Executive Summary (95-02) February 22, 1995
In recent years, policy makers have become increasingly concerned with escalating health care spending in the United States and the lack of insurance coverage for about one-seventh of the nation’s population. As a result, there has been a growing interest in health care reform. One part of the reform debate has involved health care administrative costs. Some reports have suggested that administrative costs in the United States are large relative to those in other nations and have grown even faster than overall health spending. Consequently, policy makers have become interested in reducing administrative costs as a means to fund insurance coverage for the currently uninsured or to control growth in health spending.
This report examines the impact of a variety of health reform options on health care administrative costs and overall health care spending in Minnesota. Among the reform proposals examined are several "single-payer" plans and a number of options which incorporate the MinnesotaCare reforms contained in current state law but not yet fully implemented. The report addresses the following key issues:
To assist us on this study, we hired Lewin-VHI, a national health care consulting firm. Our consultant was responsible for estimating the levels of administrative expenditures and overall health spending in Minnesota and projecting how much spending would change under alternative reform options. For those interested in the detailed methods used to make these estimates, Lewin-VHI’s report is available from our office.
In 1994, health spending in Minnesota was an estimated $15.8 billion, or roughly 13 percent of the gross state product. About three-quarters of the health spending was in three categories: hospital care (31 percent), physician care (26 percent), and long-term care (15 percent). The remaining 28 percent of expenditures was split among the following categories: insurer administration, other professional services, dental care, prescription drugs, public health and other costs, and vision care.
Health care administrative costs generally represent the transaction costs incurred in exchanging the information and resources necessary to provide health care services.
Administrative costs are incurred by private and public insurers, physicians, hospitals, employers, and government regulatory agencies. Administrative costs were an estimated $2.4 billion, or 15 percent of overall health spending, in Minnesota during 1994.
This report examines the cost implications of seven alternative reform scenarios, which attempt to achieve universal insurance coverage for Minnesotans. Three of these are single-payer plans, which include a fee-for-service delivery system for physician care and government-set hospital budgets. These tax-financed plans eliminate the payment of insurance premiums for basic health benefits and rely instead on additional state tax revenues. As a result, health insurance would no longer be provided by employers through a multitude of insurers but rather through one single government agency funded by tax revenues. The first of these reform scenarios is a Canadian-style system, in which patient cost sharing is eliminated. (Patient cost sharing refers to the payment of deductibles and copayments for services received, subject to an annual maximum. Six of the seven scenarios include patient cost sharing, although none of the options include cost sharing for persons with incomes below the poverty level.) The second scenario is a single-payer plan with cost sharing. Both of these scenarios assume Medicare patients are included in the plan. Because obtaining Congressional permission to include Medicare patients is not likely, the third scenario examined was a single-payer plan (with cost sharing) that does not include Medicare enrollees.
In addition, we examine the impact of three reform scenarios that retain insurance premium financing, exclude Medicare, and incorporate the insurance market and health delivery system reforms envisioned under MinnesotaCare. These reforms are currently in state law but are not expected to be fully implemented until 1997. Two of the scenarios assume that Minnesota implements universal coverage by requiring individuals to obtain insurance coverage, as called for in existing law. (Current law makes a commitment in principle to universal coverage by mid-1997. However, that commitment is contingent upon the availability of affordable coverage. Further legislation is necessary to implement this commitment and other aspects of the MinnesotaCare reforms.) One of these individual mandate scenarios assumes that Minnesota receives a Congressional waiver from the provisions of the Employees Retirement Income Security Act (ERISA) so that the reforms can be applied to self- insured employers. The other scenario assumes that an ERISA waiver is not granted. Both provide premium subsidies on a sliding scale to persons with incomes below 275 percent of the poverty level to help them afford insurance coverage. A third scenario implements universal coverage by requiring employers to provide insurance. The employer mandate approach includes subsidies to certain employers, as well as premium subsidies for individuals and families. Implementation of an employer mandate would require an ERISA waiver. These scenarios are an attempt to estimate the impact of MinnesotaCare reforms on health care spending, assuming that the state proceeds with plans to eventually provide universal coverage. They are an approximation at best, however, since many details, including the uniform benefits package, have not been determined.
The final scenario modeled was a tax-financed plan, in which a government agency would contract with health plans and provide Minnesotans with a choice of competing plans. The government agency would pay health plans based on the number of people choosing the plans. This hybrid model attempts to benefit from a streamlined payment system, while retaining a health insurance industry and also benefiting from the restructured delivery system envisioned under MinnesotaCare.
Our consultant estimated the effect of these reform scenarios, assuming implementation in 1997 when health spending in Minnesota is estimated to be $20.8 billion. In percentage terms, the impact of the reform scenarios on health spending is rather modest:
These estimates indicate that selecting the plan with the greatest administrative cost savings will not necessarily contain health care costs the most. The Canadian single-payer plan would reduce administrative costs the most, but it would increase overall spending almost as much as the premium-financed plans, which are expected to result in administrative cost increases. This results largely because a Canadian plan would eliminate patient cost sharing and increase health utilization by more than the $1.3 billion in administrative savings expected in 1997. In contrast, the individual mandate scenario with MinnesotaCare reforms and an ERISA waiver would increase administrative costs by $156 million and would increase overall health spending by $120 million compared with $93 million for the Canadian single- payer plan. What the individual mandate approach lacks in administrative savings is made up by requiring patient cost sharing and by realizing managed care savings from a restructured health delivery system. (These reform scenarios have more than a one-time impact on health spending. Absent any cost controls, the estimated percentage change in spending relative to a no- reform scenario is expected to remain relatively constant in future years.)
