The question posed by budget reform after the Congressional Budget and Impoundment Control Act of 1974 was never whether Congress could write a rule. Congress wrote rules in abundance across the decades that followed, in statutes that set targets, imposed caps, created scorecards, and threatened automatic cuts when political agreement failed. The harder question, and the question that decides whether budget reform controlled spending, was whether a rule written into law could change the arithmetic that the rule confronted. That arithmetic had two parts. One part arrived on the floor each year in appropriations bills that members debated, amended, and passed or failed to pass before a deadline. The other part arrived without a floor vote at all, in benefit payments determined by eligibility formulas enacted in earlier decades, in health care payments determined by program rules and medical prices, and in interest payments determined by debt accumulated in earlier years. A process that governed the first part with precision could leave the second part untouched, and a record across forty five fiscal years tests how much of the federal budget any procedural regime actually reached.

The central claim that organizes the aggregate evidence carries a short form suitable for citation. Rules bind the annual vote and not the automatic one. The sentence describes a structural division inside federal spending rather than a judgment about the sincerity of any Congress or administration. Annual appropriations for defense, for domestic agencies, for foreign operations, and for the operations of government itself require affirmative action, and a cap or a target or a sequestration threat can raise the cost of that action or block it altogether. Mandatory spending requires no comparable action. Social Security benefits, Medicare payments, Medicaid matching payments, federal retirement payments, and net interest flow under permanent law, and a budget resolution or a discretionary cap leaves the formulas that generate those payments exactly as earlier statutes left them. A PAYGO scorecard can record the cost of a tax reduction and threaten a sequester in response, and Congress can also waive the scorecard, shift the debit beyond the budget window, or designate the measure an emergency, each by a vote taken under the same authority that created the rule. The aggregate record therefore has to be read at the level of totals, because totals reveal whether binding the voted portion of the budget was sufficient to control a budget in which the unvoted portion grew decade after decade.

A fair test also requires a baseline stated before any verdict. In fiscal 1974, the fiscal year in which the 1974 statute was enacted on July 12, 1974, the federal government ran a small deficit and held a debt burden that stood at a postwar low as a share of the economy. If procedural reform controlled spending, the decades after 1974 should show deficits contained near balance, debt stabilized or falling as a share of gross domestic product, and departures from balance explained by wars or recessions rather than by ordinary operation of law in ordinary years. If procedural reform failed at the level of totals, the same decades should show deficits in most years regardless of which regime governed and debt rising across regimes of opposite design. The evidence assembled from named federal series supports the second description, and the strength of that support lies in the consistency of the totals across five distinct procedural regimes that shared almost nothing except the division between voted spending and automatic spending.

The Aggregate Record Across Five Regimes

The endpoint comparison states the outcome in its most compressed form. In the Congressional Budget Office historical data series published in the appendix to The Budget and Economic Outlook: 2019 to 2029, federal debt held by the public stood at 23.2 percent of gross domestic product at the end of fiscal 1974. In the Congressional Budget Office record for fiscal 2019, published in the Monthly Budget Review summary for fiscal 2019 and carried in The Budget and Economic Outlook: 2020 to 2030, federal debt held by the public stood at 79.2 percent of gross domestic product at the end of fiscal 2019, up from 77.4 percent at the end of fiscal 2018. The Office of Management and Budget Historical Tables series, Table 7.1, records the same category of debt at the end of each fiscal year and supports the same direction and scale of change, from roughly a quarter of the economy in fiscal 1974 to roughly four fifths of the economy in fiscal 2019. A ratio that more than tripled across forty five years did not rise in a straight line. The ratio climbed during the nineteen eighties, fell during the second half of the nineteen nineties, and then climbed across the financial crisis and the long expansion that followed. A movement with that shape cannot be assigned to a single statute or a single economic event, and can be assigned to a persistent gap between receipts and outlays that no procedural regime closed for longer than four consecutive fiscal years.

Debt held by the public is the correct measure for the question at hand, and the definition explains part of the mechanism. The category counts Treasury securities held by investors outside the federal government, including individuals, corporations, state and local governments, foreign governments, and the Federal Reserve, and the category excludes securities held by federal trust funds and other government accounts. Borrowing from the public competes for savings, carries interest paid to holders outside the government, and must be refinanced in credit markets. Intragovernmental holdings record obligations between parts of the same government, and an increase in those holdings does not place the same claim on private savings. When the Congressional Budget Office and the Office of Management and Budget report debt held by the public as a share of gross domestic product, the reported ratio therefore measures the accumulated result of unified deficits against the size of the economy available to carry the resulting interest cost. A government can stabilize that ratio while running modest deficits, provided economic growth outpaces the effective interest rate on the debt by a sufficient margin. A government that runs deficits near or above the growth rate of the economy, year after year, watches the ratio rise even in years without war or recession.

The annual flow behind the accumulated stock is equally stark. In the Congressional Budget Office historical data series, the federal budget recorded a deficit in every fiscal year from fiscal 1974 to fiscal 1997, a surplus in fiscal 1998, fiscal 1999, fiscal 2000, and fiscal 2001, and a deficit in every fiscal year from fiscal 2002 to fiscal 2019. Counted across the full span that begins with enactment of the Congressional Budget and Impoundment Control Act of 1974 and ends with the last completed fiscal year before April 15, 2020, the totals give forty two deficit years against four surplus years. The Congressional Budget Office historical data express the same record as shares of gross domestic product, with a deficit of 0.4 percent of gross domestic product in fiscal 1974, a surplus of 0.8 percent in fiscal 1998, a surplus of 1.3 percent in fiscal 1999, a surplus of 2.3 percent in fiscal 2000, and a deficit of 4.6 percent of gross domestic product in fiscal 2019 on a deficit of 984 billion dollars recorded in the Congressional Budget Office Monthly Budget Review summary for fiscal 2019. No procedural regime in the span produced balance as a normal condition. Balance appeared once, held for four years, and disappeared under legislated tax reductions, legislated spending increases, military operations, and an economic slowdown that began in 2001.

Did budget reform produce sustained balance?

No procedural regime produced sustained balance through fiscal 2019. The Congressional Budget Office historical data record deficits from fiscal 1974 to fiscal 1997 and from fiscal 2002 to fiscal 2019, with surpluses confined to fiscal 1998, fiscal 1999, fiscal 2000, and fiscal 2001 alone within that span.

The first regime, the original framework of the Congressional Budget and Impoundment Control Act of 1974, deserves a careful statement because the aggregate failure belongs to the framework in a limited sense only. The 1974 statute created the House and Senate Budget Committees, created the Congressional Budget Office, established budget resolutions, reconciliation, and a timetable that set the start of the fiscal year on October 1, and imposed controls on presidential impoundment through the rescission and deferral procedures in Title X of the act. The statute built a process for deciding totals without enacting a rule that totals had to balance, did not cap discretionary appropriations in nominal dollars, and did not subject mandatory programs or tax law to an enforcement mechanism with automatic consequences. Enforcement depended on points of order that a majority could waive. The Congressional Research Service record on appropriations timeliness captures the practical result in one dimension: all regular appropriations bills were completed before the fiscal year began in only four fiscal years across the framework and the regimes that succeeded the framework, fiscal 1977, fiscal 1989, fiscal 1995, and fiscal 1997. A timetable met four times in more than four decades regulated the form of annual action far more reliably than the size of annual action, and the size of automatic action lay outside the timetable altogether.

The second regime, the Balanced Budget and Emergency Deficit Control Act of 1985, known as Gramm-Rudman-Hollings, and the revised version enacted in 1987, attempted to correct the weakness of pure process by naming the outcome in advance. Fixed deficit targets declined on a statutory schedule toward balance, and a sequestration mechanism threatened across the board cuts divided between defense and domestic programs when projected deficits exceeded the targets. The targets were never met on the terms the statute set. The record of evasion is a record of technique: optimistic economic and technical assumptions lowered projected deficits on paper, payment dates were shifted across the boundary between fiscal years, programs and entities were moved off budget, and the targets themselves were amended when compliance approached. The Supreme Court decision in Bowsher v. Synar in 1986 removed the original enforcement officer from the mechanism on separation of powers grounds, and the 1987 revision reassigned the function to the Office of Management and Budget, a repair that preserved the form of automatic enforcement while leaving every accounting technique available for continued use. The regime ended in replacement rather than in success. A target regime demonstrated that a rule aimed at the deficit itself could be defeated by control over the forecast, the calendar, and the definition of the budget to which the target applied.

The third regime, built by the Omnibus Budget Reconciliation Act of 1990 and known through the Budget Enforcement Act of 1990, abandoned fixed deficit targets in favor of controls on legislated action. Discretionary spending received caps, while new mandatory spending and tax legislation received a PAYGO requirement enforced by a scorecard and a threatened sequester of mandatory programs. The 1990 regime governed decisions rather than outcomes, on the stated reasoning that Congress could control votes on appropriations and on new legislation, while economic performance, caseloads, and interest costs lay beyond the reach of a vote in any single session. The regime coincided with the only movement toward balance in the entire span, and attribution for that movement requires competing analyses stated side by side without resolution. The Congressional Research Service account of the 1990 agreement records a summit agreement of about 500 billion dollars in deficit reduction and an enacted value of about 482 billion dollars, combining revenue increases, discretionary restraint, and mandatory savings. Analysts at the Congressional Budget Office and at the Office of Management and Budget also assign weight to the Omnibus Budget Reconciliation Act of 1993, which extended the caps and the PAYGO requirement, to a capital gains driven surge in individual income tax receipts during the second half of the decade, and to a post Cold War drawdown that carried defense outlays from roughly 5 percent of gross domestic product in fiscal 1990 to roughly 3 percent of gross domestic product in fiscal 2000. Each named analysis identifies a real contributor, and no named analysis in the record establishes a sole cause.

The surplus interval itself, read at the level of mechanics, confirms the distinction between protecting a path and creating a path. Debt held by the public fell as a share of gross domestic product during the surplus years, from 41.7 percent in fiscal 1998 to 38.3 percent in fiscal 1999 to 33.7 percent in fiscal 2000 in the Congressional Budget Office historical data series, because nominal debt held by the public declined in absolute terms while the economy grew. Those declines required receipts to exceed outlays in fact, not in projection, and receipts exceeded outlays because revenue collections rose faster than the economy while discretionary outlays grew slowly under caps that Congress increasingly supplemented with emergency designations outside the caps. The mechanism that later dissolved the surplus was equally concrete. The Economic Growth and Tax Relief Reconciliation Act of 2001 reduced revenue on a schedule that extended across the decade, military operations in Afghanistan from 2001 and in Iraq from 2003 were financed substantially outside regular appropriations discipline in the early years, an economic slowdown reduced receipts from the fiscal 2000 peak, and the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 added a mandatory benefit whose costs fell largely beyond the budget window used at enactment. The Budget Enforcement Act framework expired after fiscal 2002, and deficits returned in fiscal 2002 and persisted through the expansion of the middle decade.

The fourth regime arrived in two parts that are often treated as one. The Statutory PAYGO Act of 2010, Public Law 111-139, restored a scorecard for mandatory spending and tax legislation, with a sequester of nonexempt mandatory programs as the threatened consequence, a cap of 4 percent on any Medicare reduction under a PAYGO sequester, exclusions listed in section 4 of the act, and emergency designations under section 4(g) of the act as routes around the scorecard. The Budget Control Act of 2011 then imposed statutory caps on discretionary appropriations for fiscal 2012 to fiscal 2021 and triggered a sequestration that executed on March 1, 2013 in the amount of 85.3 billion dollars, with Medicare reductions capped at 2 percent. The Congressional Budget Office record establishes a fact that defines the limits of scorecard enforcement: no sequestration was ever triggered under the Statutory PAYGO Act of 2010 through April 15, 2020. When the Tax Cuts and Jobs Act of 2017, Public Law 115-97, accumulated scorecard debits large enough to threaten a sequester, section 5002 of Public Law 115-96, enacted on December 22, 2017, excluded the act from the statutory PAYGO scorecards, and the Bipartisan Budget Act of 2018, Public Law 115-123, then reset all remaining scorecard balances to zero. A rule that Congress can reset by majority vote binds only for as long as a majority prefers the rule to the legislation that the rule obstructs.

The discretionary caps under the Budget Control Act of 2011 bound more tightly than the scorecard, and the evidence of binding is also evidence of the boundary of the binding. Congress raised the caps by statute four times, in the Bipartisan Budget Act of 2013, the Bipartisan Budget Act of 2015, the Bipartisan Budget Act of 2018, and the Bipartisan Budget Act of 2019. Funding for overseas contingency operations sat outside the caps, and adjustments for that category totaled 723 billion dollars from fiscal 2012 to fiscal 2019 in the series record. The sequestration of March 1, 2013 did execute, and discretionary outlays fell as a share of the economy, from 9.1 percent of gross domestic product in fiscal 2010 to 8.7 percent of gross domestic product in fiscal 2011 to 6.3 percent of gross domestic product in fiscal 2019 in the Congressional Budget Office and Office of Management and Budget Appendix F series. That decline is the clearest aggregate achievement that any regime in the span can claim, and the achievement is confined to the one category of spending that requires an annual vote. During the same decade, the fiscal 2019 deficit reached 4.6 percent of gross domestic product in the Congressional Budget Office record and debt held by the public rose to 79.2 percent of gross domestic product, because mandatory programs and interest grew on schedules that the caps did not govern.

