Nonprofit and for-profit providers in Japan's at-home care industry: evidence on quality of service and household choice

Tags: Japan, nonprofits, nonprofit providers, hypothesis, quality of service, insurance scheme, Japanese government, Care Receiver, Government of Japan, Annual Household Income, Haruko Noguchi, annual income, Satoshi Shimizutani, Cabinet Office, Care Level, natural logarithm, government, care for the elderly, service providers, number of households, provider change, households, Xi Zi, nonprofit provider, Economic and Social Research Institute, Price Policy Division, Noguchi, Economic and Social Research Institute Discussion Papers, Mean Std, Public Reputation, nonprofit services, Total Household Assets, Relationship, Care Manager, Acquaintance, services provided, Qn, COE Program Institute of Economic Research Hitotsubashi University, Hitotsubashi University Research Unit, nonprofit sector, quality services, Qp, Hitotsubashi University
Content: Hi-Stat Discussion Paper Series No.73 Nonprofit and For-profit Providers in Japan's At-home Care Industry: Evidence on Quality of Service and Household Choice Haruko Noguchi and Satoshi Shimizutani February 2005 Hitotsubashi University Research Unit for Statistical Analysis in social sciences A 21st-Century COE Program Institute of Economic Research Hitotsubashi University Kunitachi, Tokyo, 186-8603 Japan http://hi-stat.ier.hit-u.ac.jp/
Nonprofit and For-profit Providers in Japan's At-home Care Industry: Evidence on Quality of Service and Household Choice By Haruko Noguchi and Satoshi Shimizutani1 Abstract In 2000, government deregulation along with the introduction of the long-term insurance scheme for the first time allowed for-profit providers of at-home care for the elderly to compete directly with nonprofit operators. According to the contract failure hypothesis, we would expect consumers to prefer nonprofit providers over their for-profit counterparts as a result of information asymmetry and non-distributional constraints. This study takes advantage of household level data to examine whether households' choice of care provider is biased toward nonprofits. We find that nonprofit providers to command a larger market share, but this is at least partly explained by having operated in the market longer and by continuing restrictions in medical and institutional care that confer various advantages on nonprofit providers. However, we do find that user with better knowledge of providers tend to favor for-profit providers, suggesting that measures to reduce information asymmetries may help to provide a more level playing field. 1 This is a revised version of our earlier paper (Noguchi and Shimizutani 2002). We would like to thank seminar participants at ESRI, Ralph Paprzycki especially the anonymous referees for their useful comments. We also express our appreciation for the permission to use the micro-level data collected by the Price Policy Division and ESRI. The views expressed in this paper do not necessarily represent those of the Economic and Social Research Institute or of the Japanese government. All errors are our own. Please address correspondence to: Haruko Noguchi, Faculty of Social Science, Toyo-Eiwa University (Address: 32 Miho-cho, Midori-ku, Yokohama, Kanagawa, 226-0015, Japan; Email: [email protected]) and Satoshi Shimizutani (Corresponding author), Institute of Economic Research, Hitotsubashi University (Address: 2-1, Naka, Kunitachi-shi, Tokyo, Japan 186-8603; Email: [email protected]).
