Medical Savings Accounts will not advance Canadian health care objectives
=========================================================================

* Jeremiah Hurley

In his review of the literature on Medical Savings Accounts (MSAs) and the implications for the integration of MSAs into Canada's health care system1 (see page 159 of this issue), Samuel Shortt rightly emphasizes that MSAs must be assessed, not against a predominately privately financed system such as the one in Singapore, or a fragmented, multi-payer system such as the one in the United States, but rather against Canada's publicly financed system and the objectives set out for it. Evidence of “success” in these other contexts may still imply failure against Canadian goals.

MSA-based financing would represent a radical departure from Canada's current system of health care finance. Under publicly financed MSAs, the government would provide individuals and families with a lump sum of money annually to be spent (paying full price) on purchasing health care services. This would be supplemented with comprehensive, universal catastrophic health insurance for severe illnesses. Those who do not spend the annual allotment would be able to accumulate funds over time that could be spent on a broader range of goods and services. Thus, MSAs are designed to give people greater choice and control of health care services, provide them with an incentive to use fewer services and encourage them to shop around with their MSA funds, thereby inducing competition among health care providers. In economic terms, these are “demand-side” controls.

Like Shortt, I believe that MSAs are unlikely to advance key Canadian policy goals with respect to expenditure control and health system equity. Demand-side controls have historically been used extensively and found wanting: they do not lead to effective expenditure control, they generate widespread inefficiencies, and they are incompatible with equity in the financing and utilization of health care services.

For well-understood reasons, health care markets do not operate the same way most markets for ordinary consumer goods do.2,3 Contrary to predictions based on standard market theory (e.g., that competition will lead to lower physician fees) competition has frequently been associated with higher fees.2 Similarly, competition among hospitals has been associated with wasteful duplication and increased spending on new technologies.4,5,6,7 There is, therefore, nothing anomalous in the fact that MSAs have not controlled health care expenditures in Singapore — Singapore is simply one of many settings where such competition has failed.

We also know that user fees do, in general, reduce health care utilization. For a number of reasons, however, the reductions in utilization under MSAs are likely to be small and insufficient to reduce public health care expenditures. First, as Shortt notes, in a system with a physician shortage and associated unmet patient demands, reduced utilization by some people will be at least partially offset by increased utilization by others whose demands are currently unrealized. In a system with no physician shortage, reductions in patient demand would result in decreased physician incomes, which physicians may partially offset by inducing demand for services, as was observed in the most comprehensive study of user fees for physician services in Canada.8

Second, the vast majority of health care expenditures arise among a small number of people with very high utilization.9 Utilization by such people would be little affected by MSAs, because their expenses would be covered by their catastrophic insurance. 

Third, under a system of voluntary MSAs, a design most commonly discussed for Canada, risk selection (a well-established feature of voluntary insurance markets) would likely cause public expenditures to increase. MSAs are most attractive to people who are relatively healthy and who use few services, whereas traditional, comprehensive, free care is most attractive to people with high health care needs. The public sector therefore continues to provide coverage to high-risk individuals (as in the current system) while also making MSA contributions to those who use few services. 

Finally, in the unlikely event that, at the end of all this, reduced overall utilization does lead to lower overall health care expenditures, under many MSA designs the savings are not captured by the public sector. Rather, the savings accrue to individuals and private insurers. On the basis of well-documented dynamics of insurance markets and financing mechanisms, therefore, it is entirely plausible that, compared with the current system of financing health care in Canada, MSAs could lead to reduced utilization, reduced system efficiency, reduced equity and increased public expenditure. These and other effects are discussed more extensively, with an illustrative example, elsewhere.10,11

One incontrovertible benefit of MSAs would appear to be increased choice and access: Canadians could choose to spend their government-provided MSA funds on whichever health care services they most valued. But if MSA spending on services not currently insured counts toward meeting the deductible on the catastrophic policy, this is simply an expansion of insurance coverage that will increase public spending accordingly and that could more effectively be achieved through straightforward public financing of these services. If such MSA spending does not count toward meeting the deductible, then the “increased choice” for many Canadians would be more illusionary than real. Those relatively healthy Canadians who use few currently insured services but who do use other uninsured services would stand to benefit; however, those less healthy Canadians who exhaust their MSAs using currently insured services would be no better off, and might even be worse off. Healthy (and wealthy) people would benefit, ill (and often poor) people would not.

MSAs cannot deliver on all that they promise in Canada's publicly financed health care system, and they cannot do so for reasons that are well understood.

𝛃 See related articles pages [143](http://www.cmaj.ca/lookup/volpage/167/143), [151](http://www.cmaj.ca/lookup/volpage/167/151), [159](http://www.cmaj.ca/lookup/volpage/167/159) and [173](http://www.cmaj.ca/lookup/volpage/167/173)

## Footnotes

*   *Competing interests:* None declared.

## References

1.  1. Shortt SED. Medical Savings Accounts in publicly funded health care systems: enthusiasm versus evidence. CMAJ 2002;167(2):159-62.
    
    [Abstract/FREE Full Text](http://www.cmaj.ca/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoiY21haiI7czo1OiJyZXNpZCI7czo5OiIxNjcvMi8xNTkiO3M6NDoiYXRvbSI7czoyMDoiL2NtYWovMTY3LzIvMTUyLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

2.  2. Evans RG. *Strained mercy: the economics of Canadian medicare*. Toronto: Buttersworth; 1984.
    
    

3.  3. Hurley J. An overview of the normative economics of the health sector. In: Culyer AJ, Newhouse JP, editors. *Handbook of health economics*. Amsterdam: Elesevier Science; 2000. p. 56-118.
    
    

4.  4. Folland S, Goodman A, Stano M. *The economics of health and health care*. Upper Saddle River (NJ): Prentice-Hall; 2001.
    
    

5.  5. Robinson J, Luft H. The impact of hospital market structure on patient volume, average length of stay, and the cost of care. *J* Health Econ 1985;4:333-56.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1016/0167-6296(85)90012-8&link_type=DOI) 

6.  6. Benjamini Y, Gafni A. The diffusion of medical technology. Socioecon Plann Sci 1986;20:69-74.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1016/0038-0121(86)90029-7&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=10276533&link_type=MED&atom=%2Fcmaj%2F167%2F2%2F152.atom) 

7.  7. Kessler D, McClellan M. *Is hospital competition socially wasteful?* Cambridge (MA): National Bureau of Economic Research; 1999.
    
    

8.  8. Beck G, Horne J. Utilization of publicly insured health services in Saskatchewan, during and after copayment. Med Care 1980;8:787-806.
    
    

9.  9. Forget EL, Deber R, Roos LL. Medical Savings Accounts: Will they reduce costs? CMAJ 2002;167(2):143-7.
    
    [Abstract/FREE Full Text](http://www.cmaj.ca/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoiY21haiI7czo1OiJyZXNpZCI7czo5OiIxNjcvMi8xNDMiO3M6NDoiYXRvbSI7czoyMDoiL2NtYWovMTY3LzIvMTUyLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

10. 10. Hurley J. *Medical savings accounts in publicly financed health care systems: What do we know*? [CHEPA Working Paper 02-11]. Hamilton (ON): McMaster University Centre for Health Economics and Policy Analysis; 2001.
    
    

11. 11. Hurley J. Medical savings accounts: not the right prescription. *Policy Options* In press.