From Gramm-Rudman to PAYGO, a timeline of fiscal rules that bent

Imagine a man who is told by his doctor to stop eating pie. So he writes a note and tapes it to the refrigerator: "No pie after Friday." On Friday he amends the note. "No pie after next Friday." The following week he adds a clause: "Except in emergencies, and a Tuesday can be an emergency." Then he appoints a committee to review the note. Forty years later he is still eating pie, he has a very impressive collection of notes, and every one of them was signed with great solemnity.

That is the history of federal fiscal rules. It is not a story of villains. Most of the men and women who wrote these rules meant them. It is a story about a structural problem: a Congress that makes the rule is the same Congress that has to obey it, and a rule that Congress can change with a simple majority is a suggestion with a title. This post gives the dates, in order, so you can check each one yourself.

The claim, and what would count as proof

Let me define terms. A fiscal rule is a procedure or limit that Congress adopts to constrain its own taxing, spending or borrowing: a deficit target, a spending cap, a requirement that new programs be paid for. A rule bends when it is waived, repealed, suspended, reset, evaded by a definition, or struck down, before it has forced the outcome it was written to force.

My claim has three parts. One: since 1985 Congress has enacted at least half a dozen major fiscal rules. Two: in nearly every case, the rule was relaxed, waived or allowed to lapse in the years when it would have required a painful choice. Three: the exceptions, the periods when the rules seem to have worked, share a feature that tells us something important about what actually restrains spending, and it is not the rule.

Here is the fair test. If I were wrong, you would find a rule that bound Congress against its will, in a year when the vote to break it was available and the political cost of breaking it was low. As we walk through the record, look for that rule. I will tell you honestly where the record comes closest.

1985 to 1987: the rule with a trigger

December 12, 1985: Gramm-Rudman-Hollings

The Balanced Budget and Emergency Deficit Control Act of 1985, named for Senators Phil Gramm, Warren Rudman and Ernest Hollings, set a declining schedule of deficit targets that was supposed to reach zero by fiscal year 1991. If Congress and the President failed to meet a target, the law ordered across-the-board cuts called a sequester, split roughly evenly between defense and domestic programs, with some programs such as Social Security exempt.

The design was clever. The sequester was meant to be so ugly that neither side would risk it. It was a threat aimed at Congress by Congress. Note the weakness already: a target is a number in a statute, and a later statute can change the number.

July 7, 1986: Bowsher v. Synar

The trigger had a flaw that had nothing to do with economics. The act gave the final calculation of the cuts to the Comptroller General, the head of the Government Accountability Office, then called the General Accounting Office. The Comptroller General is removable by Congress through a joint resolution, which meant Congress held the lever over an official who was executing the law. In Bowsher v. Synar, the Supreme Court held 7 to 2, in an opinion by Chief Justice Burger, that this violated the separation of powers: Congress cannot keep removal power over an officer who exercises executive functions.

I will say plainly that the Court got this one right, and that the Founders would have recognized the point at once. Madison's argument in Federalist No. 47 is that the accumulation of legislative, executive and judicial powers in the same hands is the very definition of tyranny. The irony is that the deficit hawks who wrote the law were trying to defend the Treasury and tripped over the structure of the Constitution. The lesson is worth keeping: a rule that depends on a constitutionally defective enforcer is a rule with a hole in the floor.

September 29, 1987: the revision

After the decision, Congress rewrote the law. The Balanced Budget and Emergency Deficit Control Reaffirmation Act of 1987 moved the trigger calculation to the Office of Management and Budget, an executive agency, and it did something else that matters more: it moved the date. The target of a zero deficit slid from fiscal 1991 to fiscal 1993. This is the first appearance of the pattern we will see again and again. When the deadline approaches, the deadline gets amended.

1990 to 2002: the rules that seemed to work

November 5, 1990: the Budget Enforcement Act

The 1990 budget agreement, signed by President George H. W. Bush after he broke his own pledge not to raise taxes, abandoned the fixed deficit targets. In their place, the Budget Enforcement Act set two things. First, caps on discretionary spending, the part of the budget Congress appropriates each year. Second, pay-as-you-go, universally shortened to PAYGO: any new law that increased mandatory spending or cut taxes had to be offset by spending cuts or tax increases elsewhere, or the sequester would apply.

Notice the change in philosophy. The designers learned from Gramm-Rudman that it is silly to promise a fixed deficit number, because the economy, not Congress, controls much of the deficit. They tied the rules to what Congress actually votes on. That is a smarter rule. It is also an easier one to get around, because it constrains only changes to the law. A program growing on autopilot under existing law, such as an entitlement whose spending rises with enrollment and health costs, is not a "new law", so PAYGO did not touch it.

