Skip to main content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.


The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Medicare, Accountability, and Structural Reform

Publication Date
Sep 4, 2000

Medicare, Accountability, and Structural Reform

Full Report

[ Main Page of Report | Contents of Report ]

In this report to the Secretary, I discuss several alternative ways to administer the Medicare managed care program, including a reconstituted HCFA(now known as CMS), a single-headed agency outside of HCFA(now known as CMS), and an independent board or commission. I set the stage by developing an analytical framework for thinking about bureaucratic accountability and by reviewing empirical evidence on the consequences of administrative reorganizations and the establishment of boards or commissions.


Members of Congress and other interested parties have expressed frustration and disappointment over the slow progress of the Medicare Plus Choice program, which makes managed care options available to senior citizens. Since its creation in 1997, the Medicare Plus Choice program has evoked mixed but often wary reactions from the managed care community and from the elderly. After an initial flurry of excitement, many managed care plans have withdrawn entirely from participation in Medicare Plus Choice or sharply curtailed their involvement (HIAA 2000). And the penetration rate for Medicare managed care (16%) pales in comparison to the penetration rate for Medicaid managed care (54%) or for the population at large (71%).

Critics of the Health Care Financing Administration(now known as Centers for Medicare and Medicaid Services(CMS)) (HCFA(now known as CMS)) have blamed HCFA(now known as CMS) for these disappointments. They contend that HCFA(now known as CMS) favors the fee-for-service sector over managed care and that HCFA(now known as CMS) throttles the managed care industry with unnecessary rules, regulations, and restrictions. For example, Rep. Thomas Bliley (R.-Va.) has charged that HCFA(now known as CMS) micromanages the Medicare program in excruciating detail. (Harris 2000) Critics also contend that HCFA(now known as CMS) suffers from a conflict of interest that makes it difficult for the agency to deal fairly and objectively with the managed care industry (Butler 1999).

Such criticisms of HCFA(now known as CMS) may well be too harsh. The sluggish response of the managed care industry to the Medicare Plus Choice program is due primarily to the changing economics of the health care market and to statutory requirements beyond HCFA(now known as CMS)s control. The Medicare Plus Choice program was unveiled at a time when managed care was proving less attractive financially to many companies (Lemov 2000). In addition, many of HCFA(now known as CMS)s policies are direct extensions of congressional mandates. For example, the rates HCFA(now known as CMS) allows Medicare managed care companies to charge are lower than companies would wish primarily because the Balanced Budget Act of 1997 requires this (Scanlon 1999). Clearly, HCFA(now known as CMS) enjoys less discretion in its dealings with Medicare contractors than the Office of Personnel Management (OPM) does in its dealings with contractors who participate in the Federal Employees Health Benefits Program (FEHBP), as Fuchs and Potetz (2000: 41) have noted. In these and other areas, HCFA(now known as CMS)s policies reflect congressional policies, which Congress may wish to rethink.


One idea that has generated considerable interest in recent months is the possibility of creating a new Medicare board to advise HCFA(now known as CMS) on Medicare policy or to assume control over the Medicare managed care elements of HCFA(now known as CMS)s portfolio. For example, Senators John Breaux (D.-La.) and Bill Frist (R.-Tenn.) originally proposed legislation creating a Medicare Board that would strip Medicare managed care from HCFA(now known as CMS)s jurisdiction (Breaux 2000). In contrast, one can easily imagine a more modest reform: a Medicare Advisory Board that would offer recommendations to HCFA(now known as CMS) on managed care issues, fee-for-service issues, or both.

Neither idea is unprecedented. During the Progressive Era Congress frequently created independent regulatory commissions to handle such technical matters as rate-setting, market entry and exit, etc. In the Nixon era, Congress created a Postal Rate Commission (1971), a Consumer Product Safety Commission (1972), and a Federal Election Commission (1974), even though Nixon preferred single-headed agencies to commissions (Eisner 1993: 128). In the Clinton era, Congress created a Social Security Advisory Board (1994), an Internal Revenue Service Oversight Board (1998), a Competitive Pricing Advisory Commission (1997), and other bodies.


These are by no means the only options if one seeks to improve the performance of the Medicare program generally or the Medicare Plus Choice program in particular. Other options include enhanced oversight by Congress, stronger internal review by the HHS Inspector General, administrative reorganization within HHS, or administrative reorganization within HCFA(now known as CMS) itself.

Congressional oversight is a widely used tool for monitoring and improving bureaucratic performance. Congressional oversight became more frequent and more vigorous during the early 1970s, judging by the number of oversight hearings and the mass media attention they generated (Aberbach 1990). In 1993, Congress substantially enhanced its oversight capacity by enacting the Government Performance and Results Act (GPRA), which requires federal agencies to produce strategic plans, performance plans, and ultimately performance reports, all on an annual basis. As of March 2000, the building blocks for GPRA are all in place.

