Basics of Fiscal Policy and Politics

The Congressional Budget and Impoundment Control Act of 1974 (Budget Act) created the modern congressional budget process.3 Since then, amendments to the Budget Act have capped spending, skewed taxes, imposed across-the-board cuts, and — briefly — balanced budgets. Budget rules have been variously observed, waived, disregarded, allowed to expire, renewed, or enforced with a vengeance. Over time, the focus has largely turned away from the amounts of the deficit and the debt and toward the constraints that Congress has placed on itself.Reference Schick4 Meanwhile, federal spending has increased substantially, with health programs such as Medicare and Medicaid the major sources of growth.

Congress has increasingly turned to the budget process as a way of sidestepping a Senate filibuster, a legislative stalling tactic that can be ended only if 60 senators agree. Senate rules make an exception to the filibuster for consideration of the budget, demonstrating its legislative primacy. This makes the budget an almost irresistible vehicle for acting on any controversial measure.5 Senate rules also define what constitutes a budget item — essentially a change to mandatory spending or revenues — and disallow anything else. In turn, this alters the legislative and political calculus, privileging money measures and consigning all other provisions to endless waiting.

Federal budgets are divided into discretionary spending, mandatory spending, and tax spending.Reference Westmoreland6 For budget and legal purposes, it is presumed that discretionary spending will not be renewed unless Congress acts to do so, generally through annual appropriations. By contrast, mandatory spending is money that has been promised in a permanent provision of law; Social Security, Medicare, and Medicaid are the largest mandatory spending programs. Tax spending (through deductions, credits, exclusions, etc.) is revenue that is forgone by the federal government for a specific reason. Tax spending has grown to be larger than all of discretionary spending (at least before COVID-19), but it is rarely listed as part of the federal budget.

While both mandatory spending and tax spending are generally based in permanent law, the amount of money needed to meet these statutory obligations will vary from year to year. A model known as the “baseline” projects what current law requires be spent in the future, with Congress relying on the nonpartisan Congressional Budget Office (CBO). The baseline is required to be built on the assumption that current statutory provisions will be implemented as written, however unlikely that may seem from a political perspective. The baseline projects program costs and revenues, as well as the annual federal deficit and the cumulative federal debt, and generally includes the coming year and nine years into the future (the “tenyear snapshot”).Reference Kamin7

Congress has adopted progressively stricter procedural and statutory limits on its own ability to increase spending or cut taxes. For discretionary spending, Congress has created an annual global cap; no new spending for one program can be provided unless other programs are cut from their previous levels. Legislation that would exceed the cap is not “in order;” this objection can be waived, but doing so can incite opposition from both those opposed to the substantive legislation and those who are committed to budget restraint. There is also a statutory doomsday machine — the sequester — that requires the OMB to impose across-the-board, pro rata cuts in most appropriated programs to bring total discretionary spending down to the cap.

For mandatory spending, Congress adopted the “Pay-as-You-go” (PAYGO) requirement, a parliamentary rule prohibiting the consideration of any mandatory spending or revenue legislation that would make the deficit worse. Since 2010, PAYGO has also been enforced by a statutory sequester mechanism: at the end of each year, the OMB adds up the costs of all new mandatory spending and revenues and subtracts any offsetting savings. If all such legislation in the aggregate is projected to add to the deficit, the OMB is required to impose across-the-board cuts. Several programs are exempted (e.g., Social Security, Medicaid, and the Children's Health Insurance Program) while others have limits on cuts (notably no more than 4% of Medicare). Because recent Congresses have been extremely reluctant to raise taxes, PAYGO requirements have been met mainly by cutting existing mandatory spending programs, in essence robbing Peter to finance Paul.

To make the PAYGO limits workable, how new legislation affects baseline commitments must be estimated. This “scorekeeping” is done by the CBO (sometimes in tandem with the Joint Committee on Taxation). Although the CBO does credible work under intense pressure, no one (including the CBO) believes that its scores will ultimately be borne out as correct over time. The CBO score is akin to an umpire's call in a sporting event: It is deemed to be right and is accepted by participants so that the action can go on.