1. Introduction Do consumers prefer nonprofit or for-profit services? This classical and frequently debated issue is of considerable interest both for academics and policy-makers. It became of very practical interest in Japan following the introduction of public long-term care insurance and deregulation of the market for at-home care for the elderly in the spring of 2000 (Mitchell, Piggott and Shimizutani 2004). While for-profit operators had been allowed to offer their services previously, users were not reimbursed, meaning that only the wealthy were able to afford them and the market was accordingly small. Everyone else was assigned to nonprofit providers as part of the government welfare policy. The introduction of the long-term insurance scheme for the first time pitched for-profit and nonprofit providers head to head, with users free to choose and having to make copayments of 10 percent of the cost, no matter what type of provider they chose (as long as it is government-approved). The government's policy was based on the expectation that the entry of for-profits would contribute toward making the market more competitive. However, it should be noted that this outcome is not a foregone conclusion. If consumers' preferences are biased toward nonprofit services, market participation of for-profits will have little effect on competition. One possible explanation of this potential outcome is the "contract failure" hypothesis (Hansmann 1980): If consumers cannot observe the quality of services in advance, they may prefer to choose nonprofit providers. It may be perfectly rational for them to do so because of "non-distributional constraints," i.e. that fact that nonprofit operators are not allowed to distribute any operating surpluses among their owners and hence face fewer incentives to raise such surpluses by lowering the quality of their service. In contrast, for-profits, by their very nature, may be inclined to raise profits by lowering the quality of their service. This risk of opportunistic behavior by for-profits provided the justification for the
traditional solution in Japan to allow only nonprofit enterprises to operate in the market. However, in the face of a rapidly aging population, the Japanese government was forced to change the entry policy in order to overcome the bottleneck in the supply of care for the elderly. Almost five years have passed since the introduction of the new entry policy, and wide-ranging reform of the public insurance scheme for the elderly is to be implemented in the spring of 2006. In spite of this innovative institutional change, there has been little research on the determinants of Japanese consumers choices regarding nonprofit and for-profit at-home care providers. The Japanese at-home long-term care market is unusually suitable for testing the "contract failure" hypothesis because government regulation stipulates that output prices must be the same for all types of providers. Service contents are also heavily regulated and standardized across providers. Service contents are also heavily regulated and standardized across providers, meaning that consumers' choice should reflect preferences regarding nonprofit and for-profit providers free of distortions that price or product differentiation would introduce. At the same time, this of course prevents us from considering the effect of pricing behavior of different types of providers and any effect this would have on consumer choice. This study takes advantage of a unique micro-level data set to test consumer preferences with regard to the different types of providers. First, we address whether the quality of services provided by nonprofit organizations is better than that of profit-seeking organizations. Several earlier studies, such as Shimizutani and Suzuki (2002) and Suzuki (2002), have examined quality differences between service providers based on suppliers' data. However, to our best knowledge, there has been little research examining the hypothesis based on demand-side data. Second, we examine the determinants of households' choice between different types of service providers. To explore these issues, micro-data from
recipients is utilized to examine the hypothesis because information asymmetry biases care users' preferences toward nonprofits, regardless of the quality of service. Concretely, we examine households' choice of provider along with an array of variables influencing the selection of type of at-home care provider. The paper proceeds as follows. Section 2 provides an overview of previous studies, most of which have been conducted in the United States. Section 3 describes the data set used in this study. Section 4 presents the specification of the model used for the empirical analysis and discusses the Estimation results. Section 5 draws policy conclusions from our empirical findings. 2. Review of Previous Studies In contrast with the situation in Japan, there are plenty of studies in Western countries that analyze consumer behavior with regard to nonprofit and for-profit organizations. The starting point of any discussion of this topic is usually the "contract failure" approach proposed by Hansmann (1980), on which most subsequent studies are based. Hansmann's discussion begins with two key concepts: asymmetry of information and non-distributional constraints. As highlighted by Akerlof (1970), quality uncertainty is inevitably due to the asymmetry of information between sellers and buyers. Under this asymmetry, consumers are not able to observe the quality of products in advance. This is especially crucial in the service sector. If this is the case, which type of provider do consumers choose to purchase services from? Suppose there are two types of providers: for-profits and nonprofits. By definition, the nonprofit sector is prohibited from paying out profits to its owners. Hence, retained profits are utilized to provide higher wages for staff, contributing to a higher quality of service. In
contrast, for-profits need to make a profit, which is distributed to stockholders, inviting opportunistic behavior to decrease costs by lowering the quality of services. This mechanism motivates consumers to choose to purchase services from nonprofits rather than from for-profits. Consumers choose nonprofit services because they believe that nonprofit suppliers provide higher quality services, though this perception may not necessarily be related to the actual quality of service delivered by either type of provider.2 Apart from the contract failure hypothesis, another explanation as to why consumers might prefer nonprofits over for-profits is that the different types of providers behave differently. While managers of for-profits are assumed to maximize profits only, managers of nonprofits may derive greater utility from improving the quality of services as well as increasing profits. Newhouse (1970) argues that managers in the not-for-profit sector have fewer incentives to lower the quality of services provided, since their performance is judged by quality, or they desire to show professional excellence or technical virtuosity by stressing quality. While some studies proceed by modeling the contract failure hypothesis (Easley and O'Hara (1983, 1988), Chillemi and Gui (1991), Hirth (1999)), there are relatively few empirical studies that directly examine consumer choice with regard to different types of providers. One exception is the study by Holtmann and Ullmann (1993), and in what follows we use their relatively simple framework to analyze the choice problem faced by consumers. We assume that a consumer purchases services from nonprofits (Qn) and for-profits (Qp). Since the consumer cannot judge the quality of services in advance, the quality is uncertain. However, for simplicity, we assume that the quality of nonprofit services is certain, but this is 2 Noguchi and Shimizutani (2002) confirm that there is a nonprofit wage premium in the Japanese elderly care market. Although this is interpreted to imply higher quality, the belief that higher costs imply higher quality is not always well founded. Reduced incentives for nonprofits to enforce contracts and develop internal controls may result in economically inefficient organization. In this case, higher costs do not necessarily mean better quality (Jensen and Meckling (1976)). Shimizutani and Suzuki (2002) show that, indeed, the quality of services offered by nonprofits is not always higher than that offered by for-profits.
