The Rise of Reform
In early August 1991, Harris Wofford's advisers gathered in a conference room in Philadelphia to discuss the future of his senatorial campaign. The special election, only three months away, promised almost certain defeat for the little-known Democratic senator--and certain victory for his seasoned Republican opponent, former Pennsylvania governor and former U.S. attorney general Richard Thornburgh. Just months earlier, Wofford had been appointed to the Senate by Governor Robert Casey following the untimely death of the Republican incumbent, John Heinz. Now, surveys of the Pennsylvania electorate showed the liberal intellectual roughly 40 percentage points behind Thornburgh, validating the prevailing consensus that Wofford's brief tenure in the Senate would soon be brought to a close.
To those meeting in Philadelphia, however, the race was far from over. In addition to Wofford, the room contained the two principal architects of the campaign, James Carville and Paul Begala; the campaign's pollster, Mike Donilan; and a number of other campaign aides. Donilan passed around the findings of a two-question survey that he had conducted to assess voter's feelings about national health insurance--an issue that Wofford had been emphasizing in his campaign. The two questions were virtually identical. Each asked respondents to choose between Wofford and Thornburgh on the basis of a brief list of each candidate's qualifications. The second question, however, included a single caveat: Wofford wanted to enact national health insurance; Thornburgh did not. By a more than 3-to-1 ratio, voters asked the first question preferred Thornburgh. When national health insurance was mentioned, however, the relative position of the two candidates dramatically reversed, with Wofford skyrocketing from more than a 40-point deficit in the polls to an almost 10-point lead. As Begala would later tell reporters, national health care reform was the policy issue that could "turn goat spit into gasoline."(1)
The significance of Donilan's poll was not lost on Carville, who argued vigorously for making national health care reform the central theme of the campaign. As one participant in the meeting recounted, Carville saw the issue as the only real vehicle for recapturing the race: "Carville said, `I don't think we can win this election any other way. We don't have a hope in hell against [Thornburgh] with the time we have, the name recognition we have, and the money we have. [National health care reform] is what we have, and it's definitely the long bomb, the Hail Mary. But it's a pretty damn good Hail Mary.'"(2)
The others in attendance agreed. Wofford would campaign on national health care reform.
Wofford won Pennsylvania's U.S. Senate race with a startling 55 percent of the vote--a landslide by Pennsylvania standards. But it was how he won that was truly significant. Seemingly overnight, health care reform was an issue that won votes. Post-election polls revealed that for half of Pennsylvania's voters "national health insurance" was among the top two concerns reflected in their voting decision; for almost a quarter of voters it was the top concern.(3) Although national health insurance was expected to appeal to the low-income residents of the Rustbelt and Appalachian portions of the state, Wofford also defeated Thornburgh in Philadelphia's suburban counties, where no fewer than two-thirds of the residents were registered Republicans. Furthermore, he performed much better than expected in the overwhelmingly Republican regions of northeastern Pennsylvania and made significant inroads among the state's most loyal Republicans, the Pennsylvania Germans.(4) The message of the election seemed clear: no candidate could afford to appear indifferent to a policy issue of such broad public concern.
The Senate race in Pennsylvania was not the first time that health care reform appeared on the American political agenda. Comprehensive health care reform received widespread attention at many other points in American history: during the Progressive Era and the New Deal; under Presidents Truman, Kennedy, and Johnson; and in the 1970s. In none of these periods, however, did public pressure for reform play a critical role in spurring political leaders to action. In the Progressive Era, reformers looked not to the public for inspiration but to the social policies of Germany and Britain. During the New Deal, the Truman years, and the 1960s, the impetus for reform largely came from within the executive branch. And although attention to health care reform was more widespread in the 1970s, it was still primarily the province of national political leaders, labor unions, and public interest groups.
By and large, then, past reformers found that their cause evoked sentiments among the citizenry that were lukewarm and superficial. While polls showed that many Americans were receptive to the idea of national health insurance, public support was never strong enough to allow reformers to overcome the cultural, political, and institutional barriers that lay in their path. Most people expressed little interest in the plight of the uninsured or, for that matter, in the rising cost of health care. To the majority of Americans, the system seemed to work.
Viewed in this light, the public discontent manifested in the Pennsylvania election begs an explanation. Why was the campaign strategy chosen by Wofford and his tacticians so effective? To answer this question, we need to understand how a medical system that once muted public concerns became the source of such widespread dissatisfaction.
