychosocial factors; for instance, the crowd-out effect of other financing mechanisms which cover part of LTC costs (Brown, Coe, & Finkelstein, 2007), the availability of tax subsidies (Courtemanche & He, 2009), mutual perceived risks of LTC needs of married couples (Caro, Porell, & Kwan, 2011), and the bequest motives of older persons (Lockwood, 2012) have been found to influence the demand for LTC insurance in the West.

In a recent study, Brown et al. (2012) constructed a more comprehensive framework and found that older Americans’ demand for private LTC insurance is explained by four factors: preferences and beliefs (i.e., bequest motives and beliefs about the need for care), substitutes for insurance (e.g., family financial support and social welfare), substitutes for formal care (e.g., preference for formal care, care provided by family), and features of insurance (e.g., high premiums, etc.). Specifically, individuals’ anticipation of their LTC needs in their old age and bequest motives could increase the demand for insurance because they may want to avoid the catastrophic expenditure related to LTC so that they can leave more to their children. Other studies have elucidated that crowding-out by informal care (Costa-Font & Courbage, 2015) and public care (Brown & Finkelstein, 2009) may also undermine the demand for private LTC insurance.

Gender is another characteristic which may affect the demand for LTC insurance. Due to their longer life expectancy, the estimated average remaining years of disability are greater for women than men. As such, older women are also more likely to use formal LTC services than older men (Brown & Finkelstein, 2009). All these differences may lead to gender difference in the willingness to pay for LTC insurance.

Despite the value of the knowledge gained, the existing literature has three major weaknesses that will be addressed in this study. First, one potential explanation for low LTC insurance demand, which has been largely neglected in the literature, is that consumers may lack the necessary financial literacy and forward-looking personality to make complicated financial decisions with respect to LTC insurance. Specifically, a thorough examination of an insurance product requires forward-planning and the ability to perform complex estimation in order to determine the expected value of insurance. A number of recent studies have found that cognitive capacity, numeracy, and the ability to make decisions with limited data are significantly associated with the decision to purchase LTC insurance (Lin & Prince, 2016). In this study, the role of financial literacy and planning personality in willingness to pay for LTC insurance will be examined in the context of Hong Kong.

Second, another crucial factor to which the literature has not paid adequate attention is how current living arrangements affect LTC insurance demand, although this has been found to be significantly associated with anticipated need of LTC services (He & Chou, 2017). If most elders have supportive living arrangements (e.g., living with adult children), they might prefer to age in the community rather than in institutional settings; this preference, in turn, may alter decisions regarding the financing of LTC services. The effect of current living arrangements on individuals’ LTC insurance demand will be examined in this study too.

Third, our knowledge regarding intergenerational differences in LTC insurance demand remains limited. While many within the current cohort of older adults have been struggling with their LTC bills, middle-aged adults will face the same challenges in due course, probably at an even greater magnitude because of rising life expectancy and costs. Furthermore, empirical evidence suggests that individuals’ financial planning and retirement planning behaviors also appear to differ by birth cohort (He & Chou, 2017; Robison, Shugrue, Fortinsky, & Gruman, 2014). These differences may result from the fact that the future cohort of older adults tends to have higher educational attainment, higher income and assets, a lower marriage rate, fewer children, less access to informal caregivers, and better health status (He & Chou, 2017). While it may be too late to introduce risk-pooling arrangements for most of the current cohort of older adults, initiating policy interventions targeting the middle-aged cohort and increasing their financial preparedness would be not only more realistic but also cost-effective.

In sum, the multifaceted nature of LTC insurance demand has not been examined comprehensively. We intend to fill these gaps by identifying a wider range of individual characteristics that are predicted to affect willingness to pay for private LTC insurance among both middle-aged and older persons. We seek to unravel individuals’ decisions from the following aspects: age, gender, living arrangements, financial literacy, planning personality, and attitudes toward LTC needs and insurance.

