No statute in the modern federal code is studied more often for its procedure than for its substance, and the Tax Reform Act of 1986 is the reason. Enacted as Public Law 99-514 and signed on October 22, 1986, the law collapsed fourteen individual brackets into two nominal rates, repealed the investment tax credit, ended the sixty percent capital gains exclusion, phased out the deduction for consumer interest, and replaced an accelerated depreciation system with a longer-lived one. Yet the durable lesson of the measure is not any single provision. It is the design constraint under which every provision was chosen: the bill had to raise the same revenue, preserve the same distribution of the burden across income groups, and cut marginal rates, all at once. Those three simultaneous demands turned lawmaking into a closed system, a zero-sum trade in which every percentage point of rate reduction had to be purchased with an eliminated preference, and that closed system is what this guide explains.

The One Test for this article is direct. A reader who finishes it should be able to explain the design constraint, meaning revenue neutrality plus distributional neutrality plus rate reduction, and should be able to name what was traded away to buy the lowest top individual rate in half a century. The answer to the second half is a list: the consumer interest deduction, the state and local sales tax deduction, the capital gains exclusion, the investment tax credit, unrestricted IRA deductibility for workers covered by employer plans, accelerated depreciation, and the passive loss shelters that had sustained a mass-market tax avoidance industry. The answer to the first half is a mechanism: because the bill could not lose revenue and could not shift the burden between income groups, the only remaining source of funds for lower rates was the tax base itself, so preferences were sold to buy rates. Everything in the statute follows from that mechanism, and everything in this guide is organized to make the mechanism visible.

The Statutory Identity of the Tax Reform Act of 1986

The formal identity of the statute is compact and worth stating precisely, because researchers encounter it in citations and legislative histories that assume familiarity. The short title is the Tax Reform Act of 1986. The public law number is Public Law 99-514, enacted by the 99th Congress. The Statutes at Large citation is 100 Stat. 2085. The bill that became law was H.R. 3838, and the President who signed it on October 22, 1986 was Ronald Reagan. Those five data points, short title, public law number, Statutes at Large page, bill number, and signing date, are the standard identifiers used in court opinions, Treasury regulations, and Joint Committee on Taxation publications whenever the act is discussed as enacted text rather than as political history.

Section 2 of the act contains a provision that explains why every citation to the federal tax code uses the 1986 date. Before 1986, the codified body of internal revenue law was titled the Internal Revenue Code of 1954, reflecting the last comprehensive reorganization. Section 2 redesignated that body as the Internal Revenue Code of 1986. The redesignation was not a substantive amendment to any particular section. It was a renaming of the entire code, which is why practitioners cite sections like 26 U.S.C. 469 or 26 U.S.C. 168 as provisions of the Internal Revenue Code of 1986 even when those sections were later amended by other statutes. The code retains the 1986 designation, so the title of the act appears in virtually every federal tax citation written since its enactment.

The scale of the act matches the ambition of that renaming. The enrolled bill ran to nearly one thousand pages in the Statutes at Large, beginning at page 2085 of volume 100, and it amended hundreds of sections of the code. Titles of the act addressed individual income taxation, corporate taxation, the alternative minimum tax, depreciation, tax-exempt entities, pensions, insurance, foreign income, and transition rules, among others. A statute pillar article cannot narrate every title in equal depth, and this one does not attempt it. The organizing principle here is the closed system: the provisions that matter most are the ones that demonstrate how the three constraints shaped the text, because those are the provisions that explain why the act looks the way it does and why no later Congress has reproduced it.

One clarification belongs at the outset because it corrects the most common misdescription of the statute. The Tax Reform Act of 1986 is frequently summarized as a tax cut, and the summary is wrong in the specific sense that matters. On the individual side, the act reduced statutory marginal rates sharply while broadening the base that those rates applied to, and the Joint Committee on Taxation scored the individual title as roughly revenue neutral over the budget window, meaning it neither raised nor lowered individual receipts in the aggregate. On the corporate side, the act cut the statutory rate from 46 percent to 34 percent while broadening the corporate base so aggressively that aggregate corporate liability rose by approximately 120 billion dollars over fiscal years 1987 to 1991, according to the Joint Committee on Taxation Blue Book designated JCS-10-87, with individual liability falling by a roughly similar amount. A reader who absorbs only one surprise from this guide should absorb this one: the most celebrated rate-cutting statute of the late twentieth century was, for corporations, a tax increase, and the increase was not an accident but the arithmetic consequence of the design constraint.

The Bill as a Vehicle: H.R. 3838 and the Origination Rule

The designation H.R. 3838 carries constitutional significance that a statute guide should explain. The Origination Clause of the Constitution provides that all bills for raising revenue shall originate in the House of Representatives, and tax legislation therefore moves through Congress as House-numbered bills even when the substantive impetus comes from the administration or the Senate. H.R. 3838 was the House vehicle: reported by the Committee on Ways and Means, passed by the House, amended by the Senate Committee on Finance and the full Senate, reconciled in conference, and enacted as Public Law 99-514. The bill number is thus not a mere identifier but a marker of the constitutional path the legislation was required to travel.

The 99th Congress, second session, was the legislative setting. The Congress that began in January 1985 and ended in January 1987 handled the bill across two calendar years, with House action concentrated in 1985 and Senate action and conference in 1986. The enrolled bill that the President signed on October 22, 1986 was among the longest revenue measures ever enacted to that point, running to nearly one thousand pages in the Statutes at Large from page 2085 of volume 100. Its length reflected both the breadth of the base broadening, which touched hundreds of code sections, and the density of the transition rules, which consumed a large share of the pages. The 99th Congress sat in the middle of the Reagan presidency, with a Republican Senate and a Democratic House, a division that forced the triple constraint to be genuinely negotiated rather than imposed, since neither party could dictate the ledger and both had to accept the discipline of the score.

The statute’s organization into titles follows the standard pattern for major tax legislation. Each title addresses a subject area, individual income tax, corporate income tax, alternative minimum tax, depreciation, pensions, tax-exempt entities, insurance, foreign income, and so on, and each title amends the corresponding sections of the Internal Revenue Code. Researchers navigating the act should work from the title structure rather than reading sequentially: the provision of interest will be found in the title governing its subject, and the Joint Committee’s Blue Book explanation follows the same title order. The effective-date provisions, typically found at the end of each title or subtitle, are as important as the substantive amendments, because the phase-in schedules determine when each change actually applied.

One further identifier deserves mention for researchers tracing the legislative history. The conference report that reconciled the House and Senate versions is the definitive statement of what the enacting Congress intended where the two chambers differed, and courts and the Treasury cite it routinely in interpreting ambiguous provisions. The report’s explanations of the rate settlement, the transition rules, and the effective dates are the primary legislative history for the statute’s most contested provisions. This pillar article does not narrate the conference bargaining, which belongs to the passage history of the act, but readers interpreting the enacted text will find the conference report the indispensable companion to the Statutes at Large text.

The Three Constraints That Designed the Statute

Every durable explanation of the 1986 act begins with the three simultaneous constraints, because the constraints determined the contents before any particular provision was drafted. Revenue neutrality meant the legislation could not reduce federal receipts over the budget window used for scoring. Distributional neutrality meant it could not shift the overall tax burden among income classes; the share of total liability borne by each group was to remain approximately what it had been under prior law. Rate reduction was the political objective that justified the exercise: statutory marginal rates, especially the top individual rate, were to fall substantially. Any two of these demands can be satisfied without much ingenuity. All three together admit only one solution, which is base broadening, the elimination or curtailment of deductions, exclusions, credits, and timing preferences so that a wider base taxed at lower rates yields the same revenue in the same proportions.

How did revenue neutrality force base broadening?

Revenue neutrality meant the bill had to fund every rate cut from inside the bill, with no outside source permitted. Cutting the top individual rate from fifty percent to twenty-eight percent removed enormous projected receipts, and the only way to replace that sum without raising another rate was to tax income prior law had excluded, deferred, or sheltered.

Each eliminated preference therefore functioned as currency, and the rate schedule that emerged was simply the set of prices those preferences could buy.

Revenue neutrality sounds like an accounting nicety, but in the legislative process it functioned as a rationing device. The Congressional Budget Office and the Joint Committee on Taxation scored each proposal, and the score determined whether a provision could survive. A rate cut that cost a large sum in projected receipts over the window needed an offset that raised a similar sum. No offset, no cut. This turned the committee rooms into a kind of market: members arrived with rate reductions they wanted for their constituents and had to purchase them with repeals, curtailments, and base broadeners they were willing to defend against the constituencies losing them. The discipline was self enforcing. A member who proposed a popular cut without naming the offset was asking colleagues to absorb the cost, and colleagues who would later need offsets of their own had no incentive to agree.

The logic deserves emphasis because it reverses the usual direction of legislative bargaining. In ordinary tax legislation, members add provisions and the revenue cost is whatever results, with deficits or offsetting measures handled separately. Under the three-constraint design, the revenue total was fixed first, so adding a preference required either raising a rate or eliminating a different preference of equal value. That arithmetic discipline is what made the coalition possible: supporters of lower rates could not defect to protect a favored preference without visibly raising someone else’s rate, and defenders of preferences could not demand retention without naming the rate increase that retention required. The constraint converted every parochial demand into a visible trade, and the visibility is what held the bargain together through enactment.

Distributional neutrality reinforced the discipline by blocking the easiest escape route. Without it, a bill could achieve revenue neutrality while cutting rates by shifting the burden downward, broadening the base at the bottom while narrowing it at the top. The distributional constraint forbade that move, or at least forbade it in the aggregate scoring that the Joint Committee on Taxation performed. This is why the act paired its rate cuts with such aggressive increases in the standard deduction, the personal exemption, and the earned income tax credit: those provisions removed millions of low-income filers from the rolls and offset the base broadening that would otherwise have fallen hardest on them. The Congressional Research Service analysis designated RL34498 examined the surcharge mechanism discussed below and treated its design as evidence of distributional intent, while later assessments, including work by John Witte published in 1991, concluded that distributional neutrality was not fully achieved in practice. The design constraint and the realized outcome are different things, and the statute should be understood as an attempt at the constraint rather than a perfect execution of it.

The third constraint, rate reduction, supplied the political energy. A bill that merely broadened the base at existing rates would have been a tax increase by another name, and no coalition would have assembled for it. The promise of dramatically lower statutory rates, the lowest top individual rate in half a century, gave members something to sell to constituents and gave the administration its headline. But the promise was credible only because the first two constraints made it arithmetic rather than aspirational: the rates fell to exactly the level the eliminated preferences could purchase, and no further. Understanding the act therefore requires holding all three constraints in mind simultaneously, because any account that mentions only the rate cuts is describing the price tag while ignoring the purchase.

The rate history gives the half-century framing its weight. The top individual rate stood at 91 percent through the early 1960s, fell to 70 percent in the mid-1960s, and stayed there until the 1981 act cut it to 50 percent. The 1986 act’s 28 percent was thus not merely lower than its predecessor. It was lower than any top rate since the 1930s, a break with the entire postwar fiscal order. The half-century framing mattered politically because it let the coalition describe the achievement as historic rather than incremental, which in turn justified the historic scale of the repeals. A 22-point cut from 50 to 28 required a 22-point purchase. Had the drafters aimed for 35 percent, the ledger would have been smaller, the repeals fewer, and the coalition easier to assemble but harder to distinguish from ordinary tax cutting. The 28 percent figure was the forcing device: only a rate that low could justify repeals that broad, and only repeals that broad could fund a rate that low. The number and the ledger chose each other.

There is a further subtlety in the rate history worth noting. The 1981 cut from 70 to 50 percent had been enacted without offsetting base broadeners, which meant it lost revenue by design. The 1986 cut from 50 to 28 percent was enacted with offsets that made it revenue neutral by design. The two cuts look similar on a chart of top rates and differ completely in fiscal character. The first was a decision to collect less. The second was a decision to collect differently. Confusing the two is the most common analytical error in popular accounts of the era’s tax policy, and this guide’s insistence on the ledger is meant to prevent it. The 1986 rate is historic not because it is low but because it was bought.

The discipline also shaped the order in which decisions were made. Because the score was the currency, the committees resolved the revenue questions before they resolved the rate questions. A member could not know how far the top rate could fall until the base broadeners had been priced. This sequencing explains why the bill’s history reads as a long argument about deductions and credits followed by a comparatively swift agreement on rates. The rates were the residual: whatever the broadeners bought, the rates became. The 28 percent figure was not chosen first and funded later. It emerged from the ledger once the repeals had been tallied, which is why contemporaries described the final rate as discovered rather than decided.

How the Constraints Were Enforced: Scoring Inside the Closed System

A constraint that cannot be measured cannot bind, and the three constraints of the 1986 act bound because they were measured continuously, by a single scorekeeper, against a fixed baseline. That scorekeeper was the Joint Committee on Taxation, the nonpartisan staff body that produces Congress’s official revenue estimates for tax legislation. Every provision in the bill, and every amendment offered to it, was translated by the Joint Committee staff into a dollar figure representing the change in federal receipts relative to what prior law would have collected over the five-year budget window covering fiscal years 1987 through 1991. Revenue neutrality was not a slogan or a hope. It was a running total, recomputed with each alteration, and the total had to balance when the bill was finished.

Who kept the books while the closed system bargained?

The Joint Committee on Taxation produced the official revenue estimates that determined whether each provision survived, the Congressional Budget Office supplied the macroeconomic backdrop, and the Treasury’s Office of Tax Analysis produced the distributional tables that checked the committee’s work. Their numbers were trusted, so the fixed total bound every trade.

These institutions did not make policy, but their numbers made policy possible, because the triple constraint was meaningless without a trusted score. A revenue neutrality rule enforced by partisan estimates would have collapsed into argument about whose numbers to believe. The Joint Committee’s institutional credibility was the load-bearing wall of the whole structure. The scoring conventions themselves shaped the substance. Static scoring, which assumed taxpayer behavior stayed fixed, determined the official price of each provision. Everyone involved knew the assumption was false: taxpayers would change their realizations, their borrowing, their organizational forms, and their reporting in response to the new rules. But static scoring was the agreed fiction that made bargaining possible, because it gave every provision a single price that all parties accepted. Dynamic estimates, which would have tried to model the behavioral responses, would have produced a range of prices and endless argument about which model to trust. The drafters chose the fiction of fixity over the reality of dispute, and the choice was rational: a negotiation needs a common unit of account more than it needs a true one. Martin Feldstein, in National Bureau of Economic Research working paper 17531, later found the static neutrality intact and the behavioral offsets real, vindicating the choice as a legislative matter even as it complicated the economics.

The baseline against which everything was measured deserves explanation, because the choice of baseline determined what counted as a tax increase or a tax cut. The Joint Committee measured each provision against a current-law baseline: the receipts that would have occurred if Congress had done nothing and prior law had continued unchanged. A provision that repealed a deduction therefore scored as a revenue gain equal to the additional tax collected under the new, broader base, while a provision that cut a rate scored as a revenue loss equal to the tax no longer collected at the old, higher rate. The bill was revenue neutral when the gains and losses summed to approximately zero over the window. This arithmetic is what made the trade table literal rather than metaphorical: each row’s revenue figure was a Joint Committee estimate, and the rate schedule was set where the sum of the rows permitted.