Beginning in 1976 Congress created offices of the inspector general within all of the Cabinet departments. Such offices are charged with investigating waste, fraud, and abuse at HHS and other administrative agencies. The inspectors general enjoy considerable autonomy within the agencies they investigate and, more broadly, within the executive branch (Light 1993).

Another possible solution is administrative reorganization within HHS. For example, Rep. Bill Thomas (R.-Calif.) has proposed creating a Medicare Benefits Administration (MBA), within HHS, which would be responsible for Medicare Plus Choice and a new prescription drug program. HCFA(now known as CMS) would continue to handle the fee-for-service Medicare program, and the Secretary of HHS would arbitrate any disputes that arise.

A final possibility is administrative reorganization within HCFA(now known as CMS) itself. In 1995, HCFA(now known as CMS) eliminated its Office of Managed Care, creating instead three units that were designed to handle payment, policy, and contracting functions, respectively. On June 8, 2000, HCFA(now known as CMS) proposed consolidating these functions so that there will once again be a single office within HCFA(now known as CMS) that handles Medicare managed care issues.


One way to think about bureaucratic accountability is that it is inversely related to bureaucratic independence (Garcia 2000). Thus independence can be purchased at a price, and that price is a loss of accountability. In some respects, that is how both the Progressives and President Franklin Roosevelt viewed accountability, though they ultimately tilted in different directions, with the Progressives favoring greater independence, as exemplified by independent regulatory commissions, and Roosevelt favoring greater accountability, as epitomized by the Brownlow Commissions emphasis on strict adherence to Weberian principles of command and control (Moe and Gilmour 1995; Behn 2000).

There are, however, two problems with this way of thinking. First, it is by no means clear that there is a zero-sum relationship between accountability and independence. Conceivably, one might enhance bureaucratic independence without eroding accountability, by substituting one kind of accountability (e.g., ex post) for another (e.g., ex ante). Second, it is quite clear that accountability is not a unidimensional concept. To speak of accountability without specifying the kind of accountability one means is misleading. Ultimately, we need to think of accountability as a multidimensional concept. The question is not whether we will have more or less accountability but rather what kind of accountability we are willing to diminish and what kind of accountability we wish to enhance.

Consider, for example, a framework developed by Barbara Romzek (1998), which distinguishes between high and low autonomy, external and internal control. According to Romzek, such distinctions yield four types of accountability: legal, political, hierarchical, and professional. Legal accountability involves low agency autonomy, external control; political accountability involves high agency autonomy, external control; hierarchical accountability involves low agency autonomy, internal control; and professional autonomy involves high agency autonomy, internal control.






(Balanced Budget Act of 1997; Medicare Review Board, as proposed)


(Government Performance and Results Act of 1993; Social Security Advisory Board, created 1994)


(HCFA(now known as CMS) reports to Secretary of HHS;MBA and HCFA(now known as CMS) report to Secretary of HHS, as proposed)


(National Performance Review; Office of the Actuary, HCFA(now known as CMS))


HMOs Mandates
Hospitals Accreditation Can Substitute for Licensing


In Table 1 I offer some applications of Romzeks typology that are potentially relevant to the Medicare reform debate. The status quo exemplifies hierarchical control, with the HCFA(now known as CMS) administrator reporting to the Secretary of HHS, who in turn reports to the President. The proposed creation (or re-creation) of a separate managed care office within HCFA(now known as CMS) would not disturb that paradigm but rather would work within it. Existing legal controls such as the Balanced Budget Act of 1997 and proposals to create a separate Medicare Board with operational responsibilities exemplify legal control, though the mechanisms are rather different. The common denominator here is that both limit HCFA(now known as CMS)s autonomy considerably, with critical decisions being made by an external body (either Congress or an independent board). The Government Performance and Results Act of 1993 is an excellent example of political control, as is the Social Security Advisory Board created in 1994. The former strengthens accountability to Congress, while the latter strengthens accountability to an independent board. Despite these differences, neither poses a serious threat to the line agencys discretion. Finally, the National Performance Review, championed by Vice President Al Gore, illustrates professional accountability, because it relies on strong professional norms within the civil service to achieve such goals as efficiency and customer service. The Office of the Actuary within HCFA(now known as CMS) also illustrates professional accountability, because that office promotes good accounting principles and practices from within HCFA(now known as CMS), without eroding the agencys discretion.

Romzeks typology sharpens and clarifies the debate over accountability by highlighting two key dimensions of accountability (degree of autonomy, source of control) and by introducing a useful vocabulary for discussing accountability: legal vs. political vs. hierarchical vs. professional. Useful though it is, however, Romzeks typology is incomplete.