THINGS FALL APART
With the massive growth in private insurance coverage following World War II and the passage of Medicare and Medicaid in 1965, the task of mobilizing broad popular support for universal health insurance became deeply problematic.(5) The vast majority of Americans were now covered by private and public health insurance, and most were largely unaware of its cost because they paid for it through discreet tax increases and slower real wage growth. Beginning in the early 1970s, however, the rate of increase in real wages decelerated and inflation in the medical sector accelerated. The result was that premiums for employer-provided health insurance began to absorb a higher percentage of the growth in real wages. Indeed, from 1973 to 1989, health insurance premiums paid by employers accounted for more than half the rise in real total compensation per full-time employee.(6)
The cost of health insurance to working Americans also grew more visible. Starting in the late 1970s, many corporations attempted to shift the growing burden of medical spending onto workers. This shift first took the form of insurance deductibles: between 1979 and 1984, the percentage of large corporations requiring their employees to pay deductibles grew almost fourfold, from 14 to 52 percent.(7) Shifting costs from employers to workers provided only temporary relief, however, and business began to search for new ways to moderate medical inflation. The simplest solution, of course, was to stop offering insurance altogether, which many new entrants and small firms did, especially in the growing service and retail sectors of the economy. For larger employers, eliminating or cutting back on dependent coverage became increasingly attractive. Many larger firms also began to pay employee claims themselves, thus limiting the extent to which they subsidized the health care expenses of other firms' employees. Under the terms of the 1974 Employee Retirement Income Security Act, self-insurance held the additional attraction of exempting firms from state taxes on insurance premiums and state-mandated medical benefits.
The increasing cost-sensitivity of employers also further encouraged, sometimes inadvertently, a whole series of trends already under way in insurance marketing and claims management. First among them was the new vigor with which private insurers were seeking out employer groups with lower than average health care costs. Until the end of World War II, subscribers in a given geographic region generally paid the same "community" rate for group policies. This was the prevailing practice, for example, among the nonprofit Blue Cross/Blue Shield plans that dominated the group insurance market until the 1950s. In the early postwar period, however, smaller commercial insurers began to enter the insurance market, undercutting the Blues by offering low-risk groups less expensive policies. By matching premiums to the expected actuarial risk of subscribers, a practice known as "experience rating," commercial insurers eliminated many of the cross-subsidies--from healthy to ill and from young to old--inherent in community rating. Soon the Blues were experience rating too.
Experience rating runs counter to the goal of social insurance, which is to spread risk across the whole of society.(8) Taken to its natural extreme, the logic of experience rating would have left many high-risk individuals unable to afford insurance at all. Yet the early effect of experience rating was moderated by two factors: health insurance was for most firms still an incidental expense, and insurers were unable to predict actuarial risk with any real accuracy. Beginning in the 1970s, however, all that changed. Strapped by rising medical costs, firms began to seek out less expensive insurance policies, and, responding to their pleas, private insurers refined their methods for rating employer groups and avoiding those with the highest health risks.
As firms self-insured in growing numbers, moreover, the market for private insurance began to shrink. By 1988 most of the financial risk of insuring employees had shifted from insurance companies to employers.(9) In general, firms that failed to self-insure did not have enough employees to pool the risk of illness or injury. Compared with large employers, these firms were characterized by higher levels of employee turnover, more variable workforce health, and lower rates of insurance coverage. As the insurance market contracted and fragmented, therefore, rates of coverage among employer groups grew more uneven and the degree of uncertainty surrounding the health and tenure of employees increased. In this environment, the ability of insurers to accurately predict the risk of potential subscribers became crucial. If insurers had accepted all firms on an equal basis, firms that expected to incur higher than average health care costs would have been more likely to purchase coverage than those that did not. To prevent such "adverse selection," insurers resorted to increasingly sophisticated forms of underwriting to screen out or limit the coverage of costly, high-risk applicants.
The increasing pervasiveness and stringency of medical underwriting represented another major change in the insurance market of the 1980s. Underwriting is performed for all forms of insurance, but for health insurance it takes on a uniquely sinister cast. Health care costs are not evenly distributed across the population; in any given year, approximately 20 percent of the population incurs about 80 percent of total medical expenses.(10) As a consequence, it is very much to a health insurer's advantage to screen out subscribers who are likely members of this high-cost group. The object of medical underwriting is thus to identify and avoid insuring those who need protection most.
The first phase of underwriting generally occurs at the level of the firm. Insurers are wary of insuring firms in professions characterized by seasonal employment or a high employee turnover rate (such as restaurants and hotels) or those that present a high risk of occupational illness or injury (such as mining or logging). A firm that does not fall into such "redlined" professions may still be refused insurance if an insurer classifies one or more of its employees as high risk. In some cases, insurers will offer such firms coverage only after excluding high-risk employees. Even after enrolling in a health plan, most subscribers are denied coverage for "pre-existing conditions," usually for a period of six months to two years. When pre-existing condition exclusions expire, insurers frequently raise the premium rate charged to the employer group to offset the predictable increase in claims--a process known as "durational rating." As rates rise, many firms then switch to another insurer, which is likely to impose a new set of pre-existing condition exclusions on the group.
The interplay of pre-existing condition exclusions and durational rating generates substantial turnover in the insurance market, particularly the small group market. As firms seek out lower premiums, insurers constantly step up their underwriting efforts to protect themselves against adverse selection. Because of this churning effect, significant gaps in insurance coverage are a fact of life for employees of small firms. And for all workers, leaving a job can mean forgoing necessary coverage in the future.
