CAMBRIDGE -- This year’s presidential election campaign in the United States is certainly unique. Donald Trump has shaken up the way a campaign is run, how a nominee communicates with voters, and the Republican Party’s platform, with many of his positions deviating from GOP tradition. But, on tax policy, Trump has toed the party line -- and that’s not a good thing.

Of course, any assessment of Trump’s declared positions risks being rendered meaningless within a few hours. He changes his positions with head-spinning frequency, repeatedly disavowing statements soon after making them. After two days of insisting, absurdly, that President Barack Obama “founded” the Islamic State, he finally tried to play the whole thing off as mere sarcasm.

Even for a more typical candidate, promises made during a campaign rarely match actions taken once in office. During his presidential campaign in 2000, George W. Bush promised to renounce nation-building adventures abroad, to maintain fiscal discipline, and to treat greenhouses gases as pollutants under the Clean Air Act. Once in office, he did precisely the opposite.

Nonetheless, we cannot simply ignore candidates’ policy promises. Otherwise, the national discussion would focus entirely on the current week’s poll results, which is not particularly valuable for helping voters make informed decisions.

This is why it is important to examine Trump’s recent tax proposal, which has four salient features -- all of which amount to tax cuts for the rich.

First, Trump proposes to abolish the estate tax -- a long-cherished Republican goal. Trump, like his predecessors, tries to hide the fact that only the very wealthy would benefit from this change, because the tax does not apply to estates worth less than $5.45 million for an individual and $10.9 million for a married couple.

Second, Trump wants to cut America’s corporate tax rate sharply, to 15 percent. In theory, there is some merit to reducing the rate, which, at 35 percent, is one of the world’s highest (and probably spurs companies to keep profits overseas). But, as most tax policy experts will tell you, a reduction in the overall rate should be accompanied by moves to broaden the base. In particular, the US should abolish deductions, such as those designed to encourage corporate debt and oil drilling.

Third, Trump would reduce the highest rate for personal income tax from 39.6 percent to 33 percent. This is significant, but not nearly as large as the previous proposal to cut the rate to 25 percent -- which, according to independent analysts, would have led to about $10 trillion in revenue losses over the course of just one decade.

Finally, Trump has proposed tax deductions for average childcare costs. That might seem widely beneficial, but it would help only those in high enough tax brackets to itemize deductions -- mostly households with annual income above $75,000 (median US household income was $54,462 in 2015).

The question of whether Trump’s tax policies would benefit him directly remains unanswered, because, unlike all other presidential candidates since Richard Nixon, he has refused to release his income-tax records. What is certain is that Trump’s tax policies are fiscally irresponsible. Indeed, there is no indication how any of his tax policies would be financed, and every indication that they would sharply expand the budget deficit.

The budget balance equals government outlays minus revenues. It’s a simple equation, but one that Republican presidential candidates, including Trump, have seemingly had trouble understanding since 1980. They propose large specific tax cuts, without specific spending cuts, and yet claim they will reduce the deficit. Instead, they have produced record increases in budget deficits, particularly during the terms of Ronald Reagan and George W. Bush.

To be sure, Trump’s budget plans are still too vague -- particularly with respect to discretionary spending, Social Security, and Medicare -- to arrive at an informed estimate of their actual impact on the federal deficit and national debt. But he may face pressure to become more specific as the election approaches. And, judging by his tendency to offer pie-in-the-sky ideas, rather than realistic policies, it seems likely that he will respond to that pressure with more manipulation.

Fortunately for Trump, the self-proclaimed “fiscal conservatives” who preceded him have already established four magic tricks to use when making impossible fiscal promises -- tricks that he is likely to imitate.

-- The “Magic Asterisk.” The candidate promises to cut taxes while balancing the budget with as-yet-unidentified spending cuts. (This is basically what Trump is already doing, as he has specified no spending cuts.)

-- The “Rosy Scenario.” The candidate (or incumbent) forecasts an increase in tax receipts, based on forecasts that national income will grow faster. One of the three economists who have signed on as advisers to the Trump campaign has suggested that, with Trump in the White House, the rate of GDP growth will somehow double.

-- The “Laffer Proposition.” Reagan, Bush, and others declared that their tax cuts would spur so much economic activity that total tax revenue would actually increase. This proposition has been disproved many times, and the economic advisers to those presidents have disavowed it. Still, the temptation to square the budgetary circle with this claim may prove too strong for Trump to resist.

-- “Starve the Beast.” After the other justifications for big tax cuts were proved wrong, Reagan and Bush fell back on the theory that the decline in tax revenue was actually a good thing, because it would force Congress to approve spending cuts. Again, the key to this gambit is that the president never actually gets around to proposing the cuts that Congress is supposed to approve.

Perhaps it is inevitable for candidates to ignore the real-world constraints of domestic politics or international circumstances as they try to sell themselves to voters. But they shouldn’t be able to ignore the constraints of arithmetic, especially not when the same sleight of hand has been tried before -- with profoundly adverse consequences.

By Jeffrey Frankel

Jeffrey Frankel is professor of capital formation and growth at Harvard University. -- Ed.

(Project Syndicate)