Consider, for example, the distinction between internal and external sources of control. It is easy enough to distinguish between directives from the Secretarys office (internal control) and directives from the Congress (external control). But what about investigations by the HHS Inspector General? I would characterize the Inspector Generals office as a counterbureaucracy because it is a relatively durable government agency whose principal mission is to monitor, criticize, and improve the performance of other government agencies. (Gormley 1996: 276) Although Romzek would classify the Inspector Generals office as an external control, that is not altogether satisfactory, as it is external to HCFA(now known as CMS) but internal to HHS. Such an agency is better viewed as falling somewhere in between the internal and external poles (see Table 2).






internal   external
(Secretary’s office) (Inspector General) (Congress)




catalytic hortatory coercive
(Social Security Advisory Board;GPRA)   (Medicare Review Board, proposed)




ex ante   ex post
(Medicare Review Board, proposed;Social Security Advisory Board)   (Congressional oversight hearings;GAO reports)




process   results
(Administrative Procedure Act)   (Government Performance & Results Act)


Or consider the distinction between high and low autonomy. At one extreme, one can think of catalytic controls so gentle that they run the risk of having no impact on administrative agency behavior; at the other extreme, one can imagine coercive controls so rigid that they unwittingly encourage dysfunctional behavior (Gormley 1989). An advisory board such as the Social Security Advisory Board illustrates a catalytic control because it can merely make recommendations to the Social Security Administrator; a Medicare Board with operational responsibilities would represent a coercive control because it would literally substitute its judgment for HCFA(now known as CMS)s. But what about congressional oversight? In fact, congressional oversight is stronger than a mere catalyst but weaker than outright coercion. It involves bargaining and persuasion, with the threat of punishment if acceptable bargains arent struck. I would classify it, for the most part, as a hortatory control. At the moment, GPRA is a catalytic control because it encourages administrative agencies to shift their focus from outputs to outcomes without penalizing those agencies that fail to do so. Should financial rewards and penalties begin to flow from GPRA, then it too would become a hortatory control.

Another important distinction, not captured by Romzeks typology, is that between ex ante (before the fact) and ex post (after the fact) controls. Congressional oversight tends to be ex post or retrospective in nature. Because it is so difficult to specify in advance what a bureaucracy needs to do, Congress often relies on oversight hearings to generate feedback on policy implementation. Political scientists sometimes refer to this as fire alarm oversight because it depends upon citizens and interest groups to sound the alarm when a program isnt working as intended (McCubbins and Schwartz 1984). Structural reforms such as the creation of a regulatory board, an oversight board, or an advisory board, represent an alternative approach. By creating such a board, Congress tries to ensure in advance that certain interests will be represented or favored when important decisions are made. Political scientists sometimes refer to this as deck stacking because it stacks the deck in favor of certain interests (McCubbins et al. 1989).

A final distinction, also not tapped by Romzek, is that between process-oriented and results- oriented controls. Traditional oversight mechanisms, both ex ante and ex post, have been process- oriented in that they have sought to ensure that administrative agencies adhere to certain procedures when formulating rules or enforcing the law (e.g., environmental impact statements, regulatory impact analyses, or due process requirements under the Administrative Procedures Act). More recently, Congress has sought to influence agency accomplishments directly and not through such mediating variables as administrative procedures. GPRA is a broad-scale effort to do this. A related approach encourages administrative agencies to emphasize performance in their dealings with private contractors. For example, the Michigan Legislature has authorized that states health agency to award more clients to Medicaid HMOs with better track records, as measured by their performance on the Health Plan and Employer Data and Information Set (HEDIS).


Although Congress has created a number of boards and commissions in recent years, those agencies differ markedly from their Progressive Era predecessors. During the Progressive Era and extending into the New Deal, Congress created independent regulatory commissions with actual authority to make binding decisions of considerable importance (the FTC, the FCC, the FPC, etc.). The hope was that these agencies would be independent of politics, especially executive branch politics. In contrast, the boards and commissions created in recent years have usually been advisory in nature. Furthermore, they exemplify what Romzek calls political accountability. Far from being independent of politics, they were designed to ensure that certain interests would be effectively represented into the future.

Consider, for example, the Social Security Advisory Board, created by statute in 1994. By law, the Board consists of seven members, including three appointed by the President and confirmed by the Senate, two appointed by the House, and two appointed by the Senate. Staggered terms ensure some degree of continuity and independence. The statute offers little explicit guidance on who should be appointed, other than to say that they should be persons of integrity, impartiality, and good judgment.