LTC in Hong Kong

Hong Kong is one of the fastest aging societies in the world, with highest life expectancy at birth and very low fertility rate (United Nations Department of Economic and Social Affairs, 2015). Citizens aged 65 and over numbered 1.12 million (or 15.3% of total population) in 2015, and this figure is projected to increase to 2.51 million (or 30.6% of total population) by 2043 (Census and Statistics Department, 2015). Under the current system in Hong Kong, the financing options available to most individuals in need of LTC services are limited to the Comprehensive Social Security Assistance Scheme (CSSA, a means-tested, unconditional cash transfer program), direct subsidies from the government, private savings, and unpaid caregivers. Approximately 80% of those receiving residential care services are recipients of the CSSA, the territory’s social safety net (Elderly Commission, 2017). Therefore, public finance is the largest payer for LTC expenditures in Hong Kong.

In a recent report released by the Elderly Commission (2017), 96% and 90% of costs for home-based and center-based community LTC services, respectively, are directly subsidized by the government while older persons living in subsidized residential care services only pay about 10% to 20% of the bill. Hong Kong’s reliance on the public purse to fund LTC has raised serious concerns as to whether the current financing arrangements are sustainable. Without financing reform, the demographic shift will result in rapid escalation of public expenditure and significantly undermine the financial sustainability of the LTC system.

Current expenditures on LTC in Hong Kong are substantial and are expected to increase fast in the coming decades, from 1.4% of GDP in 2004 to 3.0% by 2036 (Chung et al., 2009). This growth will be accelerated by a rapidly aging population. The Hong Kong Special Administrative Region (SAR) Government has been seeking alternative funding arrangements, including co-payment and private insurance, to diversify the financing base of its LTC system. Private-sector initiatives are appealing in Hong Kong because they resonate with the territory’s long-held belief in “big market, small government.”

Traditionally, family care plays the key role in providing LTC service in Hong Kong. However, the changes in family structures as well as the deterioration of traditional values such as filial piety have made family an increasingly unreliable source of care (Lum et al., 2016). The government subsidizes nongovernmental organizations and private facilities to provide institutional care to the elderly. Users pay nominal fees to service providers that are heavily subsidized by the government. Reliable standard of service and low cost make subvented homes preferred by almost all older people, but unfortunately, very few of them can be allocated with a bed once they need, due to the shortage of subsidized facilities and long waiting list. By August 31, 2016, there were a total of 34,749 applicants being waitlisted for various types of subsidized residential care services; the waiting time was 38 months for subvented homes and contract homes (Social Welfare Department, 2016). The dual unavailability of unpaid family caregivers and subsidized residential services essentially means that the elderly would have to resort to the private sector for LTC services.

High-quality private LTC services are available in Hong Kong, but their costs are substantial even for the middle classes. The annual cost of a room in a quality private nursing home is over HK$120,000 (depending on the user’s level of disability, the fee can double or even triple), while the median income of a two-person household was HK$228,000 in 2016 (Hong Kong Consumer Council, 2015). Consequently, paying for private LTC represents one of the largest financial risks facing Hong Kong’s aged population. Despite this expenditure risk, the importance of LTC insurance is not yet recognized by the general public. The commercial sector is also reluctant to develop such products because there are too many unknowns regarding potential take-up rate and the financial risks involved (Chou, Chow, & Chi, 2005). At the same time, however, Hong Kong has one of the most fledged insurance markets in Asia, with the per capita insurance premium standing at high levels, and insurance penetration rate reaching 16.3% in 2016. It is very common for Hong Kong people to own private insurance policies (Hong Kong Trade Development Council, 2017). In 2016, the number of in-force individual life insurance policies alone was close to 12 million, with office premiums amounting US$385 billion (Hong Kong Insurance Authority, 2017). Approximately 170 commercial insurers have business in the Hong Kong market, operating a full range of products including life, health, accident, motor vehicle, property damage, and other insurance products.

Method

This study used a telephone survey to investigate the demand for LTC insurance among middle-aged and older adults in Hong Kong. Conducted by the Public Opinion Program of the University of Hong Kong in February 2016, the survey included a total of 40 items that took participants approximately 15 min to complete. Ethical approval was obtained from the Human Research Ethics Committee of the authors’ university. Verbal consent was sought from participants before the survey commenced.