Process failure accompanied aggregate failure. The longest lapse in federal funding in the span ran from December 22, 2018 to January 25, 2019, a period of thirty five days on an elapsed day count, during the operation of statutory discretionary caps that had been raised for fiscal 2019 by the Bipartisan Budget Act of 2018 months before the lapse began. The caps limited aggregate amounts for categories defined in law. The caps did not allocate funds among the regular appropriations bills, did not resolve disputes over individual provisions, and did not compel agreement before October 1. A ceiling on totals and a functioning annual process proved to be separate achievements, and budget reform delivered the first far more often than the second.

A serious account must also state the strongest version of the contrary reading, because the aggregate record contains real instances of rules altering behavior at the margin. Discretionary caps in the nineteen nineties forced tradeoffs inside the capped total, and the PAYGO rule of the same decade raised the procedural cost of tax and mandatory legislation enough that offsets became a normal part of drafting rather than an afterthought. The Budget Control Act caps, before the successive increases, held discretionary spending below the path implied by inflation adjusted continuation of fiscal 2011 levels, and the March 1, 2013 sequestration demonstrated that an automatic mechanism could execute when the trigger was drafted to operate without a further vote to start the cuts. Gramm-Rudman-Hollings placed deficit totals at the center of budget debate for half a decade and supplied the political setting in which the 1990 agreement became possible. Each instance supports a narrow proposition: a rule can change the price of a voted decision and can hold a voted category below an unconstrained path for a period of years. None of the instances supports the broader proposition that a procedural regime controlled federal spending as a whole, because mandatory spending and interest together accounted for a majority of outlays in every year of the final decade in the Congressional Budget Office and Office of Management and Budget series.

The aggregate finding can therefore be stated without qualification about the totals and with qualification about the causes. Across the original 1974 framework, the Gramm-Rudman-Hollings target regime, the Budget Enforcement Act regime of caps and PAYGO rules, the interval without statutory enforcement after fiscal 2002, and the combined Statutory PAYGO Act of 2010 and Budget Control Act of 2011 regime, the federal government recorded deficits in forty two of forty six fiscal years, produced four consecutive surpluses exactly once, and increased debt held by the public from 23.2 percent of gross domestic product in fiscal 1974 to 79.2 percent of gross domestic product in fiscal 2019 in the Congressional Budget Office historical data series. Causes divide among legislated tax policy, legislated benefit policy, wars, recessions, demographic change in retirement and health programs, and medical cost growth, and analysts named in the record assign different weights to each cause in each decade. Every regime governed the annual appropriations vote with some instrument, and every regime left the automatic growth of mandatory spending and interest to separate substantive legislation that the budget process could record, score, and obstruct at the margin, and could not substitute for. Rules bind the annual vote and not the automatic one, and forty five years of totals demonstrate the cost of the distinction in the only units that compound: percentage points of the national economy converted into debt.

What Coincided With the Surplus Years

The federal budget recorded a surplus in fiscal 1998, in fiscal 1999, in fiscal 2000 and in fiscal 2001, and the Office of Management and Budget Historical Tables series recorded a deficit in every other fiscal year from fiscal 1974 through fiscal 2019. That four year run therefore stood alone in a span of more than half a century bounded on one side by the last surplus of fiscal 1969 and on the other side by an unbroken sequence of deficits running from fiscal 2002 through fiscal 2019. The Office of Management and Budget Historical Tables series recorded a deficit of 290 billion dollars in fiscal 1992, a deficit of 22 billion dollars in fiscal 1997, a surplus of 69 billion dollars in fiscal 1998, a surplus of 126 billion dollars in fiscal 1999, a surplus of 236 billion dollars in fiscal 2000 and a surplus of 128 billion dollars in fiscal 2001. A movement of that size in six fiscal years invites a search for causes, and the record supplies several candidates that operated at the same time. Deficit reduction legislation enacted in 1990 and in 1993 changed the legislated path of revenues and spending. A new enforcement regime of discretionary caps and pay-as-you-go discipline, created by the Budget Enforcement Act of 1990 and extended in modified form by legislation in 1993 and in 1997, governed legislated changes during the decade. Revenues rose faster than forecasters had projected, with capital gains receipts running ahead of forecasts during the second half of the decade. Defense spending fell as a share of the economy after the end of the Cold War. Economic growth proceeded without a recession from March 1991 through March 2001, the interval bounded by the trough and peak months identified by the National Bureau of Economic Research. Each element coincided with the surpluses, and coincidence is the limit of what the chronological record alone establishes. The analytical dispute examined in Finding Two concerns weight, and named analysts assigned the weights differently.

The first candidate was legislation. The Omnibus Budget Reconciliation Act of 1990, Public Law 101-508, enacted on November 5, 1990, embodied the agreement reached at the 1990 budget summit. The summit agreement was recorded at about 500 billion dollars of deficit reduction over five years and the enacted effect at about 482 billion dollars, and Congressional Budget Office estimates reproduced in a Congressional Research Service compilation support the enacted figure in detail. That compilation distributed the 482 billion dollar total for fiscal 1991 to fiscal 1995 across revenue increases of 158 billion dollars, mandatory spending reductions of 75 billion dollars, discretionary spending reductions of 190 billion dollars and debt service savings of 59 billion dollars. The composition matters for attribution, because the largest single component was discretionary restraint delivered through caps. The Omnibus Budget Reconciliation Act of 1993, Public Law 103-66, enacted on August 10, 1993, supplied the second package. The same compilation placed the five year effect for fiscal 1994 to fiscal 1998 at 433 billion dollars, distributed across revenue increases of 241 billion dollars, mandatory spending reductions of 77 billion dollars, discretionary spending reductions of 69 billion dollars and debt service savings of 47 billion dollars. Contemporary public discussion often described the 1993 package as a 500 billion dollar plan, and the Office of Management and Budget projected deficit reduction of 504.8 billion dollars for the 1993 legislation, while the Congressional Budget Office projected 433 billion dollars. The gap between the two official estimates exceeded the entire surplus recorded in fiscal 1998, and attribution built on either estimate begins from a different starting point.

The enforcement regime supplied the mechanism that connected the two packages to annual decisions. The Budget Enforcement Act of 1990 replaced fixed deficit targets under the Gramm-Rudman-Hollings framework with two controls on legislated action, a set of caps on discretionary appropriations and a pay-as-you-go requirement for legislation affecting revenues and mandatory spending. A Brookings Institution analysis of the surplus, published in 1999 under the title A Surplus, If We Can Keep It, described the contrast in functional terms, observing that the 1990 law regulated the parts of the budget that the president and Congress directly controlled and left changes driven by economic conditions and by the cost of existing entitlement programs outside the enforcement trigger. That design distinction runs through the surplus period and through the wider record, because rules bind the annual vote and not the automatic one. Caps governed appropriations bills that Congress passed each year, and pay-as-you-go discipline governed new legislation, while spending under permanent law for programs such as Social Security, Medicare and Medicaid followed eligibility rules and demographic and health cost trends that no annual cap restrained. Alice Rivlin, who served as director of the Office of Management and Budget from 1994 to 1996 after serving as the first director of the Congressional Budget Office, presented the 1990 framework as an effective foundation in a Brookings Institution commentary published on November 16, 2010. Rivlin described the effort that produced balance as bipartisan, traced the sequence to the Budget Enforcement Act of 1990 enacted by a Democratic Congress and signed by President George H. W. Bush, and then described work with a Democratic Congress and later a Republican Congress during the Clinton administration that brought the deficit down and converted the deficit into a surplus for four years. An International Monetary Fund profile of Rivlin published in Finance and Development in June 2004 recorded a broader attribution in Rivlin, in which the transformation from deficits above 200 billion dollars to substantial surpluses reflected the efforts of both political parties and the extraordinary performance of the economy. The Rivlin account therefore assigned causal standing both to rules and legislation and to economic performance, without reducing the outcome to a single instrument.

The revenue surge supplied the second candidate, and the forecast record establishes that the surge surprised official estimators. The Brookings Institution analysis of 1999 recorded that the Office of Management and Budget had projected a fiscal 1998 deficit of 339 billion dollars in 1993 and that the Congressional Budget Office had projected a fiscal 1998 deficit of 357 billion dollars in 1993, against an outcome of a 69 billion dollar surplus in the Office of Management and Budget Historical Tables series for fiscal 1998. The same Brookings analysis recorded that the Congressional Budget Office in 1993 had projected cumulative deficits of 1.5 trillion dollars over the following five years under policies then in place, while actual cumulative deficits amounted to less than a third of that projected total. Forecast errors of that magnitude mean that legislated savings cannot by themselves explain the distance between projection and outcome, because a large part of the distance opened through developments that forecasters classified as economic and technical revisions. The Congressional Budget Office decomposition into legislative, economic and technical categories became the principal technical ground on which the attribution debate proceeded. In a paper prepared for a volume on American economic policy in the 1990s, Jeffrey Liebman of Harvard University applied the Congressional Budget Office record to the interval between January 1993 and 1998 and reported that two broad forces accounted for the improvement relative to the January 1993 forecast, namely legislation enacted by Democratic Congresses and improvements in the economy that the Congressional Budget Office had not foreseen in early 1993. The Liebman account also reported a Congressional Budget Office finding that more than half of the deficit reduction from the 1993 budget legislation came from lower spending and less than half from higher revenues, a finding that complicates partisan versions in both directions.

Capital gains formed the most distinctive part of the revenue surprise. A Senate Budget Committee Bulletin issued in February 2001, summarizing Congressional Budget Office material presented to the committee, recorded that capital gains realizations had contributed to rising surpluses in the preceding several years, that realizations had in large part surprised the Congressional Budget Office and other forecasters, and that most projections rested on a historical relationship between capital gains realizations and gross domestic product. The same bulletin recorded that realized capital gains had boosted receipts substantially during the preceding decade, that realizations had nearly doubled over two years as a share of individual income taxes, and that the Congressional Budget Office had calculated that capital gains realizations accounted for nearly 30 percent of the growth in individual income tax receipts in excess of nominal gross domestic product growth between 1995 and 1998. The Brookings Institution analysis of 1999 placed the broader revenue movement in proportional terms, recording that federal receipts rose from 18.2 percent of gross domestic product in 1990 to 20.5 percent of gross domestic product in 1998, an increase that the analysis valued at 190 billion dollars of additional revenue, and recording further that retention of the 1989 tax structure would have left no surplus for discussion. A Center on Budget and Policy Priorities tabulation drawn from the Office of Management and Budget Historical Tables and published in 2003 recorded a similar climb on a later data vintage, with federal receipts at 17.6 percent of gross domestic product in fiscal 1993 and at 20.8 percent in fiscal 2000.

The defense drawdown supplied the third candidate, and the proportional record is unambiguous even where causal interpretation diverges. A Mercatus Center study by David R. Henderson, a professor of economics at the Naval Postgraduate School, titled United States Federal Budget Restraint in the 1990s, A Success Story and published in 2015, recorded that federal spending fell from 21.9 percent of gross domestic product in 1990 to 18.2 percent of gross domestic product in 2000 and that defense accounted for 61.2 percent of the reduction in total spending as a share of the economy across the decade. The Henderson study recorded defense spending at 5.2 percent of gross domestic product in 1990 and at 3.0 percent of gross domestic product in 2000. The Brookings Institution analysis of 1999 approached the same movement in inflation adjusted dollar terms, recording that defense outlays in 1998 stood almost 100 billion dollars below the level of a decade earlier after adjustment for inflation, and the analysis added a counterfactual judgment that a continuation of the Cold War would probably have left no surplus. The drawdown followed the dissolution of the Soviet Union in December 1991. Part of the defense saving reduced the deficit and part financed domestic appropriations within the caps, and the division between those uses varied by year, which is one reason that a given defense figure supports different attribution conclusions in different accounts.

Economic performance formed the background condition for each candidate and a separate candidate in several accounts. The Brookings Institution analysis of 1999 recorded that the economy completed a seventh consecutive year of growth in 1998, that 13 million jobs were added during that growth interval, and that inflation averaged less than 3 percent across the interval. The same analysis nevertheless resisted a purely economic explanation, observing that the expansion of the 1980s, measured by growth rates from the end of the 1982 recession to the onset of the 1990 to 1991 recession, outperformed the expansion of the 1990s on that metric, while producing persistent deficits rather than surpluses. The contrast directed attention toward the structure of growth and toward policy settings, because growth concentrated in asset markets and in high income earnings generates more federal revenue per unit of gross domestic product growth than growth concentrated in wages near the median, given a progressive rate structure and given the 1990 and 1993 rate increases on upper incomes. In that account, legislation and economic structure interacted, because a rate schedule directed at upper incomes in 1990 and in 1993 met an income distribution and an asset market that amplified the yield of the schedule beyond official projections. Alan Auerbach of the University of California, Berkeley and William Gale of the Brookings Institution examined the durability of that yield in a 1999 study and in updates published by the Brookings Institution in 2000, and the Auerbach and Gale analysis treated a large part of the revenue surge as an assumption rather than as a secure foundation. Auerbach and Gale reported that more than 84 percent of the surge in income tax revenues relative to gross domestic product from 1994 to 1999 was assumed to be permanent in official projections, and the analysis questioned that assumption, since a surge linked to asset prices and to realizations behavior could reverse when asset prices fell. Auerbach and Gale also reported that more than 400 billion dollars of the projected on budget surplus in official forecasts arose from accumulations in trust funds for Medicare and for military and civilian pensions, balances that represented future obligations to workers rather than resources available for unrelated purposes, and the analysis concluded that plausible adjustments to discretionary spending assumptions and to revenue projections, combined with preservation of pension and health reserves, would more than eliminate any surplus available for tax cuts over the following ten years.

Did the Budget Rules Alone Produce the Surpluses?

The surplus years coincided with three conditions operating together: deficit reduction legislation enacted in 1990 and 1993 under enforceable caps and PAYGO rules, an exceptional revenue surge driven in part by capital gains, and a sustained decline in defense spending after the Cold War ended.