Distributional neutrality was measured with a parallel apparatus. The Joint Committee and the Treasury compared the percentage change in after-tax income across income classes under the bill versus under prior law, asking whether each group’s share of the total burden remained approximately unchanged. A provision that broadened the base at the bottom while cutting rates at the top would have failed this test visibly, because the lower groups would have shown larger percentage reductions in after-tax income than the upper groups. The bubble surcharge, the zero bracket expansion, and the earned income credit expansion were all calibrated against these distributional tables, adjusted until the percentage changes lined up across the distribution. Like the revenue total, the distributional table was recomputed as provisions changed, so any amendment that shifted the burden had to be offset by an amendment that shifted it back. The tables gave the negotiators a shared picture of the distribution, and the shared picture let them bargain over it. Without the tables, distributional neutrality would have been a slogan. With them, it was a constraint with teeth, enforced draft by draft until the numbers balanced.

The five-year window shaped the bill in ways that are visible throughout its transition provisions. Because neutrality was measured over fiscal years 1987 through 1991, provisions whose revenue effects were front-loaded or back-loaded could be used to balance the total even when their long-run effects differed. The phaseout schedules, the 1987 interim rate structure, and the declining passive-loss transition percentages all moved revenue across years within the window, and the drafters understood exactly how much each timing choice contributed to the five-year sum. This is one reason the statute is so dense with effective dates: the dates were not merely administrative conveniences but scoring instruments, calibrated to make the window balance.

The enforcement mechanism in the committee room followed directly from the measurement. Because every amendment carried a Joint Committee score, any member proposing to restore a preference had to identify, in the same amendment or a companion one, the offsetting revenue: a smaller rate cut, a different preference eliminated, or a narrower version of the restoration. There was no procedural rule that formally required this; the requirement emerged from the arithmetic itself, because an unpaid-for amendment would have broken the revenue total that the leadership had committed to hold. Members internalized the discipline quickly. The closed system did not need a parliamentarian to enforce it. The scorekeeper’s running total was enforcement enough, and the total was public within the room, so every trade was visible to every participant.

The Individual Rate Collapse

The most visible product of the three constraints was the individual rate schedule. Prior law imposed fourteen nonzero statutory brackets with rates climbing to fifty percent, a structure in which the marginal rate changed at many income thresholds and the top rate applied to a relatively narrow slice of high-income filers. The fourteen-rate schedule had not been designed. It had accreted: the 1954 code had consolidated the wartime rate structure, and each subsequent decade had added layers, surcharges, surtaxes, and bracket adjustments enacted for the fiscal needs of their moments. The 1986 act replaced that graduated structure with two nominal rates, fifteen percent and twenty-eight percent, effective for tax years beginning in 1988, with a transitional graduated schedule applying for tax year 1987. The top statutory rate thus fell from fifty percent to twenty-eight percent, a reduction of twenty-two percentage points that represented the lowest top individual rate in half a century.

For tax year 1987, the act provided an interim graduated schedule rather than the full two-rate structure, smoothing the transition from the fourteen-bracket system to the new schedule. The 1987 schedule carried five brackets with a top rate of 38.5 percent, a midpoint between the old 50 percent and the new 28 percent. The interim year served two purposes. It softened the transition for taxpayers whose planning had assumed the old schedule, and it distributed the revenue effects across two fiscal years in a way that helped the scoring. The existence of the 1987 transition schedule is a useful reminder that the celebrated two-rate structure did not spring into being overnight; it was phased in, like nearly everything else in the act, through transition rules that softened the first-year impact. The 1987 tables were printed, used once, and never used again.

Why 28 percent, rather than 30 or 25? The answer is the ledger, not aesthetics. As the base broadeners were priced, the negotiators could see how far the top rate could fall while holding revenue and distribution steady. Each additional repeal bought another point or two of rate reduction. The process stopped at 28 because the politically available repeals were exhausted: the remaining preferences, mortgage interest, charitable gifts, state income taxes, had constituencies that could not be taken on. Had the coalition been willing to repeal those, the rate could have fallen further. Had it repealed less, the rate would have stayed higher. The 28 percent figure is thus a precise measure of what the 1986 political market would bear: the point where the marginal repeal’s political cost exceeded the marginal rate cut’s political benefit. Reading the rate as a price, rather than as a policy preference, is the key to the whole statute. The drafters did not choose 28 percent and then find the money. They found the money and discovered 28 percent.

The rate collapse changed the economics of every deduction the act left standing. Under a 50 percent top rate, a 1,000 dollar deduction was worth 500 dollars to a top-bracket taxpayer. Under a 28 percent top rate, the same deduction was worth 280 dollars. This arithmetic is the hidden engine of the whole reform: lower rates automatically reduced the value of remaining preferences, which meant the base broadening did double duty. The repeals raised revenue directly, and the rate cuts reduced the cost of the preferences that survived. Defenders of the surviving deductions, such as the mortgage interest deduction and the charitable contribution deduction, accepted this devaluation as part of the bargain. Their provisions lived, but each was worth less than before, and the lower rates were the compensation. The rate was thus self-reinforcing. It reduced the value of the preferences it had not repealed, which reduced the incentive to invent new ones, which protected the base the repeals had built.

The symbolism of the number reinforced its function. Twenty-eight was low enough to be historic, round enough to be memorable, and close enough to the corporate 34 percent to preserve a rough relationship between the two systems. It was also low enough to change behavior in the ways the drafters wanted: at 28 percent, the payoff to shelter engineering fell dramatically, because a deduction was worth only 28 cents on the dollar instead of 50. A two-rate structure also changed the relationship between the individual tax and the corporate tax, a subject taken up in the corporate section below. For individuals, the headline result was a schedule that could be printed on a postcard and a set of phase-outs that could not. The postcard version, 15 and 28, traveled the country in speeches. The bubble, the 1987 interim schedule, and the devalued deductions did the quiet work.

The collapse from fourteen brackets to two nominal rates is the provision that gives the act its reputation for simplification, and the reputation needs the qualification developed later in this guide. Reducing the number of statutory brackets did make the rate schedule itself easier to describe, and it eliminated the fine gradations that had characterized the prior code. But the number of brackets is only one dimension of complexity, and the act added complexity along several others, including the passive activity loss rules, the expanded alternative minimum tax, and the dense transition provisions that governed the phase-in years. The bracket collapse was real and substantial. Whether it simplified the code as experienced by filers is a separate question with a less flattering answer.

Worked Arithmetic: Reading the New Schedule

The rate schedule becomes concrete when traced through the arithmetic of actual filers, and the exercise is worth doing because it shows how the zero bracket, the two nominal rates, and the bubble interacted as a single machine. Consider first a married couple filing jointly in 1988 with forty thousand dollars of gross income and no itemized deductions. The standard deduction for joint filers was five thousand dollars, and two personal exemptions at one thousand nine hundred fifty dollars each sheltered three thousand nine hundred dollars more, so taxable income was thirty-one thousand one hundred dollars. The first portion of that taxable income was taxed at fifteen percent and the remainder at twenty-eight percent, because forty thousand dollars of gross income sat well below the bubble band that began at seventy-one thousand nine hundred dollars for joint filers. The couple’s marginal rate, the rate on the next dollar earned, was twenty-eight percent, while the average rate on the full forty thousand was far lower, pulled down by the zero bracket and the fifteen percent bracket beneath the margin.

Now move the same couple to one hundred thousand dollars of gross income in 1988. After the five thousand dollar standard deduction and three thousand nine hundred dollars of exemptions, taxable income was ninety-one thousand one hundred dollars, which placed the couple inside the bubble band running from seventy-one thousand nine hundred dollars to one hundred forty-nine thousand two hundred fifty dollars. Within that band, the five percent surcharge stacked onto the twenty-eight percent rate, so the marginal rate on the next dollar was thirty-three percent. The couple paid fifteen percent on the bottom slice of taxable income, twenty-eight percent through the middle, and thirty-three percent at the margin, a structure with three effective rates despite the statute’s two nominal brackets. The example shows why the bubble could not be ignored in any honest description of incentives: for this couple, the tax price of additional earnings was thirty-three cents on the dollar, not twenty-eight.

Move the couple once more, to two hundred thousand dollars of gross income. Taxable income after the standard deduction and exemptions was one hundred ninety-one thousand one hundred dollars, above the top of the bubble band at one hundred forty-nine thousand two hundred fifty dollars. The five percent surcharge applied only within the band, phasing out the benefit of the fifteen percent bracket and the exemptions, and once those benefits were fully phased out the surcharge ended. The marginal rate on the next dollar was therefore twenty-eight percent, lower than the thirty-three percent faced by the couple at one hundred thousand dollars. This is the famous inversion: the marginal rate rose with income through the bubble and then fell at the top, a shape produced entirely by the distributional constraint’s demand that the fifteen percent bracket’s benefit be recaptured from upper-middle-income filers.

A final example shows the zero bracket’s power at the bottom. Take a single filer in 1988 with eight thousand dollars of gross income and no itemized deductions. The single standard deduction was three thousand dollars and one personal exemption sheltered one thousand nine hundred fifty dollars, for a combined zero bracket of four thousand nine hundred fifty dollars. Taxable income was three thousand fifty dollars, taxed at fifteen percent. But consider the same filer with four thousand dollars of gross income: the zero bracket of four thousand nine hundred fifty dollars exceeded gross income entirely, so taxable income was zero and no liability arose. The administration’s statement that about six million filers would leave the rolls described millions of cases like this one, where the enlarged zero bracket swallowed the filer’s entire income. The example also shows the interaction the drafters intended: base broadening might have subjected more of this filer’s income to tax, but the zero bracket ensured the newly taxable income still fell below the liability threshold.

These examples use the 1988 parameters because 1988 was the first year the full two-rate structure applied. For 1987, the interim graduated schedule produced different arithmetic, with more brackets and higher top rates stepping down toward the 1988 destination. The transition-year complexity is part of the simplification story told later in this guide: a filer comparing 1986, 1987, and 1988 returns encountered three different rate structures in three consecutive years, which is not most people’s idea of a simpler code. The worked arithmetic for 1988 nevertheless remains the essential exhibit, because it is the year the design is visible in its finished form: a zero bracket expanded for distributional neutrality, two nominal rates purchased by base broadening, and a bubble band where the distributional constraint left its fingerprint on the marginal rate.

The Thirty-Three Percent Bubble, Explained

The bubble warrants its own section because it concentrates so many of the act’s design tensions into a single mechanism. At first glance, a rate schedule that rises to thirty-three percent in the middle and then falls back to twenty-eight percent at the top looks like a drafting error or a political compromise that escaped editing. It was neither. It was the price of satisfying all three constraints simultaneously within a two-rate structure.

The arithmetic works as follows. Distributional neutrality required that high-income filers not receive a windfall from the rate collapse, but the fifteen percent bottom bracket conferred a benefit on every filer, including those at the top, because the first dollars of everyone’s taxable income were taxed at fifteen percent rather than at the higher rates that a single-rate system would have implied. Similarly, the increased personal exemption sheltered income for filers at all levels. To keep the distribution of the burden unchanged, the statute had to recapture those benefits from upper-middle-income filers, and the five percent surcharge was the recapture instrument. By phasing out the value of the fifteen percent bracket and the exemptions over a defined income band, the surcharge ensured that the effective benefit of the low bracket accrued only to filers below the band.

The 1988 bands, seventy-one thousand nine hundred dollars to one hundred forty-nine thousand two hundred fifty dollars for joint returns, forty-seven thousand fifty dollars to ninety-seven thousand six hundred twenty dollars for single returns, and sixty-seven thousand two hundred dollars to one hundred thirty-four thousand nine hundred thirty dollars for heads of household, were indexed for inflation in subsequent years, so the nominal thresholds moved with the price level rather than remaining fixed. Indexation mattered because the Economic Recovery Tax Act of 1981 had introduced indexation of brackets, and the 1986 act preserved that protection against bracket creep. A bubble defined in nominal dollars without indexation would have expanded its reach every year as inflation pushed filers into it; with indexation, the band maintained its real position in the income distribution.

The bubble also illustrates the difference between statutory rates and effective marginal rates, a distinction the act’s drafters understood and its popularizers often ignored. Headlines reported two brackets, fifteen and twenty-eight, and many summaries still describe the act that way. But any filer with income inside the phase-out band faced a thirty-three percent marginal rate on the next dollar earned, a rate higher than the top nominal rate. The distinction matters for any serious account of incentives under the act, because marginal rates, not nominal brackets, determine the tax consequence of earning additional income. The act’s true marginal rate structure had four zones, not two, and the bubble was the zone where the distributional constraint bit hardest. A joint filer in 1988 with taxable income of 100,000 dollars sat inside the band and faced 33 percent on the marginal dollar, while a neighbor at 160,000 dollars, above the band, faced 28 percent, a pattern invisible on the printed rate table itself.

Whether the surcharge succeeded in its distributional purpose is a question for the distributional section below. Here the point is structural: the bubble demonstrates that distributional neutrality was not a rhetorical flourish but an operative design parameter, one that forced the drafters to build a phase-out mechanism into the rate schedule itself. A bill without the distributional constraint could have used a clean two-rate schedule and accepted the resulting shift in the burden. The 1986 act could not, so it built the bubble, and the bubble’s ungainliness is the visible scar of the constraint.

The Zero Bracket Grows: Standard Deduction and Personal Exemption

If the rate collapse was the headline, the expansion of the zero bracket, the amount of income shielded from tax entirely, was the distributional engine. The act increased the standard deduction substantially and nearly doubled the personal exemption, and the administration stated that about six million low-income filers would be removed from the tax rolls entirely, a figure attributed to President Reagan in the administration’s presentation of the bill. Those two provisions did much of the work of distributional neutrality, offsetting the base broadening that would otherwise have increased liability for filers at the bottom of the income distribution.

The standard deduction for 1988 was set at five thousand dollars for joint filers, four thousand four hundred dollars for heads of household, three thousand dollars for single filers, and two thousand five hundred dollars for married filers filing separately, with indexation for inflation in later years. These levels represented large increases over the prior-law zero bracket amount, and they had two related effects. First, they reduced taxable income directly for every filer who claimed the standard deduction rather than itemizing. Second, by raising the threshold at which itemizing became worthwhile, they reduced the number of filers who itemized deductions at all, which simplified filing for millions of households even as other parts of the act added complexity elsewhere.

The personal exemption followed a stepped schedule: one thousand eighty dollars for 1986, one thousand nine hundred dollars for 1987, one thousand nine hundred fifty dollars for 1988, and two thousand dollars for 1989 and later years, indexed for inflation after reaching the two thousand dollar level. The near doubling from 1986 to 1987 was the single largest one-year increase in the exemption’s history, and its interaction with the rate cuts is worth tracing. Under prior law, an exemption deducted at a fifty percent marginal rate was worth fifty cents on the dollar to a top-bracket filer; under the new law, the same exemption deducted at twenty-eight percent was worth twenty-eight cents. The exemption’s value fell for high-bracket filers even as its nominal amount rose, which is one of the quiet ways the act shifted the composition of the burden while holding the distribution steady in the aggregate.

The six million figure belongs to President Reagan, who cited it as the human measure of the reform. The mechanism behind it was arithmetic, not discretion: when the sum of the standard deduction and the personal exemptions for a household exceeded the household’s income, no tax was owed and no return was required. A family of four in 1988, for example, faced a filing threshold built from the 5,000 dollar joint standard deduction plus four 1,950 dollar exemptions, a combined 12,800 dollars of income shielded before the 15 percent rate applied to a single dollar. Millions of households whose incomes fell below the new thresholds simply dropped out of the income tax system. The provision was the bottom half of the distributional bargain: the act took preferences from affluent households and used part of the proceeds to lift poor households out of the tax entirely.

Why did indexing the new thresholds matter?

