Don’t Believe Trump’s Tax Math — Or Anyone Else’s

This is In Real Terms, a weekly column analyzing the latest economic news. Comments? Criticisms? Ideas for future columns? Email me, or drop a note in the comments.

Let’s get this out of the way right off the top: Donald Trump says his new tax-cut plan will pay for itself. It won’t. Trump’s plan would reduce government revenue, most likely by trillions of dollars.

Figuring out how many trillion, though, is tricky. Despite decades of research and debate, there is still no consensus among economists on the best way to analyze how changes in tax and spending policy will affect the economy. That leaves the door open for candidates such as Trump to pick whichever estimate best suits their political needs.

In a speech in Manhattan last week, Trump laid out what he promised would be a “bold, ambitious, forward-looking plan” to revive the U.S. economy. Parts of it were certainly ambitious: Trump pledged to create 25 million jobs over the next 10 years — a figure, that, as Neil Irwin explained in The New York Times, would likely require dragging grandparents out of retirement (or perhaps bringing in more immigrants). Otherwise, there simply aren’t 25 million Americans looking for jobs.

When it comes to taxes, though, Trump was less bold. Or, at least, less bold than he used to be. The tax plan that Trump described in his economic speech represented a pullback from his original plan, released a year earlier, which would have slashed taxes by around $10 trillion over 10 years. The new plan would cut taxes by less than half as much. Trump’s new plan looks a lot more like the one proposed early this year by Paul Ryan and other Republican leaders in the House of Representatives. Trump would cut the top tax rate to 33 percent, from nearly 40 percent now. (He originally proposed cutting the top rate to 25 percent.) He would more than double the standard deduction, to$15,000 for individuals and $30,000 for couples, which would mean that millions of families would no longer owe any income taxes at all. (Although again, this was less ambitious than Trump’s earlier proposal.) And it would check two items off Republicans’ long-term wish list: cutting the tax rate for businesses and eliminating the estate tax on inherited income. But while Trump’s plan is cheaper than his old one, it still cuts taxes by a lot:$4.4 trillion, according to the campaign. Trump promised to pay for those cuts without raising the deficit and without cutting spending on defense or entitlement programs such as Social Security and Medicare. That means his plan would require big cuts to the perhaps 25 percent of federal spending that’s left over.1 Trump hasn’t provided much detail on those cuts; he said he could cut $800 billion from the budget through “simple, common-sense reforms.” He also said he could save another$1.8 trillion through “trade, energy and regulation reform” but has provided no evidence to support his claims.

Even if you take those promises at face value, though (and you shouldn’t), that still leaves some $1.8 trillion in tax cuts that Trump has to find a way to pay for. Except that Trump doesn’t think he needs to do anything at all: He argues that the tax cuts will spur economic growth, which will boost tax revenue, which will in turn make that$1.8 trillion budget hole disappear.

That claim might sound far-fetched, and it probably is. But it’s rooted in an element of truth: Tax policies have economic consequences that aren’t accounted for in traditional assessments of their costs and benefits. A big tax cut such as Trump’s, for example, would put more money in the pockets of consumers, leading them to spend more. That extra spending would probably lead businesses to hire more workers, creating jobs and driving up wages as companies competed for workers. And since those wages would be taxed, the government would end up taking in more revenue, partly offsetting the impact of the tax cut on the budget. All else equal, a “$4.4 trillion tax cut” would cost the government less than$4.4 trillion.

Of course, all else isn’t equal. All that extra spending could also drive up inflation, which could lead the Federal Reserve to raise interest rates, slowing the economy back down. Lower taxes require either less government spending or larger deficits, both of which have their own economic effects. And there are countless other variables: If the U.S. cuts its corporate tax rate, would other countries follow suit in order to stay competitive? If the federal government taxes its citizens less, would that prompt some states to tax them more?

Accounting for such “dynamic effects” is hard, and economists disagree about how to do it. Trump’s numbers are based on a model from the conservative Tax Foundation. The Joint Committee on Taxation — the congressional committee tasked with estimating the cost of tax proposals — uses an altogether different approach. And the Tax Policy Center, a centrist think tank, on Friday introduced two more models that try to account for the economic impact of taxes in totally different ways. And picking a model is only the start: An interactive graphic from the University of Pennsylvania team that built one of the TPC models shows how even small tweaks to the assumptions underlying its model can lead to big differences in the results decades down the road.