No named analysis in the record assembled for Finding Two assigns the entire outcome to the rules alone, and no named analysis assigns the rules no role at all, and the space between those positions contains the debate. On one side of the space, the Brookings Institution analysis of 1999 argued that policy choices carried weight equal to economic forces, presenting the liquidation of the deficit as a product of decisions in the 1990s that reversed decisions of the 1980s, and supporting that presentation with the revenue calculation that retention of the 1989 tax structure would have left no surplus. Rivlin, in the International Monetary Fund profile of June 2004 and in the Brookings Institution commentary of November 16, 2010, similarly placed legislated action and enforcement rules at the center of the account while adding economic performance and bipartisan cooperation as joint conditions. On another side of the space, the Liebman application of the Congressional Budget Office decomposition divided the improvement between legislation and unforecast economic improvement without subordinating either category, and the Congressional Budget Office decomposition method itself institutionalized that division by maintaining separate legislative, economic and technical ledgers for every revision of the outlook. On a third side of the space, Auerbach and Gale emphasized fragility, treating the revenue component as partly transitory and the projected surplus as partly an artifact of cash accounting for trust funds and of assumed restraint in discretionary spending, while Elkin and Greenstein at the Center on Budget and Policy Priorities emphasized that projected surpluses outside Social Security depended on caps compliance that appropriations behavior after 1998 placed in doubt. Henderson, in the Mercatus Center study of 2015, emphasized spending restraint as the central achievement of the decade and assigned the largest share of the spending reduction to defense, a distribution that supports a restraint centered account while also locating the largest restraint in the category most directly driven by geopolitical change rather than by domestic budget rules. The accounts diverge on counterfactuals that the historical record cannot settle directly, such as the path of revenues under the 1989 tax structure, the path of defense spending under a continuing Cold War, and the path of appropriations in the absence of caps.

The chronology also limits every attribution claim. The deficit fell in each fiscal year from fiscal 1993 through fiscal 1998 in the Office of Management and Budget Historical Tables series, moving from 255 billion dollars in fiscal 1993 to 203 billion dollars in fiscal 1994, to 164 billion dollars in fiscal 1995, to 107 billion dollars in fiscal 1996, to 22 billion dollars in fiscal 1997 and then to a surplus in fiscal 1998, and that steady progression began before the revenue surge of the second half of the decade reached full force and before the Balanced Budget Act of 1997, Public Law 105-33, enacted on August 5, 1997, could have produced substantial effects. The progression is consistent with a legislated path established in 1990 and in 1993 and amplified later by revenue developments, and the progression is also consistent with an economic expansion that strengthened year by year from the trough of March 1991, and chronology alone cannot separate those readings. Debt service savings appear inside the Congressional Budget Office scoring of both packages, at 59 billion dollars for the 1990 legislation and at 47 billion dollars for the 1993 legislation in the Congressional Research Service compilation, and such savings also follow automatically from any improvement in the primary balance from any source, so attribution of the interest saving reproduces the original dispute at one remove.

Finding Two therefore records coincidence in full and attribution in dispute. The only surplus period in more than half a century followed the 1990 and 1993 deficit reduction packages, operated under a caps and pay-as-you-go regime for legislated changes, coincided with a revenue surge in which capital gains receipts ran ahead of forecasts by margins that the Congressional Budget Office quantified, and coincided with a defense drawdown that moved defense spending from roughly 5 percent of gross domestic product in 1990 to roughly 3 percent in 2000. Congressional Budget Office analyses supplied the decomposition that structured the debate. Rivlin assigned joint credit to bipartisan legislated action and to extraordinary economic performance. The Brookings Institution analysis of 1999 assigned decisive weight to policy reversal while documenting the economic boom in detail. Liebman, applying Congressional Budget Office estimates, divided credit between legislation and unforecast economic improvement and recorded that the 1993 legislation delivered more through spending restraint than through revenue. Auerbach and Gale questioned the permanence of the revenue surge and the availability of the projected surplus. Elkin and Greenstein at the Center on Budget and Policy Priorities questioned the realism of the caps assumptions inside projected surpluses. Henderson assigned the largest share of spending restraint to defense. The division among those named analyses is the finding, and Finding Two presents the division without selecting a prevailing account, because the counterfactual budgets that selection would require were never enacted and never observed.

The Category the Caps Could Reach

The third finding concerns the one category that a statutory cap can reach directly, and the record for that category differs sharply from the record for the budget as a whole. Discretionary spending is the spending controlled through annual appropriation acts. Congress votes that spending anew for each fiscal year, and no agency may spend the money after the period of availability expires unless Congress enacts a further appropriation. A cap written in statute therefore attaches to an action that must recur. The finding is that the category shrank, substantially, as a share of the economy during the periods when statutory caps were in force, and that the shrinkage occurred twice, under two different cap regimes enacted twenty one years apart, against two different economic backgrounds. That repetition gives the category finding a weight that a single episode could not carry. The claim carried through the article, that rules bind the annual vote and not the automatic one, rests heavily on this evidence, because the category evidence is the part where the constrained action and the measured outcome are the same thing.

The mechanism explains why the trace should appear in the discretionary accounts if the trace appears anywhere. Under the regime created in 1990, the Office of Management and Budget measured enacted appropriations against the statutory limits for each capped year, reported a breach after the end of a session, and triggered a sequestration of resources inside the breached category when the measurement showed an excess. Under the Budget Control Act of 2011, a comparable measurement applied to the defense and nondefense categories separately, with the first sequestration order issued on March 1, 2013, reducing budgetary resources by 85.3 billion dollars, as recorded in the delivered account of the 2011 act. In neither regime did the sanction fall on the budget in general. The sanction fell on discretionary programs, account by account, and the threat operated during the drafting of appropriations bills and during scoring by the Congressional Budget Office. Appropriators worked inside an allocation that reflected the cap, and a bill that breached the allocation invited a point of order and a sequestration after enactment. Procedural discipline applied to a recurring annual decision compounds across a decade in a way that a one time instruction cannot.

The first era supplies the longer run of evidence. The historical appendix to the Congressional Budget Office outlook of January 2019, in Table F-3 and Table F-4, sourced to the Congressional Budget Office and the Office of Management and Budget, placed discretionary outlays at 8.5 percent of gross domestic product in fiscal 1990, the fiscal year that ended as the 1990 regime was enacted, and at 8.8 percent in fiscal 1991, the first fiscal year governed by the caps of the Budget Enforcement Act. The same series then recorded a decline in nearly every successive year of the decade. Discretionary outlays measured 8.3 percent of gross domestic product in fiscal 1992 and 8.0 percent in fiscal 1993 on the Congressional Budget Office and Office of Management and Budget series in Table F-3 of that appendix, then 7.5 percent in fiscal 1994, 7.2 percent in fiscal 1995, 6.7 percent in fiscal 1996, 6.5 percent in fiscal 1997, and 6.2 percent in fiscal 1998. The series placed discretionary outlays at 6.0 percent of gross domestic product in fiscal 1999 and at 6.1 percent in fiscal 2000. Measured from the first capped year, fiscal 1991, to fiscal 1999, the category surrendered 2.8 percentage points of gross domestic product. Measured from fiscal 1990, the decline to fiscal 1999 was 2.5 percentage points. Either measurement describes a category that grew far more slowly than the economy, across caps that Congress extended in 1993 and again in 1997.

The composition of the first decline matters. Table F-4 of the historical appendix to the Congressional Budget Office outlook of January 2019, sourced to the Congressional Budget Office and the Office of Management and Budget, divided the total between defense and nondefense outlays. Defense discretionary outlays fell from 5.1 percent of gross domestic product in fiscal 1990, and 5.2 percent in fiscal 1991, to 3.0 percent in fiscal 1998 and 2.9 percent in fiscal 1999 and fiscal 2000 on that series. Nondefense discretionary outlays measured 3.4 percent of gross domestic product in fiscal 1990 and 3.5 percent in fiscal 1991 on the same series, then 3.2 percent in fiscal 1998, 3.1 percent in fiscal 1999, and 3.2 percent in fiscal 2000. The defense decline was therefore much larger, accounting for most of the movement in the total, and the post Cold War drawdown described in the surplus attribution record of the preceding articles formed a major part of the explanation for the defense figures. The nondefense movement was smaller and steadier, a few tenths of a percentage point across a decade, yet the direction was consistent with a binding constraint. Attribution remains contested between analyses that emphasize the drawdown and analyses that emphasize the statutory limits. The category finding requires only the narrower observation that the capped category, taken whole, fell substantially as a share of output during the capped years, and that observation holds on the named series under either attribution.

Nominal dollars sharpen the distinction between restraint and reduction. Table F-3 of the historical appendix to the Congressional Budget Office outlook of January 2019, sourced to the Congressional Budget Office and the Office of Management and Budget, placed discretionary outlays at 500.6 billion dollars in fiscal 1990 and at 572.1 billion dollars in fiscal 1999. The category did not shrink in dollars during the first cap era. The caps bound the growth rate of the category relative to a rapidly growing economy, and the difference between those two growth rates, compounded, produced the decline in share.

Did the caps actually reduce discretionary spending as a share of the economy?

Discretionary spending fell as a share of gross domestic product in both cap eras on the Congressional Budget Office and Office of Management and Budget series. The decline measured from fiscal 1991 to fiscal 1999 and from fiscal 2011 to fiscal 2019 occurred despite repeated cap increases and funding placed outside the caps.

The second era repeats the pattern from a higher starting point and under harsher fiscal conditions. The Congressional Budget Office and Office of Management and Budget series carried in the delivered account of the Budget Control Act placed discretionary outlays at 9.1 percent of gross domestic product in fiscal 2010 and at 8.7 percent in fiscal 2011, lifted by military operations in Iraq and Afghanistan and by funding under the American Recovery and Reinvestment Act of 2009. The historical appendix to the Congressional Budget Office outlook of January 2019, in Table F-3, sourced to the Congressional Budget Office and the Office of Management and Budget, recorded the subsequent descent year by year: 8.0 percent of gross domestic product in fiscal 2012, 7.2 percent in fiscal 2013, 6.8 percent in fiscal 2014, 6.5 percent in fiscal 2015, 6.4 percent in fiscal 2016, and 6.2 percent in both fiscal 2017 and fiscal 2018. The Congressional Budget Office long term outlook of June 2019 placed discretionary outlays at 6.3 percent of gross domestic product in fiscal 2019. From fiscal 2011 to fiscal 2019 the decline was 2.4 percentage points of gross domestic product, and from the fiscal 2010 peak the decline was 2.8 percentage points, an echo, in scale, of the decline between fiscal 1991 and fiscal 1999.

The second decline, like the first, reached both halves of the category. Table F-4 of the historical appendix to the Congressional Budget Office outlook of January 2019, sourced to the Congressional Budget Office and the Office of Management and Budget, placed defense discretionary outlays at 4.6 percent of gross domestic product in fiscal 2010 and at 4.5 percent in fiscal 2011, then at 3.1 percent in fiscal 2017 and fiscal 2018. The same series placed nondefense discretionary outlays at 4.4 percent of gross domestic product in fiscal 2010 and at 4.2 percent in fiscal 2011, then at 3.2 percent in fiscal 2017 and fiscal 2018. The delivered account of the 2011 act, drawing on the same Congressional Budget Office and Office of Management and Budget series, recorded that discretionary outlays totaled about 1.35 trillion dollars in fiscal 2011 and again totaled roughly 1.35 trillion dollars in fiscal 2019, so that a category spending about the same nominal sum eight years apart occupied a far smaller share of a larger economy and a smaller share of total federal outlays, falling from close to 40 percent of outlays in fiscal 2011 to about 30 percent in fiscal 2019. Wind down of military operations and expiration of recovery funding contributed to the early part of the descent, as Congressional Budget Office testimony of 2011 recorded in describing the elevated fiscal 2010 and fiscal 2011 levels. The descent continued through fiscal 2018, after those temporary factors had receded, under caps in statute and a sequestration that had operated once, in March 2013.

The contrast with the mandatory side of the budget gives the finding analytical force. Mandatory spending does not pass through the annual appropriations process. Eligibility rules and benefit formulas in permanent law determine the outlays, with the total driven by the number of eligible beneficiaries and by the cost per beneficiary that the formula incorporates. The Congressional Budget Office outlook of January 2019 attributed growth in the largest mandatory components to the aging of the population and to growth in health care spending per beneficiary, forces that operate through demographics and medical costs rather than through an annual vote. Table F-5 of the historical appendix to that outlook, sourced to the Congressional Budget Office and the Office of Management and Budget, placed Social Security outlays at 4.1 percent of gross domestic product in fiscal 1990. The Congressional Budget Office baseline tables of August 2019 placed Social Security at 4.9 percent of gross domestic product in fiscal 2019 and the major health care programs, principally Medicare, Medicaid, the Children’s Health Insurance Program, and marketplace subsidies under the Affordable Care Act, at 5.3 percent of gross domestic product in fiscal 2019. No cap on appropriations touched either figure, because neither figure originated in an appropriation.

The fiscal 2019 composition states the structural result in a single year. The Congressional Budget Office baseline tables of August 2019 divided fiscal 2019 outlays into mandatory spending of 12.8 percent of gross domestic product, discretionary spending of 6.3 percent, and net interest of 1.8 percent, against total outlays of 20.8 percent of gross domestic product. Mandatory spending, on that accounting, was roughly double the capped category. Net interest grew from accumulated debt rather than from any annual decision, and the Congressional Budget Office update of August 2019 estimated net interest at 372 billion dollars in fiscal 2019, up from 325 billion dollars in fiscal 2018. A ceiling that held the discretionary category at 6.3 percent of gross domestic product therefore governed less than one third of total outlays in fiscal 2019, while the remaining share moved under formulas, eligibility, demographics, health costs, and interest on past borrowing. The mandatory sequestration attached to the Budget Control Act reached only a limited slice of that larger share, exempting Social Security, Medicaid, and most low income programs entirely and limiting the reduction for Medicare to 2 percent of payments, as the delivered account of the 2011 act recorded.