1993 and August 5, 1997: extensions and the Balanced Budget Act

Congress extended the caps and PAYGO in 1993 and again through the Balanced Budget Act of 1997, signed by President Clinton on August 5, 1997. This is the period critics of my thesis point to, and they have a point. The federal government ran budget surpluses in fiscal years 1998 through 2001.

Here is what the record supports, and what it does not. The rules existed, and the surpluses existed. But the surpluses also coincided with a Republican Congress after 1994 that was determined to hold spending growth down, a President willing to negotiate, a historic stock market and capital gains boom that poured revenue into the Treasury, and a post-Cold War reduction in defense outlays. Correlation is not causation. A rule can take credit for an outcome that politics produced. I will return to that, because it is the most important point in this post.

September 30, 2002: PAYGO lapses

The statutory PAYGO and the discretionary caps expired at the end of fiscal 2002. Congress did not renew them. No dramatic vote killed them. They simply reached the date written in the statute, and a Congress that had once been proud of them let the date pass. In the years that followed, the deficit returned and the debt grew. Here is the pattern in its purest form: the rule had an expiration date, and nobody in power wanted to pay the price of extending it.

2010 to 2023: the second generation

February 12, 2010: the Statutory Pay-As-You-Go Act

Democrats, then in control of both chambers of Congress, revived PAYGO in a new statute signed by President Obama on February 12, 2010. The mechanism was familiar: Congress's new mandatory spending and tax cuts were recorded on scorecards, and if the scorecards showed a net deficit increase at the end of a session, an automatic sequester of mandatory programs would follow.

The act was bent almost immediately, and the way it was bent is instructive. Congress is allowed to pass later laws that wipe the scorecard clean. Repeatedly, it did. When a bill that would have triggered the sequester came due, a different bill appeared that cleared the balance. A rule that can be erased by the body it constrains, on a simple majority, is the fiscal equivalent of a speed limit enforced by the driver.

August 2, 2011: the Budget Control Act

The most serious attempt since 1990 came out of the 2011 debt-limit standoff (if you want the history of that mechanism, see the debt ceiling began in 1917 to make borrowing easier, not harder). The Budget Control Act, signed on August 2, 2011, did three things. It raised the debt ceiling. It set caps on discretionary spending for fiscal years 2012 through 2021. And it created a joint committee, the so-called supercommittee, charged with finding a further deficit reduction. If the committee failed, a sequester would automatically cut defense and nondefense spending by equal amounts.

The committee failed. The sequester was scheduled, delayed by the American Taxpayer Relief Act at the start of 2013, and took effect on March 1, 2013. For one year, one of these rules worked as designed: discretionary spending was held down in a way that Washington had not accepted voluntarily in a generation. Defense hawks hated the sequester. Domestic spending advocates hated it. That, if you are a deficit hawk, was the point.

December 2013 to August 2019: the deals that lifted the caps

Then came the bending. The Bipartisan Budget Act of 2013, the Murray-Ryan deal, reduced the pain of the caps for fiscal 2014 and 2015, in exchange for savings farther out. A second Bipartisan Budget Act in November 2015 raised the caps again. A third, signed in February 2018, raised them by far more for two years. A fourth, in August 2019, raised them again and suspended the debt limit.

The method deserves a name. Each deal is a trade: both parties agree to raise the caps on the spending each cares about, defense for Republicans and nondefense for Democrats, and the price is paid by borrowing. Because the caps were the main binding limit in the law, each relief package cut deeper than any appropriator's budget fight could have. A cap that every faction wants lifted and no faction wants to defend is a cap on borrowed time.

June 3, 2023: the Fiscal Responsibility Act

In 2023, a divided Congress passed another debt-limit suspension, the Fiscal Responsibility Act, signed on June 3, 2023. It set spending caps for fiscal 2024 and 2025 and suspended the debt limit into January 2025. Supporters call it the most restraint negotiated in a decade. Critics call it a deadline-driven agreement that left the largest programs untouched. Both are partly right. I will let the reader judge the history from the pattern: the caps were again tied to a debt-limit vote, the thing most certain to force a deal, and they were again short in duration.

What the timeline proves

Now the argument. I will number the premises.

One. Every major rule since 1985 was enacted by a Congress and was subject to amendment by a later Congress under ordinary majority rules.

Two. In the cases where a rule was clearly about to force a hard choice, the response was almost always to change the rule rather than the spending: Gramm-Rudman's targets were pushed back in 1987, the 1990 framework was allowed to lapse in 2002, the 2010 PAYGO scorecard was wiped, and the 2011 caps were lifted in 2013, 2015, 2018 and 2019.