Without indexation, inflation would have eroded the standard deduction, the personal exemption, and the bracket thresholds every year, pushing taxpayers into higher brackets on unchanged real income. The phenomenon, called bracket creep, would have repealed the distributional bargain within a few years. Indexing locked the 1988 choices in real terms, so the zero bracket and the credit kept their value.

Congress had addressed bracket creep partially in 1981 by indexing brackets, but the 1986 act extended and regularized indexing across the new structure, including the standard deduction, the exemption, and the earned income credit parameters. The effect was to lock in the distributional choices the act made. Without indexing, inflation would have quietly repealed the six-million achievement within a few years by dragging the thresholds down in real terms. With indexing, the achievement compounded. The drafters understood this, and the indexing provisions are among the least controversial and most durable parts of the statute. The choice had a fiscal cost, because bracket creep had been a reliable source of automatic revenue growth, and giving it up meant the ledger had to find the revenue elsewhere. The drafters paid that cost because the alternative, letting inflation quietly undo the rate cuts, would have made the bargain a depreciating asset.

The Preferences That Survived

A guide organized around trades must also account for the trades that never happened. The 1986 act left several of the code’s largest preferences intact, and each survival has an explanation rooted in the same constraint that explains the repeals. The mortgage interest deduction for qualified residence interest survived, though its value fell as rates fell. The charitable contribution deduction survived. The deduction for state and local income taxes and real property taxes survived. The exclusion for employer-provided health insurance survived. Together, these survivors represented an enormous amount of forgone revenue, larger in the aggregate than some of the repeals the act is famous for. Their survival is not an oversight in the design. It is the design’s boundary, the line where the political cost of the next repeal exceeded the political value of the next rate cut.

Why did the mortgage interest deduction survive the 1986 bargain?

Repeal raised large revenue at a coalition cost the drafters could not pay. Every itemizing homeowner formed the constituency, too numerous, too sympathetic to take on inside a bill already asking homeowners to absorb the consumer interest phase-out and the passive loss rules. The yield-to-cost ratio failed, so the preference lived, devalued by lower rates but intact.

The charitable deduction survived for a related but distinct reason: its defenders included not just donors but the entire nonprofit sector, universities, hospitals, religious institutions, and cultural organizations that depended on deductible giving. That coalition could mobilize opposition in every congressional district, which made the deduction one of the most expensive preferences to name as an offset. The drafters priced the repeal, saw the coalition it would awaken, and left the provision alone. The decision reveals a hierarchy within the constraint. Revenue neutrality required offsets, but not every conceivable offset was available at any price. Some preferences were protected by constituencies whose retaliation would have destroyed the coalition, and the drafters treated those preferences as outside the tradable set. The ledger contained only the trades that could be executed without killing the negotiation.

The state and local income tax deduction survived because its repeal would have functioned as a tax increase concentrated in high-tax states, with a clear geographic incidence that members from those states could not accept. The sales tax deduction, by contrast, was repealable precisely because its constituency was smaller and less organized. The distinction between the two components of the state and local deduction is thus a miniature of the whole act’s politics: repeal what the market will bear, keep what it will not, and let the distributional tables confirm the choice. The income tax deduction’s survival also protected a fiscal federalism bargain the drafters did not want to reopen. Removing the federal subsidy for state income taxation would have pressured states to cut their own taxes or shift to other levies, a disruption with consequences far beyond the revenue score. The drafters declined the disruption.

The employer health insurance exclusion survived almost without debate, which is remarkable given its size. The explanation is timing and focus. The 1986 negotiation was about rates and the individual base broadeners that could fund them. Health policy was a separate domain with its own committees, its own constituencies, and its own impending battles. Opening the exclusion would have imported the entire health care lobby into a negotiation that was already at the edge of manageability. The drafters made a deliberate decision to leave health-related preferences for another day. The exclusion’s survival is a reminder that the closed system was closed by choice: the drafters defined the boundaries of the tradable set, and they drew those boundaries to keep the coalition assemblable.

The act also declined to move toward a consumption tax base, a road some reformers had advocated. Proposals to replace the income tax with a national sales tax or a value-added tax circulated in the reform debate, and the Treasury’s own studies had examined consumption-tax alternatives. The enacted statute remained firmly an income tax, albeit a broader-based one with lower rates. The choice reflected both the constraints and the politics: a consumption tax would have violated distributional neutrality as scored, since consumption taxes fall more heavily on lower-income households that spend a larger share of income, and it would have required dismantling the existing administrative apparatus. Base broadening within the income tax was the radicalism the coalition could sustain; changing the tax base itself was not.

The payroll tax was untouched. The act reformed the income tax exclusively, leaving Social Security and Medicare payroll taxes, their rates, their wage bases, and their benefit formulas exactly as prior law had set them. This boundary is worth stating because later discussions sometimes conflate the 1986 income tax reform with the broader fiscal picture of the decade. The distributional neutrality constraint applied to the income tax as scored; it did not extend to the combined burden of income and payroll taxes, a limitation that matters for any assessment of the act’s overall distributional record, since payroll taxes fall most heavily on wage income below the taxable maximum.

Finally, the act left the fundamental architecture of filing untouched. Filing statuses, the joint return, withholding, estimated payments, and the April filing deadline all continued as before. The drafters concentrated their limited political capital on the rate-for-base trade and left the machinery alone. The pattern confirms the closed-system logic once more: every provision in the bill had to earn its place by contributing to the three constraints, and administrative reorganization contributed nothing to revenue neutrality, distributional neutrality, or rate reduction. What did not serve the constraints did not make the bill, whether as an elimination or as a reform.

Taken together, the survivors define the act’s limits more honestly than the repeals define its ambitions. The repeals show what the constraint could compel. The survivors show what it could not. A reader who wants to understand why later base-broadening efforts stalled can start here: the preferences that survived 1986 were the ones with the strongest protection, and they have only grown stronger as their constituencies learned from the 1986 near-miss. The ledger of 1986 harvested the available repeals. What remained was, by definition, the unavailable.

Consumer Interest: Phased Out, Not Erased

The deduction for consumer interest, the interest paid on credit card balances, auto loans, and other personal borrowing, was one of the most widely claimed itemized deductions under prior law, and its treatment in the 1986 act illustrates the statute’s characteristic use of transition rules. The act did not eliminate this deduction in a single stroke. It phased the deduction out over four years. For 1987, 65 percent of consumer interest remained deductible. For 1988, the figure was 40 percent. For 1989, 20 percent. For 1990, 10 percent. For 1991 and later years, the deduction was gone. Describing this as an immediate elimination would misstate the statute; the phaseout was deliberate, giving borrowers several years to adjust their financing before the deduction disappeared entirely. The phase-out percentages are worth memorizing because they recur, with different numbers, in the passive loss transition, and because they show the drafters buying time for adjustment. The revenue followed the mirror image of the decay: a modest gain in 1987, growing each year until the full annual value arrived in 1991.

The policy logic of the phaseout reflected the base-broadening imperative. Under prior law, the deductibility of consumer interest subsidized borrowing for consumption, and the subsidy was worth more to filers in higher brackets, because a deduction’s value equals the taxpayer’s marginal rate. With the top rate falling from fifty percent to twenty-eight percent, the value of every remaining deduction fell as well, which softened the blow of repeal for high-bracket borrowers even as it reduced the subsidy’s efficiency rationale. The drafters treated consumer interest as the paradigmatic preference to sacrifice: widely used enough to raise substantial revenue, but difficult to defend as a matter of tax policy once the rate-for-base trade was on the table, since no neutral income tax would subsidize consumption borrowing. The Treasury proposals that fed the bill argued that deductible consumer borrowing subsidized consumption and distorted the choice between saving and spending.

The phaseout schedule also demonstrates how transition rules functioned as political shock absorbers throughout the act. Immediate repeal would have imposed the full revenue gain in the first year while concentrating the behavioral adjustment, as borrowers rushed to restructure debt, into a single tax year. The stepped phaseout spread both the revenue and the adjustment across four years, which made the provision easier to defend and easier to score. Households adjusted by shifting toward home equity borrowing, whose interest remained deductible as qualified residence interest, a substitution the drafters anticipated and accepted. The phase-out is a case study in how a base broadener works through behavior rather than through the statute’s face: the law changed the price of debt, and households repriced their balance sheets in response.

The Sales Tax Deduction and the Preserved Core

The act’s treatment of the deduction for state and local taxes is frequently misdescribed, so precision matters here. Prior law allowed itemizers to deduct state and local income taxes, general sales taxes, real property taxes, and personal property taxes, with a rule that filers claiming the sales tax deduction could not also claim the income tax deduction. The 1986 act disallowed the deduction for state and local general sales taxes. It retained the deductions for state and local income taxes and for real and personal property taxes. To say that the act eliminated the SALT deduction is therefore wrong; it eliminated one component of a multi-part deduction while preserving the rest.

The distinction mattered enormously to the politics of the provision. The sales tax deduction was the component most difficult to defend on administrability grounds, because claiming it required either saving receipts or using optional tables, and it was the component whose repeal raised revenue with the least concentrated geographic pain. Repealing the hardest-to-verify component while keeping the verifiable ones let the drafters claim both revenue and simplification, a combination that helped sell the provision inside the coalition. Retaining the income and property tax deductions preserved the federal offset for the largest state and local revenue instruments, which limited the provision’s impact on high-tax states relative to what full repeal would have inflicted. The compromise, repeal the sales tax piece and keep the rest, was characteristic of the closed-system bargaining: each preference was weighed not only for its revenue yield but for the coalition cost of its elimination, and the sales tax deduction was the piece whose yield-to-cost ratio made it expendable.

The geography of the sales tax repeal shows how a single line in the statute could split the country. In states with no income tax, such as Texas, Florida, and Washington, itemizers had relied on the sales tax deduction as their principal state levy write-off. Losing it meant losing the entire state tax benefit, while itemizers in New York or California kept their income tax deductions and noticed a smaller change. The provision thus transferred a slice of the federal subsidy for state taxation from no-income-tax states to income-tax states, a result no one defended as policy and everyone accepted as arithmetic. Defenders of the repeal argued that the sales tax deduction subsidized state decisions to tax consumption rather than income, distorting state fiscal choices. Defenders of the deduction replied that the federal code should be neutral among state tax structures, not a thumb on the scale for the income tax. The statute sided with the repeal, booked the revenue, and left the neutrality argument for later Congresses, which would revisit the state and local deduction repeatedly in the decades that followed.

The revenue raised by the sales tax repeal contributed to the general pool that funded rate reduction, rather than being earmarked to any particular bracket. This is worth stating because popular accounts sometimes imply a one-to-one mapping between each repealed preference and a specific rate cut, as if the consumer interest deduction bought the fifteen percent bracket and the sales tax deduction bought something else. The statute worked nothing like that. All base broadening flowed into a single revenue pool, and the rate schedule was set at the level the total pool could sustain. The trade table below presents each preference alongside its revenue contribution and its constituency precisely to show the aggregate character of the exchange: many losers, one pool, one schedule.

Capital Gains: The Sixty Percent Exclusion Ends

Perhaps no single provision better illustrates the closed-system logic than the repeal of the sixty percent exclusion for long-term capital gains. Under prior law, only forty percent of a long-term capital gain was included in taxable income, which at a fifty percent top rate produced a maximum effective rate of twenty percent on gains. The 1986 act repealed the exclusion entirely, so that capital gains were taxed as ordinary income, effective for 1987, with a transitional cap softening the first year. With the top individual rate at twenty-eight percent, the maximum rate on gains rose from twenty percent to twenty-eight percent even as the top rate on ordinary income fell from fifty percent to twenty-eight percent. The provision was simultaneously a rate cut for wages and a rate increase for gains, and that duality was the point.

The repeal raised substantial revenue, among the largest single contributions to the rate-reduction pool, and it represented a philosophical commitment to the comprehensive income base: the drafters treated the exclusion as the paradigmatic special preference, a carve-out that violated horizontal equity by taxing two dollars of economic income differently depending on its source. Eliminating it moved the code toward the principle that a dollar is a dollar regardless of how it is earned, which was the intellectual foundation of the entire base-broadening enterprise.

The constituency that lost the exclusion, investors realizing long-term gains, was among the most politically potent groups affected by the act, which makes the repeal a revealing test of the closed-system discipline. In ordinary legislation, a provision opposed by organized investor interests would have been softened or dropped. Under the three constraints, softening it would have required either raising the top rate above twenty-eight percent or finding an equivalent sum elsewhere, and neither alternative commanded a majority. The exclusion’s repeal therefore demonstrates the constraint’s binding force: it eliminated a preference whose defenders were powerful precisely because the arithmetic left no room for sentiment.

The arguments around the repeal were the most fully developed of any in the bill, and they deserve attribution rather than summary judgment. Supporters of repeal, including the Treasury reform proposals, argued that the exclusion encouraged the recharacterization of ordinary income as capital gain, spawned an industry of conversion transactions, and violated the neutrality the act was trying to achieve between different kinds of income. Supporters of a preferential rate, including many in the investment community, argued that taxing gains at full rates discouraged risk-taking, locked investors into appreciated positions, and taxed inflationary gains that represented no real increase in purchasing power. The statute chose the Treasury position, and the choice was load-bearing: without the revenue from taxing gains as ordinary income, the 28 percent rate could not have been bought.

The realization wave of late 1986 demonstrates how taxpayers price statutory deadlines. Because the exclusion disappeared for gains realized after 1986, taxpayers with appreciated assets had a powerful incentive to sell before year end. In the final months of 1986, holders of appreciated stock, real estate, and business interests rushed to sell before the exclusion disappeared, producing one of the largest concentrated realization events in the history of the income tax. Brokers reported record volume. The wave was entirely rational: a taxpayer facing a higher rate on gains in January had every reason to realize in December. The drafters had anticipated the rush and built it into the score, treating the first-year bulge as a transitional feature rather than a permanent yield. The honesty of that treatment is worth noting. A less disciplined process might have booked the bulge as recurring revenue and spent it on permanent rate cuts, creating a fiscal hole when realizations returned to normal. The 1986 scorekeepers separated the transitional from the permanent, and the ledger distinguished them. In the 1986 act, the dates were negotiated as carefully as the rates, and the lesson for readers of the statute is that effective dates are policy.

The Investment Tax Credit Repeal

The investment tax credit, which had allowed businesses to claim a credit against tax liability for a percentage of the cost of qualifying equipment, was repealed by the 1986 act, generally for property placed in service after 1985, with transition relief for property subject to binding contracts and certain other commitments. The credit had been the centerpiece of business investment incentives since its introduction, and its repeal was among the most consequential business provisions in the statute.

The credit’s repeal must be understood alongside the depreciation changes discussed below, because the two provisions were functional substitutes. Under prior law, businesses received both accelerated depreciation and an investment credit on the same equipment, a combination that could drive the effective tax rate on new investment to zero or below. The drafters viewed this stacking as the paradigmatic distortion: it subsidized equipment investment relative to other uses of capital, encouraged tax-motivated transactions whose economics depended on the subsidy, and narrowed the corporate base so severely that the statutory rate had to remain high to raise the required revenue. Repealing the credit while lengthening depreciation was the corporate analogue of the individual rate-for-base trade: the statutory rate fell from forty-six percent to thirty-four percent, but the effective burden on new investment rose for the assets that had benefited most from the stacked incentives.

The credit’s history made the repeal easier to take. Congress had enacted the investment tax credit in 1962, suspended it, restored it, increased it, and suspended it again across the following two decades, treating it as a countercyclical dial rather than a permanent feature. Each suspension taught the business community that the credit could vanish, and each restoration taught the same community to discount its permanence. By 1986 the credit’s credibility as a lasting incentive was already damaged, which reduced the political cost of repealing it permanently. A preference that Congress has turned off before is easier to turn off for good than one it has never touched. The drafters understood this depreciation of political capital and spent it.