That arithmetic explains the divergence at the center of the article. The deficit rose during the second cap era even as the capped category fell. The Congressional Budget Office monthly budget review for fiscal 2019 recorded a deficit of 585 billion dollars in fiscal 2016 and 665 billion dollars in fiscal 2017, then 779 billion dollars in fiscal 2018 and 984 billion dollars in fiscal 2019, equal to 4.6 percent of gross domestic product in fiscal 2019. Discretionary outlays fell as a share of gross domestic product across the same span, from 6.5 percent in fiscal 2015 to 6.3 percent in fiscal 2019 on the Congressional Budget Office and Office of Management and Budget series, with the intervening years at 6.4 percent in fiscal 2016 and 6.2 percent in fiscal 2017 and fiscal 2018. A shrinking category and a growing deficit coexisted because the deficit is a residual of total outlays and total revenues, and the forces enlarging the residual sat outside the category the caps constrained. Revenue legislation enacted in December 2017 as Public Law 115-97 reduced revenues relative to the prior baseline, as scored by the staff of the Joint Committee on Taxation and by the Congressional Budget Office in April 2018, and the 2011 act contained no mechanism connecting the ceilings to a revenue decision. The two findings are compatible: each describes a different object, and the distance between the objects measures the design limit.

Honest treatment of the category finding requires the erosion routes to be stated at full strength, because the caps that produced the measured decline were not the caps as originally enacted in either era. In the second era, Congress and the President raised the statutory limits four times through the Bipartisan Budget Act of 2013, Public Law 113-67, the Bipartisan Budget Act of 2015, Public Law 114-74, the Bipartisan Budget Act of 2018, Public Law 115-123, and the Bipartisan Budget Act of 2019, Public Law 116-37. Each enactment lifted the defense and nondefense ceilings for the two fiscal years the enactment covered, and appropriations for those years were written to the raised limits rather than to the limits in the Budget Control Act as enacted in August 2011. The sequestration level therefore operated in full statutory form principally in fiscal 2013, and the later years of the era ran under negotiated ceilings. In the first era, erosion took a different form, through adjustments for amounts designated as emergencies and through supplemental appropriations examined in a Congressional Budget Office study of March 2001. Emergency designation in the first era and negotiated cap increases in the second era both softened the constraint relative to the enacted text.

Funding placed outside the caps supplies the second erosion route, and the second era gives that route a precise dollar measure. The delivered account of the Budget Control Act recorded 723 billion dollars of overseas contingency operations adjustments from fiscal 2012 to fiscal 2019, funding for military operations and related activities that sat outside the discretionary caps through a statutory adjustment. Appropriations carrying that designation did not count against the defense ceiling, did not require an offset inside the capped category, and did not expose other defense accounts to sequestration. The route was lawful, visible in the budget documents, and large enough to matter in any assessment of defense restraint, because a material share of operational funding during the capped years never passed through the ceiling at all. A parallel route operated through emergency designations for domestic purposes under the adjustment provisions of the 2011 act. Erosion reduced the restraint relative to a counterfactual of unamended caps.

The qualification does not erase the measured decline, for a reason grounded in the construction of the series itself. The discretionary outlay figures in Table F-3 and Table F-4 of the historical appendix to the Congressional Budget Office outlook of January 2019 record outlays, and outlays include spending financed through overseas contingency operations funding and through emergency designations, because the classification follows the appropriations process rather than the cap adjustment. Money appropriated outside the caps appears in the discretionary total when the money is spent. The decline from 8.7 percent of gross domestic product in fiscal 2011 to 6.3 percent in fiscal 2019 on the Congressional Budget Office long term outlook figure for fiscal 2019, and the decline from 8.8 percent in fiscal 1991 to 6.0 percent in fiscal 1999 on Table F-3, therefore already count the eroded spending. Appropriations written to raised ceilings from fiscal 2014 to fiscal 2019 likewise produced outlays that fell or held steady as a share of gross domestic product on the Table F-3 series, because economic growth outpaced the raised appropriations as well as the original path.

A further qualification concerns causation, and the evidence supports a bounded rather than an exclusive claim. Both cap eras contained long expansions, defense drawdowns contributed heavily in each era, and expiring temporary funding contributed at the start of the second descent. Against those contributions stands the institutional fact that appropriations require affirmative enactment every year, that the caps supplied a statutory reference point for the size of that enactment, that sequestration supplied a consequence for exceeding the reference point, and that the two descents align in time with the two cap regimes rather than with the uncapped decade between fiscal 2002 and fiscal 2011, when discretionary outlays rose from 6.8 percent of gross domestic product in fiscal 2002 to 9.1 percent in fiscal 2010 on Table F-3 of the historical appendix to the Congressional Budget Office outlook of January 2019. Alignment in time does not prove causation in full. Alignment across two regimes and two sets of economic conditions, combined with a mechanism connecting the statute to the annual decision, supports the narrower conclusion that the caps contributed materially to the restraint and that the contribution survived repeated statutory weakening.

Finding Three, stated in full, is therefore a finding about reach. A statutory cap reached the category that Congress votes every year, and that category fell substantially as a share of the economy in fiscal 1991 to fiscal 1999 and again in fiscal 2011 to fiscal 2019, on the Congressional Budget Office and Office of Management and Budget series, even after cap raising enactments in 2013, 2015, 2018, and 2019 and even after 723 billion dollars of overseas contingency funding from fiscal 2012 to fiscal 2019 is counted inside the category total. The same cap reached no formula in permanent law, no eligibility rule, no demographic trend, and no interest payment on accumulated debt, and the mandatory and interest shares of the budget moved under those other forces while the capped category contracted. The deficit grew while the capped category shrank, in the second era, because the deficit answers to the whole budget and the caps answered to a third of the budget. That asymmetry is the strongest evidence for the proposition that a budget rule binds the decision the rule is attached to, and for the limit of the proposition, because that decision was never the decision that determined the deficit.

The Rule That Operates as a Default

The Statutory Pay-As-You-Go Act of 2010, enacted on February 12, 2010 as Public Law 111-139, restored a statutory pay-as-you-go requirement a decade after the earlier statutory regime had lapsed. The Congressional Research Service summary of the act, Report R41157, describes a scorekeeping system administered by the Office of Management and Budget. Under that system, the budgetary effects of legislation affecting direct spending and revenues are recorded on two cumulative scorecards, one covering five years and one covering ten years, with the effects for each measure averaged across the years of the relevant period and with effects in the current year combined into the budget year. If either scorecard shows a debit for the budget year at the end of a session of Congress, the act directs a sequestration of nonexempt direct spending sufficient to eliminate the debit. Section 11 of the act preserves a list of exempt programs, and reductions to Medicare under a pay-as-you-go sequestration order are capped at 4 percent. The design placed an automatic consequence behind a legislative ledger: a measure that increased the deficit did not need to be defeated for the statute to operate.

The record from February 2010 to April 15, 2020 shows a different sequence in practice. Legislation that increased the deficit was enacted repeatedly during that period, and the end of session arithmetic repeatedly arrived at a balance that required no sequestration order, because intervening statutes had removed, excluded, shifted, or reset the debits before the arithmetic was performed. The Congressional Budget Office record confirms that no sequestration was ever triggered under the act through April 2020. That outcome was not the product of a single waiver applied to a single measure. It was the product of several distinct routes, each written into statute, each used on identified dates, and each leaving the underlying requirement in force for the next measure. The requirement survived every episode intact as law. The debits did not survive. The requirement operated as a default, not as a constraint.

The first route was built into the 2010 act at enactment. Section 4(g) of the act provides that amounts resulting from a provision designated as an emergency requirement are not counted as budgetary effects and therefore are not recorded on the scorecards. The Congressional Research Service summary in Report R41157 describes the accompanying procedure, under which an emergency designation in the Senate is protected by a point of order that requires the affirmative vote of three fifths of the Senate to waive. Sections 4(c) and 7 of the act provide current policy adjustments under which the costs of specified policies are excluded from the scorecards within stated limits, covering legislation affecting Medicare physician payments, the estate and gift tax, the alternative minimum tax, and certain tax reductions first enacted in 2001 and 2003. Section 4(d) excludes additional items, including debt service costs. The act also reaches only on-budget effects, so the direct spending and revenues of the Social Security trust funds and the Postal Service Fund sit outside the ledger by definition. Each exclusion is stated in the text of Public Law 111-139, and each narrows the set of deficit increases to which the sequestration consequence can ever attach.

The emergency route was used within weeks of enactment and at a scale that defined the first year of the regime. The Office of Management and Budget 2010 Statutory Pay-As-You-Go Annual Report, published on January 20, 2011, records five laws enacted during the second session of the 111th Congress that contained provisions designated as emergency requirements under the act: the Temporary Extension Act of 2010, Public Law 111-144, enacted on March 2, 2010; the Continuing Extension Act of 2010, Public Law 111-157, enacted on April 15, 2010; the Unemployment Compensation Extension Act of 2010, Public Law 111-205, enacted on July 22, 2010; an act extending a deadline for Social Services Block Grant expenditures, Public Law 111-285; and the Tax Relief, Unemployment Insurance Reauthorization, and Job Creation Act of 2010, Public Law 111-312, enacted on December 17, 2010. In the Office of Management and Budget 2010 annual report series, the total costs excluded through emergency designations were 570.1 billion dollars over the period from 2010 to 2015 and 545.1 billion dollars over the period from 2010 to 2020.

The current policy adjustments operated alongside the emergency designations in the same first year. In the Office of Management and Budget 2010 annual report series, the total costs excluded from the scorecards through current policy adjustments were 436.4 billion dollars over the period from 2010 to 2015 and 433.8 billion dollars over the period from 2010 to 2020. The same report records that the laws enacted during the second session of the 111th Congress were estimated, before emergency designations, current policy adjustments, and other adjustments, to increase the on-budget deficit by 899.4 billion dollars over the period from 2010 to 2015 and by 820.1 billion dollars over the period from 2010 to 2020, and that a net of 883.8 billion dollars in costs over the period from 2010 to 2020 was enacted but excluded from the scorecards. Eight laws of that session contained provisions subject to current policy adjustments, as listed in the Office of Management and Budget report: Public Laws 111-144, 111-148, 111-157, 111-192, 111-240, 111-286, 111-309, and 111-312.

Was a Sequestration Order Ever Issued Under the Act?

No. The Office of Management and Budget annual reports for each session from 2010 through 2019 recorded no debit requiring an order, and the Congressional Budget Office record confirms that sequestration was never triggered under the act through April 2020.

The second route is a directed exclusion written into a later statute, instructing that the budgetary effects of a measure, or of specified divisions of a measure, shall not be entered on either scorecard maintained under section 4(d) of the 2010 act. The clearest instance concerns the largest single debit of the period. The Congressional Research Service, drawing on estimates by the Congressional Budget Office and the staff of the Joint Committee on Taxation, records that the Tax Cuts and Jobs Act, Public Law 115-97, enacted on December 22, 2017, was projected at enactment to increase the deficit by 1.46 trillion dollars over the period from 2018 to 2027. The Congressional Budget Office advised in November 2017 that, absent subsequent legislation to offset the increase or waive the recordation of the effects, the Office of Management and Budget would be required to issue a sequestration order reducing spending in fiscal 2018 by 136 billion dollars, with reductions to Medicare limited to four percentage points, or roughly 25 billion dollars for fiscal 2018 in the Congressional Budget Office estimate of November 2017, and with the remaining amount drawn from a pool of nonexempt accounts estimated at between 85 billion dollars and 90 billion dollars of available resources in each year of the period from 2018 to 2027. No such order issued. Section 5002 of Public Law 115-96, the Further Continuing Appropriations Act, 2018, enacted on December 22, 2017 on the same day as Public Law 115-97, excluded the Tax Cuts and Jobs Act entirely from the scorecards, as recorded in the Office of Management and Budget 2017 Statutory Pay-As-You-Go Annual Report published on January 26, 2018.

Directed exclusions then recurred as a standard closing provision in major measures during 2018 and 2019. The Consolidated Appropriations Act, 2018, Public Law 115-141, enacted on March 23, 2018, provides in section 701 of Division M that the budgetary effects of that division and each succeeding division shall not be entered on either pay-as-you-go scorecard. The VA MISSION Act of 2018, Public Law 115-182, enacted on June 6, 2018, provides in section 512 that the budgetary effects of the act shall not be entered on the statutory or Senate pay-as-you-go scorecards. The FAA Reauthorization Act of 2018, Public Law 115-254, enacted on October 5, 2018, was recorded among the laws exempted from the scorecard in whole or in part in the same scorecard history, as was the Substance Use-Disorder Prevention that Promotes Opioid Recovery and Treatment for Patients and Communities Act, Public Law 115-271, enacted in October 2018. In the following session, the Office of Management and Budget 2019 Statutory Pay-As-You-Go Annual Report, published on January 27, 2020, records that two laws were excluded entirely from the scorecards, the Never Forget the Heroes Act, Public Law 116-34, enacted on July 29, 2019, whose section 4 directs that its budgetary effects shall not be entered on either scorecard, and the Bipartisan Budget Act of 2019, Public Law 116-37, enacted on August 2, 2019, and that budgetary effects in four further laws were excluded in part, including the Consolidated Appropriations Act, 2019, Public Law 116-6, and the Further Consolidated Appropriations Act, 2020, Public Law 116-94. The exclusion sentence had become a routine component of major legislation, effective at enactment and requiring no justification under the 2010 act beyond enactment itself.