Three. The periods of real restraint, 1998 to 2001 and the first sequester year, were also periods in which the political parties had a reason to cooperate on restraint: divided government with an active reformist Congress, or a crisis. The rule did not create that will. The will made the rule workable.

Therefore. A fiscal rule is a thermometer, not a thermostat. It can tell you how hot the room is. It cannot turn down the heat unless somebody in the room already wants it turned down.

The Constitution, interestingly, hands Congress the heat control directly. Article I, Section 8 gives Congress the power "To borrow Money on the credit of the United States," and Article I, Section 9 says "No Money shall be drawn from the Treasury, but in Consequence of Appropriations made by Law." The framers did not build a second lock on the door. They built a lock, put the key in the hands of a body elected every two years, and trusted voters to notice. The record above is what happens when voters do not notice, or are persuaded that noticing is somebody else's job.

The best defense of the rules, answered fairly

A fair-minded defender of these laws would say the following. Rules change the default. Before Gramm-Rudman, a deficit was nobody's fault. After it, the deficit had an official number and an official culprit. Before PAYGO, a tax cut or a new entitlement was free. After it, every proposal arrived with a cost and a counter-offer. The 2011 caps did cut discretionary spending, and by most accounts the spending would have been higher without them. Even a rule that is waived forces a vote on the waiver, and a vote on a waiver is a vote with a name attached.

All of that is true, and I accept it. A rule that bends is better than no rule if it bends slowly. The 2011 caps and the 1990 PAYGO each bought real years of restraint.

But look at what the defense concedes. It concedes that rules help because they raise the visible cost of overspending, which is exactly the function that informed voters and honest budget scoring serve. If the value of a rule lies in making the cost visible, then the real fix is visibility: honest baselines (see baseline budgeting, where a smaller increase gets reported as a cut), cost estimates that a citizen can read (see Read a CBO cost estimate yourself before you believe the headline), and a public that can see the debt for what it is (see Seven myths about the national debt that Washington finds convenient). Rules built on top of dishonest numbers do not rescue the numbers.

The other possible answer is structural. If ordinary statutes cannot bind a future Congress, then perhaps a stronger instrument could: a constitutional amendment. I take that up, with the conservative doubts on both sides, in a balanced budget amendment, the case for it and the conservative doubts. And the political cost of the pie, so to speak, appears in the arithmetic of do the arithmetic on $40 trillion and the interest bill it carries, where the interest alone is now a line item no cap has touched.

The total public debt stood at $40.56 trillion on November 5, 2026, against $38.05 trillion a year earlier, according to the Treasury's Debt to the Penny. Subtract: 40.56 minus 38.05 is 2.51. In one year the debt grew by $2.51 trillion, about 6.6 percent. Every rule in this timeline was written to prevent exactly that arithmetic, and the arithmetic arrived anyway.

What the timeline says about what conservatives should do

I do not conclude that rules are useless. I conclude that they are insufficient, and that conservatives should say so, because we are the ones who most need the truth about it.

If rules do not bind, then the fight has to be won where it is always won: in the appropriations process, in elections, in the willingness of a committee chairman to say no to a colleague's project. It also has to be won in public, which means teaching voters what the rules were and how they were waived. A Congress that knows its waivers will be listed in the order they happened is a Congress that waives with more reluctance.

That is why this post is a timeline. A date is a hard thing. Anyone can check that Gramm-Rudman was signed on December 12, 1985, that Bowsher v. Synar was decided on July 7, 1986, that the 1990 act became law on November 5, and that the Budget Control Act was signed on August 2, 2011. Anyone can look up how many times the caps were raised afterward. Citizens who know the dates are harder to fool when the next rule is announced with the usual fanfare.

The case in brief

1. Since 1985 Congress has written Gramm-Rudman-Hollings, the Budget Enforcement Act, the Balanced Budget Act of 1997, a revived PAYGO, and the Budget Control Act, each meant to force itself to borrow less.
2. Each was weakened, waived, allowed to expire or struck at the point where it would have bitten: Bowsher v. Synar removed the original trigger, the targets slid in 1987, PAYGO lapsed in 2002, the 2010 scorecards were wiped, and the 2011 caps were lifted in 2013, 2015, 2018 and 2019.
3. The best years for restraint, 1998 to 2001 and the first sequester year, depended on a political will that the rules did not create.
4. A rule can raise the visible cost of overspending, and that has value, but it cannot substitute for voters and legislators who want the cost to be paid.
5. The remedies are honest scoring, public memory of the waivers, and a constitutional restraint taken seriously, not another statute with a sunset date.

What you can do: pick one dated entry above, look up the actual vote on Congress.gov, and write your own representative's name next to it, then ask the office whether they would vote to waive the next rule.