The leasing industry’s adjustment illustrates how repeals propagate through the financial system. Because the credit had been valuable mainly to profitable firms with tax liability to offset, an entire industry had grown up to transfer the credit’s value: profitable lessors bought equipment, claimed the credit, and passed part of the value to lessees through lower rents. The repeal destroyed the economics of those transactions overnight, stranding structures built around a preference that no longer existed. Transition relief for binding contracts softened the immediate damage, but the industry’s reason for being was gone. The episode is a standing warning about building business models on temporary tax preferences.

The constituency effects were concentrated in capital-intensive industries, equipment manufacturers, and the leasing industry that had grown up around the credit’s transferability. The repeal’s revenue contribution was large, one of the principal sources of the corporate base broadening that made the act a net corporate tax increase. Like the capital gains repeal, the credit’s elimination tested the closed-system discipline against organized opposition, and like the capital gains repeal, it survived because the arithmetic of the constraints admitted no alternative that could hold the coalition.

IRA Deductibility Narrowed, Not Eliminated

The act’s treatment of individual retirement accounts is another provision routinely misdescribed, and the correction matters. The statute did not eliminate IRA deductibility. It limited the deduction for workers who were active participants in employer-sponsored retirement plans, phasing the deduction out over adjusted gross income ranges of twenty-five thousand dollars to thirty-five thousand dollars for single filers and forty thousand dollars to fifty thousand dollars for joint filers. Workers not covered by employer plans retained full deductibility regardless of income, and all workers retained the ability to make nondeductible contributions whose earnings accrued tax-deferred.

The limitation reflected a judgment about the provision’s efficiency as a savings incentive. Under prior law, the universal deduction subsidized contributions by high-income filers who would likely have saved in any case, while providing its largest per-dollar subsidy to those in the highest brackets. By targeting the deduction to workers without employer-plan coverage, the act concentrated the incentive where the drafters believed it would generate new saving rather than merely reward saving that would have occurred anyway. Whether that judgment was empirically correct is debated, but the structure of the limitation reveals the design philosophy: preferences were to be pruned to their most defensible core, and the revenue saved was to fund lower rates for everyone.

The income thresholds were not indexed in the original statute, which meant that inflation would gradually expand the population subject to the limitation, a slow-motion base broadening built into the provision’s design. This non-indexation contrasts with the indexation of the standard deduction, personal exemption, and earned income credit, and the contrast was deliberate: provisions that served distributional neutrality were protected against inflation, while provisions that limited preferences were allowed to grow more restrictive in real terms. The asymmetry is a small but telling illustration of how thoroughly the distributional constraint permeated the drafting.

Depreciation Lengthened: MACRS Replaces ACRS

The Accelerated Cost Recovery System, enacted in 1981, had allowed businesses to depreciate assets over recovery periods far shorter than their economic lives, with generous front-loading that produced large early deductions. The 1986 act replaced ACRS with the Modified Accelerated Cost Recovery System, which lengthened recovery periods and slowed the pattern of deductions. Nonresidential real property, for example, moved to a thirty-one and one-half year straight-line recovery period, a dramatic lengthening from the nineteen-year period available under prior law. Residential rental property moved to 27.5 years, and equipment categories were similarly stretched, though less extremely, and the overall effect was to defer deductions into later years, raising the present value of tax liability on new investment.

The depreciation change was the largest single source of corporate base broadening in the act, and it interacted with the investment credit repeal to transform the taxation of capital investment. Under the old regime of short lives plus the credit, the code strongly favored tangible investment over other uses of funds; under MACRS without the credit, the bias was greatly reduced, though the system remained accelerated relative to true economic depreciation. The drafters accepted that some acceleration was administratively necessary, since measuring true economic depreciation asset by asset would have been unworkable, but they moved the system substantially toward neutrality between asset types and between investment and other expenditures. Because deductions taken later are worth less in present value than deductions taken sooner, the change raised the effective tax burden on new investment, and the revenue effect was large and back-loaded, pushing deductions into later years and raising taxable income in the early years of the budget window. That timing suited the drafters: the corporate rate cut needed early-year offsets, and slower depreciation delivered them.

Real estate felt the change most acutely, because real property had enjoyed the shortest recovery periods relative to economic life and therefore the largest acceleration benefit. The lengthening of real property lives, combined with the passive activity loss rules discussed next, ended the economics of the tax-shelter syndications that had channeled billions of dollars into real estate partnerships whose primary product was tax losses. The depreciation title thus did double duty: it raised revenue directly through deferred deductions, and it destroyed the raw material, artificial early losses, from which the shelter industry manufactured its product.

The Passive Loss Rules and the End of the Shelter Industry

New section 469, the passive activity loss rules, was the provision that ended the mass-market tax shelter industry, and it deserves the detailed treatment its consequences warrant. Under prior law, investors could use losses from tax shelter partnerships, most commonly real estate syndications, oil and gas deals, and equipment leasing arrangements, to offset wage and portfolio income, sheltering unrelated earnings from tax. The shelters were marketed nationally, often through brokerage firms, and they represented the most visible abuse of the base-narrowing provisions the act was designed to eliminate.

How did the passive loss rules end the shelter industry?

Section 469 provided that losses from passive activities, generally businesses in which the taxpayer did not materially participate, plus all rental activities, could offset only passive income. By quarantining shelter losses inside the passive basket, the statute destroyed their value to wage earners, since a loss usable only against passive income was worthless to an investor earning salaries and dividends.

Suspended losses that could not be used in the current year were carried forward to offset future passive income, or allowed in full when the taxpayer disposed of the entire interest in the activity.

Section 469 drew a categorical line that prior anti-abuse doctrines had failed to hold. Before 1986, the Internal Revenue Service and the courts attacked shelters transaction by transaction, challenging economic substance and profit motive case by case, and the shelter promoters stayed ahead by designing new structures faster than the government could litigate them. The passive loss rules abandoned the transactional approach in favor of a definitional one: instead of asking whether a particular deal had substance, the statute asked whether the taxpayer materially participated, and if not, the losses were quarantined regardless of the deal’s economics. This categorical design is what made the provision effective where decades of litigation had failed, and it is why the shelter industry collapsed rather than adapted.

The statute provided transition relief under which a declining percentage of pre-enactment passive losses remained deductible against non-passive income: sixty-five percent for 1987, forty percent for 1988, twenty percent for 1989, and ten percent for 1990, with the quarantine fully effective thereafter. It also provided a twenty-five thousand dollar exception for rental real estate activities in which the taxpayer actively participated, allowing that amount of rental losses to offset non-passive income. The exception preserved a space for the small landlord, the individual who owned and managed a few rental properties, while denying shelter benefits to the passive limited partners who had been the shelter industry’s customers. The 25,000 dollar offset for actively participating landlords was not derived from any theory of optimal taxation. It was the price of the votes of members whose districts contained small landlords, a constituency sympathetic enough to protect and small enough to afford. The distinction between active participation and material participation, fine but consequential, became one of the most interpreted concepts in the regulations that followed.

Administering section 469 required the Treasury to define the boundaries of the new world, and the regulations became a small library. Material participation was defined through hours tests, facts-and-circumstances tests, and lookback rules. Activities had to be grouped and regrouped under rules that determined which losses could offset which income. Each rule was a reasonable answer to a real question, and together they formed a compliance burden that fell on exactly the taxpayers the provision was meant to spare: the small business owner with a side activity, the landlord with two properties, the professional with a part-time venture. The shelter promoters the provision targeted simply left the business. The complexity stayed behind for everyone else.

The economic consequences extended well beyond the tax bar. Real estate markets that had been propped up by tax-motivated investment, particularly in overbuilt Sun Belt markets, lost a major source of demand, and the resulting distress contributed to the savings and loan difficulties of the late 1980s, a connection explored in the companion article on what the act produced. The provision thus illustrates a general lesson of base broadening: preferences do not merely reduce revenue; they allocate capital, and removing them reallocates it, sometimes painfully. The drafters understood this, which is why the transition percentages declined gradually rather than dropping to zero immediately, but the reallocation was the point, not an unintended side effect.

The Second Tier of Individual Repeals

The famous individual base broadeners, consumer interest, the sales tax deduction, the capital gains exclusion, the investment credit, and the passive loss rules, dominate every summary, but the act’s individual title contained a second tier of repeals and curtailments that collectively raised substantial revenue and illustrated the drafters’ thoroughness. Each of these provisions was small beside the giants, yet each followed the same logic: identify income that prior law had sheltered or a deduction that subsidized a favored activity, remove it, and contribute the revenue to the rate pool. The second tier shows the closed system operating at retail, provision by provision, where the political cost of each repeal was weighed individually.

Income averaging was repealed outright. Under prior law, taxpayers whose income spiked in a single year could average the spike over several years, reducing the tax on fluctuating earnings. The provision had been defended as relief for taxpayers with volatile incomes, such as authors, athletes, and commissioned salespeople, but it was complex to compute and its benefits flowed disproportionately to high-income filers experiencing one-time gains. With the rate schedule collapsing to two brackets, the drafters judged that the remaining graduation no longer justified a separate averaging mechanism, and the repeal’s revenue, modest beside the giants but real, entered the pool. The repeal also simplified the return for the minority of filers who had used the averaging schedule, a rare case where base broadening and simplification pointed in the same direction.

The two-earner deduction was repealed. Prior law had allowed married couples with two earners to deduct a percentage of the lower-earning spouse’s wages, up to ten percent, as rough justice for the marriage penalty embedded in the joint rate schedule. The deduction’s repeal was paired with the rate collapse: with only two brackets and a wide fifteen percent band, the marriage penalty shrank substantially, reducing the need for the offsetting deduction. The repeal raised revenue from two-earner couples while the lower rates reduced their liability, and the net effect for most such couples was a reduction, which is why the provision’s elimination provoked less resistance than its revenue contribution might suggest. The episode illustrates how rate reduction and base broadening could offset each other within a single filer population, leaving the distribution approximately unchanged while simplifying the computation.

Unemployment compensation became fully taxable. Prior law had excluded a portion of unemployment benefits from income, phasing the exclusion out as income rose. The act removed the exclusion entirely, treating unemployment benefits like wages for tax purposes. The rationale was horizontal equity: two filers with the same total income should face the same tax regardless of whether the income arrived as wages or benefits. The change fell on unemployed workers, a constituency with little lobbying presence, which made it enactable within the closed system even though it would have been difficult to defend as a standalone measure. The provision is a reminder that base broadening’s burden did not fall exclusively on the affluent; it fell wherever the drafters found sheltered income, and the distributional constraint operated on the aggregate, not on each provision.

Scholarship and fellowship exclusions were narrowed. Under prior law, scholarships and fellowships were broadly excludable; the act limited the exclusion to degree candidates and to amounts applied to tuition, fees, books, and required supplies, making stipends for room, board, and living expenses taxable and ending the exclusion for non-degree recipients. Graduate teaching and research assistants, whose stipends had often been treated as excludable fellowships, found a portion of their support newly taxable. The provision raised revenue from a population, students and universities, that rarely appears in tax-reform narratives, and it demonstrated the drafters’ willingness to follow the comprehensive-income principle into politically sympathetic territory.

Taken together, the second tier of individual broadeners shows the closed system at its most methodical. None of these repeals could have funded a rate cut on its own; collectively, they contributed billions to the pool. None was the subject of a national debate; collectively, they touched millions of filers. The drafters’ willingness to spend political capital on small, unglamorous repeals is one of the strongest pieces of evidence that the constraints were genuinely binding: a bill that needed only the appearance of base broadening would have stopped with the giants, but a bill that needed the revenue had to comb the code to its corners.

Closing the Shifting Channels: The Kiddie Tax and Trust Rules

Base broadening addresses income that the code excludes; anti-shifting rules address income that taxpayers move. A tax system with graduated rates invites filers to shift income to family members in lower brackets, and prior law had tolerated several shifting channels that the 1986 act closed. These provisions raised less revenue than the headline repeals, but they were essential to the integrity of the new rate structure: a twenty-eight percent top rate means little if high-bracket filers can freely reassign their income to children taxed at fifteen percent. The shifting rules therefore functioned as the rate schedule’s enforcement arm.

The kiddie tax was the most prominent of these provisions. New rules provided that the unearned income of a child under fourteen, above a modest statutory threshold, would be taxed at the parents’ marginal rate rather than at the child’s own lower rate. Before the act, parents had shifted investment income to children through custodial accounts and outright gifts, so that dividends and interest that would have been taxed at the parents’ fifty percent rate were instead taxed at the child’s minimal rate, sometimes zero. The kiddie tax defeated the shifting by applying the parents’ rate to the child’s unearned income above the threshold, while leaving the child’s earned income, such as wages from a summer job, taxed at the child’s own rate. The age cutoff reflected a judgment about control: young children were presumed to be conduits for parental shifting, while older teenagers were treated as more independent economic actors.

The provision’s design reveals the drafters’ attention to behavioral detail. By taxing only unearned income at the parents’ rate, the statute preserved the incentive for children to work, since wages remained taxed at the child’s rate. By setting the threshold at a modest level, it exempted the small custodial accounts of ordinary families while capturing the large shifting arrangements of affluent ones. And by applying the parents’ marginal rate rather than a flat penalty rate, it integrated the child’s income into the family’s overall tax picture, which was the economically coherent result. The kiddie tax has been amended many times since, with the age threshold and the mechanics revised in later legislation, but the 1986 architecture, unearned income above a threshold taxed at the parents’ rate, established the template.

The grantor trust rules were tightened in parallel. Prior law had allowed grantors to create short-term trusts, the well-known Clifford trust arrangement, that shifted income to lower-bracket beneficiaries while the grantor retained a reversionary interest taking effect after a specified term. The act curtailed these arrangements by tightening the conditions under which a grantor would not be treated as the owner of trust income, making it harder to combine a retained reversion with shifted taxation. The changes were technical, turning on the definitions of reversionary interests, powers to control beneficial enjoyment, and administrative powers exercisable for the grantor’s benefit, but their purpose was straightforward: to ensure that income could not be assigned to a lower bracket while the economic substance of ownership remained with the higher-bracket grantor.

These anti-shifting provisions interacted with the rate collapse in a way that amplified their importance. Under the old fourteen-bracket system, shifting income from a fifty percent bracket to a fifteen percent bracket saved thirty-five cents on the dollar, an enormous incentive that the old code policed imperfectly. Under the new two-rate system, the maximum saving from shifting fell to thirteen cents on the dollar, the difference between twenty-eight and fifteen percent, which reduced the incentive substantially. The kiddie tax and trust tightening thus operated in an environment where the underlying rate structure already discouraged shifting; the provisions closed the remaining channels rather than fighting the main battle. This sequencing, first reduce the rate differential, then close the shifting channels, was the coherent order of operations, and it is one reason the provisions succeeded where earlier piecemeal anti-abuse efforts had failed.

The shifting rules also illustrate a general principle of the act’s design: every rate cut required a corresponding integrity measure, because lower rates are only as real as the base they apply to. A twenty-eight percent top rate that high-bracket filers could avoid through family shifting would have been a fiction, and the distributional neutrality constraint would have failed as shifted income escaped the upper brackets. The kiddie tax and the trust rules were therefore not peripheral technical amendments but load-bearing elements of the closed system, small in revenue and large in structural necessity.