The third route is a reset, under which a later statute sets the accumulated balances on both scorecards to zero and removes debits that earlier measures had placed on the ledger. The Bipartisan Budget Act of 2018, Public Law 115-123, enacted on February 9, 2018, nullified the balances on the five-year and ten-year scorecards as of the date of enactment, a step recorded in the scorecard history maintained for the House Committee on the Budget as a wiping of the scorecard. Debits that had survived the exclusion routes used in 2017 ceased to count for any future sequestration determination once Public Law 115-123 took effect. The Bipartisan Budget Act of 2019, Public Law 116-37, enacted on August 2, 2019, repeated the step, and section 102 of that act set the scorecard balances established by the Statutory Pay-As-You-Go Act of 2010 to zero, as recorded in the Office of Management and Budget 2019 annual report of January 27, 2020. A reset differs from an exclusion in one respect only. An exclusion keeps a debit off the ledger from the start, and a reset removes a debit that the ledger had already recorded. Both end at the same place, a budget year balance of zero, and both are accomplished by a provision in a later public law.

The fourth route is a shift, under which a later statute deducts the debit for the budget year from both scorecards and adds the same debit to the scorecards for a later year, deferring the sequestration determination rather than eliminating the amount. The Further Additional Continuing Appropriations Act, 2019, Public Law 116-5, enacted on January 25, 2019, provides in section 104 that any debit on the scorecard for 2019 shall be deducted from the scorecard in 2019 and added to the scorecard in 2020. The Further Continuing Appropriations Act, 2020, and Further Health Extenders Act of 2019, Public Law 116-69, enacted on November 21, 2019, provides in section 1801 both an exemption of its own budgetary effects from the Statutory Pay-As-You-Go Act of 2010 and a direction that any debit on the statutory scorecard for 2020 remaining at the end of the first session of the 116th Congress be moved to the scorecard for 2021. The Office of Management and Budget 2019 annual report of January 27, 2020 states the combined result of the 2019 session in plain terms: the budget year balance on each scorecard was zero because Public Law 116-37 had removed all balances included on the scorecards at the time of its enactment and Public Law 116-69 had shifted the debits on both scorecards from fiscal 2020 to fiscal 2021. A shift preserves the debit in form and removes the consequence in fact for the year in which the consequence would have fallen due.

The emergency route also remained in use at the end of the period. The Coronavirus Aid, Relief, and Economic Security Act, Public Law 116-136, enacted on March 27, 2020, provides in section 6001 of Division F that the amounts provided under the act are designated as an emergency requirement pursuant to section 4(g) of the Statutory Pay-As-You-Go Act of 2010. The designation placed the budgetary effects of the largest measure enacted during the period from January to April 2020 outside the scorecards by the mechanism the 2010 act itself supplied, eighteen days before the April 15, 2020 vantage that closes the record for this finding. Taken together, the four routes cover every major deficit-increasing measure of the decade that the record identifies, and the sequestration order that the statute directs in the presence of a budget year debit was issued on no occasion from February 12, 2010 to April 15, 2020.

The explanation lies in the structure of the requirement. A statutory pay-as-you-go debit has no consequence until the end of a session, and the consequence is a statute-dependent calculation performed on balances that later statutes may lawfully alter at any point before the calculation is made. Every route described in this finding travels through the same gate: a provision enacted through the ordinary procedures that govern any other provision of comparable importance. Directed exclusions, resets, and shifts in enacted appropriations and budget acts faced no threshold beyond enactment of the vehicle that carried them. The sequestration base reinforces the same conclusion. Section 11 of the 2010 act exempts major direct spending programs from reduction, Medicare reductions are capped at 4 percent, and the Congressional Budget Office estimate of November 2017 for the Tax Cuts and Jobs Act recorded a required fiscal 2018 reduction of 136 billion dollars against an available pool of nonexempt resources estimated at between 85 billion dollars and 90 billion dollars a year, an arithmetic gap that meant the order contemplated by the statute could not have been executed in full even if the debit had remained on the scorecard. A constraint that cannot be executed in full, against a base narrowed by exemption, behind a ledger that later statutes may rewrite, functions in practice as the opening position for the next negotiation. Rules bind the annual vote and not the automatic one, and the waiver record is the evidence: the votes that set the requirement aside were taken, on dated occasions, by the same legislative process that the requirement was meant to govern.

The longer pattern places the 2010 record between two contrasting designs. The Balanced Budget and Emergency Deficit Control Act of 1985, Public Law 99-177, enacted on December 12, 1985, set maximum deficit targets declining to balance by fiscal 1991 and attached sequestration to the level of the deficit itself, an outcome that economic conditions, revenue performance, and estimation error could move independently of any single legislative decision. The targets were not met on the original schedule. The Balanced Budget and Emergency Deficit Control Reaffirmation Act of 1987, Public Law 100-119, revised the targets upward and extended the date for balance to fiscal 1993, and a Congressional Research Service account of spending limits records the fiscal 1990 target as an instance, raised from 36 billion dollars under the 1985 act to 100 billion dollars under the 1987 act. The record of the 1985 regime adds the accompanying techniques of the period: optimistic assumptions, payment date shifting, and transfers of programs off budget, applied while the targets were amended rather than met. Amendment was the characteristic response of the 1985 regime because the object of enforcement, the deficit total, could not be brought within the target by scorekeeping alone.

The regime enacted in 1990 differed in design, and the record differed with the design. The Budget Enforcement Act of 1990, Title XIII of Public Law 101-508, replaced fixed deficit targets with statutory limits on discretionary spending and a pay-as-you-go requirement applied to new direct spending and revenue legislation, tying sequestration to the recorded effects of enacted measures rather than to the deficit total. The Congressional Research Service account of the 1990 change states the reason in terms of accountability: the earlier process held the legislative and executive branches answerable for an outcome affected by factors beyond direct control, and the 1990 process confined the ledger to the consequences of legislation. That regime operated through the surplus years, fiscal 1998, fiscal 1999, fiscal 2000, and fiscal 2001, the only surplus years of the period from fiscal 1974 to fiscal 2019, and its procedures were extended in 1993 and again in 1997 rather than rewritten at the first point of pressure. The terminal episode of the earlier regime belongs to the same record of legislative adjustment, since Public Law 107-312, enacted on December 2, 2002, set the accumulated pay-as-you-go balances to zero and thereby prevented a sequestration for fiscal 2003, as the scorecard history maintained for the House Committee on the Budget records, but that adjustment arrived at the expiry of the procedures after a decade of operation, rather than as a recurring feature of each session. The 2010 act adopted the legislative ledger design of 1990 and added broader exits: the emergency designation of section 4(g), the current policy adjustments of sections 4(c) and 7, and the exclusions of section 4, each present in the text from February 12, 2010, and each used from March 2010 onward.

Finding Four follows from the dated record alone. Between February 12, 2010 and April 15, 2020, the emergency designations of 2010 in Public Laws 111-144, 111-157, 111-205, 111-285, and 111-312, the current policy adjustments applied to eight laws of the second session of the 111th Congress, the exclusion of the Tax Cuts and Jobs Act by section 5002 of Public Law 115-96 on December 22, 2017, the directed exclusions in Public Laws 115-141, 115-182, 115-254, 116-34, and 116-37, the resets in Public Law 115-123 on February 9, 2018 and Public Law 116-37 on August 2, 2019, the shifts in Public Law 116-5 on January 25, 2019 and Public Law 116-69 on November 21, 2019, and the emergency designation in Public Law 116-136 on March 27, 2020 each set aside, deferred, or kept off the scorecards the deficit effects of enacted legislation. No sequestration order issued under the act in that period, as the Congressional Budget Office record and the annual reports of the Office of Management and Budget for each session from 2010 to 2019 agree. The default governed the form of enactment, adding a designation, an exclusion sentence, a reset, or a shift to measures that increased the deficit. The default did not govern the substance of enactment, and the deficit effects of the measures became law on the dates listed in this finding regardless of the ledger.

Regular Order and the Lapse Record

Finding Five concerns process rather than totals. The Congressional Budget and Impoundment Control Act of 1974, Public Law 93-344, posed a testable proposition: a statutory timetable, a pair of Budget Committees, a Congressional Budget Office, and a sequence of enforceable steps could make the federal government decide annual spending on schedule, in separate bills, after a prior agreement on aggregates. The Congressional Research Service record gives a clear answer on the timing half of that design. Congress completed all of the regular appropriations acts before the start of the fiscal year in only four fiscal years after the framework took effect, fiscal 1977, fiscal 1989, fiscal 1995, and fiscal 1997, according to the Congressional Research Service record in Continuing Resolutions: Overview of Components and Practices. In every other fiscal year from fiscal 1977 to fiscal 2020, action on at least some regular bills ran past October 1, and interim funding, consolidated packaging, or a lapse filled the space between the calendar and the enacted result.

The timetable that failed deserves a precise description. Section 300 of the Congressional Budget Act of 1974, codified at 2 U.S.C. 631 and reproduced in the Congressional Research Service report The Congressional Budget Process Timeline, set the sequence in compressed form: presidential budget by the first Monday in February; Congressional Budget Office outlook report by February 15; committee views and estimates within six weeks of the presidential submission; Senate Budget Committee report of the budget resolution by April 1; completion of the budget resolution by April 15; House consideration of appropriations bills after May 15; House Appropriations Committee report of the last regular bill by June 10; completion of reconciliation legislation by June 15; completion of House action on appropriations bills by June 30; and the start of the fiscal year on October 1. The same statute moved the start of the fiscal year from July 1 to October 1, lengthening the period available for action by three months, according to the Congressional Research Service report Continuing Resolutions: Overview of Components and Practices.

Regular order had a substantive meaning inside that calendar. Appropriations for most federal agencies arrive through separate regular bills, each produced by an Appropriations subcommittee with a defined jurisdiction, each measured against an allocation derived from the budget resolution, and each subject to amendment, conference, and enactment as a distinct law. The Congressional Research Service table in Continuing Resolutions: Overview of Components and Practices records thirteen regular bills under the subcommittee structure of the 95th through the 108th Congresses, eleven bills during the 109th Congress after a subcommittee realignment, and twelve bills from the 110th Congress onward. A process that finishes through one consolidated measure, or through a resolution that continues a prior rate of spending, performs the constitutional function of appropriating money while discarding the deliberative structure that the 1974 design attached to the function. Few years satisfied both tests, because the four on-time years recorded by the Congressional Research Service are the only years from fiscal 1977 to fiscal 2020 in which the full set of regular bills preceded October 1.

Did the Timetable Impose Discipline on Appropriations Timing?

No. The Congressional Research Service record shows that all regular appropriations acts reached enactment before October 1 in only four fiscal years from fiscal 1977 through fiscal 2019, while continuing resolutions supplied interim funding in most other years and funding gaps interrupted operations in twenty separate episodes during the same period.

The four exceptions repay attention because each exception proves attainable a sequence that surrounding years abandoned. Fiscal 1977 closed with all thirteen regular bills enacted on time, although the Congressional Research Service record in Continuing Resolutions: Overview of Components and Practices notes that two continuing resolutions were enacted for fiscal 1977 to fund certain unauthorized programs dropped from the regular bills. Fiscal 1989, fiscal 1995, and fiscal 1997 each closed with thirteen of thirteen regular bills enacted on or before October 1 and no continuing resolutions, according to the fiscal year table in the same report. After fiscal 1997, the table records no fiscal year through fiscal 2020 in which the full set arrived by the deadline. Partial completion became the recurring pattern: five regular bills enacted on or before October 1 for fiscal 2019, three for fiscal 2009, one for fiscal 2017, and zero for fiscal 2002, fiscal 2003, fiscal 2008, and each fiscal year from fiscal 2011 through fiscal 2016.

The continuing resolution is the instrument that converted a missed deadline into a governable condition, and the Congressional Research Service report Continuing Resolutions: Overview of Components and Practices supplies the counts. Congress enacted one or more continuing resolutions in all but three of the forty-four fiscal years from fiscal 1977 through fiscal 2020, and the total reached 188 continuing resolutions, ranging from zero to twenty-one in a single fiscal year and averaging about four per fiscal year. Fiscal 2001 required twenty-one continuing resolutions, fiscal 1996 required thirteen, and fiscal 2011 and fiscal 2002 required eight each, according to the fiscal year table in the same report. A continuing resolution typically funds agencies at a rate derived from the prior fiscal year, for a specified period, under the terms and conditions of the prior appropriations acts, with specified exceptions called anomalies where Congress adjusts the rate or a condition for a particular account. The consequences follow from the form. A rate for operations freezes allocations at a prior decision. A prohibition on new starts, a standard component described in the Congressional Research Service overview of continuing resolution components, prevents agencies from beginning projects that received no funding in the prior year.

Consolidated packaging supplied the second substitute for regular order. A consolidated or omnibus appropriations act completes the year by combining several regular bills into one measure, negotiated as a package and enacted as a single law. The Congressional Research Service funding gap chronology for fiscal 2014 illustrates the sequence: no regular bills had been enacted at the start of fiscal 2014 on October 1, 2013, a continuing resolution enacted on October 17, 2013 carried funding to January 15, 2014, and a consolidated appropriations act then funded the remainder of the fiscal year. Fiscal 2019 shows the same packaging without a lapse at the start of the year. The Congressional Research Service report Past Government Shutdowns: Key Resources records that five of the twelve regular bills for fiscal 2019 had been enacted in two consolidated measures, Public Law 115-244 and Public Law 115-245, before October 1, 2018, while the other seven bills rested on continuing resolutions. Packaging changes the decision structure. Separate bills allow separate majorities to form around separate subjects and allow a failure to attach to one bill; a package joins subjects into a single decision, compresses floor consideration into the time before a funding deadline, and moves allocation choices from subcommittee markups into negotiations over the combined text.