not the case with for-profits. Thus, we define the actual unit of for-profit services as Q*p=zQp where z is a random variable with E(z)=1 . A consumer maximizes his utility EU(Qn, Q*p) under the budget constraints pnQn+ppQp=Y and Qn 0 and Qp 0. The utility function is assumed to be increasing and concave in Qn, Qp, z. We set up the Lagrangian as follows, with as the Lagrange multiplier:
L= EU(Qn, Q*p)+ (YpnQnppQp)Qn 0 and Qp 0.
(1)
The first order conditions are:
L/ Qn=E( U/ Qn) pn 0
( L/ Qn) Qn=0 L/ Qp=E[( U/ Q*p)z] pp 0
( L/ Qp) Qp=0 L/ = YpnQnppQp =0
Therefore, If Qp=0 and Qn>0, E( U/ Qn)/ pn> E[( U/ Q*p)z]/ pp If Qp>0 and Qn>0, E( U/ Qn)/ pn= E[( U/ Q*p)z]/ pp If Qp>0 and Qn=0, E( U/ Qn)/ pn< E[( U/ Q*p)z]/ pp
Given that government regulation stipulates uniform prices in the Japanese long-term
care market (pp= pn), the choice between nonprofit and for-profit providers depends on the expected utility derived from using nonprofit providers, E( U/ Qn), and for-profit providers, E[( U/ Q*p)z]. We should note that E[( U/ Q*p) z]= z E[ U/ Qp]+cov(z, U/ Qp), where, by the concavity assumption, the second term on the right-hand side is negative.
Holtmann and Ullmann (1993) suggest that this second term can be viewed as the
marginal risk premium associated with for-profit care. In this case, uncertainty regarding the
quality of services provided by the for-profit sector diminishes the expected utility and thus
motivates consumers to choose nonprofit services. In other words, even if the average quality of for-profits and nonprofits is the same, the left-hand side is always greater than the right if the amount purchased is the same.3 Thus, we must address two issues. The first is to determine which type of provider offers a higher quality of services in practice. Earlier studies on this topic for the case of Japan have been inconclusive (see Shimizutani and Suzuki (2002) for a survey). We can then turn to the second issue: which type of provider is preferred by consumers. We examine these questions in the next section by taking advantage of our micro-level data set. 3. Description of the Data The data set used in this study is based on the micro-level data from the "Survey on Long-term Care Users." This survey was performed three times, in 2001, 2002 and 2003. The first survey was conducted from November to December 2001 by the Price Policy Division of the Cabinet Office. The households surveyed were chosen on the basis of a survey carried out by a private research company that randomly selects participants based on Japan's household registration system. The distribution of the "Survey on Long-term Care Users" resembles that of the Population Census. In the Survey, a "care receiver" is defined as an elderly person receiving care in order to cope with daily life activities. The survey only includes those living at home with family members, i.e. those living in institutions are excluded. Moreover, the survey includes both those elderly whose care is paid for by the public insurance scheme (for which approval by the local government is required) and those paying for at-home care out of their own pockets. 3 As indicated, in order to keep the "contract failure" model simple, we assume that the quality of services provided by for-profits is uncertain while that of nonprofits is certain. However, as the anonymous referees pointed out, contract failure of course is also possible in the case of nonprofits. Therefore, there remains the question of how effective nonprofit governing boards are in looking after the interests of service receivers and the public. However, addressing this topic is beyond the scope of this short paper.
