by Alan Murray
Those tough questions are the domain of a small band of professional economists who toil in anonymity in the back rooms of the Treasury Department and the congressional Joint Tax Committee. Theirs is a thankless job. Unlike lawyers in both organizations, who look forward to lucrative work in the private sector once they leave government, revenue estimators are practitioners of an arcane art for which there is little demand outside of the Washington tax-writing world. When a tax bill is under way—and seldom is a tax bill not under way in Washington—they work long hours and unbroken weekends under intense pressure, trying to tag numbers onto an endless string of proposals dreamed up by Treasury officials and by Congress’s 535 lawmakers.
Although their estimates carry an air of precision, estimators readily admit that they lean heavily on guesswork. Sometimes that guesswork is wrong, very wrong. In 1981, when asked to estimate the revenue effects of a proposal to expand IRAs, the Treasury estimators projected the measure would lose $5.5 billion over a three-year period. Instead, it lost $32 billion, six times what was estimated. Says Thomas Vasquez, who headed the Treasury’s revenue estimating staff at the time, “banks started advertising all over the place, offering things like toasters to people who opened new IRAs.” The result of that marketing effort was a flood of money into the tax-sheltered accounts.
Such enormous errors are uncommon, but they have nevertheless opened the estimators up to criticism. Supply-siders like Jack Kemp are particularly harsh in their criticism; they argue that the estimators are biased, that they understate the revenues gained from tax changes in order to thwart tax cuts. The criticisms are frequently aired as well by the editorial writers of The Wall Street Journal, who complain that the “static revenooers” fail to look at changes in behavior caused by tax changes.
These supply-side criticisms are off the mark. Estimators do in fact take into account a wide array of behavioral changes in making their estimates; but whether they do so accurately is hard to gauge. “The estimates are not static, they incorporate behavioral responses,” says former Treasury economist J. Gregory Ballentine, but “no one knows if the behavioral responses used are too large or too small.”
Whatever their flaws, the revenue estimates provide a critical discipline to the tax-writing process. Politicians would always rather cut taxes than raise them. Estimates of the effects their actions have on the budget are needed to prevent excesses. Traditionally, both the Treasury and the Joint Tax Committee have worked hard to protect those estimates from political tinkering.
In Treasury’s effort to draft a tax-reform bill, the revenue estimates became shakier than ever, but they also became more important than ever. Calculating the effects of a total overhaul of the system proved far tougher than calculating the effects of an isolated tax change. More important, previous tax bills had either raised taxes overall or lowered them. Never before had the president insisted that a tax bill be revenue neutral. It was a harsh discipline: Every proposal that lost money had to be matched by one that raised an equal amount; the tax changes had to add to zero. As the scorekeepers for this exercise, the revenue estimators wielded tremendous power and influence—and bore an inordinate burden.
In agreeing to Packwood’s fringe-benefit proposal, the department lost more than $35 billion in new revenue over five years compared to Treasury I. In modifying Treasury I’s oil-and-gas provisions, it threw out another $50 billion. Manley Johnson’s generous new system of depreciation write-offs lost an eye-popping $200 billion in comparison to the original Treasury I plan. All those losses had to be made up by raising revenue elsewhere to pay for lower rates.
The first sacrifices to the hungry revenue god were the Treasury I proposals to “index” interest payments for inflation and to allow companies to deduct half the dividends they pay out. Those two proposals, unlike most of the other reforms, actually lost revenue for the Treasury rather than raised it. Indexing interest cost the Treasury more than $40 billion over five years; the dividend deduction cost more than $100 billion. Both provisions were dear to Treasury tax reformers, but did not find supporters in the real world of business and finance. The interest indexing provision was dropped entirely, while the dividend deduction was cut back to only 10 percent.
