Willingness to pay for health insurance among rural and poor persons: Field evidence from seven micro health insurance units in India
Introduction
In view of the high out-of-pocket spending level for healthcare, and the desire to improve the effectiveness and equality of healthcare financing and the quality of the care given, policy-makers in India have turned their attention to proposals for health insurance for the poor [1]. At central government level, the National Rural Health Mission put forward proposals to support community-based health insurance by subsidizing the premium of the poor [2]. And at state level, several governments have signed agreements with commercial insurers to cover certain segments of their population for certain cost-generating illnesses.1 These recent developments occur on the backdrop of very low penetration of health insurance in India in general, and in rural areas or among people at the “Bottom-of-the-Pyramid” (BoP) in particular.
We assume that health insurance is likely to develop on a voluntary basis in India, with clients having to pay a premium. Hence, it is essential to obtain reliable information on the amounts that potential clients would be willing to pay, and the major determinants influencing this choice. For many clients – including BoP groups – health insurance is likely to develop through grassroots organizations such as “micro health insurance units” (MIUs).2 Thus, the purpose of this study is to add knowledge on the maximal WTP among rural and BoP persons in India, and identify the major determinants influencing their choice.
This study is based on evidence gathered in 2005 in seven locations where MIUs are in operation. The seven locations are situated in four Indian states: Tamil Nadu, Karnataka, Maharashtra and Bihar.3 The household (HH) survey that provided the dataset we analyze here formed part of the EU-funded project “Strengthening MIUs for the Poor in India”. To the best of our knowledge this is the first survey relating to several locations where MIUs operate, enabling us to conduct a comparative analysis between an insured cohort to an uninsured cohort and across locations.
Previous sources of information on WTP for health insurance in India include a few articles. Mathiyazhagan [3] reports the results of a survey conducted in 1998 in Karnataka where the average WTP of replies of 1000 rural HHs for a fictitious health insurance package was INR 163.48 per HH per year, with only 8% willing to pay between INR 481 and INR 600. A second source is a baseline study conducted in 2001 by Karuna Trust in one district in south Karnataka (near Mysore) and in one district in the north of the state (near Belgaum) prior to the launch of insurance operations; the average WTP per HH per year was, respectively, INR 111 and INR 290 [4]. Based on this survey, Karuna Trust fixed the premium at INR 150 for a HH of five per year. Both these surveys posed an open-ended question, which does not give respondents an anchor for the choice of WTP. A study in Delhi by Gupta [5] found an average WTP of INR 220 for adults and INR 93 for children. Finally, another scheme, Uplift Health (located in Pune, Maharashtra) conducted focus group discussions in 2002 to inquire on the WTP of prospective members, which led it to establish a premium of INR 50 per person per year (applied uniformly to all ages and both sexes) [6], [7].
These findings, reported in previous studies, followed different elicitation methods, were based on smaller total samples, and were too dated to serve as the basis for an updated estimate of WTP among rural and BoP persons. Hence the necessity to conduct this study.
Section snippets
Sampling
We conducted a HH survey in seven locations in India where MIU operate. Sampling followed a two-stage process: in the first stage, we selected seven locations purposively, from among schemes that agreed to participate, and which were located in several states (Maharashtra, Karnataka, Bihar, and Tamil Nadu). In the second stage, several villages (or urban areas) within each location were included (243 villages/areas in total), and at each village about 10 insured HHs, plus about 10 uninsured HHs
Effectiveness of the bidding game
Some scholars claimed that the bidding game may suffer from interviewer bias [28] that can sometimes be grouped with other socially indicated biases called “warm glow” [29], [30]. The bias would exist when respondents accept an amount closer to the opening bid than they would actually do in reality.
We thus wished to assess whether we had an effective bidding process. For this purpose, we compared the reported WTP values with the opening bids. In this study, the lowest quartile is willing to pay
Discussion
The median WTP values found in this study are about INR 600 per HH per year, with 25% of the studied population willing to pay INR 1000 or more (Fig. 2, Fig. 3). WTP per person per year reaches INR 213, INR 199 and INR 168, respectively, for HHs composed of one, two and four persons. The value of WTP per person drops as HH size increases, but this decrease levels-off at around INR 150, when HH size included six persons or more (Fig. 4).
It is interesting to assess WTP also in relative terms,
Conclusions
This study of WTP provides field evidence that rural and BoP population segments in India would agree to pay for health insurance at least 1.35% of median HH income per household per year, or at least 1.8% of median non-health HH expenditure per HH per year.
The nominal levels of WTP identified through this study are much higher than has been known hitherto; we submit that nominal WTP levels stated in this article are conservative estimates, bearing in mind that small HHs are willing to pay up
Acknowledgements
We gratefully acknowledge funding received by the European Union within the EU-India Economic Cross Cultural Programme (ECCP). Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) kindly provided additional funding to enlarge sample size of the household survey. We recognize input received from Parul Khanna and Neelam Mishra (BIMTECH) and Eva Heyblom (Erasmus University Rotterdam/MC). We thank Trea Laske-Aldershof, Ph.D. (Erasmus University Rotterdam/MC) and Barbara Hanel (Friedrich
References (51)
The contingent valuation method in health care
Health Policy
(1999)- et al.
Valuing health care using willingness to pay: a comparison of the payment card and dichotomous choice methods
Journal of Health Economics
(2004) Criterion and content validity of a novel structured haggling contingent valuation question format versus the bidding game and binary with follow-up format
Social Science & Medicine
(2004)Sensitivity to scale in contingent valuation: the importance of the budget constraint
Journal of Health Economics
(2005)- et al.
A comparison of the reliability of the take-it-or-leave-it and the bidding game approaches to estimating willingness-to-pay in a rural population in West Africa
Social Science & Medicine
(2003) - et al.
The feasibility of community-based health insurance in Burkina Faso
Health Policy
(2004) - et al.
With a little help from an anchor: Discussion and evidence of anchoring effects in contingent valuation
Journal of Socio-Economics
(2006) - et al.
Valuing public goods: the purchase of moral satisfaction
Journal of Environmental Economics and Management
(1992) - et al.
Investigating starting-point bias: a survey of willingness to pay for insecticide-treated nets
Social Science & Medicine
(2002) - et al.
Willingness to pay for health insurance in a developing economy. A pilot study of the informal sector of Ghana using contingent valuation
Health Policy
(1997)