Poor and working-class households would gain more modest benefits. The poorest 20 percent of Americans would see an average increase of 0.5 percent in their incomes, or about $120 a year. Households in the upper middle class, those in the 60th percentile through the 95th percentile, would pay more in taxes on average.
The Republican proposal would reduce tax rates on marginal income, eliminate the estate tax and modify the corporate tax system to allow businesses to immediately write off any capital investments while also eliminating the deduction for interest payments. The plan was published earlier this year as part of House Speaker Paul Ryan's effort to lay out a comprehensive conservative agenda amid a presidential campaign that has been lacking in detailed debates over public policy.
While some of the findings in the Tax Policy Center's analysis surely will disappoint the speaker and his Republican colleagues, the authors of the report say their forecasts are uncertain and depend on a number of assumptions about how the economy would respond to the Republican plan. And in one respect, the analysis published Friday represents an important victory for Republicans in Congress.
GOP lawmakers and conservative economists have been agitating for years for policy researchers and budget wonks to take into account the effects of changes in spending and taxes on the broader economy. Democrats have argued that calculating these effects is uncertain and can bias forecasts against increased public spending.
Friday's analyses are the Tax Policy Center's first attempt to estimate those broader effects, which are known as "dynamic" in the jargon of budgetary research.
When Republicans took control of Congress last year, they ordered the Congressional Budget Office and the Joint Committee on Taxation to begin incorporating dynamic estimates in their official forecasts for major new laws.
Those agencies, however, do not issue estimates for unofficial proposals from politicians, such as the ones presidential candidates make on the stump. The Tax Policy Center, among the most widely cited organizations of its kind outside of government, produced influential analyses of former Republican presidential nominee Mitt Romney's tax plans in 2012.
As the center adopts dynamic scoring, these estimates will work their way into the broader public debate during campaign season. The group plans to release dynamic forecasts for Democratic nominee Hillary Clinton's and GOP nominee Donald Trump's agendas in the coming weeks.
"I was skeptical about doing this," said Leonard Burman, Tax Policy Center director. But he eventually became convinced that the center could present the estimates in a transparent, nonpartisan way. In any case, once that dynamic scoring became the law of the land as far as Congress's official scorekeepers are concerned, Burman's organization risked being left out of the debate.
"That horse has left the barn," Burman said.
Douglas Elmendorf, the former director of the Congressional Budget Office who helped design the agency's models for dynamic scoring, praised the center's decision to broaden its analysis. "I think it is a very good thing for trained, objective analysts like those at the Tax Policy Center to do dynamic scoring for significant proposed changes in policy," he said.
Elmendorf, who was appointed by a Democratic Congress, said that work by his former agency and others has convinced some critics that dynamic scoring is a credible approach to forecasting. "Twenty years ago, when this issue first attracted so much attention in Washington, there was no track record of these estimates."
Others remain leery, including Jared Bernstein, former chief economist to Vice President Biden. While crediting the Tax Policy Center with serving as a fair umpire, Bernstein warned that the complexity of this kind of analysis means it can be easily manipulated to serve political ends.
"I don’t think there’s anything wrong with dynamic analysis of the type that the TPC's engaged in," he said. "The problem with dynamic scoring is that it's ripe for abuse."
Yet if Republicans have won the debate, they still might not be pleased with the results.
The Tax Policy Center produced several sets of estimates for the effect of the GOP plans, one using a mathematical model of the economy designed to estimate effects in the short term, and another with a model for effects over the long term.
The plan would have positive effects in the first several years. By 2026, the plan would increase the gross domestic product by about 1 percent, relative to the current forecast.
In the long term, however, the economy would be weaker under the Republican plan because the tax cuts would force the federal government to borrow more money to continue operating. This borrowing would add to the national debt, increasing interest rates and making it harder for businesses and households to get loans, according to the forecast. By 2036, the plan would reduce gross domestic product by 0.2 percent.
Without taking into account the short-term benefits to the economy, the GOP plan would reduce federal revenues by $3.1 trillion over the first 10 years. Taking those short-term benefits into account, the reduction in revenues would be $2.5 trillion.
The authors of the report cautioned that these forecasts could prove mistaken. Over the long term, it could turn out that greater national debt does not raise interest rates, assuming investors around the world remain willing to lend the government money at rates close to zero. It could also turn out that when taxes are lowered, workers spend more time on the job, and businesses invest more than the report predicts. If so, the authors write, the House GOP plan would have substantial and enduring benefits for the economy.
On the other hand, the opposite could be true. Interest rates could rise abruptly as the government borrows more, and tax cuts might do little to boost labor and investment. In that case, according to the report, the plan would do significant damage to the economy.
"Everything is uncertain," said Burman, the TPC director. "The best thing we can do is put the information out there."