The Earned Income Tax Credit Expansion

The earned income tax credit received its largest expansion to that point in the 1986 act, and the expansion served the same distributional function as the zero bracket provisions. The credit rate was set at fourteen percent of the first five thousand seven hundred fourteen dollars of earned income, producing a maximum credit of eight hundred dollars, with the credit phasing out at nine thousand dollars of income, and the parameters were indexed for inflation. For low-income working filers, the credit offset not only income tax liability but, because it was refundable, could produce a net payment, making it the most targeted antipoverty instrument in the individual income tax.

The credit’s expansion should be understood as part of the distributional bargain rather than as an isolated antipoverty initiative. Base broadening eliminated preferences that had disproportionately benefited middle and upper-income filers, but some base-broadening provisions reached down the income scale as well. The earned income tax credit expansion, combined with the larger standard deduction and personal exemption, ensured that the net effect at the bottom was a reduction in burden, which the distributional constraint required. The credit expansion was the work-contingent complement to the bigger standard deduction and personal exemption: households lifted off the income tax rolls by the higher filing thresholds still owed payroll taxes, and the credit offset that burden while supplementing earnings. Distributionally, the credit gave the bottom of the income scale a benefit calibrated to match the rate cuts at the top, which is why the two traveled together through the legislative process. Readers seeking the full legislative history of the credit, including its later expansions, should consult the dedicated guide to the earned income tax credit law, which traces the provision from its 1975 origins through subsequent amendments; this article treats the 1986 expansion as one element of the closed-system trade.

The interaction between the credit and the new rate schedule illustrates how the act’s pieces fit. A worker whose earnings rose into the 15 percent bracket faced the phase-out of the credit at the same time as the positive income tax rate, which created effective marginal rates higher than the statutory 15 percent suggested. The drafters were aware of the stacking effect and accepted it as the cost of targeting: a credit that phases out must phase out somewhere, and the phase-out range will always overlap some positive tax rate. The 1986 parameters set the overlap where the scoring and the distributional tables said it belonged. The 1986 expansion established the pattern of using the credit as the distributional counterweight to rate cuts, a pattern that shaped every subsequent negotiation over the provision.

Indexation of the credit was as consequential as the rate increase. Before 1986, the credit’s parameters were fixed in nominal dollars, so inflation steadily reduced its real value and narrowed the population it reached. By indexing the credit, the act ensured that the antipoverty investment would not decay, and it established the pattern, followed in later expansions, of treating the credit as a permanent, inflation-protected feature of the code rather than a temporary supplement. The 1986 expansion is therefore a hinge point in the credit’s history: the moment it became large enough and durable enough to function as a structural component of the income tax rather than an appendage.

Transition Rules as a System

The transition rules of the 1986 act are often treated as technical underbrush, the dense effective-date provisions that practitioners wade through and historians skip. That treatment misses their structural role. The transitions were the mechanism by which a bill that rewrote the taxation of investment, shelters, and capital gains in a single enactment avoided imposing the entire adjustment in a single year. Nearly every major base-broadening provision carried transition relief, and the relief followed consistent patterns: phaseout percentages that declined across the window, grandfathering for binding commitments entered before specified dates, and interim schedules that stepped rates down gradually. Taken together, the transitions formed a system, and the system reveals as much about the legislative bargain as the substantive provisions do.

The phaseout pattern is the most visible element. The consumer interest deduction declined in steps across tax years 1987 through 1990 before disappearing for 1991. The passive loss transition allowed sixty-five percent of pre-enactment losses against non-passive income in 1987, forty percent in 1988, twenty percent in 1989, and ten percent in 1990, with the full quarantine effective thereafter. The investment credit repeal and the depreciation changes carried their own graduated effective dates. The common structure, declining percentages over the same four-year span, was not coincidental. It aligned the transition with the five-year budget window, so that the revenue gain from each provision ramped up across the scored period, and it gave affected taxpayers an identical adjustment horizon across provisions, which simplified planning even as it complicated the statute.

Grandfathering for binding commitments was the second pattern. Taxpayers who had entered binding contracts, begun construction, or placed orders before the act’s key dates could generally complete those transactions under the old rules, a principle applied across the investment credit, depreciation, and several other titles. The rationale combined fairness with practicality: retroactive changes to the tax treatment of committed investments would have punished reliance on prior law and invited constitutional challenge, while prospective application preserved the incentive effects the drafters sought. The grandfather provisions were among the most lobbied sections of the bill, because the precise definition of a binding commitment determined which projects kept the old subsidies, and the conference agreement’s transition rules ran to many pages of project-specific and category-specific relief.

The interim rate schedules were the third pattern. Individuals faced the graduated 1987 schedule before the two-rate structure took effect in 1988, and corporations with straddling tax years applied blended rates for 1987 before the thirty-four percent rate applied in full. The interim schedules served the scoring function as well as the adjustment function: by stepping rates down rather than dropping them at once, they moved a portion of the rate-cut revenue loss out of the first year of the window and into later years, which eased the first-year balance of the closed system. A bill scored over fiscal years 1987 through 1991 could afford deeper ultimate rate cuts if the first year’s cut was shallower, and the drafters used that arithmetic deliberately.

The transition system had costs that belong in any honest accounting. It multiplied the statute’s length, since every transition rule required definitions, dates, exceptions, and exceptions to exceptions. It created the year-end behavioral distortions, such as the 1986 rush to realize capital gains, that complicated later evaluation of the provisions’ steady-state effects. And it gave sophisticated taxpayers and their advisers a multi-year planning horizon in which to restructure affairs around the phaseout schedules, which meant some of the base broadening’s yield leaked through anticipatory planning. The drafters accepted these costs because the alternative, immediate full effectiveness, would have concentrated the economic disruption and the political opposition into a single moment, likely a fatal one.

There is a final, less noticed function of the transitions: they made the bill’s promises credible across time. A rate cut enacted without transition relief invites the suspicion that the base broadening will be repealed before it bites, leaving only the rate cut; phased base broadening with statutory percentages locked in for each year made the future pain concrete and therefore made the present rate cut believable. Within the 1986 bargain, the transition system was the enforcement mechanism that bound the future to the present deal, and its density is the price the statute paid for credibility.

The Trade Table

The act’s ledger can be read as a table. Each row names a preference that was eliminated or curtailed, describes the revenue role it played in the bargain, names the rate reduction it funded, and identifies the constituency that lost it. The revenue descriptions below are qualitative, drawn from the Joint Committee on Taxation’s scoring as summarized in the Blue Book for the act, JCS-10-87, except where a specific aggregate is stated and attributed. The point of the table is not the precision of any single figure. It is the shape of the exchange: no rate cut in the right-hand column exists without a repeal in the left-hand column, and no repeal was enacted without a constituency attached to it.

Preference eliminated or curtailed	Revenue role in the bargain	Rate reduction it funded	Constituency that lost it
Deduction for consumer interest, phased out over 1987 through 1990 and unavailable after 1990	Moderate offset within the individual base	Helped fund the individual schedule collapse to 15 and 28 percent	Borrowers who financed cars, credit cards, and other personal debt with deductible interest; lenders who marketed deductible borrowing
Deduction for state and local sales taxes, disallowed while income and property tax deductions were retained	Moderate offset within the individual base	Helped fund the individual schedule collapse to 15 and 28 percent	Itemizers in states without an income tax who had deducted sales taxes; the provision hit hardest where the sales tax was the main state levy
Exclusion of 60 percent of long-term capital gains, repealed so gains were taxed as ordinary income beginning in 1987	Major offset within the individual base	Helped fund the drop in the top individual rate from 50 to 28 percent	Investors, business sellers, and venture financiers who had timed realizations around the exclusion
Investment tax credit, repealed outright	Major offset, concentrated in the corporate base	Helped fund the corporate rate cut from 46 to 34 percent and the individual cuts	Equipment purchasers, manufacturers, and the leasing industry built around the credit
IRA deductibility for workers covered by employer retirement plans, limited with phase-outs of 25,000 to 35,000 dollars for single filers and 40,000 to 50,000 dollars for joint filers	Modest offset within the individual base	Helped fund the individual schedule collapse to 15 and 28 percent	Middle-income savers with pension coverage who had deducted contributions without income limits
Accelerated depreciation under ACRS, replaced by the longer-lived MACRS schedules	Major offset, concentrated in the corporate base	Helped fund the corporate rate cut from 46 to 34 percent	Capital-intensive industries and commercial real estate owners who had relied on fast write-offs
Passive activity loss offsets, newly limited by section 469 with transition percentages of 65, 40, 20, and 10 for 1987 through 1990 and a 25,000 dollar rental real estate exception	Major offset within the individual base	Helped fund the drop in the top individual rate from 50 to 28 percent	Shelter promoters and high-income limited partners in partnerships marketed for their losses
Corporate preferences in the aggregate, measured by the Joint Committee on Taxation at roughly 120 billion dollars in additional corporate liability over fiscal years 1987 through 1991	The central corporate offset, per JCS-10-87	Funded the individual rate cuts, with individual liability falling by a similar aggregate amount	Corporations as a class, which received a lower statutory rate and a larger aggregate bill

A note on magnitudes is in order, because readers will want numbers the table deliberately withholds. The Joint Committee on Taxation and the Treasury scored each provision as the bill moved, and the scores shifted with every amendment, every effective-date change, and every behavioral assumption. Any single figure quoted for a provision’s revenue role is therefore a snapshot of a moving target. The qualitative labels in the table, major, moderate, and modest, reflect the provisions’ relative standing in the scoring across the drafts that mattered, and they are stable in a way the point estimates are not. The one aggregate the record states with confidence is the corporate transfer: roughly 120 billion dollars in additional corporate liability over fiscal years 1987 through 1991, matched by a similar reduction on the individual side, per the Blue Book, JCS-10-87. That figure survived because it was the headline the coalition needed. The provision-level figures were working papers, and this guide treats them as such.

Read the rows together and the design constraint becomes visible as a pattern. The individual offsets cluster around the preferences of affluent households: the gains exclusion, the shelter losses, the unlimited IRA deduction, the deductible consumer borrowing. The corporate offsets cluster around the preferences of capital-intensive business: the credit, the fast depreciation, the loss rules. Each cluster funded the rate cut applied to the same cluster, which is how the drafters satisfied distributional neutrality while cutting rates. The table is the act in miniature, and every section that follows expands one or more of its rows.

Reading the table as a whole also reveals what the drafters chose not to trade. The mortgage interest deduction, the charitable contribution deduction, and the state and local income tax deduction do not appear as rows, because no majority could be assembled to name them as offsets. Their absence is as informative as the rows’ presence. The ledger contained only the trades that could clear, and the preferences that survived did so not because they were judged virtuous but because their constituencies were too large, too organized, or too geographically dispersed to take on inside a bill that was already asking so much. A complete account of the act must include the dog that did not bark: the base that remained untaxed after the broadest base-broadening in postwar history was narrower even then than a clean theoretical base, because politics, not theory, set the boundaries.

The Corporate Side: A Rate Cut That Raised Revenue

The corporate title of the act is the provision that modern summaries most often garble, so it requires a full exposition. The statutory corporate rate fell from forty-six percent to thirty-four percent, effective for tax years beginning in 1988, with a transitional blended schedule applying to tax years straddling the change in 1987. The new rate applied through a three-bracket structure running from fifteen percent to thirty-four percent, with an additional five percent surtax on corporate income over one hundred thousand dollars that phased out the benefit of the lower brackets, creating an implicit thirty-nine percent marginal band over the phase-out range. The structure paralleled the individual bubble in miniature: nominal rates that understated the true marginal rate over a defined band, built to preserve the distribution of the burden while lowering the headline number. For 1988, the first 50,000 dollars of corporate income was taxed at 15 percent, the next 25,000 at 25 percent, and income above 75,000 at 34 percent, with the surtax recapturing the low-bracket benefit above 100,000 dollars until the marginal rate returned to 34 percent.

The rate fell twelve percentage points, but the base expanded by more than enough to offset the cut. Repeal of the investment credit, longer MACRS depreciation, the new twenty percent corporate alternative minimum tax, and curtailed business preferences added far more to taxable corporate income than the lower rate subtracted from liability. The Joint Committee on Taxation Blue Book, designated JCS-10-87, estimated that the act increased aggregate corporate tax liability by approximately one hundred twenty billion dollars over fiscal years 1987 to 1991, while reducing aggregate individual liability by a roughly similar amount. The corporate title was therefore a net tax increase, scored over the standard five-year window, and the individual title was a roughly offsetting net decrease.

The one hundred twenty billion dollar figure deserves emphasis because it contradicts the reflex that associates the 1986 act with business tax relief. Corporate leaders who supported the bill did so for the lower statutory rate and for the improved neutrality of the base, not because their aggregate liability fell; in the aggregate, it rose. The increase was not uniform across firms. Corporations that had made heavy use of the investment credit, accelerated depreciation, and shelter structures saw large increases, while corporations with few preferences and income taxed at the full statutory rate saw decreases. Capital-intensive manufacturers that had combined the investment tax credit with ACRS depreciation generally paid more. Commercial real estate owners, who had relied on fast depreciation of structures, faced some of the sharpest increases. By contrast, service businesses, retailers, and financial firms with modest depreciable investment and no credit usage generally paid less: their base barely moved while their rate fell twelve points. The distributional constraint applied within the corporate sector as well as across the individual distribution, but the aggregate result, a shift of roughly one hundred twenty billion dollars of five-year liability from individuals to corporations, was the single largest intersectoral movement in the bill.

The political economy of this shift is one of the act’s most instructive features. Conventional wisdom holds that business interests block corporate tax increases, yet the corporate increase passed as part of a bill that business groups broadly supported. The closed system explains the paradox. Because the individual rate cuts had to be funded from somewhere and the distributional constraint limited how much could be taken from individuals, the corporate base was the available reservoir, and corporate supporters accepted the increase as the price of the lower statutory rate and the cleaner base. The episode demonstrates that organized interests evaluate tax legislation as a package rather than provision by provision, and that a sufficiently attractive rate can purchase acquiescence to base broadening that would be unthinkable on its own. There was no unified business interest to assault the bill because there was no single business experience of it: each firm ran its own numbers, and enough firms came out ahead, or close enough to ahead, that the coalition of the harmed could not assemble a majority.

The mechanics of the transfer repay close attention because they show the two titles of the act functioning as one system. The corporate base broadeners, credit repeal, slower depreciation, tighter loss rules, and the corporate AMT, raised corporate receipts well above what the 34 percent rate alone would have produced. The individual provisions, rate cuts, bigger exemptions, the EITC expansion, cost individual receipts. The Blue Book netted the two: corporations paid roughly 120 billion dollars more over fiscal years 1987 through 1991 than prior law would have collected, and individuals paid roughly 120 billion less. The transfer was not a policy goal stated in the statute. It was the arithmetic result of pricing each title’s provisions separately under the single constraint that the total stay neutral. The corporate title overfunded its rate cut. The individual title underfunded its. The surplus crossed the aisle between the titles and balanced the books. The 1986 act is the only postwar tax bill in which business paid for the individual’s rate cut, and the uniqueness of that fact measures the uniqueness of the constraint that produced it.

The corporate rate cut also changed the calculus of organizational form. Before the act, the 50 percent individual top rate stood above the 46 percent corporate rate, which encouraged some businesses to retain earnings inside corporations. After the act, the 28 percent individual top rate stood well below the 34 percent corporate rate, reversing the incentive and encouraging pass-through forms such as partnerships and S corporations. The drafters anticipated the shift and accepted it as a consequence of the individual-first design: the individual rate cut was the political centerpiece, and the corporate rate was set where the corporate ledger balanced, not where organizational neutrality would have placed it. The resulting migration toward pass-throughs became one of the act’s lasting structural effects.

