Fortune issue: January 24, 2000
THE CASE FOR SHACKING UP
Why Getting Married Still Costs You
By N. Gregory Mankiw
As the presidential election heats up, so does the debate over tax policy. With taxes at a historic high as a percentage of GDP, George W. Bush has proposed a Reaganesque cut in marginal tax rates. Steve Forbes wants to replace the whole tax code with a simpler flat tax. Al Gore decries these Republican ideas as "risky tax schemes" that put much needed social programs in jeopardy.
Yet on one tax topic, these candidates speak with a single voice: the marriage penalty. It is not right, we are told, that many couples pay higher income taxes simply because they are married. The marriage penalty is antifamily. It's unfair. It has to go. But if everyone is against it, why does it exist?
The reality of this issue is, as usual, more complicated than the rhetoric. According to a study by the Congressional Budget Office, 42% of married couples pay higher taxes because of the marriage penalty, averaging 2% of their income and adding $29 billion a year to the U.S. Treasury's coffers. But 51% of married couples actually pay lower taxes by virtue of being wed. This group gets an average marriage bonus of 2.3% of their income, costing the Treasury a total of $33 billion a year. So Americans get a bit more in marriage bonuses than they pay in marriage penalties.
Whether a couple is better off married or shacked up (from a tax standpoint) depends on how earnings are split between the two partners. If a man and woman have similar incomes, their wedding will almost certainly raise their tax bill. But a marriage bonus is likely if one partner earns much more than the other, especially if only one of them has earnings. Think of it as a tax on power couples, and a subsidy to those living the Leave It to Beaver life.
This strange policy, however, is the result not of a conscious design to return us to a 1950s vision of the family but of an uneasy compromise among competing goals. When considering what an ideal tax system would look like, most people are attracted to three principles:
Marriage neutrality: A couple's tax liability shouldn't change when they wed.
Equal treatment: A couple's tax liability should depend on total income, not on what each partner makes.
Progressivity: Taxpayers with higher incomes should face higher tax rates.
Here's the problem: As matter of mathematical logic, reaching all three goals at once is impossible.
To see why, imagine two couples. Mr. and Mrs. Smith are married, and each earns $50,000 a year. Ms. Williams and Mr. Jones live together but aren't wed. Ms. Williams earns the couple's entire income of $100,000, while Mr. Jones stays at home to write the great American novel. Equal treatment says these couples should pay the same tax. Marriage neutrality says their taxes shouldn't depend on whether they are married. But this means that Ms. Williams has to pay in taxes exactly twice what each Smith would pay if unmarried. (Remember: Mr. Jones pays nothing because he has no income.) Thus, to achieve marriage neutrality and equal treatment, we have no choice but to sacrifice progressivity.
U.S. lawmakers have responded to this conundrum by abandoning the principle of marriage neutrality, leading to our current system of marriage penalties and bonuses. In other countries, the common approach is to make the individual rather than the family the taxpaying unit; this preserves marriage neutrality and progressivity but forfeits equal treatment. Among the major U.S. presidential candidates, only Steve Forbes addresses the problem: His flat tax gives up progressivity.
Most candidates, by contrast, want to achieve marriage neutrality without giving up anything in the process. Unfortunately, it can't be done. The President can hope to change the laws of taxation. But the laws of mathematics aren't so easily repealed.
N. GREGORY MANKIW is an economics professor at Harvard and the author of Principles of Economics.