Using the nomenclature outlined above, the Social Security Advisory Board exemplifies a) external controls (Board members do not serve at the pleasure of the Commissioner); b) ex ante controls (methods of selection and replacement, specified in advance, shield the Board from hierarchical pressure by the Commissioner); c) catalytic controls (the Board makes recommendations, not decisions); and d) process-oriented controls (the Board is enjoined to engage in certain activities, not to produce certain results). Although created in 1994, the Social Security Advisory Board did not actually begin to conduct business until the spring of 1996 because it took that long to get Board members appointed. Since that time, the Board appears to have played a constructive role, issuing reports on such subjects as how the SSAs disability program might be reformed and how the SSA might improve its service to the public. At least two recommendations have led to concrete steps by the Commissioner: to establish a strong Office of Policy and Research; and to develop a long-term plan for the year 2010 after conferring with various stakeholders.

The IRS Oversight Board, created by statute in 1998, embodies many of the same principles. By law, the Board consists of nine members, including the IRS Commissioner and the Secretary of the Treasury. Like the Social Security Advisory Board, the IRS Oversight Board illustrates external, ex ante, catalytic, process-oriented controls. Unlike the Social Security Advisory Board, the IRS Oversight Board was deliberately designed to ensure that certain interests are represented. Instead of designating a slot for one interest, another slot for a second interest, etc., the Congress drew up a list of skills that should be represented on the Board as a whole: management of large service organizations; customer service; federal tax laws; information technology; organizational development; the needs and concerns of taxpayers; and the needs and concerns of small businesses. Early indications are that the Clinton administration took those skills into account when drawing up a list of nominees. However, two years after the legislation was passed, the IRS Oversight Board has yet to meet, because none of the discretionary appointees has been confirmed by the Senate (a hold by one senator and a retaliatory hold by another senator have delayed Senate confirmation). Therein lies a cautionary tale: the establishment of a board is, in various ways, more problematic than the establishment of a new single- headed agency. It may take years before a new board is up and running. Then, once established, it may take a while before the board develops decisionmaking norms and routines. Even beyond that, it may be more difficult for a board to reach a consensus.

In contrast to these relatively new and largely unproven organizations, we know a good deal more about the independent regulatory commissions so popular during the first half of the 20th century. First, commissions with relatively narrow jurisdictions have sometimes been captured by the industry they were charged with regulating (Bernstein 1955). Commissions with broader jurisdictions have been less vulnerable to this pathology (Macey 1992). Second, the management of a commission requires adroit political leadership; the chair of a commission must possess certain skills that a single-headed agency does not require (Welborn 1977). Sometimes it takes years before a commission fully realizes its potential, and progress is often fitful at best (Johnson 1990).


Administrative reorganizations have become commonplace (Szanton 1981; Hult 1987; March and Olsen 1989: 69-94). Some are imposed from outside the bureaucracy; others are designed from within. Some seek to achieve genuine substantive reform; others seek to placate key constituencies. Some administrative reorganizations merge and consolidate organizations with similar or complementary functions; others disaggregate functions previously housed within a given agency. Some reorganizations endure, while others do not. As I noted some years ago (Gormley 1989: 119), administrative reorganization has become the cod liver oil of government an all-purpose cure for whatever ails the body politic.

Depending on how it is designed, a reorganization may promote either hierarchical, professional, legal, or political accountability. A key question is whether the reorganization preserves clear lines of authority between the reconstituted agency and the President. If it does, the reorganization qualifies as an effort to promote hierarchical accountability. Of course, the reorganization may not succeed in achieving that objective; or it may promote hierarchical accountability without improving performance.

The Food and Drug Administration has reorganized itself or been reorganized by others on many occasions (Brannon, 1984). Some of its reorganizations have succeeded, while others have failed. A reorganization that failed was the transfer of the Division of Biologics Standards from the National Institutes of Health to the FDA in 1972. The hope was that the fusion of the two agencies would bring greater research expertise to the FDA, greater regulatory acumen to Biologics. Unfortunately, that hope failed to materialize, resulting eventually (in or near 1988) in a decision to separate the two units into a Center for Drug Evaluation and Research and a Center for Biologics Evaluation and Research, both within the FDA. In contrast, the merger of the Bureau of Radiological Health and the Bureau of Medical Devices, both within the FDA, proved much more successful. The merger, which took place in or near 1983, enabled the reconstituted unit, known as the Center for Devices and Radiological Health, to trim its staff by taking advantage of economies of scale. According to close observers, the successful merger would not have been possible without the strong leadership of John Villforth, who served as the head of the new Center during this crucial period.

