Special Report: The Uninsured

Covering many of the uninsured need not spike health care costs

Twelve to 40 million uninsured Americans could receive health care coverage at an increase to total national health spending of 1.5% to 3.7%, or $23 billion to $57 billion a year, according to new analysis from the Lewin Group.

"The bottom line is that significant progress in covering the uninsured is possible using a variety of different approaches that run the philosophical gamut," said Robert Wood Johnson Foundation senior vice president John Lumpkin. "Our problem is not that we don’t know how to cover the uninsured, but that we lack the national will to do so."

The analysis was completed as part of the Economic and Social Research Institute’s (ESRI) Covering America project. The foundation is funding the Covering America effort.

The report did not identify a preference for any of the 10 proposals it puts forward and also didn’t address the political/policy difficulties that are likely to face any of them. Rather, it explains the cost for each proposal and the number of Americans each would cover under five broad categories of change: incremental reforms, voluntary insurance pools, proposals that replace the current tax exclusion for employer-sponsored health insurance with tax credits, proposals that require employers to pay for coverage, and a tax-financed health care system.

The report highlighted that health reform involves trade-offs among different goals. Nearly all of the proposals combine coverage expansion with other objectives such as fairness to the currently insured, limiting the amount of new federal spending, increasing consumer choice, etc. Moving toward one goal, the authors say, often means moving away from another.

"While these trade-offs mean that there is no single best or problem-free solution to cover the nation’s uninsured," said Jack Meyer, ESRI president, "these proposals are real options that could dramatically reduce the total number of people without health insurance. Many of the proposals from this project share the same building blocks as those put forth by candidates, members of Congress, and state leaders who also struggle with the same trade-offs shown in these proposals. This project makes it possible to compare very different approaches using a common methodology that can help policy-makers and private-sector leaders forge effective proposals to cover the uninsured."

At an Oct. 31 briefing on the report, Meyer made a particular point of the relatively small increase in health care spending that would be needed to cover many of the uninsured.

"The reform plans evaluated by The Lewin Group would add from 1.5% to 3.7% to the more than $1.6 trillion the United States now spends on health care," he said. "This net cost of covering the uninsured is anywhere from a fourth to a half of the annual increase in our health care spending. If we do not reform the health care system, costs will continue to soar, inappropriate medical care will continue unabated, and the number of uninsured will continue to grow. These problems feed on each other, as occurs when health spending increases well in excess of wages leads to increases in the number of uninsured."

Also noting the relatively small investment that would be needed was Center for the Study of Health System Change vice president Len Nichols, an author of one of the plans, who reacted at the ESRI briefing. "The total new spending required is relatively low," he said, "at least compared to what we’re about to spend on drugs and compared to what we’re maybe going to spend on the war before we’re done. So at the end of the day, it doesn’t look like all that much new money."

The Lewin Group report by John Shiels and Randall Haught stated that total health spending would change in response to the proposals for at least four reasons. First, previously uninsured people use more medical services once they have coverage. Second, previously insured people whose coverage is now more or less comprehensive change the amount of health care services they consume. Next, health care services are provided more or less efficiently. And, finally, provider reimbursement levels change.

"The first three contributors to new health spending are, in many ways, the most important," according to Mr. Shiels and Mr. Haught. "They represent the real resource cost of coverage expansion — the manpower, capital equipment, and other resources that are used to produce health services and are thus not available to produce other goods and services that people value. These costs represent other opportunities foregone because of the decision to expand the health sector."

The researchers said that another way to look at costs is to examine who pays the health care bill. While households ultimately bear all costs of financing health care, in the form of direct payments for health services, as part of the price of non-health goods and services, as reduced wages or other compensation, or as taxes, it is important to analyze how initial financing of the health care bill is shared among the various players. Mr. Shiels and Mr. Haught point out that even though many coverage expansion proposals produce only a relatively small increase in total health expenditures, they may cause large shifts in the distribution of health care financing among households, employers, and government for several reasons.

First, to make coverage more affordable for those who are currently uninsured, all the proposals provide subsidies to lower income people, high-risk individuals, or small or low-wage employers. The result is that the cost borne by government increases.

Also, most proposals provide government subsidies to people who already purchase health coverage without subsidies — the people who are covered by their employers or purchase coverage in the individual market but switch to the new public program or become eligible for subsidies.

