David R. Henderson  

Summers, Kotlikoff, and Mintz on the Proposed Trump Tax Cuts

Don't assume that irrationalit... Anti-Market Bias in One Senten...

I posted a few days ago on Kevin Hassett's case for the Trump tax cuts, pointing out the huge positive effect on the real wages that he claims they would have. One commenter, JFA, asked the relevant question:

The question is how did Hassett combine the evidence from the literature to come up with those estimates?

I answered that I didn't know. But I would like to know.

Another commenter, MikeW, provided a link to Larry Summers's response to Hassett. The problem is that is long on attack and medium on analysis.

Larry writes:

I am proudly guilty of asserting that it [Hassett's analysis] is some combination of dishonest, incompetent and absurd. TV does not provide space to spell out the reasons why, so I am happy to provide them here.

Unfortunately, he doesn't provide enough reasoning. He gets some of the way there, writing:
In contrast, Mr Hassett throws around the terms scientific and peer reviewed, yet there is no peer-reviewed support for his central claim that cutting the corporate tax rate from 35 per cent to 20 per cent would raise wages by $4,000 per worker.

The claim is absurd on the face of it. The cut in corporate tax rates from 35 per cent to 20 will cost slightly less than $200 billion a year. There is a legitimate debate among economists about how much the cut will benefit capital and how much it will benefit labor. Mr Hassett's "conservative" claim that the cut will raise wages by $4,000 in an economy with 150 million workers is a claim that workers will benefit by $600 billion or 300 per cent of the tax cut. To my knowledge, such a claim is unprecedented in analyses of tax incidence. Mr Hassett, though, doubles down by holding out the further possibility that wages might rise by $9,000.

Like Larry, I am stunned by this numbers too, but I don't know whether they're right. Kevin doesn't "throw around" the term "peer reviewed." He claims that his estimates are indeed based on those peer-reviewed studies. As I said, I would like to see even the back of the envelope calculations that led to them. As far as I know, Kevin hasn't provided them.

But in today's Wall Street Journal, Boston University economist and sometime Larry Summers co-author Laurence Kotlikoff and Canadian economist Jack Mintz, argue that there is a model that comes close to giving the results that Kevin Hassett claimed. The WSJ article is gated but they reference a study that's not. It's "Simulating The Republican 'Unified Framework' Tax Plan" by Seth G. Benzell, Laurence J. Kotlikoff, and Guillermo Lagarda.

Here's the abstract:

This short paper simulates the economic and revenue impacts of the Republican \Unifi ed Framework" (UF) tax plan. As in our prior study of the Republican \Better Way" plan, this study uses the Global Gaidar Model (GGM). The GGM is a global 17-region, 90-period, overlapping-generations model, which is closely calibrated to U.N. demographic and IMF fiscal data. In incorporating the entire world's capital market, the GGM is particularly well suited to studying foreign capital inflows arising from U.S. corporate tax reform. We find that, depending on the year considered, the new Republican tax plan raises GDP by between 3 and 5 percent and real wages by between 4 and 7 percent. This translates into roughly $3,500 annually, on average, per working American household. The source of the increase in U.S. output and real wages is the UF plan's reduction in the U.S. marginal eff ective
corporate tax rate from 34.6 percent to 18.6 percent. This expands the U.S. capital stock by between 12 and 20 percent depending on the year in question. Due, in good part, to the economy's expansion, the UF tax plan is essentially revenue neutral. The GGM's strong supply-side response is not due to a built in bias. Rather, it reflects the mobility of the global capital stock in response to the UF's corporate tax reform and the inefficiency of the current U.S. corporate tax, marked by its very high marginal and very low average tax rates. Indeed, cuts in personal income tax rates in the GGM produce de cits, crowd out capital and lower long-run economic welfare. And reducing the U.S. corporate tax rate in the GGM much further than under UF (e.g., fully eliminating it) necessitates personal tax increases to prevent a rise in the U.S. debt-to-GDP ratio. The main diff erence between this study and a recent Tax Policy Center (TPC) study, which predicts large de ficits from enacting the UF plan, is TPC's assumption of no major economic response to the policy. The TPC also suggests that the UF tax plan is highly regressive. Our model produces roughly similar wage gains for both its low- and high-skilled workers. This said, we share the TPC's concern that the UF plan could disproportionately benefi t the top 1 percent. This concern about fairness as well as the country's massive long-term fiscal gap suggests modifying the UF plan to include, for example, the elimination of Social Security's FICA tax ceiling, a tax onlifetime inheritances and gifts received above $5 million, or a progressive cash flow tax on consumption above $100,000.

By the way, while I share this concern about the long-term fiscal gap, I don't share their concern about fairness. They take as given that the status quo is fair. But high-income people pay a massively disproportionate share of taxes now; that is they pay a much higher percent of their income in taxes than lower- and middle-income people pay.

Comments and Sharing

COMMENTS (4 to date)
A writes:

Mihir Desai, one of the researchers cited by Hassett, also offered critiques in a twitter thread: https://twitter.com/desaimihira/status/920306122506096646

Ben writes:

Instead of saying 'Look a model shows it could be true!' I think some basic common sense is in order because a model can essentially be designed to show anything to make a political point if it's designed in a certain way.

Any company that receives a windfall in taxes and a boost in after-tax profits is not going to raise wages by 3x that windfall. The only feasible scenario any company would ever do that is if workers had an enormous amount of power, which they obviously don't.

T. A. Stout writes:

High income people do not pay a "massively disproportionate share of taxes." That statement is arguably true only of income taxes. When other taxes - consumption taxes, excise taxes, FICA taxes - are included, the current system is only mildly progressive.

poorlando writes:

T. A. Stout, with respect to federal taxes, after taking into account federal income tax credits, the bottom two quintiles pay a disproportionately small share of all federal taxes. See supplemental data in this CBO report. The CBO used to be a lot more explicit about this in past reports but now has buried it a little. Your assertion that the current system is "only mildly progressive" is false, all the more so if you take into account highly progressive state income tax regimes like California's (see page 260 of this report), where the bottom 61% pay 2% of the total income tax liability, the bottom 82% pay 11%, and the top 0.05% pay 22%.

[html fixed--Econlib Ed.]

Comments for this entry have been closed
Return to top