Baker and Darman developed a tough new “minimum tax” to help stem the drain of revenue. Minimum taxes were viewed by the Treasury’s tax purists as a sort of Band-Aid approach to tax reform; if the tax net was woven tightly enough to catch all income in the first place, there would be no need for minimum taxes. But once Baker and Darman began tearing the tax net apart, the minimum tax became necessary to ensure that no one slipped through entirely. Baker and Darman wanted to make certain those stories about millionaires and corporations escaping all taxes never surfaced again. Baker even suggested an unusual proposal requiring large companies with no other tax liabilities to pay one dollar in tax, a plan that would put a conclusive, if meaningless, end to Bob McIntyre’s inflammatory reports of profitable Fortune 500 firms that paid no taxes at all. That idea was wisely rejected, but a tough minimum tax became a critical component of the new Treasury proposal.
With revenue problems hounding them, Treasury tax writers agreed they could not afford to compromise on one of the most controversial provisions of Treasury I: elimination of the deduction for state and local taxes. Everyone understood that the provision would cause a tremendous uproar. They knew the defenders of the state and local deduction could make a much stronger philosophical argument than the defenders of most other tax preferences. Allowing people to take a deduction for the amount they paid to state and local governments hardly seemed an egregious abuse of the tax system. Nevertheless, repealing the deduction raised an enormous $150 billion over five years. If Baker and Darman gave in on that one, they felt certain their plan would fall apart. The deduction had to be repealed, they thought, in order to make the numbers add up. Forget the argument over the merits, they decided: This one was for the money.
A political calculation also lurked behind the repeal of the state and local deduction. The deduction mostly benefited residents of high-tax states like New York and did far less for residents of states with lower taxes. Baker and Darman calculated that pitting New York against the rest of the nation might be good politics and help win the tax plan’s approval. It was a risky bet, but it was a bet they felt they had to make.
Another problem facing the Treasury, along with revenue, was the distribution of the tax burden. Treasury I tried to guarantee that taxpayers of all income levels received roughly equal percentage tax cuts. The policymakers did not want to soak the rich, nor did they want to tilt the tax system further in favor of high-income taxpayers, but as Baker and Darman made their political compromises to restore some deductions, exclusions, and credits, they discovered that most of the benefits were falling to those at the top of the income scale. Many low-and lower-middle-income taxpayers, after all, did not even use deductions. The revised plan began to lean dangerously in favor of the highest-income taxpayers, a result that promised political trouble.
The distribution problem became especially acute as Darman tried behind the scenes to develop a plan with a top rate of only 30 percent. He hoped to win the support of tax-overhaul sponsor Kemp and his fellow supply-siders. Kemp insisted that Treasury I’s 35-percent top rate was too high; he felt 30 percent should be the maximum. Darman directed the Treasury staff to try to accommodate that request. Such a sharp drop in the top rate, down from 50 percent under existing law, seemed inevitably to make the plan overly generous to high-income taxpayers. While that did not bother Kemp, who was perfectly willing to allow his plan to give big cuts to those at the top, Treasury officials knew that tipping the plan toward the top of the income scale would create serious political problems.
In an attempt to avoid those problems, Darman experimented with plans that appeared to have a top rate of 30 percent, but that also placed a 5-percent surtax on some higher-income taxpayers. The Kemp-Kasten proposal used a similar tactic,
enabling the authors to claim a flat rate of 24 percent when in fact many upper-income taxpayers faced a 28-percent marginal rate. Darman’s ploy was to claim a top rate of 30 percent, while having a hidden top rate of 35 percent. The scheme was typical of the sleight of hand that Darman enjoyed, but it was a nightmare to the tax-policy experts at Treasury. They were appalled that they had to spend time studying the idea. The Treasury was usually the protector of the integrity of the tax code; now with Darman on the job, it was becoming an originator of gimmicks. If the top rate was going to be 35 percent, Pearlman and his staff thought, just make it 35 percent outright; don’t use a bizarre scheme to hide the top rate. “We wanted to be spending our time pulling things together that worked,” recalls Eugene Steuerte, “and instead we were working on these crazy ideas.”
