Response to the consultation document on Voluntary Health Insurance Scheme by the Healthcare Policy Forum

 In Healthcare, Healthcare Financing
,

Our stance on voluntary private health insurance

  • We start with recapping some of our stances on voluntary private health insurance that we explained in our response to the second-stage public consultation on healthcare reform in 2010.
  • We object subsidizing the uptake of private health insurance with public money either directly (such as through offering tax incentives) or indirectly (such as through providing funding to a high risk pool) because subsidizing with public money in this matter is unjust.
    • It is unjust because private health insurance is for those with deeper pockets only. Thus, subsidizing the uptake of private health insurance with public money amounts to subsidizing the relatively well-off to have better healthcare with public money.
  • We are not convinced by the argument that expanding the uptake of private health insurance will ease pressure on the public healthcare system. Instead, international evidence shows that private health insurance often actually harms the public system in time.
    • One reason for this is that expanding private healthcare utilization simply takes badly-needed doctors and nurses out of the public system and reduces its capacity.
  • We take the view that subsidizing the uptake of private health insurance with public money is in essence a policy of “(f)or whosoever hath, to him shall be given, and he shall have more abundance: but whosoever hath not, from him shall be taken away even that he hath”, a policy that no reasonable people should accept.
  • We are, nonetheless, sympathetic to the call for better regulation of the private healthcare insurance market so that those willing to buy private healthcare insurance will be able to purchase value-for-money insurance coverage.
  • On balance, therefore, we do not oppose any private healthcare insurance scheme aimed at providing people with value-for-money insurance coverage so long as
    • it does not involve any insurance uptake subsidization with public money, whether directly or indirectly.
    • it does not adversely affect the public healthcare system (or better still, if the scheme can be made complementary to the public system with proper infrastructure planning and human resources planning).

 

Our stance on the Voluntary Health Insurance Scheme (VHIS) proposed in the consultation document

  • Based on the considerations explained in the last section, we are unable to give our support to the Voluntary Health Insurance Scheme (VHIS) proposed in the consultation document for the following reasons:
    • the scheme involves the introduction of tax deduction for insurance premiums to encourage the uptake of insurance.
    • the scheme involves the use of public money to finance a high risk pool.
    • the scheme’s impact on the public healthcare system remains uncertain (because we find the consultation document’s projection of future utilization of private healthcare under VHIS dubious).
  • In addition, we also oppose the setting up of a high risk pool on the ground of healthy development of the healthcare insurance market (because a high risk pool is incapable of containing the risk-selection behavior of insurers, which is market destabilizing).
  • As a way forward, we propose the following:
    • no tax incentive should be offered for the uptake of health insurance.
    • risk equalisation, which can be made self-financing, should be adopted as an alternative to the high risk pool.
  • We also demand that our query on the projection of future utilization of private healthcare under the scheme be openly answered and the projection reassessed.
  • We shall explain in the sections below our query on the consultation document’s projection of future utilization of private healthcare and the rationale for instituting risk equalisation.

 

Our query on the projection of future private healthcare utilization under VHIS and possible impacts on the public healthcare system