The root of the problem is that in order to understand how taxes will affect the economy, we have to understand how the economy itself works. In a number of important ways, we don’t. Paul Romer, the incoming chief economist of the World Bank, made waves last week (at least on Twitter) with a new paper arguing that the field of macroeconomics has suffered through “more than three decades of intellectual regress.” At the Tax Policy Center event on Friday, former Fed economist Louise Sheiner ticked off a list of big questions that economists can’t answer: “We don’t understand why productivity has slowed so much. We don’t understand why labor force participation has fallen by so much. We don’t understand why companies aren’t investing now with such low interest rates.” With so many unanswered questions, Sheiner said, there is no way we can accurately estimate the impact of specific tax policies.

Trump, of course, acknowledged none of that uncertainty in his speech last week. He said his $4.4 trillion tax cut would translate to just$2.6 trillion in lost revenue, a figure based on the Tax Foundation’s model, which tends to show big, positive economic effects from tax cuts. Liberal groups have criticized the Tax Foundation model as “out of line with mainstream analysis,” and it is true that most other approaches, including both TPC models and the congressional one, show much smaller effects. It’s notable that the last time the country experimented with big tax cuts, during George W. Bush’s administration, government revenues fell far short of the most optimistic “dynamic” projections.

But the truth is that there is no single, widely accepted “mainstream” position. Until there is one, candidates — including Trump, but certainly not limited to him — will be able to keep picking whichever model they like best.

## Income growth

The Census Bureau had good news last week: After years of stagnation, median U.S. household income rose 5.2 percent in 2015, the fastest growth on record. If the census numbers are to be believed, the first five years of the economic recovery brought essentially no gains for U.S. households, but the sixth year nearly closed the gap left by the Great Recession.

But there’s reason to be a bit skeptical of the census figures. Not about the big picture; there is lots of evidence, from both public and private sources, that the economic recovery is reaching the pocketbooks of everyday American families. The long-term story told by the new census data — a brutal recession, a recovery that was slow to gain steam but that is finally taking hold — seems well founded. But the census numbers show essentially zero income growth in 2014 and then huge gains in 2015. That seems unlikely. Other sources, as Gary Burtless of the Brookings Institution wrote Friday, show income gains that were less dramatic in 2015 but that have been stronger in the recovery as a whole. That kind of slow, steady growth is more consistent with other economic data than the sudden acceleration depicted by the census figures.

Members of the Federal Reserve’s policymaking Open Market Committee will meet this week to decide whether to raise interest rates. Few economists expect them to do so amid shaky economic growth, slower hiring and signs of weakness in consumer spending. (They may also be hesitant to act so close to the presidential election.)

The real question, then, will be what the Fed signals about the future. Most measures of inflation are picking up, and incomes are rising. That would tend to lead the Fed to raise interest rates sooner rather than later, perhaps before the end of the year. But economic growth has consistently fallen short of the Fed’s expectations, and several policymakers have hinted they might be willing to wait longer than usual to raise rates in order to avoid killing the recovery prematurely. Janet Yellen’s press conference on Wednesday should give a sense of how she and her colleagues are weighing the conflicting signals.

## Last week on FiveThirtyEight

College dorms have gotten fancier, and college administrations have gotten bigger. But Temple University economist Doug Webber argued that the real driver behind rising tuitions at public universities is falling state spending on higher education.

Millions of American renters are forced to leave their homes every year. But getting more precise than “millions” is impossible because we don’t have good data on evictions. In the first part of a two-day series on housing, Andrew Flowers looked at an ambitious new effort to collect better data.

One reason there are so many evictions? The U.S. is in the middle of its worst housing affordability crisis in decades. In Day Two of his series, Andrew found that rents are rising, income growth is weak and two-thirds of poor families get no government assistance paying for housing.

## Elsewhere

Initial reports of the new income data said income in rural areas fell in 2015, even as urban and suburban households experienced strong gains. But Quoctrung Bui of The New York Times says those claims were based on a statistical quirk, and that better evidence shows rural incomes are rising, too.

Data quirks aside, however, many parts of the country really are struggling, as Binyamin Appelbaum, Patricia Cohen and Jack Healy report in The New York Times.

“Disruption” may be the buzzword in Silicon Valley, but the global economy is increasingly dominated by giant corporations. A special report in The Economist argues that’s a worrying trend for the economy.

The plunge in oil prices in 2014 was meant to be great news for the U.S. economy. But as Josh Zumbrun writes in The Wall Street Journal, new research from the Brookings Institution finds that the expected stimulus never came.

## Footnotes

1. Trump hasn’t said exactly which programs would be protected from cuts. Social Security, Medicare and Medicaid collectively account for close to half of federal spending. Defense accounts for another 16 percent, and veterans’ programs, which Trump has also pledged to spare, are another 4 percent. Interest on the national debt, which the U.S. must pay to avoid defaulting on its obligations, is another roughly 6 percent.

Ben Casselman is a senior editor and the chief economics writer for FiveThirtyEight.