Many view the 2017 Republican tax reform – The Tax Cut and Jobs Act (TCJA) – as a giveaway to the rich. “The most regressive tax cut in the past 50 years.” proclaimed one Washington Post blogger. “The most regressive tax policy change of our lifetimes,” wrote a contributor to The Hill. “A big tax cut for (primarily rich) stockholders,” opined New York Times columnist Paul Krugman. The Tax Policy Center (TPC) claimed that the top 1% would receive 82.8% of the reform’s total reduction in taxation. The Joint Committee of Taxation (JCT) also judged the reform regressive. Their analysis shows top-income households receiving tax cuts of 2.5% or larger compared to 0.5% for low-income households.
Unfortunately, the JCT’s analysis, like that of the Congressional Budget Office, other government agencies and D.C. think tanks, including the TPC, is static. It treats households as if they will live only in the current year and proxies their remaining lifetime spending, the distribution of which is the ultimate focus of inequality analysis, by their current disposal income.
This raises four major problems. The first is misclassification. The static approach classifies households as rich or poor based on current income (this year’s income). But economics tells us that for purposes of studying fiscal fairness we need to classify households not by this second’s, this minute’s, this week’s, this month’s or even this year’s income. Instead, it tells us to classify households based all on their remaining lifetime resources — their human wealth (the present value of future labor earnings) plus their non-human wealth (current net worth).
The current-year income classification means that Warren Buffett, whose wealth totals $85 billion, will be classified as being poor this year if his reported after-tax income for this year is zero or negative. This can easily be the case if Buffett’s net realized capital gains this year are negative and equal or exceed this year’s other positive taxable income.
You’d expect mistake 1 to be made by people with limited economics training. But the folks in D.C. making this mistake are well trained economists and are doing so consciously. Their reason, when I ask, is that members of Congress, not to mention the current president, are too economically illiterate to process the right answer. Hence, they need to be given what makes them comfortable, namely the wrong answer.
Mistake 2 is implicitly using current disposable (after-tax) income as a proxy for current spending. But the young, on average, save; i.e., on average, their income exceeds their spending. And the old, on average, dissave; i.e., on average, their spending exceeds their income. Given economic growth, this means that proxying spending with disposable income will overstate inequality in actual current spending, which is, itself, proxying for the right measure, remaining lifetime spending.
Mistake 3 is, indeed, using this year’s spending to proxy for remaining lifetime spending – the present value of current plus all future spending. If ignoring future spending made sense, we could study inequality by comparing the spending of different households over the next second or the next minute or the next hour or … . But just like this second’s spending doesn’t tell us much about this year’s spending, this year’s spending doesn’t tell us much about remaining lifetime spending. Certainly, there is no fixed relationship between this year’s and remaining lifetime spending. For the young, remaining lifetime spending is many times current-year spending. For the old, the multiple is far smaller.
Mistake 4 is comparing inequality of households of all ages rather than distinguishing by their age, i.e., their birth cohort. To see the concern, take a world in which everyone is absolutely identical, apart from their year of birth. Since everyone enjoys the same lifetime spending and pays the same lifetime taxes, there is no inequality. But if one compares young and old people at a point in time based on this year’s income, things will look very unequal. The young, who work, will have far higher income than do the old, who are retired. Moreover, if only wages were taxed, this perfectly equitable fiscal policy would be viewed as progressive since those with higher incomes would pay taxes and those with lower incomes would not. And if only spending by the old were taxed, the policy would be declared regressive since those with lower incomes would pay taxes whereas those with higher incomes would not.
Economic theory tells us to study inequality by comparing remaining lifetime spending among people of the same age (in the same birth cohort). And it tells us to study fiscal progressivity by examining the degree to which fiscal policy reduces inequality in remaining lifetime spending within the same cohort. My recent studywith U.C. Berkeley economist Alan Auerbach and Darryl Koehler an engineer at my software company, does just this. It runs household observations from the Federal Reserve’s 2016 Survey of Consumer Finances (SCF) through a detailed life-cycle consumption-smoothing program called The Fiscal Analzyzer (TFA). TFA determines how much each household will spend, pay in taxes and receive in government benefits in each year over the rest of their lives. These data can then be used to calculate remaining lifetime spending and remaining lifetime net taxes (taxes net of benefits). The program incorporates all federal and state taxes, including corporate income taxation, and all federal and state benefits, including food stamps, Supplemental Security Income, Social Security, Medicaid, Medicare, welfare (TANF), you name it.
Although the JCT uses older data (2013, not 2016) and their data come from the IRS, not the SCF, we’re able to roughly replicate their static analysis. But if we classify households based on remaining lifetime resources and measure spending and net taxes appropriately, on a remaining lifetime basis, the results are very different. We find very little change in the distribution of remaining lifetime spending within age cohort between the rich and the poor arising from TCJA. We also find that the top 1%, as a group, receive a disproportionately small share of the tax cut.
Take 40 year olds with different levels of remaining lifetime resources. The richest 1%, middle 20%, and poorest 20% accounted for 12.9%, 14.1%, and 6.5% of the cohort’s total remaining lifetime spending under the old tax law. Under TCJA, the respective figures are 12.9%, 14.1% and 6.4%, i.e., spending inequality is essentially unchanged. Stated differently, the rich, middle class and poor experience almost identical changes in their remaining lifetime net tax rates.
What about the share of the tax cut? Do the richest 1 percent of 40-year-olds receive 82.8% of their cohort’s total tax cut as the TPC study suggests? Far from it. The right answer is 9.7%. This is smaller than 16.6% – the share of taxes paid by the richest 1% of 40-year-olds under the old tax law. Consequently, the tax share of the top 1% in this cohort rises under the TCJA – to 16.9%. The 40-year-olds aren’t special. We find a small rise or essentially no change in the tax share of the top 1 percent for all cohorts. The pre-TCJA share of taxes paid by the top 1% is highest for the 70- to 79-year-old cohort, namely 26.4 percent. Post-TCJA it’s 26.3%. Their share of the tax cut is 28.5%, again miles below TCP’s 82.8% figure.
Why is the tax reform much fairer than many believe? First, the pro-rich corporate tax break, as measured by the JCT, is far smaller than generally perceive. Second many of TCJA’s personal income-tax provisions are quite progressive. But, third, measuring the right things in the right way matters!
All the above said, the roughly identical percentage tax cut for the rich and the poor means a far greater absolute remaining lifetime tax cut for the rich than for the poor. The reason is that the rich pay far more taxes than do the poor. Hence, if your gauge of fairness is absolute tax changes, the reform appears terribly unfair.
Moreover, our analysis doesn’t incorporate the impact of the repeal of the Obamacare mandate, which may significantly impair the ability of the poor to obtain health insurance at a reasonable price. And if the economy does not materially improve due to the TCJA’s enhanced incentives to invest in the U.S., the poor may be hurt relative to the rich via benefit cuts used to cover TCJA’s higher deficits. Such cuts are already under discussion by Republicans in D.C. Alternatively, future generations may be forced to foot the bill via future tax hikes. This will be on top of the enormous fiscal burdens we are leaving our progeny due, for the most part, to all the obligations for Social Security, Medicare, Medicaid, defense spending, etc., which Congress has carefully kept off the books.
The TCJA is far from my preferred tax reform. I give it a B-. Still, it has the potential to raise investment, output and wages. And, when correctly analyzed, it’s certainly fair based on economics’ standard fairness measure – inequality in the distribution of spending, i.e., in the shares of spending by the rich and the poor.