Conceptual Framework

To account for the variety of factors that may influence individuals’ demand for LTC insurance, Andersen’s Behavioral Model of Health Services Use was adopted as the conceptual framework. According to this model, the utilization of health-related services is determined by three groups of factors, namely, predisposing factors, enabling factors, and needs factors. Predisposing characteristics are those that exist prior to sickness but may influence behaviors related to utilization of care, such as age, gender, race, and health beliefs. Enabling factors facilitate or inhibit service utilization once sickness occurs; examples could include family support, insurance, and the like. Needs factors represent both perceived and actual needs for care that exert direct influence on service utilization (Andersen, 1995). Widely applied in empirical studies analyzing the demand for LTC insurance, the Andersen Model was adopted in this study to account for the factors shaping individuals’ intention to subscribe to private LTC insurance in Hong Kong.

Sampling

The sampling was performed in two steps. First, a list of telephone numbers was randomly selected from the latest residential phone directory as seed numbers, based on which a sampling list of phone numbers was generated by using the “plus or minus one or two” method. Second, one adult aged 40 or over who was a Cantonese-speaking Hong Kong Chinese resident was randomly selected from each household using the “next birthday” method. The sampling yielded 1,613 adults aged 40 and over with a response rate of 63.3%. The final dataset used in statistical analysis contained 1,474 observations after missing values were excluded.

Dependent Variable

Most question items used yes/no or structured scale formats. The instrument included questions measuring individuals’ sociodemographic background, financial status, attitudes toward LTC insurance, and anticipated LTC needs. With reference to a recent study by Tennyson and Yang (2014), the dependent variable that probed participants’ intention of purchasing private LTC insurance was measured by the following item: “Long-term care insurance is a type of insurance that helps to pay for a place in a nursing home or assisted living facility, or for personal care in your home. This type of care is typically separate from your regular health insurance and requires paying separate premiums. Suppose there is an insurance product which requires five years’ regular contribution but gives you financial protection against financial risks from long-term care costs, would you be interested in a long-term care policy?” In this binary variable, a value of 1 denoted “Yes, I am interested,” whereas 0 denoted “No, I am not interested.”

Predisposing Variables

Predisposing variables included age (40-64 vs. 65+), gender, current living arrangements (living with children or not), marital status (married or not), education (high school and below vs. university), and five attitudinal statements about LTC insurance (Brown et al., 2012): (a) bequest motive (i.e., It is important to leave an inheritance to my loved ones); (b) availability of family care (i.e., If I need LTC, a family member will be able to take care of me); (c) no burden on family (i.e., It is important to me that I do not create a financial burden for my family if I need LTC); (d) preference for formal care (i.e., I would prefer to receive care from a professional health aide or nurse rather than my spouse or another family member); and (e) the principle of familial obligation (i.e., It is a child’s obligation to help a parent with LTC needs). Participants were invited to indicate their endorsement of each of the attitudinal statements on a 5-point scale which was later converted into binary variables (1 = strongly agree or agree; 0 = else).

Enabling Variables

Enabling variables included (a) financial literacy, (b) planning personality, (c) the availability of savings, (d) the availability of instrumental support, (e) perceived social class, and (f) three attitudes regarding LTC insurance. Financial literacy was measured by three items developed for the 2004 Health and Retirement Study which have been widely used in national surveys of other countries (Fornero & Monticone, 2011). These three questions assessed basic knowledge of finance: knowledge related to the compound nature of interest rates, understanding of the effect of inflation on consumption power, and knowledge of risk diversification.

Participants were asked whether they considered themselves to be a planner, rated on a 5-point scale ranging from 1 = strongly agree to 5 = strongly disagree. They were classified as a planner if they strongly agreed or agreed with the statement. Availability of savings was measured by asking “Do you usually have money left over at the end of the month for discretionary spending?” with yes or no options. Perceived social class was evaluated by asking “Which social class do you think your family belongs to? Upper, upper middle, middle, lower middle, or lower class?” Participants were categorized as of high social class if they identified themselves as middle class or higher.

Three attitudes toward LTC insurance were adopted from a recent study by Brown et al. (2012): (a) paying LTC by welfare (i.e., The CSSA will cover my LTC expenses); (b) ability to pay premiums (i.e., I am concerned about my ability to pay the premiums for an LTC insurance policy); and (c) appropriate pricing of insurance (i.e., LTC insurance policies are appropriately priced given the cost of the care they cover). The last item was used to evaluate the importance of administrative expenses appended to the price of the policy, as separated from the cost of the insurance itself. All these items were rated on 5-point scale ranging from 1 = strongly agree to 5 = strongly disagree and were later on recoded as binary variables for easy interpretation (1 = strongly agree and 2 = agree recoded to 1, otherwise zero).

