Trump has been talking about this.  I am not sure what version of the idea we might end up with, but let’s consider the idea in its abstract form.  Let’s also put aside money laundering issues, and talk about “simple remittances.”

The United States has a partial monopsony power over Latino (and often other) migrants, as there are few comparable places to go.  Some may switch to Spain, or stay in their home countries, but many will have to pay the tax, though of course they may send less money back home.

If elasticities were zero (unlikely), the US government would pull in 3.5 percent of the relevant flow of remittances.  More likely, funds sent will decline, and tax revenue earned will decline as well.  The former effect will strengthen the dollar against the Latin currencies, while the latter effect may weaken the dollar through the indirect mechanism of domestic output being lower.  I think most economists would expect the dollar to strengthen on net, as in essence the tax makes it costlier to sell dollars.  (As a side effect, the tax might accelerate a transition to weirder, harder to tax forms of crypto?)

So US exporters suffer a modest amount from the stronger USD, and US consumers gain from modestly cheaper imports.  Family members back home in the receiving countries are worse off, as they are receiving less in terms of real transfers, due to the tax.

It is less clear how much the receiving countries are worse off on net.  This is easiest to see in the case of El Salvador, which has dollarized.  If fewer remittances are sent to El Salvador, other dollar holders in the country may be better off.  In this regard remittances have a partial zero-sum component.  It is not right to say they are purely zero-sum, because the remittance is “the market” sending funds to where the demand to hold those funds is highest, and furthermore El Salvador as a whole has greater net command over imports.  Still, from another point of view it is a kind of domestic inflation for El Salvador and it taxes their cash balance holders.

In any case, you also can think of this as a funny, quite indirect way of auctioning off the right to come and send money back to your family.  Gary Becker once suggested a more direct auction of entry rights, an idea broadly popular with many economists.  This particular form of the auction maintains an ongoing tax on the margin, rather than a once and for all payment up front, and thus might involve higher distortions. (on the other hand it eases credit constraints, since you do not pay up front)  As a side note, this particular form of the auction mechanism also might discourage most of all the more altruistic and family-oriented migrants to a modest degree, or encourage some to try harder to bring their families with them.  Those could be pretty small effects, but substitution effects are always worth noting.

As someone with broadly libertarian sympathies, I am strongly opposed to this tax.  I think often the best way to analyze a tax is not with traditional deadweight loss tools, but rather to ask “does this allow the government to get its paws on a whole new source of revenue?”  If it does, be very suspicious.

But if you are not libertarian in that manner, I do not see why you should hate this tax.  It harms migrants and their relatives back home, but without necessarily harming those countries on net.  And international trade economics, and economics more generally, has a long tradition of “nationalistic” points of view that focus on maximizing domestic welfare, not global welfare.  I see those pop up all the time — for decades — without people screaming bloody murder (I am myself more Parfitian on these issues of course.)

Most Democrats I know really want to raise taxes.  Many centrists feel the same way, though perhaps less strongly.  So why should they hate this tax hike so much?  My views on taxes differ, though I recognize that sometimes you have to raise taxes.

I think, at least in this case, that the broadly libertarian principles are the relevant factor here.  I do not want the US federal government getting its paws on remittances as a revenue source.  In turn, I hope other opponents of this policy — and I suspect there will be many — join me and become slightly more libertarian, and slightly more willing to focus on the question “does this allow the government to get its paws on a whole new source of revenue?”

We will see.