This formalistic process is nonetheless central to the prospects of nearly all health care legislation. Consider, for example, the comprehensive tobacco control legislation passed in 2009 after decades of biomedical research and public health advocacy.8 Division A of the law deals with “Smoking Prevention and Tobacco Control;” Division B deals with “Federal Retirement Reform.” Assume that Division A would have been effective at decreasing smoking. Taken alone, however, Division A would not have been enacted because a reduction in smoking would reduce federal tax receipts on tobacco products without significantly reducing federal Medicaid costs within the relevant 10-year time frame.9 Fortunately, an experienced House committee chair noticed that an unrelated bill had been filed to expand tax-favored retirement options by permitting the equivalent of “Roth IRAs” for federal workers. This bill had been scored favorably — though economically equivalent, paying tax at deposit rather than at withdrawal brought more revenues into the ten-year snapshot. Designating the second bill as Division B of the Tobacco Control Act kept the overall legislation in the black as a technical matter, enabling its passage.10

The ACA's Fiscal Design and Implementation

Retention of “Private” Insurance: It is universally recognized that the United States spends more private money per person on health care than any other nation. It is seldom appreciated that the United States also spends more public money per person on health care than any other nation (OECD data).11 Moreover, public and private health care expenditures in the U.S. are largely overlapping (e.g., hospital care is funded with public dollars for some people and private dollars for others), unlike countries with explicitly public systems in which private spending is typically for different services in alternative settings. Fiscal characterization, not substance, distinguishes most public from private spending — a fact best illustrated by the roughly $300 billion annually in forgone revenue to the U.S. treasury because earnings paid by private employers as health coverage rather than wages are exempt from income tax.12 This tax expenditure is incompletely captured in government accounting and, prior to the ACA, blinded most workers to the full cost of their benefits because individual reporting to the IRS was not required.

Keeping the costs of the health care system off public ledgers was a primary design principle of the ACA. There were many reasons that employer-sponsored insurance had political appeal as part of the ACA (e.g., “If you like the plan you have, you can keep it.”).13 But at least one major motivation for its retention as the framework of the ACA was to avoid the appearance in bookkeeping of transmogrifying vast amounts of private spending into public spending. In this respect, the ACA continued a tradition of cloaking health reforms in the language of private competition, regardless of whether such competition was likely. The Clinton administration's Health Security Act took this to its extreme, combining a global budgeting process with an employer-based managed competition framework in the hope that CBO would credit the global budgets for capping potential costs (which it did) but consider the mandatory flow of funds through employers to be premiums rather than payroll taxes (which it did not).14 Had the Clinton bill been enacted, it seems probable that global budgets, once surpassed, would have been deferred by future Congresses in a ritual of annual hypocrisy not unlike the persistent nonenforcement of the Sustainable Growth Rate (SGR) for physician payment under Medicare.

Budget Commitments: One might argue that the ACA could not have become law had the Great Recession not shortly preceded its enactment. The reality of recession influenced voters in the 2008 presidential election, bringing “health security” once again onto the domestic policy agenda. Economic stimulus also provided fiscal “activation energy” for health care reform, both directly and by blunting legislators’ concerns about short-term budgetary effects. The American Recovery and Reinvestment Act of 2009 (ARRA), for example, included $149 billion for health care.15

But Congressional tolerance for additional public spending was not unlimited. Leadership assured members that that the ACA would be required to comply with both parliamentary and statutory PAYGO rules, implying no net increase in the budget baseline. The legislation also had to fulfill a political promise made by President Obama that total expenditures would not exceed approximately $900 billion over ten years, regardless of offsetting cuts and new revenues.16 This pledge, perhaps unnecessary, ensured that CBO scorekeeping would have a primary role throughout the congressional debate.

Individual Mandate: The ACA's “individual mandate” to purchase health coverage did not originate with progressive policymakers but with conservative commentators, who preferred personal responsibility as an obligation of citizenship to what they considered the free-rider-encouraging, potentially jobkilling effects of burdening private businesses through an “employer mandate” — the Clinton plan's centerpiece.17 But it was an employer mandate's fiscal consequences, not its arguable employment effects, that rendered it a non-starter in the ACA. The final nail in the coffin of the Clinton plan was the CBO's decision in 1994 to regard the employer mandate as an exercise of sovereign power, and therefore to score the proposal as massively increasing both taxes and government spending. Roughly half of the U.S. population — 150 million people — typically receives health coverage through employment. Regardless of its effect on the uninsured, an employer mandate placed in the ACA would likely have been scored by the CBO as increasing the annual federal tax burden by over $1.5 trillion. By contrast, individually purchased insurance would cover only about 20 million people, reducing the necessary budget offsets to $200 billion even if the CBO applied a similar analysis (it did not). The ACA's limited employer obligations are backstop provisions mainly intended to prevent free-riding on subsidized public coverage, and did not threaten CBO scoring.

Medicaid Expansion: The ACA's dramatic expansion of Medicaid to include all poor Americans not otherwise insured — including individuals that many conservative states had declined previously to cover — represents one of the law's two most direct commitments of substantial public funds. The other large, explicit public expenditure consists of refundable tax credits for the purchase of commercial health insurance by lower-income Americans not made eligible for Medicaid. During congressional debate, however, fiscal criteria tipped the scales toward Medicaid. Because administrative costs are less and provider payments are lower, CBO scores the cost of an additional Medicaid beneficiary more favorably than a heavily subsidized purchaser of marketplace insurance. To hit fiscal targets, Congress twice increased the income limits for the Medicaid expansion, including a last-minute amendment that added another 5% — explaining why some summaries of the ACA cap eligibility at 133% of the federal poverty line, while others cite 138%.18 The overall price-tag of the Medicaid provisions eventually made it necessary, largely for budgetary reasons, to require state financial participation in the form of matching payments, which in turn led to litigation that almost declared the expansion unconstitutional (see below).

Phase-In of Major Provisions: Changes made by the ACA to the U.S. health care system may have been substantial, but they also were slow to start. President Obama signed the bill into law on March 23, 2010, but the ACA's major provisions expanding Medicaid and reconfiguring and subsidizing individually purchased insurance did not become fully effective until 2014. Some of the delay was inevitable: the Department of Health & Human Services had to draft and issue a very large number of implementing regulations, and the health insurance and health care industries had to adjust their business models to the new requirements and incentives. But a prolonged phase-in also serendipitously served fiscal politics by placing fewer years of full operation within the CBO scorekeeping window, proportionately decreasing the ACA's projected overall cost. This came at a political price: opponents of the law had a much easier time mustering their arguments and mounting litigation when a system was not yet in place and the public did not yet understand what the law contained or how they might benefit from it.

Community-Based Long-Term Care: Democrats in Congress did not embrace a comprehensive singlepayer national health insurance as Senator Ted Kennedy (D-Mass.), who died of brain cancer in the first year of the Obama presidency, would have wished. Congress did, however, incorporate one of Kennedy's last legislative causes into the ACA: the so-called CLASS Act creating a program of subsidized long-term-care insurance to fund services outside of nursing homes. In fiscal terms, the CLASS Act required 5 years of voluntary (i.e., non-tax) contributions for individuals to vest, pushing nearly all payouts beyond the scoring window and appearing to reduce the ACA's projected budget impact by $87 billion.19 Having helped secure the ACA's passage in 2010, the CLASS Act was unceremoniously repealed in 2013, its negative long-term fiscal impact having been clearly revealed simply by time rolling forward into a new scoring window.