A final important trend that was encouraged by employers in the 1980s was the movement by insurers away from traditional indemnity plans and toward managed care.(*) Before the passage of the HMO Act of 1973, the Nixon administration predicted that sixteen hundred new health maintenance organizations would be created by 1980, with 20 percent of the public enrolled.(11) Even accounting for the contradictory design of the HMO legislation, these expectations were wildly unrealistic. Indeed, when federal start-up grants for HMOs were curtailed in 1981, the actual number of HMOs fell far short of the Nixon administration's goal.(12) If the 1970s was a decade of unrealized expectations, however, the 1980s brought unprecedented growth. By 1991 there were more than five hundred HMOs in operation in the United States with a total enrollment of nearly 40 million people.(13)
Proliferating even more rapidly were more loosely organized systems of managed care, such as preferred provider organizations (PPOs) and point-of-service (POS) plans. Members of PPOs and POS plans receive treatment through a limited network of care-givers. In contrast to tightly integrated HMOs, however, physicians and practitioners who contract with these plans are not salaried employees, and patients can seek care outside the network (although coverage is less complete). While HMOs with integrated financing and delivery systems (so-called staff-model HMOs) have been shown to deliver quality care less expensively than conventional fee-for-service insurance (primarily by reducing the hospital use of their members), there is no evidence to suggest that PPOs and POS plans have significantly reduced costs--except, perhaps, to shift them onto other payers.(14) Nonetheless, PPOs and POS plans grew from their inception in the early 1980s to cover about a quarter of American workers by 1991.(15)
With employers demanding lower premiums, even fee-for-service insurance plans rushed to adopt the cost-management mechanisms pioneered by managed care. Utilization review techniques--such as the requirement that patients receive an insurer's authorization before being admitted to a hospital--quickly became part of the standard operating procedure of claims management. Although physicians complained loudly about the growing encroachments upon their clinical autonomy, utilization review held out the promise of one-time cost savings--a prospect that few employers could pass up.(16) By 1990 only 5 percent of employees in the United States remained in plans with no utilization management, down from more than 40 percent in 1987.(17)
Partly as a result of these trends, and partly due to cutbacks in Medicaid, the number of uninsured rose steeply during the 1980s, leaving an estimated 36 million Americans without health insurance in 1991.(18) Among the poor the decline in coverage reversed somewhat in the late 1980s due to an expansion of Medicaid (which covered about half of the poor in 1991). Nonetheless, the uninsured still came predominantly from low-income families. The vast majority of these families were headed by workers, who were generally self-employed or employed by small firms. In a political sense, however, the uninsured hardly formed a group at all. They were faceless and quiescent, without common ties or identification, and more than a quarter were children. So the growing number of uninsured could not by itself be expected to spur political leaders to action. Something more was needed to create significant public pressures for national health care reform. As it turned out, that something was the intrusion of rising health care costs into the medical security of the middle class.
THE MIDDLE CLASS AND NATIONAL HEALTH CARE REFORM
In the summer of 1991, as Harris Wofford climbed back from his early deficit in the polls, James Carville came across an article in Harper's magazine that immediately caught his attention. Its author was Paul Starr, a sociologist at Princeton and author of the Pulitzer Prize-winning book, The Social Transformation of American Medicine. Entitled "The Middle Class and National Health Reform," the article was a reprint of a short piece that Starr had written for the American Prospect, the fledgling liberal quarterly he coedited.
For years Starr had lamented that reformers in the United States would never be able to overcome the formidable roadblocks to national health insurance. Now, however, he saw new possibilities. Although significant political obstacles remained, "the underlying pressures for the adoption of national health reform in the 1990s" were stronger than they had been "for decades." The reason, claimed Starr, was the "increasing jeopardy of the middle class."
In the past, Starr argued, "advocates of universal insurance, like advocates of so many other liberal policies since the 1960s, found themselves ... appealing to middle-class voters to support change, not for their own good, but for the benefit of a minority--and a hazily defined, politically inaudible minority at that.... To the middle-class taxpayer, even the family of the unionized worker, or the elderly protected by Medicare, national health insurance seemed to promise too little for too much."
This was no longer the case, Starr claimed. The efforts of employers to control costs and of insurers to screen out high-risk individuals were "undoing some of the middle-class insulation from health care costs that made it so difficult to construct an alliance for health insurance reform across class lines." Thus health care reform could become the Democrats' "best campaign issue in the nineties."(19)
For obvious reasons, Carville found Starr's thesis persuasive. He called one of Wofford's legislative assistants and ordered him to "get Wofford together with this guy." In early October, Wofford and Starr met at the senator's home in Pennsylvania, and for most of the day they drifted through the house together, chatting about health care reform and its role in the campaign. According to one campaign aide, "They both saw very much eye to eye."(20)
Starr and Wofford were not mind readers of the middle class. It was no secret that public discontent with the cost of health care and the vagaries of insurance coverage was substantial and growing. The percentage of Americans who felt that American health care needed to be fundamentally changed or completely rebuilt increased steadily in the late 1980s, exceeding 90 percent in November 1991.(21) What Starr and Wofford recognized, however, was a subtle but important shift in the character of public opinion. This shift involved not only an increase in overall public support for reform, but also a growing convergence of opinion among Americans of diverse viewpoints, circumstances, and characteristics.