The most prominent proponent of administrative reorganization in contemporary American politics was President Jimmy Carter, who launched numerous reorganizations, including the creation of two new Cabinet-level departments. One of Carters legacies is the Federal Emergency Management Agency (FEMA), created in 1979. FEMA consolidated disaster relief and civil defense units scattered throughout the federal bureaucracy, including the Defense Civil Preparedness Agency (within DOD), the Federal Disaster Assistance Administration and the Flood Insurance Administration (within HUD), the U.S. Fire Administration (within Commerce), and the Federal Preparedness Administration (within GSA). FEMA was roundly criticized during the 1980s and the early 1990s for being too slow and bureaucratic in responding to natural disasters such as earthquakes, tornadoes, and floods. According to some critics, one of FEMAs problems was that it combined too many functions in one agency. One observer characterized FEMA as a check-writing agency, an intelligence agency, a social service agency and an insurance agency, with a fire administration thrown in. (Ferrara 1995: 291). Were the definitive history of FEMA written in the early 1990s, the lesson might well have been that FEMA was a failure, in part because it was expected to do too many things. More recently, however, observers have concluded that FEMA is a vibrant, innovative, proactive agency (Kettl 2000). FEMAs administrator, James Lee Witt, is credited with having transformed the agencys culture and improved the agencys performance. Thus the lesson seems to me that an agency with multiple responsibilities may be overwhelmed by the challenges it faces but that such an agency can succeed brilliantly in the hands of an able political executive who enjoys strong political support. One wonders whether Witt would have been equally successful had it been necessary for him to negotiate with a wide variety of agencies throughout the federal bureaucracy.

What have we learned about administrative reorganization and its consequences? First, it is disruptive. After discussing numerous reorganizations of law enforcement agencies in the U.S., Ferrara (1995: 39) concluded, All of these reorganizations caused tension and confusion among agency staff, altered organizational cultures, and, some observers argue, had little effect on the quality of drug law enforcement. James Q. Wilson (1989: 267-268) reached similar conclusions. Although positive long- term consequences flow from some reorganizations (Hult 1987), short-term disruption is the one certain consequence. Second, externally imposed reorganizations are less likely to succeed than internally propelled reorganizations (Frederick Mosher, cited in Wilson 1989: 265; Brannon, 1984.). Externally imposed reorganizations often seek to achieve legal or political accountability, while internally propelled reorganizations often seek to achieve hierarchical or professional accountability. Externally imposed reorganizations are less likely to be rooted in sound policy theory, less likely to take organizational culture into account, and more likely to trigger bureaucratic resistance. Third, administrative reorganizations work best when they combine organizations with similar functions (Hult 1987). They work worst when they combine organizations with dissimilar functions or when they separate organizational units with similar functions.


Proposals to restructure Medicare range in scope and in their specifics. Some would transfer responsibility for Medicare managed care from HCFA(now known as CMS) to a new organization or an existing agency. Others would allow HCFA(now known as CMS) to continue to handle these issues but would subject HCFA(now known as CMS) to greater oversight from an executive branch board, from the HHS Inspector General, or from Congress.

One option, urged by Gail Wilensky, former HCFA(now known as CMS) Administrator, is to transfer responsibility for Medicare managed care from HCFA(now known as CMS) to the Office of Personnel Management. Wilensky argues that this approach would locate responsibility for an increasingly important function in the Executive Office of the President, thus heightening accountability to the President or what Romzek would refer to as hierarchical accountability. Wilensky (2000: 7) also notes that OPM is the place that currently is involved in negotiations with private plans and with providing oversight and education about the plans.

This option would reduce HCFA(now known as CMS)s workload to what Wilensky and others regard as a more manageable level. It would eliminate any conflict of interest that may exist under current arrangements, because HCFA(now known as CMS) would not be responsible for both fee-for-service care and managed care. It would also give OPM a chance to replicate what some regard as a very successful program, namely the Federal Employees Health Benefits Program (FEHBP).

A key problem with this proposal, however, is that OPMs experience in managing a health care program for relatively healthy federal employees may not equip it to manage a much larger program aimed at the sick, the disabled, and the disadvantaged. As Cain (1999: 30) has pointed out, OPM and HCFA(now known as CMS) currently serve very different clienteles, with OPMs clients being easier to serve. The two agencies also operate under sharply different statutory arrangements. Whereas HCFA(now known as CMS)s enabling legislation is detailed, prescriptive, and voluminous, OPMs enabling legislation is brief and not very restrictive. If Congress were to apply to OPM the same kinds of tight statutory controls and tight legislative oversight that it applies to HCFA(now known as CMS), OPM might find itself developing a regulatory culture not dissimilar from HCFA(now known as CMS)s.

A second option, promoted by Rep. William Thomas (R.-Calif.), would create a new agency within HHS, the Medicare Benefits Administration (MBA), to handle Medicare managed care issues. The agency, headed by a single administrator, appointed for a five-year term, would report directly to the Secretary of HHS, just as the HCFA(now known as CMS) administrator does. HCFA(now known as CMS) would continue to be responsible for fee-for-service Medicare and for Medicaid. The Secretary of HHS would resolve disagreements between the two agencies.