Budget Reconciliation and the Byrd Rule: Perhaps the most pervasive effect of the Budget Act on the ACA's ultimate wording resulted from budget reconciliation processes. After House passage of one version of a bill and Senate passage of a different version, the expectation would be for the two bodies to confer. But the loss of Senator Kennedy's 60th vote in the Senate forced negotiations into the form of a budget package in order to avoid a filibuster. Only compromises that produced a direct change in revenues or outlays could be considered; other measures were barred by the Byrd Rule. Even the clean-up of inadvertent drafting errors was procedurally precluded. It is unlikely that “regular order” would have produced a perfectly drafted law, but Budget Act restrictions made mistakes unavoidable. As described next, some would come back to haunt health reform in court.

The ACA's Fiscal Economy Meets the Supreme Court

Dozens of lawsuits have been filed against the ACA, with three already generating decisions from the United States Supreme Court and more on their way.20 Unlike challenges to the Social Security system during the Great Depression, the incidence of litigation did not abate after initial uncertainties regarding the ACA's constitutionality were resolved by the courts. Maneuvering under the Budget Act was a major contributor to this legal morass. For fiscal reasons, the ACA allocated to potentially litigious states and private parties obligations more intuitively lodged in federal agencies and deployed convoluted and therefore contestable procedures made necessary by that apportionment of financial responsibility.

NFIB v. Sebelius (2012): It is well known that Chief Justice John Roberts “saved” the ACA from constitutional nullification in its first trip to the Supreme Court by recasting its two principal mechanisms for reducing the number of uninsured Americans — the individual mandate and the Medicaid expansion — to avoid compelling individuals or coercing states in excess of his view of federal powers.21 Neither would have been necessary absent budget constraints during the law's consideration and enactment. Had the federal government assumed the full cost of the Medicaid expansion, states would have had neither public rationale nor legal standing to sue. But hitting the declared budget targets drove Democrats in Congress to place 10% of long-term matching requirements on the states, a fiscal need enhanced by the CBO-scoring induced shift of more beneficiaries from tax-subsidized private insurance to Medicaid. This gave seven members of the Court grounds to rule that the threat of withholding all Medicaid funds from states that refused to bear this burden would violate the Tenth Amendment.

Obligating individuals to purchase private coverage also offended a majority of the Justices, who opined that Congress's power to regulate commerce did not permit Congress to compel commercial activity. Direct government funding of coverage through taxation would have rendered this mechanism unnecessary, lending a touch of irony to the Chief Justice's conclusion that the penalty for declining to purchase insurance was properly considered a tax and therefore that the private coverage provisions were within Congress's Article I authority.

Burwell v. Hobby Lobby (2014): Although the ACA's imposition of an individual rather than an employer mandate avoided most newly regulated private coverage being scored as taxation, the law placed other obligations on private employers that provoked litigation. Chief among these has been the requirement that most employer-sponsored coverage offer “essential health benefits,” including no-cost contraception as a preventive health service. Not surprisingly, mandating contraceptive benefits has been controversial among employers with religious objections, with the federal Religious Freedom Restoration Act (RFRA) providing a legal vehicle for their grievances.

In the Hobby Lobby case, the Supreme Court's conservative majority conferred standing on closely held corporations with sincerely held religious beliefs, an unexpected expansion of RFRA.22 In another touch of fiscal irony, the Court asserted that the ACA failed to use the least restrictive alternative to achieve its goal, as contraceptive coverage could have been publicly funded. The Obama administration subsequently attempted to reach an accommodation with employers that maintained contraceptive access, an approach the Trump administration summarily reversed. Both positions prompted significant litigation, some of which is ongoing and none of which would have occurred in a publicly funded system as there is no RFRA-relevant alternative to raising general revenues and giving grants and subsidies.23

King v. Burwell (2015): In King, the ACA survived another near-death experience in the courts, though one without a constitutional dimension.24 The dispute focused on poorly drafted language in the statute suggesting that federal income tax subsidies were unavailable to persons buying individual coverage in states that refused to establish their own insurance exchanges and who were relying on the federal exchange by default. The plaintiffs’ standing was based on their desire not to receive subsidies so that they could be exempt from the individual mandate. The Supreme Court rejected the argument, but it highlights the litigation risk created by the ACA's Rube-Goldberg-like financing mechanisms. A more direct, tax-funded, nationally administered system would have been less challengeable in court. Moreover, the limitations of budget reconciliation rules and the absence of a House-Senate conference increased the risk of such a drafting error remaining uncorrected.

