Long holiday weekends have always been ideal for either floating trial balloons or dumping bad news. In that vein, a Friday tweet from President Trump seems to suggest he is still considering signing an executive order indexing the capital gains tax to inflation. That would be a terrible political move for the already beleaguered president.

From a political perspective, any economic stimulus should be provided to the maximum number of people. That’s even more true in this instance, since much of the current drag on the economy could be ascribed to the president’s tariffs against imported Chinese goods. Indexing the capital gains tax would accomplish the opposite goal.

The conservative group Americans for Tax Reform recently released an analysis showing how indexation could give millions of Americans direct tax relief. Looking only at the three key Midwestern states whose voters will likely decide the 2020 election, the group found that nearly 1.1 million households in Pennsylvania claimed a capital gain in 2016, along with about 540,000 in Wisconsin and 750,000 in Michigan. Assuming that every household contains two adults and every adult is a regular voter — both highly favorable assumptions — that means between 1 million and 2 million voters in each state might get a tax cut in 2020 under indexation.

On the surface, that seems like a lot, but those numbers pale in comparison with the total number of voters. Nearly 6.2 million Pennsylvanians voted in 2016, along with 4.8 million Michiganders and nearly 3 million Wisconsinites — in each case around three times the number of voters who would benefit from indexing. By contrast, every one of those people would benefit from, for example, a payroll-tax cut. This begs the question: Why help one-third of the voters when you can help everyone?

There is an even bigger political argument against making indexation the 2020 GOP economic policy centerpiece, however: the direct personal benefit that Trump would likely receive. It’s clear even to indexation supporters that the biggest gains would go to people or entities that have held property or stocks for decades, especially those who owned such assets in the 1970s and 1980s when inflation was much higher. The Trump Organization apparently owns many such properties, including Trump Tower (purchased and built between 1979 and 1983) and Mar-a-Lago (purchased in 1985). The company apparently still holds interests in many other properties developed throughout the 1980s and 1990s. Assuming the president, the Trump Organization, or another entity in which he or his family has a personal interest, has retained ownership of those properties on a continual basis, Trump would likely reap tens of millions of dollars in tax benefits from capital gains indexation if and when those properties are sold.

In the current environment, when hostile media outlets and Democratic state attorneys general examine every move Trump makes, it is surely terrible politics to sign a legally-questionable executive order that would give the president and his family such a huge windfall. Republican economic policy, like Caesar’s wife, should be beyond suspicion that it is motivated by private interests rather than the public good. Indexation of capital gains does not pass that test.

Trump campaigned on draining the Washington swamp. Bypassing Congress to cut his own taxes by tens, if not hundreds, of millions of dollars is about as swampy as it gets. The economics of indexation might make sense in the long run, but the political optics in the short run are devastating. The president should ignore his advisers’ well-intentioned counsel and look elsewhere for economic stimulus.

Read more: