Remittances Leave the Country by the Billions. Should Washington Tax Them

Here is a policy idea that sounds simple at a rally and turns into a thicket the moment anyone asks a follow-up question: tax the money that immigrants send home.

The slogan writes itself. Money earned here, spent there. Why shouldn't Uncle Sam take a cut on the way out? Half the room cheers. The other half, which includes some of the most reliable free-market voices on the right, folds its arms. Both halves are conservatives. So this is not a post about whether the left is wrong. It is a post about a real disagreement among people who share a goal, and I am going to give both positions a fair hearing before I tell you which one I find stronger.

What a remittance is

Define the term before you argue about it.

A remittance is money that a person who lives and works in one country sends to family or others in another. In the context of immigration, it usually means wages earned in the United States and sent to relatives in the worker's country of origin: a parent, a spouse, children, siblings. The money might travel through a bank, a money-transfer company, a mobile app, a cash courier or an informal network of agents.

How big is it? The World Bank tracks remittances every year and publishes the figures in its migration and development reporting. Its numbers come largely from the balance-of-payments statistics that national central banks report, following international accounting standards, with categories for personal transfers and for compensation of employees who work abroad. Those numbers have limits, and the World Bank says so: informal channels are hard to count, and countries differ in how well they record flows. I will not quote a dollar figure here, because the numbers change every year and because the argument does not depend on one. What matters is the scale and the pattern.

The scale is large. For some countries in Central America and the Caribbean, remittances amount to a substantial share of the entire national economy, and for many developing countries the flow has for years exceeded the total of official development assistance from rich governments. That fact is the starting point for both sides of the argument. One side sees a leak. The other sees a lifeline that does more good per dollar than anything our foreign-aid agencies run.

The first conservative position: tax it, or at least track it

Here is the case for a tax, stated the way its defenders would put it.

One: it raises money for enforcement. If a country wishes to secure its border and enforce its employment laws, it needs revenue. A modest levy on outbound transfers is a user-fee-like source of funds that falls on a flow that is plainly connected to labor performed here. Supporters would earmark it for the border, for the courts that handle immigration cases or for verification systems.

Two: it discourages unlawful work. The argument is about incentives. A person who works without authorization is partly motivated by what he can send home. A tax on the transfer reduces the payoff of that work, and a transfer requirement that includes proof of identity and lawful status makes it harder for people without authorization to move money easily. Tracking, on this view, is a way of making the law bite where the incentive lives.

Three: money that leaves is not spent here. A dollar sent abroad is a dollar that does not pay rent, buy groceries or fill a gas tank in an American town. Supporters say a flow of this size is a drag on the communities where the work is done.

Four: the principle of reciprocity. Several countries that receive large remittance flows are also countries that make life difficult for American businesses or fail to cooperate on enforcement. A tax is a way of making the benefit to those governments less free.

That is a coherent position. I would add that a version of it has moved from slogan to statute. Congress has taken up the idea in recent years and enacted a limited excise tax on certain remittance transfers, aimed at transfers funded with cash, with exemptions for transfers paid from bank accounts or by certain cards. The details are in the Internal Revenue Code and in Treasury guidance, and they may change, so check the current text before relying on anything I say about the rate or the scope.

The second conservative position: leave it alone

Now the case against, from conservatives who are just as serious.

One: it is the worker's own money. The wages were earned here, the income tax was withheld or owed here and the worker is entitled to spend what is left as he sees fit. A tax on a transfer is a second tax on income that has already been taxed. We are, after all, the party that objects to double taxation. The estate tax gets the same objection, which I take up in the death tax, the family farm and the case for ending a tax on dying. A transfer tax is a tax on moving your own property from one pocket to another, and a conservative who dislikes taxes on savings and inheritance should hesitate before inventing a new one on generosity.

Two: it pushes the money underground. This is the practical heart of the objection. Remittance providers are regulated. They are subject to the Bank Secrecy Act and to the anti-money-laundering rules the Treasury's Financial Crimes Enforcement Network administers, and to a federal consumer rule issued under the Dodd-Frank Act that requires disclosures about fees and exchange rates. A transfer through a regulated company leaves a record. A tax gives people a reason to use the channels that leave no record: cash carried by couriers, informal agents who settle accounts across borders, and crypto-currency transfers. If the objective is visibility, a tax reduces it. The tax that was meant to track the money drives it into the dark.

Three: it lands on lawful workers. A transfer tax cannot tell a lawful worker from an unlawful one without a system for verifying status at every counter. Without that, the tax falls on everyone who sends money: green card holders, temporary agricultural workers, students, citizens who were born abroad and whose parents live there. The tax is not narrowly aimed. It is a levy on a population that includes many people who followed every rule.

Four: the "leak" argument is weaker than it looks. Economists point out that the worker who sends part of his pay home usually spends the rest of it here, on rent, food, transport and clothes, and that the money sent abroad does not vanish from the world economy. Dollars that go abroad come back as foreign purchases of American goods, services and financial assets. The balance of payments has to balance. A dollar sent to a family in Honduras becomes, in due course, a dollar paid for something American. It is not a loss in the way a stolen purse is a loss.

Five: development at home reduces pressure to leave. A mother in a poor country who can buy food and send a child to school on a relative's remittance is a mother who is less likely to decide the only option is to send the child north. If a tax reduces the flow, it may also reduce the family's means of staying put. Conservatives who want fewer people to cross the border illegally have a reason to care about whether the home economy offers a decent alternative, and remittances are one of the ways it does.

A worked example

Let me do the arithmetic, because policy debates go better with numbers. This is a hypothetical, labeled as such.

Picture a lawful worker who earns a wage and sends $400 a month home to his parents. Over a year that is 12 x $400 = $4,800.

• At a 1 percent tax, he pays $4 a month, or $48 a year.
• At a 3.5 percent tax, he pays $14 a month, or $168 a year.
• At a 5 percent tax, he pays $20 a month, or $240 a year.

On top of the tax sits the fee the transfer company already charges. The international target for the average cost of sending remittances, set under the United Nations' development goals, is three percent, and the actual global average has for years been above that. So a 1 percent tax is an addition of a third to the target cost and a smaller addition to the real cost.

What does the example show? That a small tax is a small burden for any one worker, and a very large burden if it is paid by millions of people across the country, which is the point of it as a revenue measure. It also shows that the incentive effect on the decision to work without authorization is weak for a small tax: no one risks a border crossing to save $48. A tax large enough to change that decision would be large enough to push money underground. Those two facts together are the weak point of the tax argument. The rate that is low enough to be tolerable is too low to deter, and the rate that is high enough to deter is high enough to evade.

What the Constitution has to say

There is little controversy about Congress's power here. Article I, Section 8, Clause 1 gives Congress the power "to lay and collect Taxes, Duties, Imposts and Excises," and a levy on a financial transaction is an excise. The Sixteenth Amendment is about taxes on incomes, not excises, and the tax does not have to be apportioned among the states by population. The question is not whether Congress can do this. The question is whether it should, and a federal system leaves that to the people's representatives. Any such law would also have to respect the ordinary constitutional limits on how the Treasury collects information, and on the privacy of financial records, so the design matters.

There is a Founding-era reason to be careful about a levy on the movement of property. The colonists' grievances, listed in the Declaration, included "imposing Taxes on us without our Consent." A tax on people who cannot vote, and who have no representative, is a harder sell to anyone who takes that grievance seriously. I say that not to end the argument, since non-citizens who live here are subject to many of our taxes, but to note that a principle of representation runs through the debate.

What would work better

A reader could fairly ask what I would do instead, so let me answer. I find the case against a broad remittance tax stronger than the case for it, for the reasons above, and I think the aims of the tax's defenders are better served in other ways:

1. Enforce the law at the workplace. The employer is the point of control. A system that makes it hard to hire without verification removes the unlawful job at its source, which is the same lesson as in The Conservative Split Over Mandatory E-Verify, Argued From Both Sides and The Employer Who Hires Off the Books Is Breaking the Law, Not Helping. If there is no unlawful job, there is no unlawful wage to send home.
2. Keep the transfers visible. The existing tools, the Bank Secrecy Act reporting, identification at larger transfers and the consumer disclosure rules, put the flow where investigators can see it. Strengthen those tools, give investigators the resources to use them and go after the networks that launder cartel proceeds under cover of ordinary remittances. A visible flow serves enforcement better than a taxed one.
3. If there must be a levy, aim it narrowly. A fee on cash-funded transfers, which are the ones that leave the least trail, is easier to defend than a charge on every transfer, because it does not hit the worker who uses a bank account and it nudges people toward the channels that leave records. Whether the actual design does that is a question for the text of the statute.
4. Compete on cost. Lowering the cost of legitimate remittances, through competition and better technology, brings more money into the visible system, and that has a stronger claim on the conservative faith in markets than a tax.
5. Hold foreign governments to their side. A country that wants the remittances to keep flowing has an interest in cooperating on returns of criminals and on border enforcement. That is a diplomatic lever, which is more precise than a tax. For the broader picture of how neighbors handle this, see How Japan, Denmark, Switzerland and Mexico Guard Their Own Borders.

The strongest remaining objection to my view

I owe you the best argument against my own conclusion.

You say the rate that deters is the rate that drives money underground. But we do not abandon other taxes because some people evade them. Cash is hard to track, yes, but a tax creates a cost and a paper trail at the point of the transaction. And if it raises revenue for enforcement, and nothing else, it has paid for itself. Opposing it because it is imperfect is the kind of reasoning that leaves every policy unenacted.

That is fair, and the narrow, cash-only version of the tax is my concession to it. If the revenue is real and the burden on lawful workers is limited, a conservative can support it as a modest measure. My objection is to treating a tax as a substitute for what actually works, enforcement at the point of hire and visibility in the financial system, and to describing a small levy as though it would stop a flow it will barely dent.

There is also a plain moral point to make before leaving the subject. Behind the figures are real people: a mother, a father, a child, a sibling who are being supported by someone far away. A policy that treats the worker as the problem will fail, because the worker is mostly doing what a decent person does. The problem is the unlawful hiring and the unsecured line, and those are the things a serious law should address.

The case in brief

1. Remittances are money sent home by workers abroad, measured by the World Bank mainly from central-bank balance-of-payments data, and they are large enough in several countries to exceed official development aid.
2. Supporters of a tax argue that it raises money for enforcement, discourages unlawful work and keeps wealth in American communities, and a limited excise on certain cash-funded transfers has been enacted.
3. Opponents argue that a tax is a second tax on income already taxed, pushes transfers into untraceable channels, lands on lawful workers and misunderstands how the balance of payments works.
4. The arithmetic shows a low rate is too small to deter unlawful work and a high rate is large enough to drive money underground, so the tax is a weak tool for its stated purposes.
5. Enforcement at the workplace, visibility through existing financial rules and competition on cost serve the same goals better, with a narrow tax on cash transfers as the most that I think a conservative can easily defend.

What you can do: read the current text of the remittance excise tax in the Internal Revenue Code and Treasury guidance, and ask your representative, through the my-representatives page, whether the money would be better spent on workplace verification and financial-crime enforcement than on a levy that evasion can avoid.