From all the observations of the "Survey of Long-term Care Users", we randomly selected households with care receivers and, among those, chose 1,300 households with only one receiver4. The research company mailed our questionnaires to these households and received responses from 1,005 households (for a response rate of 77.1 percent). The second "Survey on Long-term Care Users" was conducted by the same private research firm in October and November 2002, but this time under the aegis of the Economic and Social Research Institute. The notable merit of this survey is that it performed a similar survey on the same households as in the 2001 survey. Of the 1,005 households in the first survey, we obtained responses from 822. In 617 of these households, the elderly care receiver still lived at home with family members and received at-home care.5 In addition, we selected new households with one care receiver and added 457 households to the sample. Thus, the sample size for 2002 is 1,074 households. The third survey was also implemented by the Economic and Social Research Institute, in December 2003. The data were obtained following the same methodology as in 2002. Of the households surveyed both in 2001 and 2002, we mailed the questionnaire to 544 and obtained responses from 381. Of those first surveyed in 2002, we sent the questionnaire to 432 households and received replies from 251. In addition, we mailed the survey to 423 households which were newly chosen in 2003 and had responses from 349. Thus, the sample size for 2003 is 981 households. The respondents to the survey were the main caregivers in the household. The questionnaire covers a variety of items to establish patterns of care use over a period of 4 The screening results show that households with one care receiver account for 71.1 percent of the total, those with two care receivers account for 26.9 percent, and the remainder are households with more than two care receivers. Our survey sample is limited to those households with one care receiver since a household with two or more care receivers is very different from those with one care receiver in that the care burden is more onerous and care is possibly provided by two or more caregivers, which makes it hard to identify who cares for whom. 5 Of the 205 elderly care receivers who no longer lived at home with family members, 75 entered an unspecified type of institution, 68 died, and 29 were hospitalized.
several years, the health condition of caregivers and receivers, and household demographics. Table 1 provides summary statistics of our dataset. We pool the 2001, 2002 and 2003 survey in order to obtain a large sample size of 479 observations. Note that the survey asks respondents to indicate the type of provider they rely on for at-home care services. This is the segment on which most for-profits concentrate. The number of households in our sample that use for-profit at-home care services is 151 (31.5 percent), while 328 (68.5%) use nonprofit providers, implying that nonprofits have still a larger marker share even after the deregulation. Comparing household characteristics, we find that, users of for-profits have, on average, a higher annual income and greater total assets. Moreover, care levels are typically higher for users of nonprofit providers. We should note that the share of those who have changed providers is much higher among users of for-profits than among of nonprofits. One possible explanation of changes of provider is that households become familiar with a provider, are dissatisfied, and want to try another provider. But it should be noted that in the survey, the dominant case of provider change is from nonprofit to for-profit operators, suggesting that households with more knowledge about the quality of care tend to switch from nonprofit to for-profit providers. Finally, as was to be expected given that the market for at-home care was deregulated only in 2000, the share of those who began using their provider before that year is much higher among users of nonprofits than for-profits. Although the survey does not contain any information from suppliers on the quality of services, we do have results on care users' subjective evaluation of the care they receive. These results help us to address the first issue referred to in the previous section, namely: which type of provider offers higher quality services? We asked respondents to rate the service of their provider ("Excellent", "Above average", "Below average", "Very bad", and "Don't know") and divided the results by type of provider. The results are shown in Figure 1 and yield some interesting conclusions. Generally, the dominant answer is "above average,"
and this share is higher for nonprofits. In 2001 and 2002, the quality of for-profit providers was perceived to be marginally worse than that of nonprofit providers, though uncertainty was certainly much higher regarding the quality of for-profit providers.. But we observe a drastic change in 2003. The share of users considering the service of their provider to be "excellent" for the first time was higher for for-profits than for nonprofits, while at the same time, more users of nonprofits judged the service of their provider to be "below" average than user of for-profits. Thus, users' perception of the quality of care provided by for-profits has clearly improved, while the reverse seems to be the case for nonprofits. Figure 2 shows the reasons that motivated users' choice of a nonprofit or for-profit provider." The dominant determinant is the recommendation made by the care manager responsible for devising an individual's care plan once eligibility for insurance-funded care has been determined. A recommendation by an acquaintance played a more important role in the choice of for-profits than of nonprofits; public reputation also played a more important role in the choice of a for-profit provider than a nonprofit provider in 2001 and 2002, though this had reversed by 2003. The share of those who began using their provider before 2000 has decreased. The share of those who chose nonprofits due to anxiety over using for-profit providers is very small, suggesting that there is little apprehension that for-profit providers engage in opportunistic behavior. These results indicate that users do not perceive the quality of service delivered by either type of provider to be evidently better. In sum, according to respondents' perception, the quality of care services provided by for-profits is not necessarily worse, and the perception of these providers in fact has become more favorable. There seems to be little prejudice against profit-seeking behavior that might cause users to hesitate in using the services offered by for-profit providers. This brings us to the second issue, to be addressed in the next section: which type of provider is preferred by consumers.