Like the previous proposal, the MBA proposal would lighten HCFA(now known as CMS)s heavy load and eliminate a perceived conflict of interest. It would also promote hierarchical accountability. Unlike the OPM proposal, it would create a new bureaucracy, which has both advantages and disadvantages. On the positive side, MBA would focus exclusively on a single mission namely, that of promoting Medicare managed care. Also, it could hire professionals with the skills needed to negotiate contracts with managed care organizations. In part, that is because the proposed law would exempt the MBA from federal civil service pay scales, thus allowing the agency to hire individuals with valuable private sector experience at higher pay than would otherwise be the case. On the negative side, a new agency often takes years to get its footing (Bernstein 1955; Johnson 1990), and a single-mission agency runs the risk of capture by clientele groups (Bernstein 1955; Macey 1992).

A third proposal, originally favored by Sen. John Breaux (D.-La.) and Sen. Bill Frist (R.- Tenn.), would create a new Medicare Board to run the Medicare managed care program. Members of the seven-member Board would be appointed by the President and confirmed by the Senate, for staggered seven-year terms. The proposal, embodied in S. 1895 (the Medicare Preservation and Improvement Act), does not specify any criteria for selecting the Board members. It does, however, specify that the Board would be independent in four key respects: first, the President could not remove Board members except for neglect of duty or malfeasance in office; second, the Chair of the Board would be selected not by the President but by the Board membership; third, the Board would be free to submit reports and recommendations to Congress without first securing OMBs approval; and fourth, the Board would have guaranteed funding and would not be dependent on general revenues.

Although superficially similar to the MBA proposal, in that it would create a new agency, the Breaux-Frist proposal would undermine hierarchical accountability because it would remove Medicare managed care from direct control by the Secretary of HHS and the President. Instead control would be exercised by an independent board whose behavior would be very difficult to predict. One possibility, as Feder (2000) has noted, is narrow interest group politics, in which a Medicare Board made policy through a negotiation process among vested interests. Romzek would call such an arrangement political accountability because it involves high control but from sources external to HHS.

The principal advantage of a board is that it facilitates interest representation. If Congress wishes, it could even specify which interests should be represented. Even if Congress does not go that far, it could reasonably expect that diverse interests would be articulated. But the representation of interests on a policymaking board differs from interest representation on an advisory board. In the latter case, the worst that can happen is that policy recommendations are slow in coming, limited in scope, or inelegant in design because they reflect political compromises. In the former case, the worst that can happen is gridlock, incoherent policymaking, or interest group politics, with potentially harmful consequences for Medicare recipients, taxpayers, or both.

The latest version of the Breaux-Frist proposal would avoid some of these pitfalls by creating a single-headed agency rather than a board to handle Medicare managed care. While preferable to the original proposal, it would still weaken hierarchical accountability because the new agency, known as the Medicare Competition and Prescription Drug Agency, would be located outside of HHS. At the very least, this would create coordination problems. It would also involve the usual startup costs associated with creating a new bureaucracy.

A fourth option would be to reorganize HCFA(now known as CMS) to facilitate Medicare managed care but to leave HCFA(now known as CMS)s jurisdiction and discretion untouched. HCFA(now known as CMS), which abolished its Managed Care office in 1995, has proposed recreating such an office, to ensure one-stop-shopping for managed care providers and to facilitate integrated decisionmaking for managed care. An advantage of this approach is that it would make interactions with HCFA(now known as CMS) less time-consuming and less frustrating for the managed care industry. It would also be far less disruptive than some of the other reorganization options under consideration. In itself, however, this strategy would not solve some of the basic problems alleged by HCFA(now known as CMS)s critics: insufficient enthusiasm for managed care, insufficient staff expertise with managed care, and a regulatory culture that stifles managed care.

A fifth option would be to deal with two of the root causes of HCFA(now known as CMS)s problems: inadequate resources and congressional micromanagement. Many observers from across the political spectrum agree that HCFA(now known as CMS) lacks the resources to handle the many complex challenges it faces (Butler et al. 1999). The problem is not just a lack of money or insufficient FTEs but more importantly an inability to hire staff members with valuable private sector expertise because of civil service hiring rules and pay restrictions (Cain 1999: 33). Many observers also agree that Congress has micromanaged HCFA(now known as CMS) in unfortunate ways (Nichols 2000; Bedlin 2000). If HCFA(now known as CMS)s rates for Medicare managed care are inadequate, then it is Congress, not HCFA(now known as CMS), that is to blame. If HCFA(now known as CMS)s regulations seem unduly severe or sclerotic, then that is substantially due to detailed legislative requirements in recent health care statutes (e.g., the BBA Amendments of 1997) and in older legislation that applies to all federal agencies (e.g., the APA of 1946).