  • Statistics on hospitalization in Hong Kong can bring relief to our query.
  • According to the Census and Statistics Department’s Thematic Household Survey Report No. 50 conducted 2011, 483,300 persons had been admitted to hospital in the 12 months prior to the enumeration.
    • Among the 483,300 persons,
      • 376,900 (78.0%) were admitted to hospitals under the Hospital Authority (HA).
      • 133,600 (27.6%) were admitted to private hospitals.
    • Among the 483,300 persons who needed hospitalization, 203,900 (42.2%) were covered by private healthcare insurance (PHI) of some kind.
    • According to the consultation document, among the 203,900 persons who were covered by PHI, about 54% of their local hospital admissions pertained to the public sector (i.e., 110,106 persons’ local hospital admission pertained to the public sector) (pare. 1.14).
    • In other words, as high as 30% of those admitted to HA hospitals were covered by PHI of some kind.
  • Since, according to the consultation document, the expected uptake of insurance under VHIS is high (about 38% of population), it is not unreasonable to expect that most of the 110,106 private- insurance-covered pubic hospital inpatients will migrate their insurance to VHIS.
  • According to the consultation document as well as the consultancy report that it refers to, it is estimated that after the implementation of VHIS in 2016, a total of 120,000 procedures will be shifted out of the public system and among these, 90,000 procedures (75%) are advanced diagnostic tests (consultation document, para. 8.25 and consultancy report, p. 158).
  • Assuming that the coverage of VHIS is comprehensive and that there is no big change in the pattern of hospitalization in 2016, the aforementioned figures seem to suggest that currently, the majority of public hospital inpatients with private insurance of some kind are admitted to public hospitals for advanced diagnostic tests only. But this is highly unlikely to be the case by common understanding.
  • For the highly unlikely case to obtain, there are only two possible scenarios:
    • either that the coverage of VHIS is inadequate
      • Under this scenario, patients covered by VHIS will fall back on the public system for further treatments after receiving advanced diagnostic tests in the private sector.
      • The implication of this will be VHIS failing to provide patients with good-value-for-money protection, bearing in mind that the expected premium of VHIS is $5,300 at the age band of 50-54 and uprated to $6,900 at the age band of 60-64.
      • The consequence of this will be pressures for improving the cover of the scheme overtime (or the scheme will fail eventually).
      • With improvement in insurance coverage, the utilization of private healthcare can be expected to increase too.
    • or that the private healthcare utilization projection is in fact inaccurate
      • It may be that the coverage of VHIS is indeed adequate, only that the increase in private healthcare utilization has been underestimated.
      • The bulk of the increase in private healthcare utilization after the introduction of VHIS cannot come only from advanced diagnostic tests but from “other inpatient procedures” as well.[1]
      • The implication of this scenario will be significantly greater expansion of the private healthcare sector than suggested by the original estimation in order to carter for the increase in private healthcare activities
  • Both scenarios point to the possibility that the increase in private healthcare utilization under VHIS may be a lot greater than that reported in the consultation document, which may in turn significantly shrink the capacity of the public healthcare system as medical manpower is drawn away from the public to the private sector.
  • Hospitalization statistics can help illustrate the likely increase in private healthcare utilization under VHIS.
    • Recall that there were 110,106 pubic hospital inpatients who were covered by private insurance of some kind in 2011.
    • If 20% of these patients (numbering 22,021) migrate their insurance to VHIS and subsequently shift to private hospitals (assuming that the coverage of VHIS is adequate), this will represent a 16% increase in private hospital inpatients.
    • If 30% instead of 20% of these patients migrate, it will represent a 25% increase in private hospital inpatients.
  • Due to the lack of relevant data, we are unable to ascertain how such growths in private hospital inpatients will drain human resources from the public system. But intuition says the drain will be noticeable.
  • Since the expansion of private healthcare utilization may negatively affect the majority of Hong Kong people who will continue to rely on public healthcare, the government has the responsibility to provide satisfactory answers to the queries raised in this section and to reassess the projection of private healthcare utilization under the health insurance scheme.

 

Our proposal to set up a risk equalisation mechanism as an alternative to a high risk pool