The lowest-cost option is a single-payer plan with cost sharing, which reduces administrative costs significantly but not as much as the Canadian model. This option reduces administrative costs by an estimated $1.0 billion in 1997 ($0.9 billion if Medicare is excluded) and has a much lower expected increase in utilization than the Canadian model because it requires patients to pay deductibles and copayments.
These reform options differ considerably in the amount of additional state taxes required to implement them in 1997. The tax-financed options, including all the single-payer plans, would require a substantially greater increase in taxes ($5.8 billion to $9.5 billion) compared with the premium-financed options ($0.4 billion to $1.2 billion). (For all options, the estimated tax increases do not include the anticipated growth in state costs of public insurance programs between 1994 and 1997. In particular, they do not include the costs of expanding the MinnesotaCare subsidy program to include adults without children up to 275 percent of the poverty level. In addition, they do not consider how the state’s costs may be affected if the state receives the pending health care reform waiver from the federal government. However, this increased reliance by the single-payer plans on taxes tends to be offset by the elimination of premium payments made by employers and individuals.
The results of this study are generally consistent with national studies of health spending under alternative health reforms, although national studies do not always agree. For example, the Congressional Budget Office (CBO) estimated that a Canadian single-payer plan would increase U.S. health spending by 5 percent. This increase was the result of a 7 percent reduction in spending due to lower administrative costs and a 12 percent increase due to growth in health care utilization. In contrast, the General Accounting Office (GAO) projected a 0.4 percent reduction in overall spending resulting from a 9.1 percent drop due to lower administrative costs and a 8.7 percent increase due to increased utilization of health services. Our study for Minnesota estimated that overall spending under a Canadian-style system would increase by 0.4 percent as a result of lower administrative costs (6.5 percent) and higher utilization costs (6.9 percent). The studies differ in part because Minnesota data were used in our study. In addition, the GAO estimated greater administrative cost savings and used a different assumption about how significantly utilization would be affected by the elimination of cost sharing.
Like our study, the CBO also found that a single-payer plan with cost sharing would lower overall health spending, although it would not lower administrative costs as much as the Canadian approach. CBO estimated that the single-payer plan with cost sharing would reduce U.S. health spending by about 2 percent. Our study for Minnesota estimates a spending reduction of about 3 percent.
Estimates for the three premium-financed versions of MinnesotaCare reforms are also consistent with available research. Our study forecasts that changes in administrative costs will increase overall spending by a modest amount (about 0.7 percent). This result is consistent with the Congressional Office of Technology Assessment’s finding that administrative costs would not change much under various proposals to reform the private insurance market. Lewin-VHI’s estimates of the savings from a restructured delivery system are consistent with the best available research on health maintenance organizations (HMOs). However, our consultant suggests that the savings may take time to achieve, particularly outside of the metropolitan Twin Cities area. The eventual savings could exceed our consultant’s estimates, if Integrated Service Networks (ISNs) are able to alter medical practices even more than established HMOs or if the market share of ISNs exceeds an estimated 66 percent statewide.
Our study, like similar national studies, has some limitations. First, it is important to recognize that this study focuses only on health care spending. Reform options may have significantly different impacts on the quality of care, health outcomes, access to care, timeliness of care, technological development, and consumer choice. These considerations are of significant importance to policy makers interested in health care reform.
Second, spending estimates are sensitive to the data used, the assumptions made, and projections of general economic growth and health care cost trends. Health care analysts have an extremely difficult job predicting changes in health spending under reform, because most reforms would substantially alter the incentives of consumers, providers, insurers, and employers. Some variation in the conclusions reached by analysts can be expected, because it is difficult to forecast how people will respond to dramatically different incentives.
Third, studies generally do not attempt to take into account certain economic factors. These include possible migration effects, the economic impact of additional taxes (or premium subsidies) on work and savings incentives, the costs imposed on patients by paperwork, or the costs imposed on patients from lost time or productivity due to less timely care.
Finally, many studies including this one do not attempt to measure the impact of cost controls. Both single-payer plans and MinnesotaCare-type reforms would probably include certain types of expenditure and fee controls. The possible cost impact from such controls could be large. However, the relative effectiveness of controls under different reform options is unclear. Furthermore, controls could provide beneficial cost containment or could result in an inefficient use of resources.
A study by the Legislative Auditor of the administrative costs associated with a "single-payer" health plan was requested by Minn. Laws (1994), Ch. 625, Art. 5, Sec. 9 and Art. 14, Sec. 1, Subd. 4. On June 22, 1994, the Legislative Audit Commission directed the Program Evaluation Division to conduct an expanded study of health care administrative costs.
For a copy of the full report, entitled "Health Care Administrative Costs" (95-02), published on February 22, 1995, please call 612-296-4708, e-mail Legislative.Auditor@state.mn.us, or write to Office of the Legislative Auditor, 658 Cedar St., St. Paul, MN 55155.
Staff who worked on the study included John Yunker (project manager), Tom Walstrom, and Don Feige (intern). The study also utilized the services of Lewin-VHI, a health care policy consultant.