Needs Variables

Needs variables included the perceived likelihood of needing LTC (Robison et al., 2014) and anticipated dependence in the future. LTC needs were assessed by asking participants “What is your assessment of your chances of needing LTC (including informal or formal care at home, assisted living, or residential home care) in the future?” to which answers were given on a 4-point scale ranging from 1 = very low to 4 = very high. The anticipation of dependence in the future was captured by asking participants how far they agreed with the statement, “At some point in the future it is likely that I will no longer be able to live independently because of my health,” on a 5-point scale ranging from 1 = strongly agree to 5 = strongly disagree. Both variables were dichotomized (scores of 3 and 4 recoded as 1 for the former variable, and those of 1 and 2 as 1 for the latter).

Analytical Methods

Univariate statistics and bivariate analyses were performed first; the latter examined the relationships between predisposing, enabling, and needs variables and the intention to purchase LTC insurance. Later on, the dependent variable was regressed with logistic models on predisposing, enabling, and needs variables, in a phased fashion to estimate the effects of these three groups of factors individually and in combination.

Results

Descriptive Results

Descriptive results are presented in Table 1. Overall, one third of participants (33.3%) reported the intention to purchase LTC insurance. Of 1,474 participants, 28% were older adults, 46% were male, 62% were living with their children, 77% were married, and 17% had tertiary education credentials. More than half (57%) believed that leaving a bequest was a crucial consideration and reported that family care was available when they needed LTC; 84% felt that their LTC needs constituted a financial burden on their family; 48% preferred formal care over family care; 72% endorsed the importance of familial obligations to elderly care. Overall financial literacy was very low as only 6% of participants answered all three test items correctly; 58% considered themselves to be planners; 60% reported having money left over regularly for discretional spending; 25% had instrumental support available; 23% perceived themselves as belonging to the middle or upper classes; 17% reported that welfare covered their LTC expenses; 60% indicated that LTC insurance premiums were affordable; 34% believed that the insurance was appropriately priced. Finally, 28% expected to need LTC and 20% of these expected to become dependent in the near future.

Table 1.

Descriptive Statistics and Bivariate Analysis of Relationships Between Predisposing, Enabling, and Needs Factors and Intention of Purchasing Private LTC Insurance (n = 1,474).

Variables	% Descriptive statistics	% Interest in purchase	χ²
Total sample	% Descriptive statistics	33.3	χ²
Predisposing variables
Age cohorts			55.233**
40-64	72.3	38.7
65+	27.7	18.4
Gender			6.829**
Female	53.7	30.1
Male	46.3	36.5
Living arrangement			22.533**
Not living with children	37.6	25.6
Living with children	62.4	37.6
Marital status			5.200*
Not married	23.3	28.0
Married	76.7	34.6
Education			18.917**
High school or below	83.2	30.7
University	16.8	44.9
Attitudes
1. Bequest motive			17.398**
No	43.0	27.2
Yes	57.0	37.5
2. Availability of family care			3.092
No	45.5	30.7
Yes	54.5	35.1
3. No burden on family			49.033**
No	16.3	13.7
Yes	83.7	36.9
4. Preference for formal care			21.083**
No	51.9	27.7
Yes	48.1	39.0
5. Principle of familial obligation			1.820
No	27.9	35.8
Yes	72.1	32.1
Enabling variables
Financial literacy scores			38.936**
Score = 0	12.3	14.9
Score = 1 or 2	81.3	34.6
Score = 3	6.4	48.9
Planning personality			4.116*
No	41.8	30.2
Yes	58.2	35.2
Availability of savings			37.092**
No	40.4	24.0
Yes	59.6	39.2
Availability of instrumental support			3.470**
No	74.8	31.8
Yes	25.2	37.0
High social class			28.225**
Lower middle or below	76.9	29.5
Middle or above	23.1	45.0
Paying LTC by welfare			0.205
No	82.8	33.4
Yes	17.2	31.9
Ability to afford premium			0.035
No	40.3	32.8
Yes	59.7	33.3
Appropriate pricing of insurance			82.567**
No	65.9	25.1
Yes	34.1	48.6
Needs variables
Likelihood of needing LTC			8.655**
No	72.1	30.9
Yes	27.9	38.9
Anticipation of being dependent			13.086**
No	79.6	30.9
Yes	20.4	41.9

Note. LTC = long-term care.

p < 0.05, **p < 0.01.