Fiscal Influences on ACA “Repeal and Replace”

Political opposition to the ACA is generally credited for returning the House of Representatives to Republican control in the 2010 midterm elections, much as opposition to the Clinton health plan had done for both houses of Congress in 1994. In each instance, longstanding partisan divides on fiscal issues such as taxation, government spending, and redistribution mixed with concerns about specific health policies to generate new majority legislative agendas. In 2016, the Republicans added control of the Senate and White House, with ACA “repeal and replacement” an explicit and urgent campaign promise they were eager to fulfill. At the same time, the siren call of tax cuts and the related temptation to redirect permanent spending overcame partisanship for legislators on both sides of the aisle, putting at risk important features of the ACA.

But the rules and processes of the Budget Act complicated the actions of the ACA's opponents during repeal as they had those of its supporters during enactment. Reconciliation in particular limited the measures available to Republican legislators seeking to undo the ACA's core provisions. Spending and taxes are permitted in a budget package; insurance reforms and regulatory standards are not. A simple repeal of the ACA would have involved laws that had no effect on outlays or revenues and would have been ruled out of order in the Senate and thus subject to filibuster. Politicians who had campaigned that they would uproot the ACA found themselves able only to trim selected branches. Ultimately, this left ACA opponents without even a majority in the Senate, as confirmed in July 2017 by Senator John McCain's dramatic, late-night thumbs-down vote on so-called “skinny repeal.”25

Even more basically, the ACA's passage altered the fiscal terrain on which these battles would be fought. The ACA, its spending, and its revenues and savings were now included in the baseline for CBO scorekeeping. “Repeal and replace” would be measured against this reset baseline, creating a CBO price tag for proposed legislation. Moreover, highly visible CBO projections included both dollars saved and people rendered uninsured, because the budget effects could not be calculated without estimating how reducing penalties and subsidies would affect private decisions to obtain coverage.

Prevention and Public Health Fund (PPHF): Intended as a dramatic and permanent investment in population health — long the poor stepchild of acute care — the PPHF received an initial allocation of $500 million in mandatory spending, with programmed increases to $2 billion annually by 2015.26 Beginning in 2012, however, successive pieces of essentially bipartisan legislation raided the PPHF for other purposes such as reforming the SGR, reauthorizing the Children's Health Insurance Program, and enacting the 21st Century Cures Act. After further cuts in the 2017 Republican tax reform bill, the PPHF was left with roughly $1 billion annually, hardly a paltry sum but far less than intended. In retrospect, fiscal rules caused the original strategy to misfire. By designating the PPHF as mandatory spending, the ACA's drafters created a supposedly reliable funding base for public health and avoided the risk of politically motivated refusals to appropriate funds. However, the infusion of mandatory funding reset the spending baseline and painted a target on the PPHF's back as a source of long-term budgetary offsets for unrelated spending increases and revenue cuts.

Cost-sharing Reduction Payments and House v. Burwell: Perhaps no provision of the ACA has been more affected by fiscal ambiguity and associated maneuvering than the government's obligation to reimburse private insurers for cost-sharing amounts not permitted under law to be charged directly to low-income purchasers of “Silver”-level coverage. The Obama Administration took the position that the funding for these “cost-sharing reduction payments” was intended to be mandatory spending. The Republican- controlled House of Representatives filed suit, arguing that it was discretionary spending that had not been appropriated and therefore was not available. As with King, ambiguous or mistaken drafting likely resulted from the constraints of budget reconciliation. In 2016, a federal district court ruled that the Republican House had legal standing to sue.27 The litigation was settled in 2017, with the payments becoming discretionary. When the Trump administration eventually ceased payments, however, several insurers brought their own lawsuits for the denied funds, with some prevailing and others still pending in court. With repayment foreclosed, insurers also “loaded” the cost-sharing amounts into the premiums charged to Silver Plan enrollees. Because the ACA more broadly subsidizes premiums for low-income enrollees than it does cost-sharing obligations, the net effect was to increase the ACA's total cost to the government, an ironic outcome that a CBO estimate confirmed.28