A third substitute carried substitution to a limit: the full-year continuing resolution, under which interim funding becomes final funding for most of a fiscal year. The Congressional Research Service report Full-Year Continuing Resolutions: Frequently Asked Questions counts sixteen measures containing full-year continuing resolutions for at least one regular bill since fiscal 1977, and lists four such measures enacted since 2000. The Revised Continuing Appropriations Resolution, 2007, Public Law 110-5, enacted on February 15, 2007, carried nine of eleven regular bills for the remainder of fiscal 2007. The Full-Year Continuing Appropriations Act, 2011, enacted as Division B of Public Law 112-10 on April 15, 2011, carried eleven of twelve regular bills for fiscal 2011. The Further Continuing Appropriations Act, 2013, enacted as Division F of Public Law 113-6 on March 26, 2013, carried seven of twelve regular bills for fiscal 2013.

Failure upstream compounded failure downstream, because the budget resolution that was supposed to precede and discipline the appropriations bills also missed the statutory calendar. The Congressional Research Service report Congressional Budget Resolutions: Historical Information records that Congress adopted the budget resolution by the April 15 target date only six times from fiscal 1976 through fiscal 2015, and that Congress completed action on no budget resolution for nine fiscal years in the same period. The Congressional Research Service report Deeming Resolutions: Budget Enforcement in the Absence of a Budget Resolution, in the version updated on April 29, 2019, lists no agreement on a budget resolution for ten fiscal years through fiscal 2019, namely fiscal 1999, fiscal 2003, fiscal 2005, fiscal 2007, fiscal 2011, fiscal 2012, fiscal 2013, fiscal 2014, fiscal 2015, and fiscal 2019. In each of those years, according to the same report, one or both chambers employed a deeming resolution to supply enforceable budgetary levels in place of the missing concurrent resolution. A deeming resolution operates inside one chamber, so the joint decision that the budget resolution was designed to embody never occurs.

The lapse record measures the point at which substitution failed altogether. A funding gap occurs when authority to obligate funds expires for an account and neither a regular appropriations act nor a continuing resolution has been enacted to replace the authority, as the Congressional Research Service defines the term in Federal Funding Gaps: A Brief Overview. The legal consequence flows from the Antideficiency Act, 31 U.S.C. 1341 and following sections, which prohibits obligations in excess of or in advance of appropriations. For years before 1980, agencies tended to curtail operations during a gap while continuing to operate, according to the Congressional Research Service summary in Federal Funding Gaps: A Brief Overview. Opinions issued by Attorney General Benjamin Civiletti in 1980 and in early 1981 changed the operative reading: the opinions held that the act permits obligations during a lapse only as authorized by law, including obligations connected with the orderly termination of operations and activities involving the safety of human life or the protection of property. After the Civiletti opinions, a funding gap produced a shutdown of affected activities, with furloughs of personnel whose functions fell outside the exceptions. The Government Accountability Office report Funding Gaps Jeopardize Federal Government Operations, PAD-81-31, issued on March 3, 1981, recorded that interruptions in agency funding at the beginning of the fiscal year and operations on continuing resolutions had become the norm rather than the exception.

The Congressional Research Service count of gaps, in Federal Funding Gaps: A Brief Overview as updated on February 4, 2019, stands at twenty funding gaps from fiscal 1977 through the fiscal 2019 gap, with durations ranging from one full day to thirty-four full days on the full day counting convention of the report. The convention counts days with no available budget authority, ending on the day before new authority was enacted. The early gaps were long. The Congressional Research Service table lists a gap of ten full days in fiscal 1977, gaps of twelve, eight, and eight full days in fiscal 1978, a gap of seventeen full days in fiscal 1979, and a gap of eleven full days in fiscal 1980. Six gaps in the first four fiscal years of the framework, each between eight and seventeen full days, established that the October 1 deadline carried no operative sanction from the start of the system. The middle period produced shorter interruptions. The Congressional Research Service report Shutdown of the Federal Government: Causes, Processes, and Effects records nine funding gaps from fiscal 1981 through fiscal 1995 with durations of up to three full days, a pattern of brief expirations at the start of a fiscal year or at the expiration of a continuing resolution, each closed by enactment within days.

Fiscal 1996 broke the middle pattern and produced the longest lapse recorded to that date. Two funding gaps occurred in fiscal 1996, amounting to five full days and twenty-one full days, according to the Congressional Research Service table in Federal Funding Gaps: A Brief Overview. Budget authority for the first gap expired at the end of November 13, 1995, the five full days ran from November 14, 1995 through November 18, 1995, and a continuing resolution terminated the gap on November 19, 1995. Authority expired again at the end of December 15, 1995, and the second gap ran for twenty-one full days before termination on January 6, 1996, a duration that the Congressional Research Service report Shutdown of the Federal Government: Causes, Processes, and Effects identifies as the longest shutdown recorded through the period covered by editions of the report issued before January 2019. A long interval without a full day gap followed. The Congressional Research Service table lists no funding gap between the fiscal 1996 gap that ended on January 6, 1996 and the gap that began at the start of fiscal 2014, a span covering fiscal 1997 through fiscal 2013.

The gap at the start of fiscal 2014 restored the lapse as a feature of the system after an absence of more than seventeen years. At the beginning of fiscal 2014 on October 1, 2013, none of the regular appropriations bills had been enacted, a governmentwide funding gap occurred, and the gap concluded on October 17, 2013 after lasting sixteen full days, according to the Congressional Research Service in Federal Funding Gaps: A Brief Overview. The same report records a further gap in fiscal 2018, when a continuing resolution covering all of the regular appropriations bills expired on January 19, 2018 and the gap concluded on January 22, 2018 after lasting two full days. A second fiscal 2018 lapse in February 2018 ended within hours and produced no full day without budget authority, and the counting convention of the table therefore excludes the February 2018 lapse from the twenty gap count.

The fiscal 2019 gap exceeded every prior entry in the table. A continuing resolution covering agencies and activities funded in seven of the twelve regular bills expired on December 21, 2018, the gap began on December 22, 2018, and the gap concluded on January 25, 2019, according to the Congressional Research Service in Federal Funding Gaps: A Brief Overview and in Past Government Shutdowns: Key Resources. The February 4, 2019 edition of Federal Funding Gaps: A Brief Overview counts the lapse at thirty-four full days under the full day convention, and Past Government Shutdowns: Key Resources describes the partial shutdown as lasting thirty-five days. Both figures describe the same event under two counting conventions stated by the same institution, and the article record carries both figures with the conventions attached. The lapse was partial because five regular bills had been enacted for fiscal 2019 before October 1, 2018, and the expiration reached only the seven bills carried on continuing resolutions. The Congressional Budget Office report The Effects of the Partial Shutdown Ending in January 2019, issued on January 28, 2019, estimated that the lapse delayed eighteen billion dollars in discretionary spending, reduced real gross domestic product by eleven billion dollars across the fourth quarter of 2018 and the first quarter of 2019, and left about three billion dollars of that reduction unrecovered, with about 800,000 federal employees affected by furlough or by work without pay. A lapse, on the Congressional Budget Office accounting, does not convert unspent appropriations into savings: delayed compensation was paid after enactment of funding, and output lost during the closure was in part lost permanently.

The mechanical significance of the process finding follows from the difference between a timetable and a constraint. The deadlines in Section 300 operate as target dates backed by points of order, and a point of order can be waived, set aside by a special rule in the House, or overcome by a supermajority vote in the Senate, while no automatic consequence attaches to the arrival of October 1 with bills unenacted except the expiration of the prior appropriations themselves. A statutory cap or a sequestration mechanism, by contrast, attaches a consequence that operates without a further vote, and the contrast gives content to the claim that rules bind the annual vote and not the automatic one. The information machinery built by the same act endured through the failure of the calendar. The Congressional Budget Office delivered the February outlook reports, committees delivered views and estimates, scorekeeping continued under deemed levels in years without a budget resolution, and subcommittees produced bills in most years that later entered consolidated measures. Information arrived on schedule while decisions arrived late, and the split outcome identifies the precise locus of the failure. The 1974 framework succeeded as a system for describing the budget in common terms and failed as a system for forcing the annual decision to occur inside the period that the description assumed.

Process failure of that kind bears on spending control through identifiable channels. A year funded at a prior rate under a continuing resolution receives no fresh allocation decision, so spending in such a year reflects inertia rather than a judgment about the year in question, and a sequence of such years embeds the priorities of the last completed cycle into later years without a vote on the merits of the extension. A year completed through a consolidated package receives a decision, but a decision taken under deadline pressure, in a single vote, on a text assembled from bills that in several cases had passed neither chamber, so the discipline of separate consideration that was supposed to test each allocation disappears at enactment. A lapse interrupts the execution of enacted priorities and, on the Congressional Budget Office estimates for the fiscal 2019 lapse, destroys a portion of output without reducing obligations by a corresponding amount. A reform that could not bind the timing of the annual appropriations vote left the annual portion of spending to be decided by extension, by package, or by interruption, and each substitute weakened the connection between the aggregate limits that Congress announced and the appropriations that Congress enacted in a given year.

The Rule and Record Table

The table below compresses five procedural regimes into the constraint each regime imposed, the record of compliance each regime produced, and the route by which compliance was set aside.

Regime	Period in force	What it constrained	Whether it bound	How it was evaded or waived
The 1974 framework	Enacted in 1974, governing fiscal years from fiscal 1975 onward	Annual appropriations process, budget resolutions and impoundment procedure	Process operated, timetable bound in four fiscal years only	Points of order waived, deadlines missed, continuing resolutions and consolidated packages substituted
The 1985 deficit targets	Enacted in 1985, revised in 1987, replaced in 1990	The projected deficit total on a declining statutory schedule	Targets were never met on the terms set	Optimistic assumptions, payment date shifting, off budget moves and amendment of the targets
The 1990 caps and pay-as-you-go regime	Enacted in 1990, extended in 1993 and 1997, expired after fiscal 2002	Discretionary appropriations by caps, and new mandatory and revenue legislation by a scorecard	Discretionary share fell substantially, scorecard produced no sequester	Emergency designations, supplemental appropriations and a statutory resetting of scorecard balances in 2002
The statutory pay-as-you-go requirement restored in 2010	Enacted on February 12, 2010, in force through April 15, 2020	New mandatory spending and revenue legislation recorded on five year and ten year scorecards	No sequestration order was ever triggered under the act	Section 4 exclusions, section 4(g) emergency designations, directed exclusions, scorecard resets and debit shifts
The 2011 caps	Enacted in 2011, covering fiscal 2012 to fiscal 2021	Discretionary appropriations in defense and nondefense categories	Discretionary share fell substantially and a sequester executed on March 1, 2013	Four cap raising budget acts and overseas contingency funding placed outside the caps

Against the Cynical Verdict

The cynical verdict deserves to be stated at full strength, because a weaker version would be too easy to answer. Between the Congressional Budget and Impoundment Control Act of 1974 and the end of fiscal 2019, Congress enacted deficit targets, discretionary spending caps, two versions of statutory pay-as-you-go discipline, an automatic reduction process, and repeated revisions. The aggregate result, measured on the Congressional Budget Office historical data series, was that federal debt held by the public rose from 23.2 percent of gross domestic product in fiscal 1974 to 79.2 percent of gross domestic product at the end of fiscal 2019, a figure the Congressional Budget Office restated in Federal Debt: A Primer in March 2020. Deficits were recorded in every fiscal year of the period except fiscal 1998, fiscal 1999, fiscal 2000, and fiscal 2001. The Gramm-Rudman-Hollings targets of 1985 were missed and then amended. The statutory pay-as-you-go sequester was never carried out under either the 1990 regime or the requirement restored by the Statutory Pay-As-You-Go Act of 2010, Public Law 111-139. The caps of the Budget Control Act of 2011 were raised by four successive Bipartisan Budget Acts, in 2013, 2015, 2018, and 2019. On that record, the cynical reading concludes that budget process is theater: legislators perform concern about deficits, enact rules for the performance, and waive the rules whenever a rule would require a decision that a majority prefers to avoid. A rule that the ruled body can set aside, the argument runs, is not a constraint at all, and the rising debt constitutes the proof.

The aggregate record is accurate, and the inference drawn from it tests the rules against an outcome that most of the rules never governed. Debt held by the public is a residual. It accumulates every deficit, and every deficit combines revenues, mandatory spending driven by eligibility law and demography, discretionary spending set by annual appropriation, and net interest on debt already incurred. A cap on discretionary appropriations governs one of those components, in one category, for the years the cap covers. Judging that cap by the movement of total debt attributes to the cap the behavior of revenues, of Social Security and health program outlays, and of interest costs, none of which the cap reached. The pattern established across the regimes points in a consistent direction, and it can be stated plainly. Rules bind the annual vote and not the automatic one. A mechanism aimed at a category that Congress must affirmatively fund each year has repeatedly changed the amount funded. A mechanism aimed at the deficit as an outcome, or at mandatory spending that flows without an annual vote, has repeatedly been waived, delayed, or redesigned at the point where compliance would have required revisiting eligibility law or revenue law itself. The cynical verdict treats that distinction as an excuse. The record treats it as the finding.

Did the caps actually reduce spending?