4. Estimation and Results
Based on the discussion in Section 2, we employ the following specification to consider
the choice users make: Zi*= Xi + ui Zi =1 if Zi* 0 or - ui Xi Zi =0 if Zi* < 0 or - ui > Xi where Zi* is the likelihood of a household to choose for-profit providers, which is
unobservable. However, we can observe Zi, which is a dichotomous variable that takes 1 if a user actually chooses a for-profit facility. Xi refers to users' characteristics that might affect choice of provider and includes gender, age, care level, annual income, assets, the relationship
between care giver and receiver, and a dummy variable for using the same provider since before 2000.6 is the tendency of a household to choose for-profits. Xi also contains several
variables to proxy the asymmetry of information. It includes the number of family members,
since more members can provide more informal care and monitor outside caregivers more
intensively. It also includes whether a member of the immediate family or a relative is a
doctor or professional caregiver, since it can be assumed that such households have more
information on the quality of care. And, naturally, any households that have changed
providers are also expected to have more information on providers. ui is an error term
following
N
(0,
2 u
)
.
Table 2 shows the estimation results. The first column reports the results for the
random-effects probit model. The coefficients on household income and assets are positive
6 In the survey, respondents were provided with a choice of annual income and asset brackets which they were asked to tick. We exclude any households falling into lowest categories (less than 2 million yen annual income and 3 million yen worth of assets) and the highest (more than 20 million yen annual income and 15 million yen worth of assets).
but not statistically significant. Households where the elderly patient requires a lower care level are more inclined to choose for-profit providers. Regarding family size, larger households tend choose nonprofits rather than for-profits. Households with an acquaintance who is a medical doctor or a professional caregiver tend to choose nonprofits. Households that have changed providers are more inclined to choose for-profits. Moreover, households that began using their provider before 2000 tend to be with nonprofits. The estimation using the fixed-effects probit model confirms these results. The significant coefficient on the dummy for changing providers indicates that households with better knowledge of suppliers tend to choose for-profits, which is consistent with our discussion above. "On the other hand, the significant negative coefficient on the dummy indicating whether any acquaintance is a medical doctor or professional caregiver may be explained by the fact such individuals typically work in the provision of medical or institutional care, a field in which at present only nonprofit organizations are allowed to operate, and it can be assumed that they are more likely to recommend the same type of provider. This result is also consistent with the fact that households with elderly patients requiring higher care-levels are inclined to choose nonprofits, since nonprofit providers usually offer both institutional care and at-home care and users may be reassured to know institutional care services are available should they become necessary as a result of an accident or emergency. Finally, we found that nonprofits enjoyed an "advantage" over for-profits by virtue of having a longer history of operating in the market. In sum, the Japanese long-term care market does not seem to conform to the situation predicted by the contract failure hypothesis. Rather, two other factors seem to bias user choice towards nonprofits: First, the fact that they have operated in the market since before deregulation in 2000; and second, the fact that they are also able to offer medical and/or institutional care services, which for profit operators at present are still banned from doing.