Unlike other options, the fifth option would strengthen what Romzek calls professional accountability. It would enable HCFA(now known as CMS) to hire the right kinds of professionals to negotiate contracts with managed care firms and give those professionals sufficient autonomy to do their job well. It would ease up on procedural restrictions in favor of greater emphasis on performance or results. It would reduce the importance of the APA and increase the importance of GPRA. It would rely less on ex ante controls, more on ex post controls. It would promote a model of governance that Wilson (1989: 369), Peters (1996: 91), and others have called deregulated government. The basic idea behind deregulated government is to unleash the creative energies of the bureaucracy by eliminating congressional micromanagement and the red tape it causes and by allowing bureaucrats to function more like their private sector counterparts. Instead of a rule-bound bureaucracy, it would promote a results-oriented bureaucracy. In Wilsons (1989: 369) words, If deregulation of a market makes sense because it liberates the entrepreneurial energies of its members, then it is possible that deregulating the public sector also may help energize it.

Some of HCFA(now known as CMS)s critics (Butler 1999; Wilensky 2000) would agree with much of the preceding analysis but disagree on the remedy. From their vantage point, HCFA(now known as CMS) has too many responsibilities and its regulatory habits are too deeply entrenched to warrant continued confidence in HCFA(now known as CMS) as the custodian of Medicare managed care. But organizational cultures can change, when administrative reorganization and effective leadership by political executives coincide (Hult 1987; Ferrara 1995). The catch is that neither is likely to be sufficient in this instance without unprecedented congressional restraint. As Nichols (2000) has noted, if you really want private sector-like performance, you need to grant private sector-like discretion. Without that discretion, neither HCFA(now known as CMS) nor an MBA nor a Medicare Board is likely to succeed. With that discretion, either HCFA(now known as CMS) or an MBA might be able to offer both hierarchical and professional accountability.


There is a good deal of frustration, on Capitol Hill and elsewhere, with the performance of important bureaucracies, including HCFA(now known as CMS). Structural reform can be a useful technique for improving bureaucratic performance, though its precise consequences often surprise and disappoint reformers (Szanton 1981). The key question, then, is not whether structural reform will have some impact but rather whether it will solve the basic problems that triggered the reform. A related question is whether it will trigger other equally worrisome problems.

Students of public administration and public management often argue that a key problem with contemporary bureaucracies is poor coordination, which limits the bureaucracys ability to engage in holistic thinking and coherent policymaking (Seidman and Gilmour 1986; Wilson 1989: 264-274). That, for example, has been a recurring criticism of the U.S. EPA, whose organization into air, water, and waste divisions makes it difficult to take spillover effects from one medium to another into account (NAPA 1997: 61-74). The same criticism has led to numerous proposals to reorganize our nations law enforcement agencies (Wilson 1989: 265-267; Ferrara 1995).

It is more unusual for reformers to advocate a move in the opposite direction to take an agency that, in principle, is capable of holistic thinking and break it up. That is what a Medicare Board or a Medicare Benefits Administration or a Medicare Competition and Prescription Drug Agency would do. Such proposals would almost certainly create new coordination problems, especially a Medicare Board which would not only separate responsibility for Medicare managed care and Medicare fee-for-service care but would also require seven men and women to reach a consensus on how the former should be administered.

In pondering these issues, it is important to think carefully about the kind of accountability that needs to be enhanced if the federal government is to do a better job of meeting the health care needs of Medicare recipients. A Medicare Board or a new Medicare Advisory Committee would promote political accountability through interest representation. Measures to strengthen and deregulate HCFA(now known as CMS) would promote professional accountability by emphasizing results. The creation of a Medicare Benefits Administration within HHS would endorse hierarchical accountability, as the status quo does, but would create new coordination problems that the Secretary of HHS would have to resolve.


Aberbach, Joel. 1990. Keeping a Watchful Eye: The Politics of Congressional Oversight. Washington, D.C.: Brookings Institution.

Bedlin, Howard. 2000. Remarks at Heritage Foundation Conference on Medicare: The Crisis in Governance. Washington, D.C., June 12.

Behn, Robert. 2000. Rethinking Democratic Accountability. Washington, D.C.: Brookings Institution, forthcoming.

Bernstein, Marver. 1955. Regulating Business by Independent Commission. Princeton: Princeton University Press.

Brannon, Michael. 1984. Organizing and Reorganizing FDA. In Seventy-Fifth Anniversary Commemorative Volume of Food and Drug Law, ed. Food and Drug Law Institute. Washington, D.C.: Food and Drug Law Institute, pp. 135-173.

Breaux, John. 2000. Senator Breaux Promotes Better Way to Run Medicare Program. Press release. Washington, D.C.: U.S. Senate, May 4.

Butler, Stuart. 1999. Reorganizing the Medicare Program to Ensure a Better Program for Seniors. The Heritage Foundation Backgrounder, No. 1294, June 14.

Butler, Stuart, et al. 1999. Crisis Facing HCFA(now known as CMS) & Millions of Americans, Health Affairs 18: 1 (January/February), pp. 8-10.