The Corporate Base Broadening, Provision by Provision

The corporate rate cut from forty-six percent to thirty-four percent dominates summaries of the business title, but the base broadening that paid for it extended well beyond the investment credit repeal and MACRS depreciation already discussed. Three additional provisions illustrate the drafters’ method: each targeted a timing preference that allowed corporations to defer income or accelerate deductions, each raised revenue by moving the tax base closer to economic income, and each hit a specific industry practice that had grown up around the preference. Together they show that the corporate base broadening was not a handful of headline repeals but a systematic attack on deferral.

The uniform capitalization rules of new section 263A required businesses to capitalize certain indirect costs, including purchasing, handling, and administrative expenses allocable to production, into inventory rather than deducting them in the current period. Under prior law, many manufacturers and retailers had expensed these costs immediately while deferring the income from the goods those costs produced, a mismatch that understated current-year income. Section 263A forced the costs to travel with the goods, matching deductions to the income they helped generate. The provision raised revenue by accelerating income recognition in present-value terms, and it fell most heavily on manufacturers, wholesalers, and retailers with significant indirect production costs. Its complexity was notorious from the start: allocating indirect costs across product lines required accounting systems many mid-sized businesses did not have, and the regulations interpreting the section became a compliance industry of their own.

The long-term contract provisions limited the completed-contract method of accounting, under which contractors had deferred all income from a project until its completion, sometimes years after the work was performed. The act generally required contractors to use the percentage-of-completion method, recognizing income as work progressed, with an exception for certain small contractors and home construction contracts. The completed-contract method had allowed large contractors to bunch income into completion years and to time completions for tax advantage; the percentage-of-completion requirement smoothed recognition and raised revenue by pulling income forward. The construction industry, particularly large commercial contractors, bore the change, and the small-contractor exception preserved the old method where compliance costs would have outweighed the revenue.

The bad-debt reserve repeal ended the use of the reserve method for deducting bad debts for most taxpayers, requiring instead the specific charge-off method under which a deduction is allowed only when a particular debt becomes worthless. Under the reserve method, businesses had deducted additions to a general reserve for anticipated future defaults, effectively accelerating deductions for losses that had not yet occurred and might never occur. The repeal, with an exception for certain small banks, moved the deduction to the year of actual worthlessness, raising revenue in present-value terms by deferring the tax benefit. Lenders and businesses extending significant trade credit felt the change most directly.

These three provisions share a design philosophy worth naming: the drafters treated timing as base. A deduction taken sooner is worth more than the same deduction taken in five years, so accelerating deductions or deferring income narrows the base in present-value terms just as surely as an outright exclusion does. The corporate title’s attack on timing preferences, through uniform capitalization, percentage-of-completion accounting, and charge-off-only bad debts, therefore belonged to the same base-broadening project as the headline repeals. It also contributed to the net corporate tax increase: timing changes raise revenue within any finite budget window even when they are neutral over an infinite horizon, because the window captures the deferral’s reversal only partially. Over fiscal years 1987 through 1991, the pull-forward of income from these provisions added billions to the corporate total that the rate cut could not offset.

International provisions, though less celebrated than the domestic rate cuts, completed the corporate title. The act tightened the foreign tax credit, replacing the single overall limitation with separate limitation baskets for different categories of income, preventing taxpayers from averaging high foreign taxes on one type of income against low foreign taxes on another to maximize the credit. It tightened the rules for allocating expenses between domestic and foreign income and rewrote the provisions governing controlled foreign corporations. The separate baskets were the foreign analogue of the passive loss quarantine: just as section 469 prevented the averaging of shelter losses against wage income domestically, the baskets prevented the averaging of foreign tax rates across income categories internationally. These provisions raised revenue, protected the domestic base broadeners from avoidance, and added their own layer of complexity to multinational compliance.

The Alternative Minimum Tax, Individual and Corporate

The alternative minimum tax provisions illustrate the act’s characteristic pattern of broadening while cutting. On the individual side, the AMT rate rose from twenty percent to twenty-one percent for 1987, and the base of the tax was broadened by adding preference items and adjustments that the regular tax had eliminated or that the act newly identified. The individual AMT functioned as a backstop: filers computed liability under both the regular schedule and the AMT and paid whichever was higher, so that the elimination of preferences in the regular tax was reinforced by a parallel minimum tax that caught what the regular tax missed.

The individual AMT’s expansion had a quieter but broader effect. Under prior law, the individual AMT had touched mainly aggressive shelter users, a small population that the provision was designed to catch. The 1986 act’s broader base pulled in a larger population, including upper-middle-income filers in high-tax states whose combination of preferences tripped the parallel computation. These taxpayers had never considered themselves the targets of a minimum tax, and their first encounters with the AMT’s worksheets were a significant source of the compliance burden the act created. The drafters had a choice: narrow the AMT and lose revenue, or broaden it and complicate millions of returns. The ledger required the revenue, so they broadened it. The decision illustrates the hierarchy of the act’s values. Revenue neutrality outranked simplicity wherever the two conflicted, and they conflicted often.

On the corporate side, the act repealed the existing add-on minimum tax and replaced it with a broad-based corporate alternative minimum tax imposed at twenty percent. The new corporate AMT was not an add-on to the regular tax but a parallel system: a corporation computed its liability under both systems and paid the higher. For tax years 1987 through 1989, the corporate AMT included a book-income adjustment, meaning that a portion of the difference between a corporation’s financial statement income and its taxable income was added back in computing minimum tax liability. The book-income adjustment was among the most controversial provisions in the corporate title, because it effectively taxed income as reported to shareholders, importing financial accounting concepts into the tax base. By tying a portion of the tax base to income reported on financial statements, the adjustment created pressure to manage book income with tax consequences in mind, and it drew criticism for punishing honest reporting and blurring the line between two systems meant to serve different purposes. Defenders replied that book income was the hardest base to manipulate for AMT purposes and that the adjustment was temporary, a bridge to keep the corporate revenue gain intact while the permanent AMT rules took hold. Its inclusion reflected the drafters’ determination that the minimum tax have real bite: without it, corporations could have continued to report robust profits to investors while showing minimal taxable income, which was precisely the outcome the base-broadening project was designed to end. The corporate AMT was one of the major corporate base broadeners, contributing billions to the revenue pool that funded the rate reduction from forty-six to thirty-four percent.

The AMT provisions complicate the simplification narrative in a specific way. The regular tax was simplified along the dimension of brackets, but the AMT added a second, parallel computation that millions of filers had to perform to determine whether it applied. For any filer near the AMT threshold, the effective complexity of the code was the regular tax plus the AMT, not the regular tax alone. The expansion of the AMT’s reach, driven by the broader base and the non-indexation of its exemption in the original design, meant that over time the minimum tax threatened to become the de facto tax system for upper-middle-income filers, a development that later Congresses addressed through repeated patches. The 1986 act thus planted a complexity time bomb even as it defused others. The AMT was the enforcement arm of the ledger: a base broadener that taxpayers could avoid through remaining preferences was not a reliable offset, and only a reliable offset could buy a rate cut.

The Rest of the Statute: Titles Beyond the Headlines

The individual and corporate income tax titles dominate every account of the 1986 act, but the statute’s nearly one thousand pages in the Statutes at Large contain titles on pensions, tax-exempt bonds, insurance, and foreign income whose provisions affected millions of filers and billions of dollars. A pillar article cannot treat each exhaustively, and this one will not try. What it can do is map the outlying titles briefly, showing how each reflected the same base-broadening philosophy applied to a different corner of the code, so that readers understand the act’s full scope without mistaking this survey for a specialist treatment.

The pension title tightened the rules for tax-favored retirement saving beyond the IRA limitation already discussed. It capped the amount workers could shelter through salary reduction to 401(k) plans, strengthened the nondiscrimination rules that require retirement plans to cover rank-and-file workers and not merely highly compensated executives, extending coverage testing and tightening the definitions that had allowed plans to skew benefits upward, and imposed new restrictions on plan loans, early distributions, and the integration of private pensions with Social Security benefits. The philosophy matched the individual title’s: tax-favored saving was to be preserved but pruned to its most defensible core, with the revenue from the pruning contributed to the general rate-reduction pool. Workers who had maximized salary deferrals under the old limits faced a direct reduction in sheltered saving, while the nondiscrimination tightening extended plan coverage to workers previously excluded.

The tax-exempt bond title curtailed the use of tax-exempt financing for private purposes. Under prior law, states and municipalities had issued tax-exempt bonds whose proceeds financed private business facilities, sports stadiums, and other projects with only an attenuated public purpose, allowing private borrowers to capture the federal tax subsidy embedded in the exemption. The act imposed volume caps on private activity bonds, tightened the definitions of qualifying purposes, and restricted arbitrage practices by which issuers profited from investing bond proceeds at taxable yields. The provisions raised revenue by shrinking the universe of tax-exempt interest, which broadened the base of taxable investment income, and they redirected the subsidy toward traditional governmental purposes. Municipal issuers and the developers who had relied on private activity financing constituted the affected constituency, and the title’s complexity, with its caps, carryforwards, and purpose tests, became a lasting feature of municipal finance practice.

The insurance title rewrote the taxation of life insurance companies and insurance products. It tightened the definition of life insurance for tax purposes to prevent investment products from masquerading as insurance to capture the inside buildup exclusion, imposed new reserve computation rules on insurers, and curtailed the deduction for policyholder dividends in ways that raised the industry’s effective burden. The provisions reflected the drafters’ judgment that the insurance industry’s tax treatment had drifted far from the economics of its business, with reserves and product definitions manipulated to minimize taxable income. The industry’s opposition was intense and organized, which makes the title’s enactment another demonstration of the closed system’s binding force: the revenue was needed, the constraint admitted no substitute, and the opposition was outvoted.

These titles share the act’s signature move: identify a preference or timing advantage, curtail it, score the revenue, and spend the revenue on lower rates. None of them was the political centerpiece, and none appears in popular summaries, but their combined revenue contribution was essential to the arithmetic. A reader who understands only the individual rate collapse and the corporate rate cut understands the bill’s shape but not its weight. The outlying titles supplied much of the weight, and their obscurity is itself instructive: the closed system worked by aggregating dozens of unglamorous base broadeners into a single pool, so that no single provision had to carry the political load alone.

Distributional Neutrality and Who Paid

The distributional constraint requires its own accounting, because the statute’s claim to preserve the distribution of the burden is both central to its design and contested in its achievement. The constraint, as scored by the Joint Committee on Taxation, required that the percentage change in after-tax income be roughly equal across income classes, so that no group systematically gained or lost relative to the pre-reform baseline. The instruments of distributional neutrality were the zero bracket expansion, the earned income credit expansion, the bubble surcharge, and the calibration of base broadening to fall most heavily on preferences used disproportionately by upper-income filers.

Named analyses of the distributional outcome reach different conclusions, and the differences turn on assumptions that should be stated explicitly. The Congressional Research Service analysis designated RL34498 examined the surcharge mechanism and treated its phase-out design as evidence that distributional preservation was an operative goal of the drafting, not merely a rhetorical claim. Work by John Witte published in 1991 concluded that distributional neutrality was not fully successful in practice, finding that the realized distribution of the burden shifted in ways the static scoring had not predicted, particularly once behavioral responses and the timing effects of the transition were taken into account. Martin Feldstein’s National Bureau of Economic Research working paper 17531 examined the revenue question and found that the act was revenue neutral on a static scoring basis, meaning the Joint Committee’s conventional estimates showed no net revenue change, while behavioral responses to the new incentives partially offset the static estimates in ways that complicated the neutrality assessment.

These analyses are not contradictory so much as they are answers to different questions. Static scoring, which assumes no behavioral change, showed the act meeting its revenue and distributional targets by construction, because the drafters calibrated the provisions to the static estimates. Dynamic reality, in which taxpayers accelerated gains into 1986, restructured shelters, shifted income across the transition years, and adjusted investment to the new depreciation rules, produced outcomes that diverged from the static baseline. The divergence does not show that the constraint was meaningless; it shows that any constraint defined against a static baseline will be imperfectly realized in a world where taxpayers respond to incentives. The 1986 act is the clearest available case study of that general proposition, because its constraints were stated more explicitly than those of any comparable statute.

A further distinction clarifies why the analyses diverge: statutory incidence versus economic incidence. The distributional tables measured statutory incidence, the liability computed on tax returns by income class. Economic incidence, who ultimately bore the burden after markets adjusted, could differ. The corporate tax increase, for example, appeared in the tables as a burden on the corporate sector, but economists debate how much of any corporate tax is borne by shareholders through lower returns, by workers through lower wages, or by consumers through higher prices. The repeal of the investment credit raised the cost of capital for equipment investment, and the resulting reduction in investment could have depressed wages over time, shifting the true burden toward labor. None of the contemporary distributional analyses attempted to trace these general-equilibrium effects comprehensively; they measured the tax as written, not the economy as adjusted. Readers should therefore treat the distributional record as an account of the statute’s design and its first-order effects, not as a complete welfare analysis.

Baseline choice is the second source of divergence. Every distributional comparison requires a counterfactual, the distribution that would have obtained without the act, and the choice of counterfactual shapes the result. Comparing the act to prior law as it stood in 1986 yields one answer; comparing it to prior law as it would have evolved, with inflation, bracket creep in the unindexed provisions, and the scheduled growth of preferences, yields another. Witte’s finding that neutrality was not fully achieved in practice rests partly on baseline and timing choices that differ from the Joint Committee’s, which is why the neutral presentation reports both the design intent and the contested realization rather than declaring a winner. The statute’s ambition was explicit and measurable; its achievement was partial and debated, and both halves of that sentence belong in the record.

The neutrality rules for this article require that distributional effects be reported through named analyses with their assumptions stated, and that no rate structure be characterized as fair or unfair in the article’s own voice. The analyses above satisfy that requirement: the CRS analysis describes the surcharge’s intended function, Witte’s work reports the imperfect realization, and Feldstein’s paper addresses the static-dynamic distinction. Readers seeking to evaluate the act’s distributional record should weigh these analyses against each other with attention to their differing baselines, and should resist the temptation to treat any single estimate as the definitive verdict.

The Redesignation’s Afterlife: Citing the 1986 Code

Section 2’s redesignation of the Internal Revenue Code of 1954 as the Internal Revenue Code of 1986 is a single sentence with a long afterlife, and it repays attention because it shapes how every tax researcher encounters the statute. The redesignation did not change the substance of any section. It changed the name of the container, which meant that every citation to the federal tax laws written after October 22, 1986 pointed to a code bearing the act’s date. Court opinions, Treasury regulations, revenue rulings, and practitioner treatises all cite provisions as sections of the Internal Revenue Code of 1986, abbreviated I.R.C., followed by the section number: 26 U.S.C. 1 for the individual rate schedule, 26 U.S.C. 11 for the corporate rates, 26 U.S.C. 168 for MACRS depreciation, 26 U.S.C. 469 for the passive activity loss rules.

The persistence of the 1986 designation through decades of subsequent amendment is itself a historical fact worth noting. Later Congresses enacted major tax legislation in 1990, 1993, 1997, 2001, 2003, 2017, and other years, each amending many sections of the code, yet none saw fit to redesignate the code again. The 1986 name therefore functions as a permanent marker of the last comprehensive reorganization, a dating convention that tells the researcher the code’s architecture dates to the reform even where its details have been repeatedly revised. A student encountering a citation to the Internal Revenue Code of 1986 for a provision enacted in 2017 should understand that the date refers to the code’s naming, not the provision’s origin.