5. Conclusion This study takes advantage of original survey data to examine whether households' choice of at-home care service providers is biased toward nonprofit operators, as the contract failure hypothesis predicts. Our empirical findings can be summarized as follows. In our survey, for-profit providers account for about 40 percent of at-home care, suggesting that for-profit operators have met with acceptance following deregulation of market entry along with the introduction of the public long-term care insurance scheme. In this sense, for-profits have been well received by those who need them following deregulation and the introduction of the public long-term care insurance. As regards preferences with respect to nonprofit and for-profit providers, households with an elderly patient requiring higher care levels or with an acquaintance who is a medical doctor or professional caregiver are more inclined to choose nonprofits. Both results are associated with the fact that, as incumbents, only nonprofits offer medical or institutional care. In addition, nonprofits enjoy an advantage by virtue of being longer in the market. In this sense, proprietary providers are disadvantaged. On the other hand, households with better knowledge of providers tend to choose for-profits, which suggests that reducing information asymmetry of may help to create a more level playing field. Before the introduction of the long-term care insurance scheme in 2000, the private provision of health care services was generally viewed with suspicion, based on the fear that profit-orientation would lead to opportunistic behavior. However, the growth and success of the for-profit sector have allayed such fear and this might be an opportune moment to reassess the role and status of the nonprofit sector since equal market access has now be achieved and for-profits can compete on an equal footing with nonprofits. Traditionally, nonprofits use operating surpluses to cover deficits, grow or reduce bottlenecks, provide Community Services
such as charitable care, research, education. (Schlesinger et al.(1996)). With prices and contents of services for at-home care regulated by the government, the role of charitable organizations in this industry should be reconsidered. In this context, one important issue that would need to be scrutinized is the tax status of nonprofit organizations. Like in the U.S., as entities serving the public interest, nonprofits in Japan enjoy tax exemptions. However, recent research from the U.S. suggests that in order to justify this special treatment, the social benefit provided by nonprofits should be at least equal to the benefits provided by profit-seeking firm (community benefits and taxes paid) plus the profit these firms generate for their owners (Claxton et al. 1997; Nicholson et al. 2000). Future research should examine these issues in Japan in order to allow an assessment of the optimal division of labor between different types of providers and the tax treatment of nonprofit operators in order achieve this. Such research could also help to inform policies regarding other markets where nonprofits are still dominant.
References Akerlof, George A. (1970). "The Market for `Lemons': Quality Uncertainty and the Market Mechanism." Quarterly Journal of Economics, vol. 84(3), pp.488-500. Chillemi, Ottorino and Gui, Benedetto (1991). "Uninformed Customers and Nonprofit Organization: Modeling `Contract Failure' Theory." Economics Letters, vol. 35(1), pp.5-8. Cabinet Office (2002). "Kaigo Service Shijo no Isso no Koritsuka no tame ni [Report on the Long-term Care Market]" (in Japanese). Government of Japan; available online: . Claxton, Gary, Feder, Judith, Shactman, David and Altman, Stuart (1997). "public policy Issues in Nonprofit Conversions: An Overview," Health Affairs, vol. 16(2), pp. 9-28. Easley, David and O'Hara, Maureen (1983). "The Economic Role of the Nonprofit Firm." Bell Journal of Economics, vol. 14(2), pp.531-538. ­­­ and ­­­ (1988). "Contracts and Asymmetric Information in the Theory of the Firm." Journal of Economic Behavior and Organization, vol. 9(3), pp.229-246. Gertler, Paul (1989). "Subsidies, Quality and the Regulation of Nursing Homes." Journal of Public Economics, vol. 38(1), pp.33-52. ­­­ (1992). "Medicaid and the Cost of Improving Access to Nursing Home Care." Review of Economics and Statistics, vol. 74(2), pp.338-345. Hansmann, Henry (1980). "The Role of Nonprofit Enterprise." Yale Law Journal, vol. 89(5), pp.835-901. Holtmann, A.G., and Ullmann, S.G. (1993). "transaction costs, Uncertainty, and Not-for Profit Organizations: The Case of Nursing Homes." in A. Ben-Ner and B. Gui,