Despite the complaints from the Treasury staff, Darman continued to pursue the plan, searching for a way to satisfy Kemp’s demand for a 30-percent top rate, at least in appearance, if not in fact. He even presented a version of his 30-percent plan to President Reagan. But the White House ruled out the idea as fraudulent, never guessing the strange ploy would later be picked up, dusted off, and made the law of the land.
On April 23, Baker, Darman, and Pearlman made the familiar trip across East Executive Avenue to begin briefing the president on the new tax plan. It was a strange reversal of roles from five months earlier. Regan was now sitting on the president’s side of the conference table, along with aides Kingon and Dawson, who also had moved to the White House from Treasury. Baker and Darman, on the other hand, were in the hot seat across from the president.
There was clear tension across the table between Regan and Baker, and between their aides. When they were in the White House, Baker and Darman had felt a certain amount of contempt for Regan, who often spoke before thinking. Regan, for his part, had resented the haughty White House officials who frequently managed to scuttle his efforts. Since the job switch, the animosity had only increased. Baker had been praised in the press for his pragmatic approach to the Treasury job; Regan had been pounded for some slip-ups as chief of staff. At the time of the meeting, the new White House team was under attack for its decision to have President Reagan, as part of his upcoming trip to Europe, visit a cemetery at Bitburg where members of the SS—the deadly Nazi elite guard during World War II—were buried. Now Regan and his aides had to watch Baker take responsibility for reviving Treasury I from the dead. They believed that they were the true fathers of tax reform; they had plunged ahead when Baker still thought the effort was foolish. Now they had to take a back seat to Baker, the born-again reformer.
During the first day’s discussion, the president agreed to Baker’s recommendation to reverse some of the most controversial provisions of Treasury I. The exemption for ministers’ housing allowances was reinstated, as well as the exemption for veterans’ disability benefits. The unusual, Packwood-proposed compromise on fringe benefits was discussed and accepted by the president. Baker also proposed easing Treasury I’s tough restrictions on the deduction of business meals. The original plan limited the deduction to $10 for breakfast, $15 for lunch, and $25 for dinner. Baker’s plan put a $25 limit on full deductibility for all meals, prompting Regan to quip, “There’ll be a lot of expensive breakfasts going on.”
The Treasury officials also told the president they wanted to stick with the original proposal to end the deduction for state and local taxes, and the president agreed.
Not until the subject of charitable deductions arose did the president balk. Treasury I allowed those deductions only to the extent that the contributions exceeded 2 percent of a person’s income; Baker proposed lowering that to 1 percent. But in cutting back government spending, the president had frequently emphasized the importance of private giving. He overturned the Treasury recommendation and asked that the full charitable deduction for those who itemize be left intact.
The Treasury team returned to the White House on the twenty-fourth and again on the thirtieth of April, briefing the president on the most controversial elements of the plan. The president agreed to drop interest indexing and dividend relief and to keep generous depreciation write-offs. He accepted most of the details of the plan, and by the time he left for his trip to Europe, the Treasury team felt they had won his approval for their proposal.
While the president was gone, however, a new problem arose, the first and largest of many revenue estimating errors that would dog the tax-writing process. Treasury’s tax experts discovered they had made a mistake, and it was a whopper. The plan they briefed the president on, it turned out, was not revenue neutral at all. In fact, corrected estimates showed the plan would add $150 billion to the nation’s budget deficit over the next five years!
Shock waves from the mistake reverberated around the Treasury Department. Estimates of the revenues raised or lost by tax changes are always subject to error; but this was far bigger than any previous error. It was a hole so large that it threatened to sink the entire tax-reform vessel. The president had signed on to a tax plan that did not work.
The mistake and its aftereffects sparked new antipathy between Pearlman and his tax staff on the one side, and Baker, Darman, and Johnson on the other. Both Darman and Johnson suspected the tax staff might have come up with the error in an effort to force a move back to tough, Treasury I-style depreciation write-offs. “You would think if you were going to brief the president, you would take the time to check and recheck the numbers before you let it get over there,” Johnson said. The atmosphere at the Treasury grew tense and frantic.
