This study was co-authored by R Street Associate Fellow Neil Gilbert.


Marriage rates are down in the United States. Indeed, they are down throughout almost all of the West. For years, analysts have criticized tax and safety-net penalties for marriage, which they fear have contributed to this trend. After a number of ameliorative adjustments to the tax code in the 1990s and early 2000s, interest waned, partly because doing more seemed too expensive and technically challenging. At the time, little attention was paid to the marriage penalties (or bonuses) embedded in means-tested social-welfare benefits.

This analysis addresses the growing problem of marriage penalties created by the increased size and coverage of means-tested social-welfare benefits. Depending on the relationship between cohabiters (whether or not they have children in common and whether or not they share food or utility expenses) and their combined and relative earnings, getting married can result in bonuses of as much as 11 percent of their combined income or penalties of more than about 32 percent of their combined income. For example, in Arkansas (one of the states with the highest marriage penalties) if a nonparent marries a parent with two children, and they have a 50/50 split of $40,000 in combined earnings (counting benefits, a total income of $41,892), they would lose approximately $13,248 in annual means-tested benefits, or 32 percent of total household income.

Importantly, these penalties can be avoided by cohabiters who fail to disclose to authorities either that they are biological parents of the children in the household or that they share food and utility costs, a key point to which we will return.

The future of marriage is of concern not just due to nostalgia for the “good old days.” Analysts on both the left and the right believe that, all things being equal, getting and staying married is the most effective way to avoid poverty and the best way to raise children. For example, research by a team of economists from Harvard and the University of California at Berkeley suggests the fraction of children in single-parent families is the strongest and most robust predictor of upward mobility – even more than minority-group affiliation.

However, all things are rarely equal. A heavy-handed effort to promote marriage would not be in the best interests of many couples or their children. A strong marriage-promotion policy could be offensive to couples who do not wish to marry or can’t do so. It also could be offensive to those who advocate women’s complete economic freedom from men. Moreover, a serious attempt at such social engineering could cause a host of unintended consequences and some serious mischief.

Thus, while marriage in general is a social good, the prudent policy for government (as opposed to civil society) is neutrality. That is, government should get out of the way of marriage by minimizing marriage penalties. The issue is not how government policies could be used to encourage marriage, but rather, how policies might be modified to “neutralize” their potentially negative effects on decisions to marry. In addressing this issue, we will estimate the marriage penalties currently embedded in the U.S. tax code and, especially, in means-tested social-welfare programs; examine the implications of these penalties in the changing context of modern marriage and cohabitation; and analyze various options for limiting their potentially negative consequences.

We believe most people know what is best for themselves (at least better than does government), and we think it best to let the future of marriage rest on the individual decisions of millions of American couples. That is why we are concerned about marriage penalties embedded in the U.S. tax code and, especially, in means-tested social-welfare programs.