  • We recognize the importance of instituting a mechanism to stabilize a healthcare insurance market where community rating and open enrollment are mandated.
    • Major market destabilizing factors under the conditions of community rating and open enrollment include:
      • a continuous exit of insurers with an adverse risk profile due to bankruptcy (because insurers can neither charge a higher premium nor reject any high-risk subscribers).
      • a continuous exit of low-risk enrollees due to a rapid increase in premiums (since in order to cover their losses, insurers with an adverse risk profile will demand an increase in premiums that is higher than it will otherwise be; in other words, low-risk insurance enrollees will be charged a premium higher than it will otherwise be and will therefore be driven out of the market).
    • Because of these factors, the market may collapse eventually.
  • We also recognize that a high risk pool is one tool for stabilizing the market and that the pool can be instituted in a way such that it is financed by the health insurance industry itself without the need to involve any public funding.
  • We nonetheless do not propose this option as an alternative.
  • Instead, we propose risk equalisation because it can better deter the risk-selection behavior of insurers. In our view, the high risk pool mechanism is not capable of addressing the problem of risk selection.
    • Risk selection refers to the profit maximization behavior or strategies on the part of insurers to select low-risk enrollees (cream-skimming or cherry-picking) and reject high-risk enrollees (lemon-dropping).
    • For individual insurers, risk selection is a more effective strategy for maximizing profits than improving efficiency through, say, lowering the costs of healthcare or of operation.
    • Between lemon-dropping and cherry-picking, the former is more effective than the latter for maximizing profits because “on average the per capita predictable losses on the ‘lemons’ are higher than the per capita predictable profits on the ‘cherries’”.[2]
    • Under the arrangement of a high risk pool, one obvious risk-selection strategy is to drop the lemons to the high risk pool.
      • We are of the view that the consultation document has underestimated insurers’ incentive to drop lemons to the high risk pool.
      • The document argues that as long as insurers can charge a premium loading rate commensurate with the extra risks that they take on, insurers will have the incentive to keep the policies of higher-risk individuals under their own portfolio so that they can retain the profits on these policies.
      • This argument indicates that the document has missed the important point mentioned above, namely, that “on average the per capita predictable losses on the ‘lemons’ are higher than the per capita predictable profits on the ‘cherries’”.
      • Insurers’ incentive to drop lemons, in particular those with an adverse risk profit, to the high risk pool are actually higher than the document expects.
      • The consequence of this lemon-dropping behavior is adverse impact on the financial sustainability of the high risk pool.
      • Thus, the financial commitment of the government to the pool may be higher than originally estimated.
    • Apart from the aforementioned lemon-dropping strategy, some other risk-selection strategies are also available even when open enrollment is mandated. (See Appendix 1 for strategies of risk selection in an insurance market mandating community rating and open enrollment.)
    • All in all, apart from affecting the financial sustainability of the pool, the adverse impacts of risk selection on the insurance market include:
      • risk selection threatening good quality care for the chronically ill.
      • possible reduction in the efficiency of the insurance industry.
      • reducing the high risks’ affordability of health insurance.
      • the insurance market remaining unstable.
      • individual insurers gaining by risk selection while society as a whole gains nothing, with any resources used for selection representing a welfare loss.[3]
  • We are not aware that the consultation document has proposed any measures to address these potential adverse impacts of risk selection.
  • Risk equalisation can deter insurers’ risk-selection behavior because it aims to equalise the risk profiles among insurers.[4]
  • In a nutshell, the risk profiles among insurers are equalised in the following way:
    • In general, it is supposed that the risk profile of each insurer should be equal to that of the population. For example, if age and gender are considered as the relevant risk factors and if x% of the market are within the “50–59 year old male” risk group, then each insurer should in principle have x% of their membership being 50–59 year old male.
    • Because of this consideration, insurers who have a favorable risk profile are required to compensate insurers with an adverse risk profile by paying the latter a risk equalisation payment.
    • In other words, insurers who have a favorable risk profile will be penalized, and this will in turn dis-incentivize insurers to risk-select.
    • A central fund (sometimes called a solidarity fund) is usually set up to facilitate the equalisation payment transfers. Under this arrangement, insurers pay their risk equalisation contributions to and receive their risk equalisation compensations from the fund.
    • The amount an insurer receives from or pays to the central fund is calculated in this way:
      • Firstly, an insurer’s “expected expenditures” under the assumption that it has a risk profile equal to that of the population is calculated. One formula for calculating the insurer’s “expected expenditures” is based on the insurer’s own business/operation costs.
      • The “expected expenditures” are then compared with the insurer’s “actual expenditures” resulting from its own actual risk profile.
      • The difference between the “expected expenditures” and the “actual expenditures”, subject to an adjustment to ensure that the central fund is self-financing (i.e. to ensure that the total payments to the fund is equal to the total outlays of the fund), is the amount that an insurer will receive from or will pay to the fund. (Appendix 2 provides a mathematical outline of how the risk equalisation transfers are calculated.)
    • A final note on using an insurer’s own business/operation costs for calculating its “expected expenditures” is in order.
      • Using an insurer’s own business/operation costs can promote efficiency and cost containment of the health insurance market.
      • This is because the calculation method allows an insurer to retain the profit gains resulting from its own effort to improve the efficiency of its business.
      • It can be shown that the payment paid to a recipient insurer (i.e. an insurer with an adverse profile of risk) will not diminish if it becomes more efficient.
      • Likewise, the payment paid to the central fund by a contributing insurer (i.e. an insurer with a favorable profile of risk) will decrease if it becomes more efficient.[5]
  • To sum up, risk equalisation not only can deter insurers’ risk-selection behavior but can also help promote efficiency and cost containment of the market.
  • Because of these advantages of risk equalisation over the high risk pool, we urge the government to seriously consider this mechanism for stabilizing the health insurance market.