Cain, Harry. 1999. Moving Medicare to the FEHBP Model, Or How to Make an Elephant Fly, Health Affairs 18: 4 (July/August), pp. 25-39.

Eisner, Marc. 1993. Regulatory Politics in Transition. Baltimore: Johns Hopkins University Press.

Feder, Judith. 2000. Testimony Before the Senate Finance Committee. Washington, D.C., May 4.

Ferrara, Joseph. 1995. Defense, Disasters, and Drugs: Understanding Policy Implementation. Ph.D. Dissertation, Government Department, Georgetown University, December 15.

Fuchs, Beth and Lisa Potetz. 2000. A Framework for Comparing Incremental and Premium Support Approaches. In Marilyn Moon, ed., Competition with Constraints: Challenges Facing Medicare Reform. Washington, D.C.: Urban Institute, February, pp. 7-80.

Garcia, Rogelio. 2000. Statement Before the Senate Finance Committee. Washington, D.C.: May 4.

Gormley, William, Jr. 1996. Counterbureaucracies in Theory and Practice, Administration and Society 28: 3 (November), pp. 275-298.

Gormley, William, Jr. 1989. Taming the Bureaucracy: Muscles, Prayers, and Other Strategies. Princeton: Princeton University Press.

Harris, Katherine. 2000. Members Say Congress, HCFA(now known as CMS) Share Blame for Medicare Management Problems, Health Care Daily Report 5: 125 (June 28).

Health Insurance Association of America. 2000. Inadequate Funding, Excessive Red Tape Push Medicare HMO Program on ?Code Blue. Washington, D.C.: press release, June 8.

Hult, Karen. 1987. Agency Merger and Bureaucratic Redesign. Pittsburgh: University of Pittsburgh Press.

Johnson, Cathy. 1990. New Wine in New Bottles: The Case of the CPSC, Public Administration Review 50: 1 (January/February), pp. 74-81.

Kettl, Donald. 2000. Reinventing Reinvention: Reforming the Federal Bureaucracy in the New Century. LaFollette Policy Report 11: 1 (Spring), pp. 8-10. Madison, Wisc.: Robert La Follette Institute of Public Affairs.

Lemov, Penelope. 2000. The HMO Laid Low, Governing, June, pp. 32-36.

Light, Paul. 1993. Monitoring Government: Inspectors General and the Search for Accountability. Washington, D.C.: Brookings Institution.

Macey, Jonathan. 1992. Organizational Design and Political Control of Administrative Agencies, Journal of Law, Economics, and Organization 8: 1 (March), pp. 93-110.

March, James and Johan Olsen. 1989. Rediscovering Institutions: The Organizational Basis of Politics. New York: The Free Press.

McCubbins, Mathew and Thomas Schwartz. 1984. Congressional Oversight Overlooked: Police Patrols Versus Fire Alarms, American Journal of Political Science 28: 1 (February), pp. 165-179.

McCubbins, Mathew, Roger Noll, and Barry Weingast. 1989. Structure and Process, Politics and Policy: Administrative Arrangements and the Political Control of Agencies, Virginia Law Review 75, pp. 431-482.

Moe, Ronald and Robert Gilmour. 1995. Rediscovering Principles of Public Administration; The Neglected Foundation of Public Law, Public Administration Review 55: 2 (March/April), pp. 135-146.

Nichols, Len. 2000. Remarks at Heritage Foundation Conference on Medicare: The Crisis in Governance. Washington, D.C., June 12.

Peters, B. Guy. 1996. The Future of Governing: Four Emerging Models. Lawrence, Kan.: University of Kansas Press.

Romzek, Barbara. 1998. Where the Buck Stops: Accountability in Reformed Public Organizations. In Transforming Government: Lessons from the Reinvention Laboratories, ed. By Patricia Ingraham, James Thompson, and Ronald Sanders. San Francisco: Jossey-Bass Publishers, pp. 193-219.

Scanlon, William. 1999. Medicare+Choice: Impact of 1997 Balanced Budget Act Payment Reforms on Beneficiaries and Plans. Statement Before the Committee on Finance, U.S. Senate. Washington, D.C.: June 9.

Seidman, Harold and Robert Gilmour. 1986. Politics, Position, and Power, 4th ed. New York: Oxford University Press.

Szanton, Peter, ed. 1981. Federal Reorganization: What Have We Learned? Chatham, N.J.: Chatham House.

Welborn, David. 1977. Governance of Federal Regulatory Agencies. Knoxville, Tenn.: University of Tennessee Press.

Wilensky, Gail. 2000. HCFA(now known as CMS)s Governance and Medicare Reform. Testimony Presented to the Committee on Finance, U.S. Senate, May 4.

Wilson, James Q. 1989. Bureaucracy: What Government Agencies Do and Why They Do It. New York: Basic Books.