Like public opinion about other major policy issues, general public sentiments about health care have remained relatively constant over the last half-century.(22) Support for medical spending and for the government to "do more" on health care, for example, has been consistently high, with large majorities (60 to 80 percent) favoring increased government spending and assistance. Along with public support for other government initiatives, support for government spending on medical care reached a low point between the mid-1970s and the early 1980s as the economy soured and antigovernment sentiments prevailed. Unlike other domestic programs, however, health care programs rapidly regained public support during the 1980s. By 1987 they were even slightly more popular than they had been in the mid-1970s. Interestingly, neither programs targeted to the poor nor general domestic programs regained the levels of public support they had enjoyed in the 1970s. Indeed, after examining national polling data collected between 1975 and 1989, Mark Schlesinger and Taeku Lee conclude that "only for health care have Americans become distinctly more supportive of government in the late-1980s than in the mid-1970s."(23)
Why did government involvement in health care become more popular relative to other government initiatives? The most likely answer is that health care, in comparison with other domestic policy priorities, became less socially divisive between the mid-1970s and the late 1980s. Although support for government involvement in medical care varies by income, age, education, gender, race, and degree of political involvement, these differences are substantially smaller for health care than for other areas of government policy and may in fact have narrowed slightly over the past fifteen years. More important, federal health initiatives do not evoke the same negative public association as do other redistributive policies, such as antipoverty programs. They are not strongly identified with racial minorities or with the economically disadvantaged, and federal involvement in health care does not appear to be linked in the public's mind to a wider role for government in other areas of American society.(24)
In a number of important ways, then, health care appears to be different from other domestic policy priorities. It is not only more popular than antipoverty programs and general domestic initiatives, but also characterized by greater consensus between rich and poor, old and young. Yet support for greater government involvement in health care does not necessarily translate into support for comprehensive health care reform. Was there a parallel shift in public attitudes toward reform?
As figure 1.1 shows, the increase in public support for comprehensive reform in the years leading up to Wofford's victory was indeed considerable. In 1986 only 15 percent of Americans believed that the health care system needed to be "completely rebuilt," while more than a quarter felt it needed only "minor changes." By the time of the Pennsylvania election, the number of Americans who felt a complete rebuilding of the system was necessary had risen to 42 percent, while those who wanted only minor changes had fallen to 6 percent.(25)
The increase in public support for comprehensive reform in the years leading up to the Pennsylvania election was also quite rapid. Between October 1989 and November 1991, the number of Americans expressing support for a complete restructuring of the health care system jumped from 23 to 42 percent, while those supporting minor changes dropped from 21 to 6 percent. All told, by November 1991 more than 90 percent of Americans believed the health care system needed to be fundamentally changed or completely rebuilt--roughly 20 percent more than had felt that way two years earlier.
What explains this rapid increase in public discontent? Recent research on the dynamics of public opinion suggests that public sentiments mainly shift in response to observable changes in external conditions or to changing patterns of media coverage.(26) I have already outlined some of the changes in the structure and performance of the medical sector that might have prompted public concern--the increasingly visible burden of medical costs, the rising number of uninsured, the increasing stringency and pervasiveness of medical underwriting--but these trends had been under way for more than a decade before the sudden jump in public discontent occurred. No doubt they had some cumulative effect, but if that were the whole story, public support for reform would have built gradually over the course of the decade, rather than increase sharply between 1989 and 1991. Instead, the sudden increase in public discontent appears to have been linked to the national economic downturn that began in 1989 and worsened through 1991.(27) By promoting a sense of economic insecurity, the recession heightened public anxiety about the escalating cost of medical care and the fragility of employer-sponsored health insurance. Moreover, the recession was harder on white-collar and professional workers than were other recent economic downturns, forcing many middle-income Americans who had taken their personal health care arrangements for granted to face the prospect of losing their health insurance or having their coverage cut back. According to the public opinion data assembled by Schlesinger and Lee, all the factors associated with economic slumps--unemployment, disenchantment with the business community, and economic anxiety--increase public support for government involvement in health care, both in absolute terms and relative to other domestic initiatives.(28)
If the recession precipitated the rapid change in public opinion that polls picked up between 1989 and 1991, this would go a long way toward explaining why support for fundamental reform jumped so quickly and dramatically at the same time that Americans remained deeply ambivalent about its proper scope and direction. Americans expressed dissatisfaction with the medical system as a whole, but they remained overwhelmingly satisfied with the quality and accessibility--if not the cost--of the medical care they and their families received. In addition, the extent to which Americans were willing to sacrifice to enable reform remained unclear. Although most Americans supported an expansion of coverage to the uninsured, opinion surveys showed limited public support for restrictions on personal medical care decisions or increased taxation to fund reform. Moreover, the public remained ambivalent about the proper role of government in a reformed system. Polls indicated that Americans were convinced that government had to be part of the solution but at the same time fearful that it would botch the effort.(29)
Finally, public dissatisfaction with the status quo did not embody anything approaching a latent public consensus about the direction reform should take. In opinion surveys, Americans appeared quite ready to embrace any alternative to present insurance arrangements, including a national health insurance system funded through taxes. But when respondents were asked to choose among competing reform approaches, public support tended to split fairly evenly among the alternatives, with levels of support varying greatly with the wording of the survey questions.