The redesignation also created a small but persistent research trap. Because the code kept its 1986 name, provisions are sometimes described as dating to 1986 when they were in fact added later, and the 1986 act is sometimes credited with provisions it merely renumbered or left untouched. Careful researchers distinguish between the code’s title and the enactment date of the specific section, using the Statutes at Large citation or the public law history to date the provision itself. The distinction matters for the passive loss rules, which genuinely date to the 1986 act, versus provisions like the 1993 rate brackets, which live in the 1986-named code but were enacted seven years later.

There is a final irony in the redesignation worth recording. The act renamed the code to mark a comprehensive revision, and the name’s persistence testifies to the revision’s architectural durability: the two-rate structure is gone, the bubble is gone, and dozens of titles have been rewritten, yet the skeleton the act built, the section numbering, the organization into subtitles, the placement of the major operative provisions, remains the framework within which all later amendments operate. The redesignation outlasted the substance it was meant to commemorate, which is either a tribute to the drafters’ structural work or a comment on later Congresses’ preference for amendment over reorganization. Either way, the 1986 in every tax citation is the statute’s most widely seen legacy, encountered far more often than any of its provisions.

The Simplification That Wasn’t

The most persistent misconception about the 1986 act is that it simplified the tax code, and the misconception persists because it contains a grain of truth surrounded by a larger falsehood. The grain of truth is the bracket collapse: fourteen rates became two nominal rates, and the rate schedule became easier to describe. The falsehood is the inference that a simpler rate schedule meant a simpler tax system. The act removed shelters and brackets while adding the passive activity loss rules with their material participation tests, an expanded individual alternative minimum tax, a new corporate alternative minimum tax with a book-income adjustment, and hundreds of pages of transition rules governing the phase-in years. Complexity migrated rather than disappeared, moving from the rate schedule into the base definition, the minimum tax, and the effective-date provisions.

Did the Tax Reform Act of 1986 simplify the tax code?

The act simplified the rate schedule while complicating nearly everything else, and the net filer experience was negative within four years. It collapsed fourteen brackets into two nominal rates, genuine simplification of one dimension, but added passive loss rules, an expanded individual AMT, a new corporate AMT, and dense transition provisions.

The evidence on perceived complexity is direct and unflattering. The Gallup Poll Monthly survey of March 1990 asked filers whether the tax system had become more or less complicated, and the share answering less complicated fell from nineteen percent to twelve percent between the pre-reform and post-reform readings, while the share answering more complicated rose from seventeen percent to thirty-one percent. Within four years of enactment, the public’s assessment of the system’s complexity had moved decisively in the wrong direction. Filing-time studies tell the same story in harder data: research by Marsha Blumenthal and Joel Slemrod published in 1992 estimated that average household filing time rose from 21.7 hours to 27.4 hours across the reform period, an increase of more than one quarter, driven by the new computations the act required.

The structure of the two-rate schedule itself contained the seeds of perceived complexity. The 33 percent bubble, the 1987 interim schedule, and the phase-outs of the IRA deduction and the rental loss exception were all conceptually simple to a drafter and operationally demanding to a filer. Each required the taxpayer to track income across thresholds, apply percentages that changed by year, and coordinate provisions that interacted. The bubble in particular punished the intuition that a two-rate system would be easy: a filer whose income crossed into the surcharge band needed to understand why the marginal rate rose and then fell, a pattern that no postcard could explain. The drafters knew the bubble was inelegant. They kept it because distributional neutrality required it, which is another instance of the design constraint overriding aesthetic preferences. The statute optimized for the constraint, not for the filer.

The professional tax preparation industry’s growth after 1986 provides a market verdict on the complexity question. If the act had simplified compliance, demand for paid preparers and tax software should have eased. Instead, the industry expanded, selling exactly the expertise the new provisions required: passive activity grouping, AMT computation, transitional schedule application, and the coordination of phase-outs. Markets do not lie about difficulty. When millions of filers pay for help with a supposedly simpler system, the system is not simpler for them. The postcard rate schedule made a good speech. The worksheets made a good living for preparers.

The Treasury’s regulatory output tells the same story from the government’s side. Implementing section 469, the corporate AMT, the revised sourcing rules, and the dozens of transition provisions required regulations that ran to thousands of pages. Each regulation answered genuine questions the statute had left open, and each added a layer of interpretation between the filer and the law. A simple statute needs few regulations. The 1986 act needed many, because its provisions were intricate even where its rates were plain. The regulation count is not a criticism of the Treasury, which did the job the statute assigned. It is a measure of the statute’s true complexity, taken at the source.

The legislative aftermath confirmed the verdict. In 1990, just four years after the act’s celebrated two-rate structure took effect, the Omnibus Budget Reconciliation Act of 1990 added a third individual bracket at thirty-one percent, breaking the two-rate schedule the 1986 act had established. Further brackets followed in later legislation, and the two-rate structure that had symbolized the reform survived only from 1988 through 1990. A simplification that lasts four years before its central achievement is repealed was not a simplification in any durable sense; it was a temporary rearrangement, and the code’s subsequent growth in length and complexity has only widened the gap between the legend and the record.

The distinction between the rate schedule and the tax base deserves one final emphasis, because it is the conceptual key to the whole simplification debate. A rate schedule is a small object: a handful of brackets and thresholds that can be printed on a single page. The tax base is an enormous object: the thousands of definitions, inclusions, exclusions, timing rules, and anti-abuse provisions that determine what number the rates apply to. The 1986 act simplified the small object dramatically and complicated the large object substantially, and filers experience the large object far more intensely than the small one. Computing which bracket applies takes seconds; determining whether a loss is passive, whether the alternative minimum tax applies, and how a phaseout schedule affects the current year takes hours. The Gallup and filing-time evidence measures the large object, which is why it moved against the reform even as the small object improved.

There is also a conceptual point worth making about what simplification could have meant. Simplification and targeting are in tension. A simple tax applies the same rules to everyone, which means its burden falls where the base lies, not where fairness would place it. A targeted tax places the burden where fairness suggests, which means it needs the rules that distinguish cases, and distinguishing cases is what complexity is. The 1986 act chose targeting, because distributional neutrality required it, and accepted the complexity that targeting entailed. Reformers who promise both radical simplification and careful targeting are promising to square this circle. The 1986 record suggests the circle does not square: the act simplified what it could, rates and shelters, and complicated what it had to, phase-outs and parallel taxes, and the net experience got harder. Anyone proposing to do better needs a theory of which complexity is dispensable, and the 1986 ledger is the best available catalog of what each complexity cost and what it bought.

The Closed System as Procedure

The series thesis thread for this cluster holds that a procedural constraint adopted at the outset determines what a statute can contain, and no statute in the series demonstrates the thesis more clearly than the Tax Reform Act of 1986. The three constraints were not substantive policy choices about any particular deduction or rate. They were procedural rules about the shape the bill was allowed to take: it had to be revenue neutral, distributionally neutral, and rate-reducing, and any provision that violated those parameters was out of order regardless of its merits. Those rules determined the contents more powerfully than any substantive debate did, because they defined the universe of feasible bargains before bargaining began.

The closed-system constraint: the 1986 act worked because revenue and distributional neutrality turned tax reform into a zero-sum trade in which no member could take a rate cut without naming the preference that paid for it, and every subsequent tax bill has abandoned that constraint, which is why none has reproduced the coalition.

The claim’s second half requires careful handling, because it concerns legislation beyond this article’s scope. The comparison with the 2017 tax legislation, which cut rates without the revenue and distributional constraints and therefore without the closed-system bargaining, is developed in the dedicated comparison article, and readers interested in that contrast should consult the 1986 versus 2017 comparison rather than expecting a full treatment here. The procedural point stands on its own: the 1986 coalition was a product of the constraint, and the constraint was a choice, adopted at the outset, about what kind of bill would be permitted. Different constraints would have produced a different statute, and the absence of the constraints in later bills produced different politics. Subsequent tax bills abandoned the constraint in different ways. Some cut rates and paid for the cuts with borrowing, abandoning revenue neutrality. Some shifted the distribution of the burden deliberately, abandoning distributional neutrality. Each abandonment removed a veto point, which made passage easier, and removed the market, which made the 1986 kind of coalition impossible. Without a fixed total, there is no need to name an offset. Without the need to name an offset, there is no trade. Without the trade, there is no coalition of the kind the 1986 act assembled.

The committee route through which the constraints operated is itself part of the procedure story. The bill moved through the House Committee on Ways and Means and the Senate Committee on Finance, the two tax-writing committees whose jurisdiction over revenue measures is established by chamber rules, and the markup process in those committees is where the closed-system arithmetic was actually enforced. Readers unfamiliar with how tax bills move through Congress, including the roles of committee markup, the Committee of the Whole, and conference, will find the mechanics explained in the series guide to how tax bills move through Congress. The narrative of how this particular bill survived its near-deaths belongs to the passage history companion, which this pillar article deliberately does not retell; the division of labor between the statute guide and the passage history is part of the cluster’s design, and each article links to the other rather than duplicating it.

What the pillar article must establish, and what this section has argued, is that the procedure determined the text. The two-rate schedule, the bubble, the passive loss rules, the corporate AMT, the phaseout schedules, and the trade table’s every row are all artifacts of the three constraints operating through the committee process. A different procedure, one that permitted revenue loss or distributional shifting, would have produced a bill with higher rates, fewer repealed preferences, or both. The statute is the constraints made visible, and that is why it serves as the cluster hub: every specialist article in the group, on passage, on impact, on the credit expansion, and on the later comparison, is ultimately an elaboration of something the constraints determined.

The counterfactual sharpens the point. Imagine the same Congress attempting rate reduction without revenue neutrality: the bill would have cut the top rate to twenty-eight percent, retained the most defensible preferences, and financed the difference with borrowing, producing a larger deficit and a shorter, simpler statute. Imagine it without distributional neutrality: the bill would have broadened the base at the bottom, cut rates at the top, and skipped the bubble, the zero bracket expansion, and the credit expansion entirely, producing a cleaner rate schedule and a regressive shift in the burden. Imagine it without the rate-reduction objective: the bill would have been a pure base-broadening revenue measure, politically impossible, and it would never have been introduced. Each counterfactual removes one constraint and produces a recognizable, ordinary tax bill. Only the combination of all three produced the strange, specific, historically singular text that was enacted.

This is what the series thesis means by a procedural constraint determining contents. The constraints did not merely filter proposals; they generated them, because the drafters worked backward from the required totals to the provisions that could produce them. The bubble was not proposed and then checked against distributional neutrality; it was invented because distributional neutrality demanded a recapture mechanism and the two-rate structure left no other place to put it. The passive loss quarantine was not selected from a menu of anti-abuse options; it was designed because revenue neutrality required the shelter revenue and no narrower provision could raise it. The transition percentages were not chosen for their elegance; they were calibrated to the five-year window. In each case the procedure specified the problem and the provision was the solution the procedure admitted.

The claim also explains the act’s peculiar immunity to the usual legislative pathologies. In ordinary tax legislation, each provision is fought on its own merits, and the bill becomes a collection of independent victories and defeats. In the 1986 act, no provision was fought on its own merits, because no provision could survive without its offset and no offset could survive without its provision. Attacking the gains repeal meant attacking the 28 percent rate it funded. Defending the credit meant defending the higher rate its retention required. The linkage disciplined both sides: opponents of a repeal had to propose an alternative offset or accept the rate consequence, and proponents of a cut had to name their repeal or lose the cut. This mutual hostage-taking is what made the bill cohere. It is also what made it so difficult to amend afterward without unraveling the whole, a difficulty later Congresses discovered when they tried to adjust single provisions and found the ledger resisting.

The constraint lens also answers the question the trade table implicitly poses: why these preferences died while others survived. The drafters selected repeal targets by a yield-to-cost ratio, the revenue a preference’s elimination would raise measured against the coalition cost of taking it on. The consumer interest deduction had a high ratio: enormous revenue, diffuse losers, no sympathetic industry. The investment credit had a lower ratio, concentrated industrial losers, but its revenue was indispensable to the corporate arithmetic, so it died anyway. The mortgage interest deduction had the lowest ratio of all: large revenue in theory, but a constituency, homeowners, realtors, and builders, whose opposition could have destroyed the bill, so it survived. The charitable deduction survived on a similar calculation with a public-purpose rationale attached. The pattern is not hypocrisy or inconsistency. It is the closed system operating as a selection mechanism, ranking every preference by what its elimination would buy against what its elimination would cost, and keeping only those whose cost exceeded their yield.

That selection mechanism is the final reason the statute cannot be understood as applied tax theory. Tax theorists of every school found provisions to praise and provisions to condemn in the enacted text, which is exactly what a constraint-driven process produces: the bill was not the victory of one school but the residue of a bargain, containing the repeals the arithmetic required and the survivals the coalition demanded. The closed system did not produce the theoretically optimal tax code. It produced the code that three simultaneous constraints permitted, and the distance between those two ideals is the space in which legislative history happens. The 1986 act occupies that space more completely than any comparable statute, which is why it remains the clearest case in the series of procedure determining text.

Reading the Statute: A Study Section

A statute of this scale rewards structured study, and the structure should follow the closed system rather than the table of contents. Begin with the three constraints and the One Test: can you state the design constraint and name what was traded for the twenty-eight percent top rate. Then work through the trade table row by row, asking of each preference who lost it and what rate reduction it purchased. Then examine the rate schedule, including the bubble bands and the 1987 interim schedule, and verify that you can explain why the thirty-three percent band existed. Then turn to the corporate title and confirm that you can state the net corporate increase and its source. Finally, read the simplification evidence and practice stating the qualified version of the simplification claim, since the unqualified version is the most common error.

For organizing notes, timelines of the phase-in provisions, and citation lists for the authorities named in this guide, the legislation study notebook on VaultBook provides a free workspace designed for this kind of statutory study. The notebook format suits the 1986 act particularly well, because the act’s logic is tabular at heart: preferences in, rates out, constraints binding throughout.

Two cautions close the study guidance. First, do not treat the qualitative revenue labels in the trade table as exact accounting; they show relative scale, and the provisions interacted in ways that make isolated yields imprecise. Second, keep the scope boundaries of this pillar article in mind: passage mechanics belong to the passage history, outcome evidence belongs to the impact article, and the 2017 comparison belongs to the comparison article. This guide has linked to each rather than covering them, and the cluster is designed to be read as a set. The statute itself, Public Law 99-514, 100 Stat. 2085, remains the primary source, and every claim in this guide is ultimately answerable to its text.

Frequently Asked Questions

Q: What did the Tax Reform Act of 1986 do?

The Tax Reform Act of 1986, Public Law 99-514, restructured the federal income tax around three simultaneous constraints: revenue neutrality, distributional neutrality, and rate reduction. It collapsed fourteen individual brackets into two nominal rates of fifteen and twenty-eight percent, with a five percent surcharge creating a thirty-three percent phase-out band, and cut the top rate from fifty to twenty-eight percent. It paid for those cuts by broadening the base: phasing out the consumer interest deduction, repealing the sales tax deduction and the sixty percent capital gains exclusion, repealing the investment tax credit, limiting IRA deductibility, lengthening depreciation under MACRS, and creating the passive activity loss rules of section 469. On the corporate side it cut the rate from forty-six to thirty-four percent while broadening the base so much that aggregate corporate liability rose about one hundred twenty billion dollars over fiscal years 1987 to 1991. Section 2 redesignated the Internal Revenue Code of 1954 as the Internal Revenue Code of 1986.

Q: Which president signed the Tax Reform Act of 1986?