Eager to solve the problem neatly and quietly before the president returned, Treasury officials scrambled to find new and unobtrusive sources of revenue. Their first move was to eliminate some key rules meant to ease the transition to the new tax system. The tax writers originally had planned to phase out the repeal of the state and local tax deduction and the repeal of the investment tax credit over a few years to soften the impact on the economy. In the desperate search for revenue, they changed course and decided to repeal both of them as of January 1, 1986. The department also intended to include a rule that protected investments already under contract from the cutbacks in investment incentives. That rule was scrapped as well. It was a cynical move, since the officials knew Congress would insist upon such a “binding contract” rule, but they needed revenue badly, and were in a hurry to patch things up.
Those changes still brought in only a third of the revenue needed to fill the gaping hole. They still had to have more money—a lot more.
It was Johnson who offered the solution—a proposal he had been thinking about for some time. He worried that lowering the corporate tax rate would provide an unnecessary “windfall” to people who owned old equipment and had already used up their depreciation write-offs. That was a perverse incentive, he argued; the tax code ought to encourage new investment, not the retention of old, outmoded equipment.
To fix the incentive, Johnson proposed a totally new tax called a “windfall recapture tax” to take back some of the windfall that would be reaped by owners of old capital. It was a reasonable idea in theory, but in the face of the desperate desire for money it was turned into a behemoth. The proposal was fashioned to raise roughly $56 billion over five years—far more revenue, Treasury officials later admitted, than such a provision should actually have raised if done as dictated by theory. It called for a significant increase in the taxes of many corporations and was viewed by businessmen as a “retroactive” levy, because it was based on the depreciation write-offs they had taken in previous years. “We all knew that no one was going to understand it, and we were going to be accused of having a retroactive tax,” Pearlman said. Nevertheless, it was put into the plan. It was needed as a plug, a placeholder—something the Treasury could use in order to claim its plan was revenue neutral, even thought it did not have the slightest chance of being accepted by Congress.
When the president returned from Europe, the group gathered again at the White Hous
e on May 14. The memorandum for the meeting noted that “because of a computer programming error, the revenue was substantially understated in a memorandum on April 24.” On the president’s side of the table, jaws dropped. “I was disgusted,” says Regan, whose anger was clearly mixed with glee at the new Treasury team’s embarrassment. “I couldn’t figure out how they could possibly have made such a mistake and why they hadn’t double-checked it before they brought it over here. I had worked with those numbers for four years and we never came anywhere near a mistake like that. It was a shock!”
Confused by the turn of events, the president asked for an explanation. Baker turned to Pearlman and said sharply, “You explain it Ron, it’s your mistake.” Red-faced, Pearlman tried to say it was a computer error, but the explanation did not sit well.
The Treasury team offered White House officials two ways to deal with the huge problem. One was to increase the personal exemption to only $1,500 in 1986 and then raise it gradually to $2,000 over a few years. The White House group summarily rejected that; the $2,000 exemption had become an important pro-family symbol for the New Right Republicans. That left the other choice as the only alternative: windfall recapture. The president signed on.
The Treasury officials scheduled another meeting with the president on May 21. It was to be the last. Congressional tax writers were already beginning to complain about White House delays in releasing its plan. At the first of the year, Darman had talked of releasing the new plan on April 15, tax day, but the timetable kept slipping. Baker and Darman felt the plan had to be pushed through Congress in 1985. If it was still around in 1986 it would suffer under the strains of an election year, they feared. Rostenkowski, on the other hand, had always thought that the timetable was unrealistic; when Baker and Darman first proposed it, he and his aides had laughed at them. Now Rostenkowski’s laughter was changing to irritation as the Treasury slipped past its own deadlines. “Delay saps the cause of tax reform and raises doubts about the president’s nerve,” he complained.