These aspects of public ambivalence accord well with Daniel Yankelovich's three-stage model of opinion formation, in which members of the public become aware of a policy issue, weighthe consequences of alternative courses of action, and then arrive at a general conception of what should be done.(30) At the time of the Pennsylvania election, this last stage of opinion development--what Yankelovich terms "public judgment"--had not yet been reached. Although many Americans were anxious about the security of their health care arrangements, few understood the major options for reform being considered by policymakers or the trade-offs those options might entail. What emerged in the late 1980s and found expression in Wofford's stunning upset victory was a negative public judgment about a system of health care finance that Americans found too costly, unreliable, and inhumane. A corresponding positive judgment about the direction reform should take would prove more difficult for policymakers and the public to reach.
MEDIA COVERAGE OF HEALTH CARE REFORM
What role did the media play in drawing public and political attention to health care reform? Since Maxwell McCombs and Donald Shaw first examined the "agenda-setting function of mass media" in the 1968 presidential campaign, communications researchers have demonstrated conclusively that the media affect citizens' perceptions of issue importance.(31) There is also reason to believe that policymakers rely on media for their assessment of public concerns.(32) Whether media amplify the effects of other political actors or actively attempt to set the government agenda themselves, they can play an important role in raising public awareness about policy issues.
Media coverage of health care reform was almost nonexistent through the 1980s. Between 1980 and 1990, for example, three of the most widely read national newspapers--the Christian Science Monitor, New York Times, and Wall Street Journal--together ran a total of forty stories on health care reform or national health insurance. In not one of these years were more than fourteen stories printed, and in four--1981, 1984, 1985, and 1987--there was no coverage of the issue whatsoever. These figures stand in stark contrast to the last year of reform activity, 1979, when a total of almost seventy articles appeared in these three newspapers. As figure 1.2 shows, the total number of articles rose slightly in the latter half of the 1980s, but the real jump took place in 1991, as the total number of articles grew from five to thirty-five, only to be followed by another more dramatic rise in 1992 as the number of articles reached ninety-three.
These trends become clearer when we focus on the six years between 1987 and 1993 and expand the pool of newspapers. Figure 1.3 shows the number of articles mentioning health care reform that appeared in eight widely read newspapers between 1987 and 1993. As before, there is a slight drop in print-media attention in 1990 followed by surges of coverage in 1991 and 1992.
Juxtaposed against the public opinion data and Wofford's victory in 1991, these trends suggest that the press was reacting to, rather than driving, public opinion and political events. If this were the case, the media's role in raising public concern about health care would run counter to the traditional model of media agenda setting in which the public "learns what issues are important from the priorities of the news media."(33) In this instance, the reverse appears to have happened: media coverage of health care reform responded to, rather than precipitated, public and political concern.(34)
There are of course other plausible hypotheses. The rise in public concern and increase in news coverage occurred within a short enough period that both the public and media may have been responding to a common third factor--for example, a significant rise in medical costs or in the number of uninsured. Identifying the exact causal process at work is difficult because public opinion, media coverage, and policymaking all mutually influence one another. The concept of media agenda setting implies a one-way causal process whereby media coverage (the independent variable) changes public issue priorities (the dependent variable). Yet, as Everett Rogers and James Dearing argue, "there is undoubtedly a two-way, mutually dependent relationship between the public agenda and the media agenda in the agenda-setting process."(35) Likewise, there is clearly an interactive relationship between journalists and policymakers. The great bulk of domestic news stories focus on people in government, and media outlets are an important vehicle by which policymakers draw public attention to new policy issues.(36) At the same time, because people in government demand timely political feedback, journalists often become "stand-ins" for the national constituency, shaping policymakers' perceptions of the prevailing "national mood."(37)
The multiple paths of influence that connect the public, news media, and policymakers can make tracing an instance of agenda setting to its ultimate causal source difficult, if not impossible. In this case at least, the rise in public support for health care reform appears to have been a precursor to the increase in news media coverage, and perhaps a cause of it. But what seems most significant is that public opinion, media coverage, and political events all closely tracked one another. The sudden emergence of health care reform onto the national political agenda thus appears to be an example of an "amplifying feedback loop," in which the public, news media, and national policymakers all responded to the cues put forth by the others, creating a rapid cascade of attention to the issue.(38)
HEALTH CARE REFORM AND THE CONGRESSIONAL AGENDA
In 1991 Congress faced a growing number of health care reform proposals. Yet there remained some trepidation among policymakers about addressing the issue. The federal government's last great foray into the wilderness of health care reform, the Medicare Catastrophic Coverage Act (MCCA), had been passed in 1988 and repealed the next year after outraged upper-income senior citizens revolted against the progressive taxes it levied on Medicare beneficiaries. Also ill-fated was the Bipartisan Commission on Comprehensive Health Care (the Pepper Commission). Authorized by the MCCA to come up with a series of recommendations for health care reform, the commission had quickly descended into partisan bickering. Its final recommendations, released in 1990, had been greeted with little fanfare.