 

 

Healthcare Policy Forum (醫療政策論壇)

15 April 2015

 

 

[1] It should be reminded that according to the consultant’s original estimation, “other inpatient procedures” account for less than 10% of the increase in private healthcare utilization.

[2] Wynand P. M. M.Van de Ven and Frederik T. Schut. 2007. Risk Equalization in an Individual Health Insurance Market: The Only Escape from the Tradeoff between Affordability, Efficiency and Selection. The Netherlands as a Case Study. Stanford: Center for Advanced Study in the Behavioral Sciences, Stanford University, p. 10.

[3] The insights on the adverse effects of risk selection are drawn from ibid., pp. 10-12.

[4] The following sketch of a risk equalisation mechanism is based on Ireland’s original risk equalisation scheme introduced in 1996. It should be noted that the sketch is for illustrating the spirit and logic of risk equalisation only. Many of the details of the original Irish scheme have not been presented. Also, it does not imply that we recommend the Irish scheme per se. Our understanding of risk equalisation and the Irish scheme is based on: John Armstrong. 2010. Risk equaliation and voluntary health insurance markets: the case of Ireland. Health Policy 98(1): 15-26; John Armstrong and Francesco Paolucci. 2010. Risk equalisation in Ireland and Australia: a simulation analysis to compare outcomes. The Geneva Papers on Risk and Insurance — Issues and Practice 35(4): 521–538; John Armstrong, Francesco Paolucci, and Wynand P.M.M. van de Ven. 2010. Risk equalisation in voluntary health insurance markets. Health Policy 98(1): 1-2; John Armstrong, Francesco Paolucci, Heather McLeod, and Wynand P.M.M. van de Ven. 2010. Risk equalisation in voluntary health insurance markets: a three country comparison. Health Policy 98(1): 39-49.

[5] See the stimulation results in John Armstrong and Francesco Paolucci. 2010. Ibid., p. 534.

 

 

Endorsed by:

  1. Dr George Cautherley (Convenor, Healthcare Policy Forum; Vice-Chairman, Hong Kong Democratic Foundation)
  2. Mr Richard Tsoi (Deputy Convenor, Healthcare Policy Forum)
  3. Dr Cheung Chiu-hung, Fernando (Legislative Councillor)
  4. Mr Cheung Kwok-che (Legislative Councillor)
  5. Ms Ho Sau-lan, Cyd (Legislative Councillor)
  6. Dr Kwok Ka-ki (Legislative Councillor)
  7. Mr Lee Cheuk-yan (Legislative Councillor)
  8. Mr Leong Kah-kit, Alan (Legislative Councillor)
  9. Mr Leung Yiu-chung (Legislative Councillor)

 

 


Appendix 1


Appendix 1: Forms of risk selection under community rating and open enrollment[6]

 

[The form of risk selection under community rating and open enrollment may depend on the additional information that insurers have. Three situations can be distinguished.]