Yes. Under the caps enacted in 1990, discretionary outlays fell from 8.5 percent of gross domestic product in fiscal 1990 to 6.2 percent in fiscal 1998, on the Congressional Budget Office historical series. Under the caps enacted in 2011, the same category fell from 9.1 percent in fiscal 2010 to 6.3 percent in fiscal 2019.

The first cap regime was created by the Budget Enforcement Act of 1990, which set separate limits for defense, international, and domestic discretionary categories for fiscal 1991 to fiscal 1993, a single discretionary limit for fiscal 1994 and fiscal 1995, and an extended structure through fiscal 2002. The measured effect, recorded in full in the category finding, was a substantial fall in the discretionary share across the capped years on the Congressional Budget Office historical data series published as Appendix F of The Budget and Economic Outlook: 2019 to 2029. An analysis published by the Brookings Institution in June 1999, written while the caps were under political attack for tightness, put the same movement in inflation adjusted terms and recorded that discretionary spending between 1990 and 1998 had been cut by 11 percent in real terms. A qualification strengthens the cap finding. Much of the decline in the discretionary share came from defense, as the post-Cold War drawdown reduced defense outlays from roughly 5 percent of gross domestic product in 1990 to roughly 3 percent in 2000. The caps did not cause the Soviet collapse, and no serious account credits them with the drawdown itself. What the caps did was govern the disposition of the dividend. Without a binding ceiling, a fall in defense requirements would have left room for equivalent increases elsewhere in the discretionary budget or for tax reductions scored against the same room. The Brookings analysis of June 1999 recorded that real defense spending fell by 27 percent between 1990 and 1998 while real domestic discretionary spending rose by 18 percent across the same period, and the total nevertheless fell by 11 percent in real terms. A ceiling that permits reallocation inside a shrinking total, and prevents the total from refilling, has performed the exact function a category rule can perform.

The second cap regime repeated the pattern under less favorable conditions. The Budget Control Act of 2011, Public Law 112-25, imposed caps for fiscal 2012 to fiscal 2021, and the Congressional Budget Office estimated at enactment that the caps alone would reduce federal spending by 917 billion dollars between fiscal 2012 and fiscal 2021, with the additional automatic reduction process designed to produce at least 1.2 trillion dollars of further deficit reduction across the same period, as the Congressional Research Service summarized the estimate in its report on the act. The automatic process was triggered after the Joint Select Committee on Deficit Reduction reported no agreement, and on March 1, 2013 a sequester of 85.3 billion dollars was carried out, with Medicare reductions capped at 2 percent. That event distinguishes a cap sequester from a pay-as-you-go sequester. The pay-as-you-go sequester never occurred under either statutory regime. The cap sequester occurred, in public, on a date fixed in advance, against the category whose discipline the statute imposed. The measured outcome followed the same pattern, summarized here and recorded in full in the category finding: a fall of almost three percentage points of gross domestic product in the discretionary share across the capped years, despite four cap raising enactments and despite overseas contingency funding placed outside the caps. The override was exercised by statute, for specified years, at levels that left the category far below the fiscal 2010 share. A rule that is raised four times and nevertheless coincides with a decline of that scale in its category over eight fiscal years has not been exposed as decoration. Office of Management and Budget Director Jacob J. Lew made the narrower institutional version of this point in testimony to the House Budget Committee, arguing that the enforcement laws had imposed an essential discipline on discretionary spending by means of enforceable caps. The outlay series supports the testimony.

The second counter-reading concerns information, an effect with no waiver mechanism. Before the 1974 act, Congress possessed no institution of its own charged with estimating what pending legislation would cost. Decisions on spending and taxation were taken by separate committees acting on separate measures, with no vote on totals, no agreed baseline against which individual measures could be added together, and no congressional estimate to set against the figures supplied by the executive branch. The Bureau of the Budget, later the Office of Management and Budget, served the President, and its estimates reflected an administration program as well as an arithmetic exercise. Committees relied on agency testimony, on the revenue estimates of the staff of the Joint Committee on Taxation for tax measures, and on informal comparisons with the President’s budget. Allen Schick documented the fragmentation of that system in Congress and Money, published by the Urban Institute in 1980, and the Congressional Research Service, in its Introduction to the Federal Budget Process, described congressional action before the 1974 act as lacking any formal coordination, with measures considered separately and compared only informally to presidential proposals. The confrontation over impoundment exposed the practical consequence. A president who declined to spend appropriated funds, and who controlled the executive information about the amounts and programs affected, confronted a legislature that lacked an independent apparatus for establishing the fiscal facts of the dispute.

The Congressional Budget Office changed that environment in ways that persisted through every subsequent change of majority and administration. Created by the 1974 act and led from 1975 by its first director, Alice M. Rivlin, the office published its first five year projections in January 1976, as the office recorded in its own Brief History of the Baseline, and it built three practices into the legislative routine. The first was the baseline: a projection of spending, revenues, and deficits under existing law, prepared by the Congressional Budget Office rather than by the administration, against which every proposal could be measured on identical assumptions. The second was the formal cost estimate: for nearly every bill reported by a committee, an estimate of the change in outlays or revenues relative to that baseline, incorporating revenue estimates from the staff of the Joint Committee on Taxation, so that a vote on a measure became a vote taken with a published price attached. The third was scorekeeping: a running comparison, maintained by the budget committees with Congressional Budget Office and Office of Management and Budget estimates, between enacted and pending action and the allocations and totals set in the budget resolution. Philip G. Joyce, in The Congressional Budget Office: Honest Numbers, Power, and Policymaking, published in 2011, traced how those practices made the office a central participant in legislative bargaining without giving it a vote on any measure. Rivlin, writing for the Brookings Institution in March 2017 after service as both the first director of the Congressional Budget Office and a later director of the Office of Management and Budget, argued that the two offices performed different constitutional functions, that the Congressional Budget Office served Congress as the Office of Management and Budget served the President, and that Congress required its own team of neutral experts to do its work under the Constitution. Legislative majorities disputed individual estimates, pressed the office over particular assumptions, and drafted measures to score favorably within the ten year window, through sunsets, delayed effective dates, and phase ins whose fiscal shape only makes sense as a response to a published estimate. Drafting to defeat a score is an admission that the score governs the debate. Before 1974, there was no common score to defeat, because there was no common estimate to argue from, only rival guesses advanced by interested parties.

The third counter-reading concerns pay-as-you-go discipline, where the waiver record is genuine. Under the Statutory Pay-As-You-Go Act of 2010, the budgetary effects of enacted legislation affecting mandatory spending or revenues were recorded on five year and ten year scorecards maintained by the Office of Management and Budget, using estimates prepared by the Congressional Budget Office and the staff of the Joint Committee on Taxation. A net debit left standing at the end of a session would have triggered a sequester of non-exempt mandatory programs, with Medicare reductions capped at 4 percent. No such sequester was triggered through April 15, 2020. The Tax Cuts and Jobs Act of 2017, Public Law 115-97, was excluded from the statutory PAYGO scorecards by section 5002 of Public Law 115-96, enacted on December 22, 2017, and the Bipartisan Budget Act of 2018, Public Law 115-123, then reset all remaining scorecard balances to zero. Other legislation avoided the scorecard through the exclusions written into section 4 of the 2010 act or through emergency designations under section 4(g), which required the designation to appear in the statute itself and to be made by the President as well. A verdict that stopped at that record would call the rule a nullity. The record of behavior before enactment points the other way. Pay-as-you-go operates during drafting, not after passage. A committee that knows a measure will be scored, recorded, and published has to identify offsets, narrow eligibility, delay a start date, or accept that the measure will carry a recorded debit into a subsequent negotiation. Senator Kent Conrad, chairman of the Senate Budget Committee, argued on the Senate floor in 2007 that the Senate pay-as-you-go rule had provided a significant deterrent, preventing costly bills from being offered at all, and that measures reaching conference in that year had been paid for. Deterrence of that kind leaves no statute to cite: bills reduced, offset, or abandoned during drafting appear in no table of enacted law. The earlier statutory regime supplies the harder evidence. The Office of Management and Budget issued twelve pay-as-you-go scorecard reports covering fiscal 1992 to fiscal 2003, and each report recorded net savings or no increase in the deficit, so no pay-as-you-go sequester was issued in that period either. The Congressional Research Service attributed the absence of a sequester in the earlier years to effective compliance and in the later years to statutory intervention through directed scorekeeping and related techniques. Compliance and intervention are different phenomena, and neither resembles indifference to the rule. A waiver under the 2010 act is not an evasion of the Gramm-Rudman-Hollings kind, achieved through optimistic assumptions or payment dates shifted across a fiscal boundary. A waiver requires statutory language, a vote, a signature, and a scorecard entry recording what was done. The political cost is the difference between increasing a deficit silently, as one consequence among thousands in a large bill, and increasing it by means of a provision whose sole function is to set the discipline aside. Pay-as-you-go therefore operates as a price rather than a prohibition. Prices do not prevent purchases. Prices force the purchaser to decide, on the record, that the purchase is worth the stated cost, and a legislature that repeatedly pays that price has revealed a preference, not discovered a loophole.

The balanced budget amendment supplies the limiting case that clarifies the trade. An amendment to the Constitution would have been an outcome rule in its purest available form: outlays would not exceed receipts, subject to exceptions requiring three fifths votes of both chambers or a declaration of war, and no ordinary majority could waive the requirement by statute. On January 26, 1995, the House of Representatives passed House Joint Resolution 1 by a vote of 300 to 132, the first passage of a balanced budget amendment by the House, as the Congressional Research Service recorded in its survey of balanced budget amendments. On March 2, 1995, the Senate recorded a vote of 65 to 35 on the same proposal, one vote short of the two thirds required; the Senate Report accompanying a later version, Senate Report 105-3, explained that substantive support stood at 66 to 34, because Majority Leader Robert Dole had switched a vote to no in order to preserve the right to move for reconsideration. On reconsideration, on June 6, 1996, the Senate voted 64 to 35 and the proposal failed again. On March 4, 1997, the Senate considered Senate Joint Resolution 1 of the 105th Congress and voted 66 to 34, again one vote short of the two thirds required to propose an amendment, a tally the Congressional Research Service recorded in the same survey. No balanced budget amendment was proposed by Congress or ratified by the states, and none had been enacted by April 15, 2020. Support reaching 66 votes in a chamber of 100, twice, cannot be described as indifference to deficit discipline. Failure at the same threshold, twice, cannot be described as an accident of scheduling. Legislators who enacted caps with statutory overrides, and pay-as-you-go rules with legislated exclusions, declined by the narrowest available margin to surrender the override itself. Even the amendment text conceded the underlying difficulty, by building in supermajority escape routes for war and for specified excesses of outlays over receipts. An outcome rule strong enough to bind in a recession, a war, or a financial emergency proved to be an outcome rule that a two thirds coalition would not send to the states. Bindingness and enactability traded against each other, and the amendment record marks the point where the trade became visible.

The measured assessment follows from the three counter-readings, and it is narrower than either celebration or dismissal. Procedure shapes behavior at the margin and cannot substitute for a political consensus. At the margin, the effects were real and, in the discretionary category, measurable in percentage points of gross domestic product across two separate cap regimes separated by two decades. In the information environment, the effect was structural: after the creation of the Congressional Budget Office, legislation was debated against a common estimate rather than against rival guesses, and no subsequent Congress chose to legislate without one. In pay-as-you-go discipline, the effect was a recurring political price, paid in offsets during drafting or in open statutory waivers after it, rather than a mechanical bar. Beyond the margin, the limits were equally consistent. No rule enacted between 1974 and 2019 generated the underlying agreement about the size of government, the level of taxation, or the distribution of program reductions that a sustained reduction in debt as a share of the economy would have required. Where such an agreement was reached, in the deficit reduction packages of 1990 and 1993 operating inside a caps and pay-as-you-go regime, the rules gave the agreement durability by making departures visible and costly. Where no agreement existed, the rules recorded the absence, through waivers, cap adjustments, directed scorekeeping, and missed targets, with a precision that the pre-1974 system could not have supplied. The verdict that the record supports is therefore neither that process reform failed nor that process reform succeeded on the terms its most ambitious advocates proposed. Rules governed the category that Congress votes on each year, informed every vote with an estimate that Congress did not control and could not quietly replace, and charged a public price for deficit increasing legislation even when the price was paid. Rules did not, and by design could not, decide the fiscal questions that only majorities can decide.

Five findings converge on a single measurement problem. The aggregate record gave deficits in forty two of forty six fiscal years and a debt ratio that rose from 23.2 percent of gross domestic product in fiscal 1974 to 79.2 percent in fiscal 2019 on the Congressional Budget Office historical data series. The surplus record gave four consecutive surpluses, fiscal 1998 to fiscal 2001, under a coincidence of legislated deficit reduction, a capital gains driven revenue surge, and a post Cold War defense drawdown, with named analysts assigning different weights to each condition and no named analysis assigning the whole outcome to the rules alone.

The category record and the scorecard record point in opposite directions for a structural reason. Discretionary outlays fell substantially as a share of the economy in both cap eras on the Congressional Budget Office and Office of Management and Budget series, even after cap raising enactments and funding placed outside the caps had been counted inside the category total. The statutory pay-as-you-go scorecard, by contrast, produced no sequestration order between February 12, 2010 and April 15, 2020, because exclusions, emergency designations, resets, and shifts removed or deferred each debit before a budget year balance could require an order, a contrast examined directly in sequestration versus pay-as-you-go enforcement.