President Ronald Reagan signed the Tax Reform Act of 1986 on October 22, 1986, enacting H.R. 3838 of the 99th Congress as Public Law 99-514, cited at 100 Stat. 2085. The signing followed House and Senate passage of the conference report and represented the culmination of a legislative effort that had begun with Treasury reform proposals early in the administration. Reagan had made individual rate reduction a central domestic objective, and the administration presented the bill as delivering the lowest top individual rate in half a century while removing about six million low-income filers from the tax rolls through the expanded standard deduction, personal exemption, and earned income tax credit. The signing ceremony marked the statute as the administration’s signature domestic legislative achievement. The procedural story of how the bill reached the President’s desk, including its near-deaths in both chambers, belongs to the passage history rather than to this statute guide.

Q: Did the Tax Reform Act of 1986 raise corporate taxes?

Yes, in the aggregate, even though it cut the corporate statutory rate from forty-six percent to thirty-four percent. The Joint Committee on Taxation Blue Book designated JCS-10-87 estimated that the act increased aggregate corporate tax liability by approximately one hundred twenty billion dollars over fiscal years 1987 to 1991, while reducing aggregate individual liability by a roughly similar amount. The increase resulted from base broadening that outweighed the rate cut: repeal of the investment tax credit, lengthened depreciation under MACRS, the new twenty percent corporate alternative minimum tax with its book-income adjustment for 1987 through 1989, and the curtailment of various business preferences. The effect varied by firm. Corporations that had relied heavily on credits and accelerated depreciation saw large increases, while corporations with little preference income benefited from the lower rate. The net corporate increase is the single most counterintuitive fact in the statute and the clearest proof that the three-constraint design bound the corporate title as tightly as the individual one.

Q: Why did the Tax Reform Act rename the tax code?

Section 2 of the Tax Reform Act of 1986 redesignated the Internal Revenue Code of 1954 as the Internal Revenue Code of 1986. Before the act, the codified body of federal internal revenue law carried the 1954 date because the last comprehensive reorganization of the code had occurred in that year. Congress chose to rename the code to mark the 1986 act as a comparably comprehensive revision, even though the act amended the existing code rather than replacing it outright. The redesignation was a renaming of the entire codified body, not an amendment to any particular section, which is why provisions later added or amended by other statutes are still cited as sections of the Internal Revenue Code of 1986. The practical consequence is that every citation to the federal tax code, such as 26 U.S.C. 469 for the passive activity loss rules or 26 U.S.C. 168 for depreciation, reads to a code named for this statute.

Q: What is the public law number of the Tax Reform Act?

The public law number is Public Law 99-514, enacted by the 99th Congress. The Statutes at Large citation is 100 Stat. 2085, meaning the enrolled act begins at page 2085 of volume 100 of the United States Statutes at Large. The bill number was H.R. 3838, indicating it originated in the House of Representatives, as revenue measures must under the Origination Clause. These identifiers are the standard references used in court opinions, Treasury regulations, and Joint Committee on Taxation publications. Researchers should note that the Joint Committee’s general explanation of the act, the Blue Book designated JCS-10-87, is the authoritative secondary source for the enacted provisions and the revenue estimates, and it is organized by reference to the public law’s titles and sections.

Q: How many tax brackets did the Tax Reform Act create?

The act created two nominal individual brackets, fifteen percent and twenty-eight percent, effective for 1988, replacing fourteen nonzero statutory brackets under prior law. A five percent surcharge applied over a phase-out band of income, producing an effective marginal rate of thirty-three percent within the band, so the true marginal rate structure had four zones rather than two. In 1988 the band ran from seventy-one thousand nine hundred dollars to one hundred forty-nine thousand two hundred fifty dollars for joint filers, from forty-seven thousand fifty dollars to ninety-seven thousand six hundred twenty dollars for single filers, and from sixty-seven thousand two hundred dollars to one hundred thirty-four thousand nine hundred thirty dollars for heads of household. For tax year 1987 the act provided an interim graduated schedule rather than the two-rate structure. The two-rate schedule lasted only from 1988 through 1990; the Omnibus Budget Reconciliation Act of 1990 added a third bracket at thirty-one percent.

Q: What deductions did the Tax Reform Act eliminate?

The act phased out the deduction for consumer interest over tax years 1987 to 1990, ending it for 1991, and disallowed the deduction for state and local general sales taxes while retaining the deductions for state and local income taxes and for real and personal property taxes. It repealed the sixty percent exclusion for long-term capital gains, so gains were taxed as ordinary income effective 1987, and repealed the investment tax credit. It limited IRA deductibility for workers covered by employer plans, phasing the deduction out over twenty-five thousand to thirty-five thousand dollars of income for single filers and forty thousand to fifty thousand for joint filers. It replaced ACRS depreciation with longer-lived MACRS and created the section 469 passive activity loss rules, which ended the deduction of shelter losses against unrelated income. Each elimination contributed revenue to the pool that funded the lower rates.

Q: Was the Tax Reform Act revenue neutral?

On the static scoring basis used by the Joint Committee on Taxation, yes for the legislation as a whole. Revenue neutrality meant the bill could not reduce federal receipts over the budget window, and the individual title was scored as roughly neutral while the corporate title was scored as a net increase of about one hundred twenty billion dollars over fiscal years 1987 to 1991, offset by a similar individual decrease. Martin Feldstein’s National Bureau of Economic Research working paper 17531 examined the estimates and found the act revenue neutral on a static basis, with behavioral responses to the new incentives partially offsetting the static figures. The distinction matters: static scoring assumes taxpayers do not change behavior, while actual receipts reflected responses such as the acceleration of capital gains realizations into 1986 before the exclusion repeal took effect. Revenue neutrality was therefore achieved as scored, with real-world receipts diverging as taxpayers responded to the new rules.

Q: What was the 33 percent “bubble” in the Tax Reform Act of 1986?

The bubble was the phase-out band in which a five percent surcharge stacked on top of the twenty-eight percent rate, producing an effective marginal rate of thirty-three percent. The surcharge phased out two benefits: the value of the fifteen percent bottom bracket and the value of personal exemptions. In 1988 the band ran from seventy-one thousand nine hundred dollars to one hundred forty-nine thousand two hundred fifty dollars for joint filers, from forty-seven thousand fifty dollars to ninety-seven thousand six hundred twenty dollars for single filers, and from sixty-seven thousand two hundred dollars to one hundred thirty-four thousand nine hundred thirty dollars for heads of household, with indexation thereafter. Below the band, the marginal rate was twenty-eight percent; inside it, thirty-three percent; above it, where the phased-out benefits were exhausted, twenty-eight percent again. The Congressional Research Service analysis RL34498 treated the surcharge as a distributional instrument designed to preserve the pre-reform burden distribution within a two-rate structure.

Q: How did the Tax Reform Act phase out the deduction for consumer interest?

The act reduced the deductible share of consumer interest, meaning interest on credit cards, auto loans, and other personal borrowing, in steps across tax years 1987 to 1990, and disallowed the deduction entirely for tax years beginning in 1991. For 1987, 65 percent of consumer interest remained deductible; for 1988, 40 percent; for 1989, 20 percent; and for 1990, 10 percent. The phaseout was deliberate rather than immediate: spreading the repeal across four years gave borrowers time to restructure debt and spread the revenue gain across the budget window. The provision illustrates the statute’s consistent use of transition relief as a political shock absorber. The policy rationale was base broadening in its purest form. Deductible consumer interest subsidized borrowing for consumption, and the subsidy was worth more to higher-bracket filers since a deduction’s value equals the marginal rate. With the top rate falling to twenty-eight percent, the drafters treated the preference as expendable, converting its revenue into general rate reduction.

Q: Why did Congress repeal the investment tax credit in 1986?

The investment tax credit let businesses subtract a percentage of qualifying equipment costs directly from tax liability, making it one of the most powerful and most manipulated incentives in the code. Prior Congresses had suspended and restored it repeatedly, which taught businesses to treat it as temporary. The 1986 drafters repealed it outright, with transition relief for property under binding contract, because the revenue it freed was essential to the corporate side of the bargain. Repeal raised major corporate revenue that helped fund the rate cut from 46 to 34 percent, and the credit’s repeal is a large component of the roughly 120 billion dollar net corporate increase the Joint Committee on Taxation scored over fiscal years 1987 through 1991. Equipment makers, purchasers, and the leasing industry built around transferring the credit’s value lost the preference and accepted the lower rate as compensation, on the drafters’ calculation that a lower rate on all income outweighed a credit on some investment.

Q: How did the Tax Reform Act of 1986 change IRA deductibility?

The act limited the deduction rather than eliminating it. Workers not covered by an employer retirement plan kept full deductibility up to the statutory contribution limit, as did covered workers whose incomes fell below the phase-out ranges. For covered workers with higher incomes, the deduction phased out between 25,000 and 35,000 dollars of adjusted gross income for single filers and between 40,000 and 50,000 dollars for joint filers. Above those ranges, a covered worker could still contribute, but the contribution was nondeductible, though earnings continued to accumulate tax-deferred. The line separated workers the pension system already served from those it did not, preserving the savings incentive where no employer plan existed and recapturing the deduction where one did. The revenue raised was modest beside the act’s larger base broadeners, but the limit protected distributional neutrality by denying high-income covered workers a deduction the rate cuts would otherwise have cheapened.

Q: What did the Tax Reform Act of 1986 do to the alternative minimum tax?

On the individual side, the act raised the AMT rate from 20 to 21 percent for 1987 and broadened the AMT base by adding preference items and adjustments, so more taxpayers computed the tax and more income fell within it. The AMT had been a backstop against aggressive use of preferences; with many preferences repealed outright, the act refocused the backstop on what survived. On the corporate side, the change was structural: the act repealed the corporate add-on minimum tax and replaced it with a broad-based corporate AMT at a 20 percent rate, computed as a parallel system under which a corporation paid the higher of its regular and minimum liabilities. For 1987 through 1989, the corporate AMT included a book-income adjustment tied to financial reporting income. The expanded AMT protected the revenue score by ensuring surviving preferences could not be stacked into zero liability, at the cost of installing a second tax system many filers had never touched.

Q: How did the Tax Reform Act of 1986 change the standard deduction and personal exemption?

The act raised the standard deduction sharply and indexed it. For 1988, the figures were 5,000 dollars for joint filers, 4,400 dollars for heads of household, 3,000 dollars for single filers, and 2,500 dollars for married individuals filing separately. The personal exemption climbed a stair-step: 1,080 dollars for 1986, 1,900 dollars for 1987, 1,950 dollars for 1988, and 2,000 dollars for 1989 and later years, with indexing thereafter. Together, the larger deduction and the near-doubled exemption lifted the filing threshold above the incomes of millions of poor households; President Reagan said about 6 million low-income filers would leave the tax rolls. Indexing locked in the gains by preventing inflation from eroding the thresholds in real terms, which preserved the distributional choices the act made at the bottom of the schedule.

Q: What transitional tax schedules applied in 1987?

For individuals, a five-bracket interim schedule applied for tax year 1987, with a top rate of 38.5 percent, midway between the old 50 percent and the new 28 percent. The two-rate structure of 15 and 28 percent took full effect in 1988. For corporations, a transitional rate schedule applied for calendar year 1987 before the 34 percent top rate applied in full for 1988. The interim year softened the change for taxpayers whose planning had assumed the old schedules and distributed the revenue effects across two fiscal years, which aided the scoring. The transitions extended beyond rates: consumer interest deductions and passive loss disallowances phased in over 1987 through 1990, and the capital gains exclusion repeal carried a transitional cap for 1987. The 1987 schedules were one-year-only tables, never used again, and they remind readers that the act’s famous simplicity was a 1988 phenomenon.

Q: What happened to capital gains treatment under the Tax Reform Act of 1986?

The act repealed the 60 percent exclusion for long-term capital gains. Under prior law, individuals excluded 60 percent of qualifying gains and paid tax on the remaining 40 percent, which produced an effective top rate far below the 50 percent statutory rate. Beginning in 1987, long-term gains were taxed as ordinary income, with a transitional cap softening the first year. The maximum statutory rate on gains thus converged with the new 28 percent top rate on wages, ending the long-standing preference for gain over ordinary income. Supporters of repeal argued the exclusion encouraged conversion transactions and violated the neutrality the act sought between income types; defenders of preference argued full-rate taxation discouraged risk-taking and taxed inflationary gains. The repeal raised major individual revenue that helped fund the 28 percent rate, and the year-end 1986 realization wave it triggered became a textbook case of effective dates shaping behavior.

Q: How did the Tax Reform Act of 1986 change depreciation rules?

The act replaced the Accelerated Cost Recovery System, ACRS, with the Modified Accelerated Cost Recovery System, MACRS, assigning longer recovery periods and slower write-off methods to most asset classes. Residential rental property moved to 27.5 years and nonresidential real property to 31.5 years, while equipment schedules stayed accelerated relative to economic depreciation but ran markedly slower than under ACRS. Lengthening a write-off schedule pushes deductions into later years, which raises taxable income in the early years of the budget window, and that timing suited the drafters because the corporate rate cut needed early-year offsets. The change hit commercial real estate especially hard, since the industry had built an investment culture around ACRS write-offs, and it contributed heavily to the corporate revenue increase behind the 46 to 34 percent rate cut. Slower depreciation also removed one of the principal engines of the shelter industry, complementing the new passive loss rules.

Q: Why did the Tax Reform Act impose distributional neutrality?

Distributional neutrality blocked the easiest escape from the other two constraints. Revenue neutrality alone would have permitted a bill that cut rates and paid for them by shifting the burden downward, broadening the base at the bottom while protecting preferences at the top. By requiring that the share of total liability borne by each income group remain approximately unchanged, the constraint forced the drafters to pair base broadening with offsetting relief at the bottom, through the larger standard deduction, the doubled personal exemption, and the expanded earned income tax credit, and to recapture upper-bracket benefits through the bubble surcharge. The Congressional Research Service analysis RL34498 treated the surcharge as evidence of this distributional intent. Later work, including John Witte’s 1991 assessment, concluded the neutrality was not fully achieved in practice once behavioral responses and timing effects were accounted for.

Q: Where can the official revenue estimates for the Tax Reform Act be found?

The authoritative source is the Joint Committee on Taxation’s General Explanation of the Tax Reform Act of 1986, designated JCS-10-87 and commonly called the Blue Book. It walks through the enacted provisions title by title and presents the revenue estimates used in scoring, including the finding that aggregate corporate liability rose approximately one hundred twenty billion dollars over fiscal years 1987 to 1991 while individual liability fell by a roughly similar amount. The Blue Book is a secondary explanation rather than the statute itself; the primary source remains the enrolled act at 100 Stat. 2085. For the distributional analysis of the surcharge mechanism, the Congressional Research Service report RL34498 is the named reference. For the behavioral qualification of the static estimates, Martin Feldstein’s National Bureau of Economic Research working paper 17531 presents the revenue-neutral static finding with the behavioral offset discussion.

Q: How did the Tax Reform Act of 1986 expand the earned income tax credit?

The act gave the credit its largest expansion to that point. The credit rate rose to 14 percent of the first 5,714 dollars of earned income, producing a maximum credit of 800 dollars, with the credit phasing out as income rose toward 9,000 dollars. The act indexed the credit’s parameters against inflation. The expansion was the work-contingent complement to the bigger standard deduction and personal exemption: households lifted off the income tax rolls by the higher filing thresholds still owed payroll taxes, and the credit offset that burden while supplementing earnings. Distributionally, the credit gave the bottom of the income scale a benefit calibrated to match the rate cuts at the top, which is why the two traveled together through the legislative process. The 1986 expansion set the pattern of using the credit as the counterweight that lets rate-cutting bills hold the distribution steady.