Medical savings accounts: What story do the data tell?

Health Affairs

Volumn 25, Issue 1, 2006, Pages 256-267

  Minicozzi, Alexandra ^a  

^a Office of Tax Analysis   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

CONFERENCE PAPER; COST; HUMAN; MIDDLE AGED; SOCIAL CLASS; STATISTICS; TAX; UNITED STATES; UTILIZATION REVIEW;

HUMANS; INCOME TAX; MEDICAL SAVINGS ACCOUNTS; MIDDLE AGED; SOCIAL CLASS; UNITED STATES;

EID: 32044445095     PISSN: 02782715     EISSN: 02782715     Source Type: Journal    
DOI: 10.1377/hlthaff.25.1.256     Document Type: Conference Paper

References (9)

1. The qualified, minimum for individual and family policy deductibles is indexed by the medical component of the Consumer Price Index.
2. U.S. Government Accountability Office, Medical Savings Accounts: Findings from Insurer Survey, Pub. no. GAO/ HEHS-98-57 (Washington: GAO, December 1997);
3. and GAO, Medical Savings Accounts; Results from Surveys of Insurers, Pub. no. GAO/HEHS-99-34 (Washington: GAO, December 1998).
4. Congressional Budget Office, "A Review of Reported Employer Experiments with Medical Savings Accounts," Memorandum (Washington: CBO, January 1997);
5. American Academy of Actuaries, "Medical Savings Accounts: Cost Implications and Design Issues" (Washington: AAA, May 1995);
6. and D. Zabinski et al., "Medical Savings Accounts: Microsimulation Results from a Model with Adverse Selection," journal of Health Economics 18, no. 2 (1999): 195-218.
7. An alternative approach is a nested model where individuals first choose whether to buy health insurance and then decide their plan characteristics. The nested approach seems inappropriate because one might choose a newly introduced lean policy with tax incentives when they would have otherwise been uninsured. In fact, according to Master File data, many MSA consumers were previously uninsured.
8. The counterintuitive result that expected premium is higher for the middle-aged than for the near-elderly (ages 56-65) could be the result of omitted variable bias: policy benefit ratio and individuals covered are not controlled for, are potentially correlated with age, and may affect premium.
9. AAA, "Medical Savings Accounts."