The process record completes the pattern. All regular appropriations reached enactment before October 1 in only four fiscal years across the span on the Congressional Research Service record, while continuing resolutions, consolidated packages, and twenty funding gaps filled the space left by missed deadlines. Information did not fail in the same way. Baselines, formal cost estimates, and scorekeeping gave each vote a published price, and the practice described in how the Congressional Budget Office scores legislation made departures visible even when a majority chose to pay the political price of a waiver. A timetable without an automatic consequence missed deadlines, a cap with a sequestration consequence changed a category total, and a scorecard that later statutes could rewrite charged a price without imposing a bar.

The claim that the record supports is therefore narrow, dated, and testable against each regime in turn. Rules bind the annual vote and not the automatic one. A reform that attaches a consequence to appropriations can hold the voted category below an unconstrained path for years at a time, and a reform that records mandatory and revenue legislation can force offsets during drafting or an open waiver after drafting. No regime in the record substituted for substantive legislation on eligibility, benefit formulas, health costs, or revenue levels, and the totals moved with those substantive choices. A reader assembling the statutes, scorecards, votes, and lapse chronologies behind that verdict can keep your statute notes, citations, and case chronologies together free on VaultBook.

Frequently Asked Questions

Q: Did budget reform control spending?

Budget reform controlled one category of spending and failed to control the total. Discretionary outlays fell substantially as a share of gross domestic product during both statutory cap eras on the Congressional Budget Office and Office of Management and Budget series. Total outlays, deficits, and debt followed a different path, because mandatory spending under permanent eligibility and benefit formulas, plus net interest on accumulated debt, sat outside the caps. The Congressional Budget Office historical data record deficits in forty two of forty six fiscal years from fiscal 1974 to fiscal 2019 and debt held by the public rising from 23.2 percent of gross domestic product to 79.2 percent across that span.

Q: When was the last federal surplus after budget reform?

The last federal surplus in the record through April 15, 2020 was the surplus of fiscal 2001. The Office of Management and Budget Historical Tables series recorded surpluses in fiscal 1998, fiscal 1999, fiscal 2000, and fiscal 2001, at 69 billion dollars, 126 billion dollars, 236 billion dollars, and 128 billion dollars respectively, and deficits in every fiscal year from fiscal 2002 through fiscal 2019. Fiscal 2001 therefore closed the only run of consecutive surpluses in the period that began with enactment of the Congressional Budget and Impoundment Control Act of 1974, and no later procedural regime reproduced balance before the April 2020 vantage closed the record.

Q: Did the 1990s caps work as budget reform?

The caps worked on the category that the caps governed and within limits set by erosion routes. Discretionary outlays fell from 8.8 percent of gross domestic product in fiscal 1991 to 6.0 percent in fiscal 1999 on the Congressional Budget Office and Office of Management and Budget series, a decline that included spending financed through emergency designations. Defense drawdown after the Cold War supplied much of the decline, while nondefense discretionary outlays moved by smaller amounts. The Brookings Institution analysis of June 1999 recorded an 11 percent reduction in real discretionary spending between 1990 and 1998. The caps therefore restrained growth in annual appropriations without governing mandatory programs or revenues.

Q: Why are budget reform rules always waived?

Waiver routes were built into several regimes and added by later statutes in others. The Statutory Pay-As-You-Go Act of 2010 contained emergency designations under section 4(g), exclusions in section 4, and current policy adjustments in sections 4(c) and 7. Later statutes then directed exclusions for particular measures, reset scorecard balances to zero, or shifted debits to a later year. Each route required enacted statutory language, a vote, and a signature, so a waiver carried a public political price rather than operating as a silent evasion. Majorities paid that price when a measure mattered more to the majority than the recorded debit that the measure created.

Q: Did budget reform stop shutdowns?

Budget reform did not stop shutdowns or funding lapses. The Congressional Research Service count in Federal Funding Gaps: A Brief Overview, updated on February 4, 2019, recorded twenty funding gaps from fiscal 1977 through the fiscal 2019 gap. Durations ranged from one full day to thirty four full days under the full day counting convention of that report. The fiscal 2019 lapse began on December 22, 2018 and concluded on January 25, 2019, after five regular appropriations bills had been enacted and seven bills rested on a continuing resolution that expired. Statutory caps limited aggregate amounts without allocating funds among bills or compelling agreement before October 1.

Q: Did discretionary spending fall under budget reform caps?

Discretionary spending fell as a share of the economy under both cap regimes on the Congressional Budget Office and Office of Management and Budget series. The category measured 8.8 percent of gross domestic product in fiscal 1991 and 6.0 percent in fiscal 1999 during the first era, and 8.7 percent in fiscal 2011 and 6.3 percent in fiscal 2019 during the second era. Nominal dollars told a different story in the first era, rising from 500.6 billion dollars in fiscal 1990 to 572.1 billion dollars in fiscal 1999. The caps bound the growth rate of the category relative to the economy rather than imposing an absolute dollar reduction in each year.

Q: Has a balanced budget amendment ever passed as budget reform?

No balanced budget amendment was proposed by Congress or ratified by the states through April 15, 2020. The House of Representatives passed House Joint Resolution 1 on January 26, 1995 by 300 to 132, as the Congressional Research Service recorded in a survey of balanced budget amendments. The Senate recorded 65 to 35 on March 2, 1995, one vote short of the two thirds required, with substantive support recorded at 66 to 34 after a procedural vote switch by Majority Leader Robert Dole. Further Senate votes in June 1996 and March 1997 also fell short, leaving statutory rules with legislated overrides as the enacted form of discipline.

Q: What do economists say about budget reform rules?

Named analysts divided on weight rather than on the underlying figures. Alice Rivlin assigned joint credit to bipartisan legislated action and to extraordinary economic performance. Jeffrey Liebman, applying Congressional Budget Office decompositions, divided the surplus improvement between legislation and unforecast economic improvement. Alan Auerbach and William Gale questioned the permanence of the revenue surge and the availability of projected surpluses held in trust fund balances. David Henderson, in a Mercatus Center study published in 2015, assigned the largest share of spending restraint to defense. A Brookings Institution analysis published in 1999 argued that policy choices carried weight equal to economic forces during the surplus decade.

Q: How did discretionary caps affect the overall deficit?

Caps reduced the discretionary share of the budget without determining the deficit, because the deficit is a residual of total revenues and total outlays. During the second cap era, discretionary outlays fell as a share of gross domestic product while the deficit rose from 585 billion dollars in fiscal 2016 to 984 billion dollars in fiscal 2019 on the Congressional Budget Office record. Mandatory spending measured 12.8 percent of gross domestic product in fiscal 2019 against discretionary spending of 6.3 percent and net interest of 1.8 percent in Congressional Budget Office baseline tables. Revenue legislation enacted in December 2017 also reduced revenues relative to the prior baseline, outside any cap mechanism.

Q: What role did emergency designations play in weakening budget reform?

Emergency designations removed designated amounts from enforcement while leaving the spending in the budget totals. Under the first cap regime, appropriations designated as emergencies raised the discretionary caps by the designated amount. Under the Statutory Pay-As-You-Go Act of 2010, section 4(g) kept amounts designated as emergency requirements off the scorecards entirely. The Office of Management and Budget annual report for 2010 recorded 570.1 billion dollars of costs excluded through emergency designations over the period from 2010 to 2015. The Coronavirus Aid, Relief, and Economic Security Act, enacted on March 27, 2020, used the section 4(g) designation for the largest measure of the opening months of 2020.

Q: How did overseas contingency funding affect the discretionary caps?

Funding designated for overseas contingency operations sat outside the discretionary caps through a statutory adjustment under the Budget Control Act of 2011. Appropriations carrying that designation did not count against the defense ceiling, did not require an offset inside the capped category, and did not expose other defense accounts to sequestration. Adjustments for the category totaled 723 billion dollars from fiscal 2012 to fiscal 2019 in the series record carried in the account of the 2011 act. Outlay series nevertheless counted the spending inside discretionary totals when the money was spent, so the measured decline in the category share from fiscal 2011 to fiscal 2019 already included the amounts placed outside the caps.

Q: Why did deficit targets fail where caps produced measurable restraint?

Deficit targets punished an outcome that legislation alone did not control, while caps measured a recorded vote. A deficit moves with economic growth, unemployment, interest rates, asset prices, and forecasting error alongside tax and spending law, so a recession could breach a target after a restrained budget had been enacted. The Gramm-Rudman-Hollings targets were missed, revised upward in 1987, and met through optimistic assumptions, payment date shifting, and off budget moves before replacement in 1990. A cap breach, by contrast, traced to identifiable appropriations bills, and sequestration fell inside the breached category. That traceability gave cap enforcement a legitimacy that target enforcement lacked under the Congressional Research Service account of the 1990 change.

Q: What did the Congressional Budget Office add to budget reform?

The Congressional Budget Office supplied a common factual basis that Congress lacked before the Congressional Budget and Impoundment Control Act of 1974. Created by that act and led from 1975 by Alice Rivlin, the office published a baseline projection under existing law, formal cost estimates for nearly every bill reported by a committee, and scorekeeping comparisons against budget resolution allocations. Philip Joyce, in a history of the office published in 2011, traced how those practices made the office a central participant in legislative bargaining without a vote on any measure. Majorities disputed estimates and drafted measures to score favorably inside a budget window, behavior that confirmed the estimates governed debate.

Q: How did scorecard resets affect pay-as-you-go discipline?

A reset set accumulated scorecard balances to zero by statute, removing debits that earlier measures had recorded before any sequestration determination could act on a balance. The Bipartisan Budget Act of 2018, enacted on February 9, 2018, nullified the balances on the five year and ten year scorecards as of enactment. The Bipartisan Budget Act of 2019 repeated the step in section 102, as the Office of Management and Budget annual report of January 27, 2020 recorded. A reset differed from an exclusion in timing alone. An exclusion kept a debit off the ledger from enactment, while a reset removed a debit after recordation, and both routes ended at a budget year balance of zero.

Q: How often were appropriations finished on time under budget reform?

All regular appropriations acts reached enactment before the start of the fiscal year in only four fiscal years after the 1974 framework took effect: fiscal 1977, fiscal 1989, fiscal 1995, and fiscal 1997, on the Congressional Research Service record. Congress enacted one or more continuing resolutions in all but three of the forty four fiscal years from fiscal 1977 through fiscal 2020, with 188 continuing resolutions in total. Partial completion became the recurring pattern after fiscal 1997, including five regular bills enacted on time for fiscal 2019 and zero for each fiscal year from fiscal 2011 through fiscal 2016. Consolidated packages and full year continuing resolutions supplied the other substitutes for separate enactment.

Q: When did a budget reform sequester actually execute?

A discretionary cap sequester executed on March 1, 2013 under the Budget Control Act of 2011, reducing budgetary resources by 85.3 billion dollars, with Medicare reductions capped at 2 percent. The order followed failure by the Joint Select Committee on Deficit Reduction to report an agreement under the automatic reduction process of the 2011 act. No pay-as-you-go sequester executed under either statutory regime in the record. The earlier regime produced twelve Office of Management and Budget scorecard reports covering fiscal 1992 to fiscal 2003 that recorded net savings or no deficit increase, and the regime restored in 2010 produced no sequestration order through April 15, 2020 on the Congressional Budget Office record.

Q: How did mandatory spending growth limit the reach of budget reform?

Mandatory spending flows from eligibility rules and benefit formulas in permanent law, driven by beneficiary numbers, demographic change, and health care costs per beneficiary rather than by an annual vote. The Congressional Budget Office outlook of January 2019 attributed growth in the largest mandatory components to population aging and to growth in health care spending per beneficiary. Social Security outlays rose from 4.1 percent of gross domestic product in fiscal 1990 to 4.9 percent in fiscal 2019, and the major health care programs measured 5.3 percent in fiscal 2019 on Congressional Budget Office figures. No appropriations cap touched either figure, because neither figure originated in an appropriation.

Q: What evidence shows budget reform changed drafting behavior?

Pay-as-you-go discipline operated during drafting, before any scorecard balance could threaten a sequester. A committee expecting a measure to be scored, recorded, and published had reason to identify offsets, narrow eligibility, delay a start date, or accept a recorded debit in a later negotiation. Senator Kent Conrad argued on the Senate floor in 2007 that the Senate pay-as-you-go rule had provided a significant deterrent, preventing costly bills from being offered at all, and that measures reaching conference in that year had been paid for. Measures reduced, offset, or abandoned during drafting left no statute to cite, so deterrence appeared in the record as absence rather than as an enacted total.

Q: Why did deficits return after the surplus period?

Deficits returned in fiscal 2002 and persisted through fiscal 2019 on the Congressional Budget Office historical data. The Economic Growth and Tax Relief Reconciliation Act of 2001 reduced revenue across the following decade, military operations in Afghanistan from 2001 and in Iraq from 2003 were financed substantially outside regular appropriations discipline in the early years, and an economic slowdown reduced receipts from the fiscal 2000 peak. The Medicare Prescription Drug, Improvement, and Modernization Act of 2003 added a mandatory benefit with costs falling largely beyond the budget window used at enactment. The Budget Enforcement Act framework expired after fiscal 2002, removing the caps and scorecard that had guarded the surplus path.

Q: How should the success of budget reform be measured?

Measurement depends on matching each rule to the object that the rule governed. A discretionary cap is properly judged against discretionary outlays, a pay-as-you-go scorecard against the recorded effects of enacted mandatory and revenue legislation, and a timetable against enactment dates for regular appropriations bills. Judging any single instrument against total debt attributes revenues, mandatory formulas, demographic change, health costs, and interest on past borrowing to a mechanism that reached none of those forces. The five findings separate those objects deliberately: totals for the aggregate verdict, category series for the caps, scorecard history for pay-as-you-go enforcement, attribution analyses for the surpluses, and the lapse and continuing resolution record for process discipline.