Bivariate Results

Bivariate analysis revealed statistically significant association between the intention to buy LTC insurance and age, gender, living arrangements, marital status, education, bequest motive, burden on family, preference for formal care, the value of familial obligation, financial literacy, planning personality, availability of savings, availability of instrumental support, perceived social class, price of insurance premium, likelihood of needing LTC, and anticipated dependence, as shown in Table 1.

Multivariate Analyses

Table 2 shows the results of logistic regressions examining correlates of the intention to purchase LTC insurance. Significant chi-square statistics suggest that the set of factors in each model reliably predicted the dependent variable while the percentage of variance explained was moderate. In the phased regression analysis, the percentage of variance in the dependent variable explained by the models increased from 14.0% to 21.5%, and further to 22.7%, with the addition of enabling variables and needs variables, respectively.

Table 2.

Multivariate Predictors of Intention of Purchasing Private LTC Insurance (n = 1,473).

	Model 1		Model 2		Model 3
	OR	95% CI	OR	95% CI	OR	95% CI
Predisposing variables
Age cohorts
65+ (ref.)
40-64	2.328**	[1.729, 3.134]	2.043**	[1.493, 2.794]	2.109**	[1.537, 2.893]
Gender
Female (ref.)
Male	1.384**	[1.091, 1.756]	1.259	[0.982, 1.613]	1.280	[0.997, 1.643]
Living arrangement
Not living with children (ref.)
Living with children	1.449**	[1.112, 1.888]	1.376*	[1.046, 1.811]	1.393*	[1.057, 1.837]
Marital status
Not married (ref.)
Married	0.952	[0.749, 1.360]	0.925	[0.677, 1.265]	0.949	[0.692, 1.300]
Education
High school or below (ref.)
University	1.599**	[1.189, 2.149]	1.467*	[1.061, 2.028]	1.462*	[1.055, 2.026]
Attitudes
Bequest motive	1.320*	[1.034, 1.683]	1.172	[0.908, 1.514]	1.167	[0.902, 1.509]
Availability of family care	0.959	[0.755, 1.218]	0.793	[0.614, 1.025]	0.832	[0.643, 1.078]
No burden on family	2.999**	[2.010, 4.474]	2.536**	[1.680, 3.830]	2.493**	[1.647, 3.775]
Preference for formal care	1.621**	[1.284, 2.048]	1.430**	[1.119, 1.828]	1.332*	[1.038, 1.710]
Principle of familial obligation	0.860	[0.664, 1.113]	0.790	[0.603, 1.035]	0.786	[0.599, 1.031]
Enabling variables
Financial literacy scores
Score = 0 (ref.)
Score = 1 or 2			1.655*	[1.042, 2.631]	1.662*	[1.042, 2.649]
Score = 3			2.450**	[1.302, 4.612]	2.565**	[1.356, 4.853]
Planning personality			1.002	[0.780, 1.287]	0.971	[0.755, 1.250]
Availability of savings			1.398*	[1.072, 1.823]	1.478**	[1.129, 1.935]
Availability of instrumental support			1.211	[0.923, 1.588]	1.175	[0.894, 1.544]
High social class			1.519**	[1.1.133, 2.038]	1.542**	[1.148, 2.071]
Paying LTC by welfare			0.788	[0.571, 1.087]	0.772	[0.558, 1.070]
Ability to afford premium			1.113	[0.858, 1.443]	1.084	[0.835, 1.409]
Appropriate pricing of insurance			2.704**	[2.099, 3.485]	2.612**	[2.023, 3.372]
Needs variables
Likelihood of needing LTC					1.331*	[1.009, 1.756]
Anticipation of being dependent					1.522**	[1.124, 2.061]
Constant	0.046		0.027		0.022
Model summary chi square	156.349		246.980		262.032
(df, p value)	(10, p < .001)		(19, p < .001)		(21, p < .001)
–2 Log likelihood	1,714.882		1,624.251		1,609.199
Nagelkerke R²	14.0%		21.5%		22.7%

Note. LTC = long-term care; OR = odds ratio; CI = confidence interval.

p < 0.05, **p < 0.01.