Shared Responsibility Payments and California v. Texas: Anticipating a constitutional challenge, the ACA's drafters had emphasized in recitals the centrality of the individual mandate to the effectiveness of the law taken as a whole. Their strategy backfired: the Supreme Court rejected the Obama administration's Commerce Clause argument in NFIB, and four Justices were poised to overturn the entire ACA on the grounds that once the mandate was held unconstitutional, it could not be severed from the remainder of the law. This analysis resurfaced in late 2018, when a conservative District Judge in Dallas ruled that the 2017 Republican tax law setting the “shared responsibility payment” at zero removed the mandate's constitutional protection and rendered the entire ACA invalid.29 Following partial affirmance by a federal appeals court, the Supreme Court agreed to hear the case. When it does, it will resolve what is almost a metaphysical question about the “intent of Congress.” Budget reconciliation processes exposed the shared responsibility payment itself to elimination by simple majority vote in the Senate, but the remaining language mandating coverage — though now without a tax penalty to enforce it — was beyond the scope of budget reconciliation and therefore could not formally be repealed without a 60-vote supermajority. One might argue that if the Congress had an intent to repeal the ACA it would have waived or amended the Budget Act, the Senate Rules, or both. It did not do so.

Lessons and Implications

The ACA was the first universal coverage legislation to become law after a hundred years of trying. But its reliance on the definitions, scorekeeping artifacts, and arcane parliamentary rules of the Congressional budget process has come — dare we say it? — at a cost. Fiscalizing the legislative debate created ambiguities to be litigated and conferred standing on unlikely plaintiffs. This placed much of the ACA at the mercy of the judiciary, which is the branch of government least able to assess either health effects or budget impacts. A more straightforward legislative approach akin to Social Security and Medicare, while fraught in its fiscal politics, might have produced a less confusing and less precarious result.

The tyranny of the federal budget is troubling to those of us who care deeply about health care reform. Other areas of domestic policymaking do not have their substantive goals so often subordinated to deficit targets. The 2009 stimulus package and the 2017 tax cuts both were scored as adding billions, even trillions, in red ink to the government's accounts, but the House and Senate deemed them necessary for the economy, sidestepped PAYGO, and passed them intact.

The hard history of the ACA shows that when health reform is viewed primarily through a budgetary lens, it is health policy that suffers. Fiscal politics distorts the process and the goals of legislating — in both predictable and wildly unpredictable ways. Future health reform deliberations — such as public discussions of Medicare-for-All — should aspire to greater transparency regarding the artifice of whether a reform is labeled tax-subsidized “private insurance” or federally provided “public insurance.” Fiscal estimates should employ measures that account clearly for costs and savings in all health expenditures, not just those denominated as on-budget. Long-term investments should be valued, even if they do not produce returns within the budget snapshot.

Moreover, budgetary projections should not be the only metrics. At a minimum, Congress should assess health legislation by routine analysis of its effects on illness, life, and death — not just its effects on outlays, offsets, and debt. It would be possible, for example, for the CDC as well as the CBO to score most health proposals. A non-partisan evaluation of projected effects on morbidity and mortality would be valuable to the political process and likely more meaningful to the public.

The COVID-19 pandemic shows how dependent the nation's financial health can be on its physical health. Requiring for CBO scorekeeping reasons state financial participation in the ACA's Medicaid expansion led to litigation that resulted in a coverage gap between Medicaid eligibility and subsidized exchange coverage in non-expansion states.30 This left many suddenly unemployed Americans with no source of payment for emergency services, including COVID-19 testing and treatment, which in turn has required billions in ad hoc federal spending. Systemic underestimation of the value of public health infrastructure, including the evisceration of the ACA's Prevention and Public Health Fund, has cost both lives and dollars.

Money may not be limitless, but neither is life expectancy.