Noticeably absent from the debate over health care reform was Republican President George Bush. In his 1990 State of the Union address, the president had called upon Health and Human Services Secretary Louis Sullivan to craft a response to the issue. But dissension within the administration and between Bush and congressional Republicans over the appropriate character of that response had left the White House on the sidelines and Republican members of Congress bereft of executive leadership.
While President Bush faltered, the Democratic leadership in the Senate pushed ahead with its comprehensive health care reform initiative, Health America. In early June, the sponsors of Health America--Senators George Mitchell, Ted Kennedy, Don Riegle, and Jay Rockefeller--made an impassioned plea for support to an assembled hall of interest-group representatives and corporate lobbyists, warning them that the failure of the bill and its play-or-pay approach would open the door to more disruptive reforms.(39) But other Democratic members of the Senate and House wanted to go further. In the spring of 1991, Representative Marty Russo of Illinois, a member of the pivotal House Ways and Means Committee, introduced a single-payer proposal to replace all existing private and public insurance with a universal public program.(40)
One way to estimate changes in the level of congressional attention to health care is to count the number of hearings that were held on the topic by congressional committees and subcommittees.(41) Yet hearings have their shortcomings as a measure of agenda status. Many hearings on health care are concerned with routine or ongoing issues surrounding existing health programs, including the periodic reauthorization of programs. Many others concern topics only tangentially related to health care reform, such as medical research or the process by which new drugs are approved. As figure 1.4 reveals, the total number of congressional hearings on health care remained consistently high from 1980 through 1991, exceeding 100 hearings per year after 1983 and topping 150 hearings annually in 1987 and 1991.(42) Although the number of hearings generally increased over this twelve-year period, the level of congressional attention remained remarkably high throughout. In every year from 1980 through 1991, Congress appears to have paid a considerable amount of attention to topics surrounding health policy.(43)
The picture looks somewhat different, however, if we consider only those congressional hearings that focused on medical costs or the uninsured.(44) Figure 1.5 shows the total number of congressional hearings that were held annually on these topics over the same twelve-year period. The volume of hearings is smaller in this case, and it increases more dramatically between 1980 and 1991. The number of hearings rises erratically through the mid-1980s, jumps sharply in 1989, falls back in 1990, and then rises again in 1991.
There is a similar but more dramatic increase in another indicator of legislative activity--the volume of legislation introduced in Congress.(45) Figure 1.6 shows the total number of bills on the subject of national health care introduced in each Congress from the 96th (1979 and 1980) through the 102nd (1991 and 1992).(46) The number of bills remains fairly meager until the 100th Congress (1987 and 1988), then rises precipitously, from 13 bills in the 96th Congress (1985 and 1986) to 63 in the 100th Congress to around 100 in both the 101st (1989 and 1990) and 102nd Congresses. Perhaps most surprising is the considerable legislative activity that surfaced in the 101st Congress, well before the Pennsylvania election. Indeed, the number of bills actually fell slightly in the 102nd Congress, from 103 to 92. However, this largely reflects the large amount of legislation pertaining to the MCCA that was introduced between 1987 and 1989. When tabulated using a more narrow definition of health care reform, the number of bills in the 100th Congress showed no increase, and the number in the 101st Congress rose only modestly.(47) By this narrower definition, as figure 1.7 shows, the real jump in legislative activity took place in the 102nd Congress, during which the number of bills rose from 17 to 63. About two-thirds of these bills--41 of the 63--were introduced after the Pennsylvania election.
Taken as a whole, the pattern of congressional hearings and legislative introductions suggests two basic trends in congressional attention to health care reform. First, there was clearly a significant degree of congressional attention to health care reform even before the Pennsylvania election, with the subject beginning to rise steeply on the congressional agenda in the late 1980s. Second, like both public support for reform and media coverage of reform, congressional attention to the subject appears to have jumped sharply in the 102nd Congress (1991 and 1992), in the run-up to and immediate aftermath of the Pennsylvania election.