Birth cohort turned out to be a very powerful predictor of the dependent variable in all three models, with middle-aged adults being significantly more likely than older adults to show purchase intention. In the first model with only predisposing factors, men were more likely than women to plan to buy LTC insurance, but this significance diminished in the second model after enabling variables were entered into the regression. Living arrangements were significantly associated with the dependent variable in all three models. Specifically, participants who were living with children demonstrated greater interest in buying insurance than those who were not. Similarly, participants educated beyond high school more often planned to pay for LTC insurance than those with lower educational attainment, in all models.

As shown in Model 1, individuals who considered bequests important were more likely to demonstrate greater intention to purchase insurance than those who did not, but the association became insignificant after enabling variables were entered into the model. Two attitudes persistently predicted the dependent variable in all models: Participants who agreed or strongly agreed that their LTC needs should not become a financial burden on their families and those who stated a preference for formal care tended to demonstrate a greater interest in LTC insurance. Financial literacy score also stood out as a powerful predictor of the dependent variable in all models. Having money available for discretional spending increased the likelihood of insurance subscription. Those who identified themselves as middle or upper class and those who considered premium prices to be appropriate also demonstrated stronger interest in insurance. Finally, anticipated LTC needs and predicted dependence in the future increased participants’ likelihood of purchasing insurance.

Discussion

By adopting the Andersen Model, this study has analyzed the intention to buy private LTC insurance plans among both middle-aged and older adults in Hong Kong. A more comprehensive range of factors was included in the quantitative analysis, which revealed a battery of detailed factors underlying Hong Kong Chinese’ decision regarding their LTC financing plan. Overall, one third of participants expressed interest in owning a private LTC insurance policy. Notwithstanding the hypothetical nature of the question put to participants, this figure still reflects a fairly encouraging level of interest and conveys a positive policy signal. Even in the United States and France, where private LTC insurance coverage is the broadest, only 5% and 15%, respectively, of the population aged 40 and over owns a private policy (Colombo & Mercier, 2012).

Statistical analysis has revealed a series of nuanced findings regarding the factors associated with participants’ purchase intention, which could be regrouped into four broad sets of factors. The typical characteristics of Hong Kong middle-aged and older adults who tend to show a demand for LTC insurance include (a) being younger, better educated, of relatively high social status, and financially literate; (b) living with children but being reluctant to become a burden on family; (c) being in a better financial situation and able to afford premiums; (d) anticipation of LTC needs and dependence, and a preference for formal care.

When compared with the international literature, this study has not only reinforced some conclusions drawn in the West but also offers several divergent results. A couple of observations are worth further discussion. First, we did not find a strong effect for bequest motives on the demand for LTC insurance in Hong Kong. One plausible explanation is that individuals with strong bequest motives may prefer to self-insure and do not pay insurance premiums so as to protect bequests, hoping that some of the money set aside for LTC will remain and be reallocated as a bequest (Brown et al., 2012). This explanation can be further supported by the observation that most participants in our sample wished to avoid becoming a financial burden on family because of their LTC bills.

Second, while it is reasonable to hypothesize that people with supportive living arrangements may be in less urgent need of insurance to finance their LTC services, our analysis based in Hong Kong noticed that those who were living with their children actually showed stronger interest in insurance ownership. This could be interpreted with the same logic illustrated above; Hong Kong’s highly competitive society, coupled with the gradual weakening of family ties between older parents and adult children, has reinforced elders’ mentality of self-reliance. To the extent that they have anticipated their LTC needs and find the insurance product reasonably affordable, they are willing to purchase private insurance that helps them avoid becoming a burden on their families.

Third, approximately 40% of the participants were concerned about the affordability of premiums, but, unexpectedly, there was no statistically significant association between premium affordability and LTC insurance demand. This may be because the question about the LTC insurance product was hypothetical in nature while no indicative rate for the premium was given; participants thus may have different reference values in mind.

Fourth, counterintuitively, while only 17% of participants indicated that public assistance would cover their LTC expenditures, this did not appear to be correlated with the intention of insurance purchase. This could be explained by that participants perceived welfare as the last resort and were not aware of the current situation of LTC financing in Hong Kong. This finding echoes many previous studies conducted in the West: The existence of a public safety net that covers LTC costs may substantially crowd out individuals’ plans regarding LTC insurance ownership. The generous subsidization of LTC by the Hong Kong SAR Government essentially has the same effect, by providing needy elders a comfortable last resort.