 

First, if insurers only know that there are high- and low-risk individuals within the allowed premium-risk-groups, but they cannot ex-ante identify who are the high-risk individuals and they also don’t know what the relevant omitted risk factors are, they may structure their coverage so that the insurer is unattractive for the high-risk individuals (Newhouse, 1996)*. For example, insurers may exclude prescription drugs from coverage, they may offer a policy with a high deductible, or they may contract with a selected panel of providers who work according to strict protocols. Such an insurer is more attractive for the low-risk individuals than for the high-risk individuals within each premium-risk-group. In this way the insurers use adverse risk selection as a tool for preferred risk selection. They stimulate the different risk groups to reveal their risk. Insurers may also share financial risk with the contracted providers in a way that encourages providers to ‘drop the lemons’. As Newhouse (1982)** highlighted in an example of a “mother with an asthmatic child”, providers of care have subtle tools to encourage high cost patients to seek care elsewhere, such as keeping the patient in uncertainty about the correct diagnosis, making the patient wait for an appointment, making the patient wait in the office, being discourteous to the patient, or advising chronically ill patients to consult another physician who is “more specialized in treating their disease”.

 

Second, if insurers know that some omitted risk factors are relevant (e.g. AIDS, disability, prior utilization or hypochondria), but they cannot ex-ante identify the individuals with these characteristics, they may deter the high-risk consumers by selectively not contracting with physicians who have the best reputation of treating patients with such problems.*** Insurers also could contract with providers whose facilities have no disabled access. They may also select by the design of their supplementary health insurance (no coverage for mental health care, prescription drugs and reconstructive breast surgery).

 

Third, if insurers can ex-ante identify predictably unprofitable individuals based on certain risk characteristics, they can focus their selection strategy directly on those identifiable individuals, e.g. by providing the high risks with poor quality care or poor services (such as delayed payments of reimbursement and delayed answers to letters); by not coordinating the care for people with multiple needs; by selective advertising and direct mailing; by providing the insurance broker with incentives to advise relatively unhealthy persons to buy health insurance from another company; or by a golden handshake for unhealthy members at disenrollment, such as offering an AIDS patient a large sum of money to choose a different insurer during the next open enrollment period.

 

* Newhouse, Joesph P. 1996. Reimbursing health plans and health providers: selection versus efficiency in production. Journal of Economic Literature 34(3): 1236-1263.

** Newhouse, Joesph P. 1982. Is competition the answer? Journal of Health Economics 1(1): 109-115.

*** This in turn can discourage physicians and hospitals from acquiring such a reputation.

 

[6] The contents of this appendix are drawn from Wynand P. M. M. van de Ven and Frederik T. Schut. 2007. Risk Equalization in an Individual Health Insurance Market: The Only Escape from the Tradeoff between Affordability, Efficiency and Selection. The Netherlands as a Case Study. Stanford: Center for Advanced Study in the Behavioral Sciences, Stanford University, p. 49, Appendix 1.

 


Appendix 2


Appendix 2: Mathematical outline of risk equalisation transfers[7]

 

 

[…]

 

[…]

 

 

[7] The contents of this appendix are drawn from John Armstrong. 2010. Risk equalisation and voluntary health insurance markets: the case of Ireland. Health Policy 98(1): p. 26, Appendix A.


 

 

 

 

Facebook Comments
Recent Posts
The Professional Commons’s response on Healthcare Reform Second Stage Consultation 
Developing country park – from absurdity to reality
Taking stock of the local agricultural industry is more than a matter of market values (Chinese version only)
Our dignified autonomy: farmers for the umbrella movement (Chinese version only)
China survival tip: where in the world are Guangdong’s hazardous chemical wastes? (Chinese version only)
Political expectations of the new China-HK model in post-political reform Hong Kong (Chinese version only)

Start typing and press Enter to search

Home
Home