I’m pleased to announce the publication of my new Cato Policy Analysis, “Marriage against the State: Toward a New View of Civil Marriage.”

As I note in the introduction, it’s quite rare that Congress ever considers marriage as a policy area in its own right. There are comprehensive health care bills, defense spending bills, farm bills, and civil rights bills, but no really comprehensive marriage bills.

Of course, this might be a good thing, but one of the side effects is that marriage policy can be haphazard in the extreme. Inconsistencies and surprises abound. Marriage influences welfare, immigration, tax law, child custody and support, and many others besides.

Are all of these things legitimate? A popular view among libertarians is that the federal government, and possibly the states, should get out of the marriage business altogether. It’s an approach with much to recommend it, but I can’t entirely agree. For at least some areas of public policy, marriage represents a barrier to government meddling in your financial, family, and intimate life. In these areas, it’s an unqualified good. Marriage is often a defense against the state, and as such, it’s something libertarians ought to want.

Consider child custody. All children born to a married couple are presumed to belong to them. You don’t have to do anything special to assert your paternity (or maternity). You are presumed to have it. This is probably for the best. Inviting the government to prospectively examine married couples’ fitness as parents is one of the most corrosive things I could imagine doing to the nuclear family.

Or consider the gift-tax exemption for married couples. Husbands and wives may gift one another money or property without limits, tax-free. It’s an important part of the financial independence that we are accustomed to having in our families, and it allows a family to conduct an interdependent financial life with dignity and autonomy.

Yet this same exemption, oddly enough, can make a legal divorce cheaper than the breakup of a never-married relationship. A married couple can divide their assets, including houses, cars, and other properties, before they split up. A never-married couple will often have to pay taxes on their pre-breakup transfers — making the government in effect a third party to their relationship. No one would want this for all couples, of course, least of all libertarians.

Extricating marriage from other parts of federal law won’t be easy, either. For some fairly complicated reasons that I explain in the paper, the only way to make the income tax fully neutral with respect to marriage — and also neutral across families with unequal income distributions between spouses — is to adopt a flat tax. While I share the view of many of my Cato colleagues that a flat tax is a good idea, the marriage-related consequences of our current tax system aren’t always appreciated as a reason to move in that direction. They should be.

As a third example, consider immigration. Marriage to a citizen considerably hastens the process of immigrating legally. Even if that process were not unconscionably slow (which I think it is), we would probably still want the immigration of marriage partners to be a high priority. Immigrant spouses of citizens are clearly integrated to some extent into American society. The American spouses’ own liberty interests are clearly implicated. And, perhaps best of all for critics of immigration, immigrant spouses’ numbers are relatively small in any case.

Lastly, and because I know a lot of you probably skimmed up to this point, I do discuss same-sex marriage. One of the more common arguments against same-sex marriage is that those who have moral objections shouldn’t be forced to subsidize same-sex unions with their tax money.

Let’s grant the basic justice of the argument (and never mind that Quakers, Buddhists, and others could morally object to our enormous defense spending!). Still, it’s not well known that by the best available estimates, federal same-sex marriage would leave the government in a better fiscal position, not a worse one. A good way to channel less federal money to same-sex couples is actually… to allow them to marry.

Why is this? Well, some married couples still pay a marriage penalty, and gay and lesbian couples obviously would too. More significantly, spouses’ incomes and assets are declared in the means testing for federal welfare programs. Marriage would exclude some gays and lesbians from these programs. They may want marriage anyway, but on balance, it’s clearly not for grabbing federal dollars.

I discuss quite a few other marriage-related issues in this Policy Analysis, and even so, it’s not remotely comprehensive. My goal is to suggest a new way of thinking about marriage, one that evaluates the effects of various marriage-related policies using the individual right to form a family as the standard. Not every aspect of federal marriage policy stands up, but some of them do. Let’s let a new conversation begin.