Concluding Remarks

This study has revealed that multiple factors on both the demand and the supply sides appear to limit the demand for LTC insurance in Hong Kong simultaneously. Most of the limitations identified in the study are resolvable, however. Based on the detailed results revealed above, more educational and promotional activities targeting specific groups of middle-aged and older people could be undertaken to increase their awareness of LTC financing and acceptance of private insurance. Relevant policy instruments, such as subsidies, vouchers, and tax incentives, can be utilized to encourage purchase. In particular, encouraging middle-aged working adults to start financial preparations would certainly be a very effective strategy. The soon-to-be-older will face greater pressure to finance their possible LTC needs in their old age, in part due to the erosion of filial piety, fewer children, and weakening of the family support system. Hence, getting this generation of future elderly financially prepared is a rather pressing task.

With the increase of financial resources from the private sector, LTC service providers would face precious opportunities for development. Supportive government policies could be introduced to foster the development of service providers’ physical and professional capacities. Although this study concentrated primarily on demand-side issues, it has also generated useful implications for the supply side. Individuals’ concerns about appropriate pricing of insurance premiums stood out as a very important consideration affecting their purchase decision. Both government regulators and commercial insurers need to thoroughly assess the demand and price elasticity of multiple insurance products, be they indemnity policies or reimbursement plans.

Despite the value of this study, two limitations affect causality and responses. First, the statistical association presented has not established causality, and hence, it is subject to alternative interpretations. There may be a possibility of reverse causality for some. Second, in the survey instrument, no indicative premium rate was provided to participants as a reference value, which may have affected their revelation of preference. Therefore, the results of this study should be interpreted with caution.

Footnotes

Acknowledgements

The authors are very grateful to two anonymous reviewers and the editor for valuable comments.

Authors’ Note

Ethical approval was obtained from the Human Research Ethics Committee of the Education University of Hong Kong (Ref. 2014-2015-0373).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the Telephone Survey Funding Scheme of the Department of Asian and Policy Studies, The Education University of Hong Kong.

ORCID iD

Alex Jingwei He

References

Andersen

R. M.

(1995). Revisiting the behavioral model and access to medical care: Does it matter? Journal of Health and Social Behavior, 36(1), 1-10.

Barr

(2010). Long-term care: A suitable case for social insurance. Social Policy & Administration, 44, 359-374.

Brown

J. R.

Coe

N. B.

Finkelstein

(2007). Medicaid crowd-out of private long-term care insurance demand: Evidence from the health and retirement survey. Tax Policy and the Economy, 21, 1-34.

Brown

J. R.

Finkelstein

(2009). The private market for long-term care insurance in the United States: A review of the evidence. Journal of Risk and Insurance, 76, 5-29.

Brown

J. R.

Goda

G. S.

McGarry

(2012). Long-term care insurance demand limited by beliefs about needs, concerns about insurers, and care available from family. Health Affairs, 31, 1294-1302.

Caro

F. G.

Porell

F. W.

Kwan

(2011). Expectancies and ownership of long-term care insurance policies among older married couple. Journal of Applied Gerontology, 30, 562-586.

Census and Statistics Department. (2015). Hong Kong Population Projections 2015-2064. Retrieved from http://www.statistics.gov.hk/pub/B1120015062015XXXXB0100.pdf

Chen

Y. P.

(2003). Funding long-term care: Applications of the trade-off principle in both public and private sectors. Journal of Aging and Health, 15, 15-44.

Chou

K. L.

Chow

N. W.

Chi

(2005). A proposal for a voucher system for long-term care in Hong Kong. Journal of Aging & Social Policy, 17(2), 85-106.

10.

Chung

R. Y.

Tin

K. Y.

Cowling

B. J.

Chan

K. P.

Chan

W. M.

S. V.

Leung

G. M.

(2009). Long-term care cost drivers and expenditure projection to 2036 in Hong Kong. BMC Health Services Research, 9(1), Article 172. doi:10.1186/1472-6963-9-172

11.

Colombo

Mercier

(2012). Help wanted? Fair and sustainable financing of long-term care services. Applied Economic Perspectives and Policy, 34, 316-332.