MOMENTUM TOWARD REFORM IN CONGRESS
What explains the momentum toward health care reform in Congress in the years immediately preceding the Pennsylvania election? We have already surveyed some of the proximate causes--deteriorating conditions in American medicine, rising public anxiety, and a sharp economic downturn--but three others seem comparably important. The first was the increasing pressure public health programs were placing on state and federal budgets. By the late 1980s, the escalating cost of the two largest federal health programs, Medicare and Medicaid, was generating widespread unease in Congress. In 1982 Congress had passed legislation requiring the Health Care Financing Administration to develop new methods for paying hospitals under Medicare. By the time of the Pennsylvania election, Congress had also enacted legislation creating fee schedules for physicians serving Medicare beneficiaries. Despite these measures, however, the cost of public health programs continued to rise rapidly.(48)
A second proximate cause of health care reform's enhanced stature on the congressional agenda was the emergence in the late 1980s of a revitalized Democratic leadership in the House, centered around the chairs of the major fiscal committees and the Speaker of the House. A key reflection of the Democratic party's stronger leadership style was the aggressive legislative activism of the new Speaker, Jim Wright of Texas, who entered office in 1987 with an ambitious list of legislative goals that included catastrophic health insurance. Although it was President Reagan who first drew national attention to catastrophic health insurance with some offhand comments at a presidential news conference in January 1986, leading members of Congress quickly made the issue part of their own more far-reaching health policy agendas.(49) The return of the Senate to Democratic control in the 1986 election placed both houses of Congress in Democratic hands, and shortly thereafter Reagan's standing was further weakened by the Iran-Contra scandal. Wright and other members of the leadership were able to take advantage of Reagan's faltering stance, coupled with more than a decade of growth in the power of the Speaker and the Democratic Caucus, to pursue a partisan agenda more forcefully than the House leadership had since the beginning of Reagan's presidency. In the end, neither Wright, who resigned under a cloud of ethics violations in 1989, nor catastrophic health insurance, which was repealed the same year, proved to have much lasting impact. Yet the new assertiveness of the Democratic leadership on catastrophic health insurance and other policy issues marked the arrival of a congressional leadership more willing to define its policy commitments and use its expanded leadership prerogatives to achieve them, even in the face of presidential opposition.(50)
Finally, increased congressional attention to health care reform had a third important proximate cause: the growing demands for reform being voiced by the major stakeholders in the health policy domain. By the time of the Pennsylvania election, many of the organized interests in health care had expressed support for some kind of reform. Even the American Medical Association (AMA), the most vociferous opponent of reform in the past, had proposed an employment-based plan to improve access to medical care.(51) Other provider groups--such as the American College of Physicians and American Academy of Family Physicians--appeared willing to accept even more comprehensive reforms. Business was also beginning to countenance a greater federal role in the financing of medical care after more than a decade of futile attempts to handle costs on its own.
At first glance, this chorus of interest-group support for reform appears puzzling. As Lawrence Brown asks, "If a formidable phalanx of powerful groups throttled change in the 1980s, why did they seemingly cease to do so in the early 1990s?"(52) The answer, paradoxically, is that the seeds of interest-group discontent were sown by the very success of these powerful stakeholders in preventing substantial policy change--not just in the 1980s, but during most of this century. By the 1970s continued cost increases were driving both the government and private employers to take more active steps to rein in medical inflation. In the 1980's these efforts began to produce results but not exactly the results their architects had intended. Federal cost-containment initiatives were shifting costs onto the private sector, the growing microregulation of clinical decisions by insurers was irritating providers and patients alike, and corporate America's quest for lower insurance premiums was further segmenting the insurance market and lowering coverage rates among individuals and small firms. In the past, when the nation's capacity for absorbing rising medical spending appeared virtually limitless, stakeholders had seen little reason to challenge one another over the exact division of the spoils of medical progress. But now these same stakeholders saw the conflict as zero-sum--one group's gain came at the expense of another--and each believed that it was uniquely disadvantaged in relation to its antagonists. Accordingly, medical providers, insurers, and business all began to put forward proposals for reform that they believed would restore the proper balance.
If the new aggressiveness of the Democratic leadership and the changing stance of powerful stakeholders were important proximate causes of the increased congressional attention to health care reform, they also reflected a broader transformation of the structure of American health politics. This transformation can be viewed as the deep or underlying cause of the increased congressional attention to reform, for it significantly increased the probability of a renewed legislative push for national health care reform.