editors, The Nonprofit Sector in the Mixed Economy, Ann Arbor, University of Michigan Press. Jensen, Michael and Meckling, William (1976). "Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure," Journal of Financial Economics, vol. 3(4), pp. 305-360. Mitchell, Olivia, Piggott, John and Shimizutani, Satoshi (2004). "Aged Care Support in Japan: Perspectives and Challenges," NBER Working Paper no.10882. Newhouse, Joseph P. (1970). "Toward a Theory of Nonprofit Institutions: An Economic Model of a Hospital," American Economic Review, vol. 60 (1), pp.64-74. Nicholson, Sean, Pauly, Mark, Burns, Lawton, Baumritter, Agnieshka and Asch, David (2000). "Measuring Community Benefits Provided by For-profit and Nonprofit Hospitals," Health Affairs, vol.19 (6), pp.168-177. Noguchi, Haruko and Shimizutani, Satoshi (2002). "Earnings and Quality Differentials in For-Profit versus Nonprofit Long-Term Care: Evidence from Japan's Long-Term Care Market," Economic and Social Research Institute Discussion Papers No.17, Cabinet Office, Government of Japan. http://www.esri.go.jp/en/archive/e_dis/abstract/e_dis017-e.html ­­­ (2003). "Is Non-Profit Status a Signal of Better Quality? Micro-Level Evidence from Japan's At-home Care Industry," Economic and Social Research Institute Discussion Papers No.80, Cabinet Office, Government of Japan. http://www.esri.go.jp/en/archive/e_dis/abstract/e_dis018-e.html Schlesinger, Mark, Gary, Bradford, and Bradley, Elizabeth (1996). "Charity and Community: The Role of Nonprofit Ownership in a Managed Health Care System," Journal of Health Politics, Policy and Law, vol. 21(4), pp. 697-751. Shimizutani, Satoshi and Noguchi, Haruko (2004). Kaigo hoiku service shijo no keizai
bunseki --- micro data ni yoru jittai kaimei to seisaku teigen [An Economic Analysis of Care for Children and the Elderly in Japan: A Micro-level Investigation and Consideration of policy implications] (in Japanese), Toko-Keizai Shimpo Sha, Tokyo. June 2004. Shimizutani, Satoshi and Suzuki, Wataru (2002). "The Quality and Efficiency of At-Home Long-term Care in Japan: Evidence from Micro-level Data." Economic and Social Research Institute Discussion Papers No.18, Cabinet Office, Government of Japan. Suzuki, Wataru (2002). "Hi-eiri homon kaigo gyosha wa yuri ka? [Are Nonprofit At-Home Long-term Care Providers Advantageous? (sic)], " (in Japanese) Kikan Shakai Hosho Kenkyu [Quarterly of social security Research (sic)], vol. 38 (1), pp.74-88. National Institute of Population and Social Security Research. Weisbrod, Burton A. (1988). The Non-Profit Economy, Cambridge, Harvard University Press.
Figure 1 Subjective Evaluation of the Quality of Services
2003 Survey (%)
70 Nonprofits 60
50
40
For-profits
30
20
10
0 Excellent Above Average Below Average Very Bad
Don't Know
2002 Survey (%) 70 Nonprofits 60 50 For-profits 40 30 20 10 0 Excellent Above Average Below Average Very Bad
Don't Know
2001 Survey (%) 70 Nonprofits 60 50 For-profits 40 30 20 10 0 Excellent Above Average Below Average Very Bad
Don't Know
Figure 2 Reasons to Choose the Provider 2003 Survey Recommended by Care Manager Recommended by Acquaintance(s)
Nonprofits
For-profits
Public Reputation
Reassured to know that provider is able to provide medical services in case of illness 0
10 20
30
40
50
60
(%)
2002 Survey
Recommended by Care Manager
Recommended by Acquaintance(s)
Public Reputation
Nonprofits
For-profits
Reassured to know that provider is able to provide medical services in case of illness 0
10 20 30 40 50 60
2001 Survey
Recommended by Care Manager
Recommended by Acquaintance(s) Public Reputation
Nonprofits
For-profits
Reassured to know that provider is able to provide medical services in case of illness 0
10
20
30
40
50
60
Table 1 Descriptive Statistics
Total
Nonprofit Users
For-profit Users
No. of Obs. Mean
Std. Dev. No. of Obs. Mean
Std. Dev. No. of Obs. Mean
Std. Dev.