12.

Costa-Font

(2010). Family ties and the crowding out of long-term care insurance. Oxford Review of Economic Policy, 26, 691-712.

13.

Costa-Font

Courbage

(2015). Crowding out of long-term care insurance: Evidence from European expectations data. Health Economics, 24(Suppl. 1), 74-88.

14.

Courtemanche

(2009). Tax incentives and the decision to purchase long-term care insurance. Journal of Public Economics, 93, 296-310.

15.

Elderly Commission. (2017). Elderly services programme plan. Retrieved from http://www.elderlycommission.gov.hk/en/download/library/ESPP_Final_Report_Eng.pdf

16.

Feder

J. M.

Komisar

H. L.

Friedland

R. B.

(2007). Long-term care financing: Policy options for the future. Georgetown University Long-Term Care Financing Project. Retrieved from https://hpi.georgetown.edu/ltc/papers.html

17.

Fornero

Monticone

(2011). Financial literacy and pension plan participation in Italy. Journal of Pension Economics & Finance, 10, 547-564.

18.

Gleckman

(2007, September). The role of private insurance in financing long-term care (Brief #7-13). Center for Retirement Research, Boston College. Retrieved from http://crr.bc.edu/briefs/the-role-of-private-insurance-in-financing-long-term-care/

19.

A. J.

Chou

K. L.

(2017). Long-term care service needs and planning for the future: A study of middle-aged and older adults in Hong Kong. Ageing & Society, 39(2), 221-253.

20.

Hong Kong Consumer Council. (2015). Huge difference among residential care homes charge with various miscellaneous items-choice #467. Retrieved from https://www.consumer.org.hk/ws_en/news/press/elderlyhome_0915.html

21.

Hong Kong Insurance Authority. (2017). Annual long-term business statistics. Retrieved from https://www.ia.org.hk/tc/infocenter/statistics/annual_long_term_business_statistics.html

22.

Hong Kong Trade Development Council. (2017). Insurance industry in Hong Kong. Retrieved from http://hong-kong-economy-research.hktdc.com/business-news/article/Hong-Kong-Industry-Profiles/Insurance-Industry-in-Hong-Kong/hkip/en/1/1X000000/1X003UWM.htm

23.

Lin

Prince

J. T.

(2016). Determinants of private long-term care insurance purchase in response to the partnership program. Health Services Research, 51, 687-703.

24.

Lockwood

L. M.

(2012). Bequest motives and the annuity puzzle. Review of Economic Dynamics, 15, 226-243.

25.

Lum

T. Y. S.

Yan

E. C. W.

A. H. Y.

Shum

M. H. Y.

Wong

G. H. Y.

Lau

M. M. Y.

Wang

(2016). Measuring filial piety in the 21st century: Development, factor structure, and reliability of the 10-item Contemporary Filial Piety Scale. Journal of Applied Gerontology, 35, 1235-1247.

26.

McGarry

Temkin-Greener

(2014). Role of race and ethnicity in private long-term care insurance ownership. The Gerontologist, 54, 1001-1012.

27.

Robison

Shugrue

Fortinsky

R. H.

Gruman

(2014). Long-term supports and services planning for the future: Implications from a statewide survey of baby boomers and older adults. The Gerontologist, 54, 297-313.

28.

Sloan

F. A.

Norton

E. C.

(1997). Adverse selection, bequests, crowding out, and private demand for insurance: Evidence from the long-term care insurance market. Journal of Risk and Uncertainty, 15, 201-219.

29.

Social Welfare Department. (2016). Waiting list for subsidized residential care services for the elderly. Retrieved from https://www.swd.gov.hk/doc/elderly/ercs2/LTC%20statistics%20HP-Eng(201607).pdf

30.

Tennyson

Yang

H. K.

(2014). The role of life experience in long-term care insurance decisions. Journal of Economic Psychology, 42, 175-188.

31.

United Nations Department of Economic and Social Affairs. (2015). World population prospects (2015 revision). Retrieved from https://esa.un.org/unpd/wpp/publications/files/key_findings_wpp_2015.pdf

32.

Wiener

J. M.

(2010). Long-term care: Getting on the agenda and knowing what to propose. Medical Care Research and Review, 67(4 Suppl.), 126S-140S.