The proposals assign initial costs quite differently to households, employers, and government, based on the trade-offs authors are willing to make and the objectives they seek to promote in addition to expanding coverage. Thus, household payments of health premiums and out-of-pocket health costs decline between $3 billion and $187 billion a year; for employers, the impact on health spending ranges between a $69 billion increase and a $77 billion annual savings; state and local health care spending would drop under most of the proposals, by as much as $28 billion and go up under a few of the plans by as much as $6 billion a year; and federal costs would grow between $34 billion and $552 billion annually.

The Lewin study said that in deciding how to structure major reforms, policy-makers must balance a number of goals, which could include covering a large number of people who otherwise would be uninsured, avoiding unfair treatment of those currently insured, limiting the extent of new federal spending, limiting growth in total national health spending, increasing consumer choices among health plans and health care providers, improving quality of care, and limiting the amount of federal spending going to people who already have coverage.

"Frequently, pursuing one goal involves trading off another," explained Mr. Shield and Mr. Haught. "For example, a number of proposals deliberately incur additional federal costs to accomplish goals other than expanded coverage."

Thus, some tax credit proposals shift hundreds of billions of dollars in annual premium costs from currently insured, lower-income households to the federal government. This "horizontal equity" provides the same level of subsidy to all similarly situated low-income individuals, including both the uninsured and those who previously purchased coverage on their own. In addition, some proposals seek to limit future growth in national health care spending and to increase consumer choice through insurance pools that offer multiple health plan options, with federal premium subsidies, on terms that give consumers incentives to select less expensive coverage. Because such pools are not limited to the newly insured, but also serve numerous workers who previously received employer-sponsored coverage, some of the proposals, depending on how they are structured, may shift billions of dollars in annual health premium costs from employers to government.

In contrast, the authors said, other proposals prioritize the competing policy goal of limiting new government spending to just the uninsured, whenever possible. By reducing the shift of current private-sector costs to the public sector, such proposals spend fewer federal dollars on each newly insured person. "In sum," added Mr. Shiels and Mr. Haught, "there is no problem-free solution to the problem of the uninsured. Policy-makers must inevitably resolve trade-offs among competing, desirable objectives. The analysis suggests that policy-makers may face a basic trade-off between preserving a voluntary and nonguaranteed system of health insurance and maximizing the increased coverage that results from reform. Automatic enrollment mechanisms appear to increase the coverage gains from voluntary and nonguaranteed systems, but even policy-makers incorporating auto-enrollment strategies must decide whether to seek the additional increment of coverage that would result from mandates or legal guarantees of health coverage."

Here is the basic scoring Mr. Shiels and Mr. Haught developed for each of the 10 plans (listed by the last name of the lead author):

Feder — cover about 12 million uninsured. Federal costs less offsets would be about $34.1 billion.

Pauly — cover about 20.5 million people with a net federal cost of $89.7 billion.

Singer — cover 11.8 million people with new federal spending net of offsets of $102.8 billion.

Gruber — cover 14.5 million people with new federal spending of $190.5 billion.

Holahan — cover 15.2 million people with new federal spending of $127.4 billion.

Hacker — cover 37 million people with new federal spending of $241.9 billion. Average cost to firms that do not now offer coverage is estimated at $1,000 per worker. Firms that currently offer insurance would see savings averaging about $409 per worker.

Weil — cover 37 million people with new federal spending of $160.9 billion. Average cost to firms that don’t offer coverage now is estimated at $1,000 per worker. Firms that currently offer insurance would see savings averaging about $22 per worker.

Wicks — cover 40.3 million people, achieving nearly universal coverage, with new federal costs less offsets of $230.8 billion.

Butler — cover 26.9 million people with new federal spending of $236.1 billion.

Kronick — cover virtually all Americans, except 1.6 million undocumented immigrants and lower-income nonworkers who are difficult to reach, with new federal spending of $551.7 billion, funded with payroll taxes and an increase in the personal income tax. Health spending in firms currently providing insurance would increase by about $197 per worker, while the cost to firms that do not now offer coverage would be about $1,760 per worker. While this proposal has the highest level of new federal spending, it has the lowest level of spending per new person covered.

Mr. Shiels and Mr. Haught said that many of the proposals considered in their study have never been attempted on a broad scale in this country, and there are few data on the likely outcomes of such programs. Also, programs that substantially restructure the health care financing system fundamentally could change consumer, employer, and provider incentives in ways that could have a significant impact on program costs.

(For more information, go to: www.rwjf.org; and www.esresearch.org.)