Mark Peterson describes this transformation as a shift from an "iron triangle," in which powerful private interests had privileged access to an oligarchically organized Congress, into a more open "policy network," in which power was widely dispersed among interest groups and individual legislators.(53) The point of departure for Peterson's analysis is the early postwar period, when the medical lobby and a cross-party conservative coalition in Congress effectively blocked consideration of President Truman's proposal for national health insurance. Led by the AMA, the organized interests opposing Truman's proposal were unified and powerful--more than a match for the scattered band of supporters of the bill. By the 1960s and the debate over Medicare, the interest-group community had grown more polarized and its opposing camps more evenly paired, leading one observer to describe the Medicare struggle as a classic illustration of "class-conflict politics."(54) But the crucial shift occurred in the two decades following Medicare's passage, as new "stake-challengers" entered the political arena and old stakeholders had increasing difficulty maintaining their traditional alliance. In 1945 the dominant groups in the health policy domain had been profit and nonprofit sector associations, with organized labor and citizen groups serving as the main political counterweight. From 1945 to 1985, and especially after 1970, the proportion of citizen groups and mixed profit-nonprofit organizations rose, while the proportion of profit-sector groups declined. More important, newer groups were more likely than their older counterparts to endorse increased government support for health care, even in sectors historically antagonistic to government involvement.(55)
Not only did the composition of interest groups in the health policy domain change in the decades after World War II, the aggregate number of groups increased as well. Since the 1960s, there has been a dramatic rise in the number of interest groups operating in all federal policy areas, but in none more so than health. The number of interest groups active in health and social policy rose from 674 in 1960 to almost 4,000 in 1990--an almost six-fold increase. In comparison, the number of groups in all other policy areas increased by about a factor of four over the same thirty-year period.(56)
Paralleling the transformation of interests in the health policy domain were dramatic changes in the organization of Congress. The oligarchic Congress of the Truman era, dominated by conservative committee chairs, soon ran afoul of the liberal tide of the 1960s. In 1964 liberal Democrats tamed the Rules Committee by reinstating the twenty-one day rule, which prevented the committee from stalling bills for more than three weeks. During the 1960s, congressional staffs grew significantly, providing members with new independent resources for policy advocacy. And beginning in 1971, the Democratic Caucus instituted a series of rule changes that expanded the number of subcommittees and increased their power vis-a-vis full committees.
These institutional reforms ushered in a new congressional order, more supportive of legislative entrepreneurship and less conducive to the formation of entrenched policy subsystems than the oligarchy before it. Even rank-and-file members now had the resources and discretion to play a leadership role on policy issues, and many more committees and subcommittees participated in the making of health policy. In the House, for example, health care issues were divided among no fewer than seven major committees. And for the first time, all but one of the House and Senate committees and subcommittees with a significant jurisdictional interest in health care reform were chaired by members who supported reform.(57)
This dispersion of leadership among congressional committees is evident in the distribution of congressional hearings on health care. From 1980 to 1991, hearings on health care were divided among more congressional committees than hearings on any other major policy issue before Congress. Furthermore, no single committee enjoyed clear jurisdictional dominance. The committees with the largest share of hearings--the House Select Committee on Aging and the Senate Labor and Human Resources Committee--each accounted for only around one-quarter of the total number of hearings held in their respective parent bodies. Four committees in the House and three in the Senate held more than one hundred hearings each between 1980 and 1991. No other policy area is characterized by this degree of jurisdictional fragmentation.(58)
The effects of congressional reforms on the prospects for achieving national health care reform were not uniformly favorable. Although the reforms broke the legislative stranglehold of baronial committee chairs, and with it the privileged influence of the medical lobby, they also made it difficult for congressional chairs and party leaders to construct coalitions and broker legislative compromise.(59) But the effects of these changes on the prospects for addressing national health care reform were not so ambiguous. Never before had so many members of Congress been poised to play a leadership role on the issue, or possessed the analytic resources needed to develop serious proposals for reform.(60) By the 1990s, the structure of power in Congress and the interest-group community was uniquely hospitable to a renewed legislative campaign for national health care reform.
FNT[(*) Since the term managed care is often used loosely, I should make clear that I employ it to mean network health plans. These are plans that "restrict coverage in whole or part to services provided by a specified network or group of physicians, hospitals and other health care providers." Marilyn J. Field and Harold T. Shapiro, eds., Employment and Health Benefits: A Connection at Risk (Washington, D.C.: National Academy Press, 1993), 100. Network plans include all the plans mentioned in this chapter and several variants thereof. They do not include conventional fee-for-service insurers that exercise some oversight over physician, hospital, and patient behavior.]FNT (*) Since the term managed care is often used loosely, I should make clear that I employ it to mean network health plans. These are plans that "restrict coverage in whole or part to services provided by a specified network or group of physicians, hospitals and other health care providers." Marilyn J. Field and Harold T. Shapiro, eds., Employment and Health Benefits: A Connection at Risk (Washington, D.C.: National Academy Press, 1993), 100. Network plans include all the plans mentioned in this chapter and several variants thereof. They do not include conventional fee-for-service insurers that exercise some oversight over physician, hospital, and patient behavior.
Return to Book Description
File created: 8/7/2007
Questions and comments to: email@example.com
Princeton University Press