Dummy for For-profits
479
0.315
0.465
328
0
0
151
1
0
Annual Household Income (ten thousand yen) 446 733.408 418.135 307 711.889 408.425 139 780.935 436.571
Annual Household Income (natural logarithm) 446
6.432
0.592
307
6.401
0.594
139
6.501
0.582
Total Household Assets (ten thousand yen)
422 4,618.246 4,201.091 289 4,247.924 3,922.566 133 5,422.932 4,665.498
Total Household Assets (natural logarithm)
422
7.950
1.097
289
7.880
1.070
133
8.104
1.144
Care Level (support)
479
0.052
0.223
328
0.043
0.202
151
0.073
0.261
Care Level 1
479
0.207
0.405
328
0.186
0.390
151
0.252
0.435
Care Level 2
479
0.196
0.398
328
0.189
0.392
151
0.212
0.410
Care Level 3
479
0.154
0.362
328
0.143
0.351
151
0.179
0.384
Care Level 4
479
0.146
0.354
328
0.171
0.377
151
0.093
0.291
Care Level 5
479
0.165
0.372
328
0.186
0.390
151
0.119
0.325
Relationship (spouse=1)
479
0.102
0.303
328
0.122
0.328
151
0.060
0.238
Relationship (own parent=1)
479
0.428
0.495
328
0.412
0.493
151
0.464
0.500
Relationship (spouse's parent=1)
479
0.428
0.495
328
0.427
0.495
151
0.430
0.497
Care Receiver's Sex (female=1)
469
0.738
0.440
323
0.728
0.446
146
0.760
0.428
Care Receiver's Age
470 84.038 7.226
324 83.701 7.252
146
84.788
7.136
Number of Family Members
472
3.964
1.447
324
4.019
1.421
148
3.845
1.502
Acquaintances (Doctor or Professional Caregiver)
479
0.336
0.473
328
0.369
0.483
151
0.265
0.443
Dummy for Changing Providers
479
0.123
0.329
328
0.088
0.284
151
0.199
0.400
Dummy for Use Before 2000
475
0.221
0.415
328
0.271
0.445
147
0.109
0.313
Notes: The care levels of individual care receivers are determined by local governments and classified into six categories (support needed and 1-5) .
"Relationship" refers to relationship of the care receiver to the caregiver. "Dummy for Changing Providers" takes 1 for those households that
have changed their providers. "Dummy for Use Before 2000" takes 1 for those households that began using the current provider before 2000.
Table 2 Estimation Results
Dependent Variable :
Random Effects Model
For-profits=1 and nonprofits=0
Coefficient Std. Dev.
Annual Household Income (natural logarithm 0.233
0.153
Total Household Assets (natural logarithm) 0.037
0.074
Care Level (support)
0.860
0.435 *
Care Level 1
0.747
0.338 *
Care Level 2
0.457
0.336
Care Level 3
0.499
0.343
Care Level 4
-0.026
0.368
Care Level 5
-0.011
0.349
Relationship (spouse=1)
4.854
1.373 ***
Relationship (own parent=1)
5.500
1.478 ***
Relationship (spouse's parent=1)
5.547
1.497 ***
Care Receiver's Sex (female=1)
-0.133
0.187
Care Receiver's Age
0.015
0.011
Number of Family Members
-0.099
0.058 *
Acquaintances (Doctor or Professional Caregiver)
-0.418
0.154 ***
Dummy for Changing Providers
0.558
0.203 ***
Dummy for Use Before 2000
-0.612
0.195 ***
Constant
-8.779
0.000
Fixed Effects Model
Coefficient Std. Dev. Marginal Effect
0.234
0.153
0.077
0.037
0.074
0.012
0.843
0.437 * 0.317
0.733
0.339 * 0.264
0.446
0.337
0.157
0.490
0.343
0.175
-0.032
0.368
-0.011
-0.027
0.351
-0.009
4.888
1.378 *** 0.869
5.535
1.483 *** 0.991
5.586
1.501 *** 0.994
-0.133
0.187
-0.044
0.016
0.011
0.005
-0.100
0.058 * -0.033
-0.415
0.154 *** -0.130
0.565
0.203 *** 0.205
-0.605
0.196 *** -0.174
-8.849
0.000
Hausman test statistics
-55.63
-55.88
Number of Observations.
401
401
Log Likelihood
-217.10721
-217.006
Pseudo R squared
0.133

Note: ***, **, and * indicate statistical significance at the 1%, 5%, and 10%-level, respectively.

File: nonprofit-and-for-profit-providers-in-japans-at-home-care-industry.pdf
Published: Tue May 1 11:07:57 2018
Pages: 23
File size: 0.14 Mb


All sleepless and light, 6 pages, 0.34 Mb

, pages, 0 Mb

, pages, 0 Mb

Escaped Alone, 11 pages, 0.31 Mb

, pages, 0 Mb
Copyright © 2018 doc.uments.com