Fox News host Bill O'Reilly falsely claimed that in Sweden, “marriage between men and women declined drastically since gay marriage was legalized there.” “These stats are irrefutable. They're government statistics,” O'Reilly insisted. In fact, recent demographics show that since Sweden passed its 1994 “Registered Partnership Law” establishing same-sex civil unions, marriage rates have increased. O'Reilly's false claim was apparently a reference to an argument by Stanley Kurtz, a research fellow at the conservative Hoover Institution. But Kurtz's analysis of the statistics has been debunked.

Despite O'Reilly's claims, “gay marriage” per se is not legal in Sweden, or in other Scandinavian countries. Those countries have versions of civil union laws that offer both heterosexual and homosexual unmarried couples most of the same rights as heterosexual married couples.

Further, contrary to O'Reilly's claim that the number of heterosexual marriages has declined in Sweden since the domestic partner law took effect in 1995, marriage rates for people aged 15-64 years rose to 7.0 per 1,000 people in 2000 from 6.0 in 1995, having fallen from 7.4 in 1990, according to figures from the Statistical Office of the European Communities, as noted in the September 2003 edition of Monthly Labor Review.

O'Reilly appeared to be referencing Kurtz's research, which one of his guests -- Sally Satel, a psychiatrist and resident scholar at the conservative American Enterprise Institute -- referred to directly. (O'Reilly claimed that “60, 65 percent of all Swedes are not married,” while Kurtz wrote that "[a]bout 60 percent of first born children in Denmark now have unmarried parents.") But in his research, Kurtz did not claim, as O'Reilly did, that marriage rates in Sweden (or other Scandinavian countries) have declined since legal recognition of same-sex couples. Rather, he claimed that such recognition has undermined the institution of marriage by encouraging cohabitation and out-of-wedlock births. In fact, even this weaker claim has been debunked.

Kurtz wrote in February 2004: “A majority of children in Sweden and Norway are born out of wedlock. Sixty percent of first-born children in Denmark have unmarried parents. Not coincidentally, these countries have had something close to full gay marriage for a decade or more.”

Here, Kurtz conflates correlation with causation. The data show that high cohabitation and out-of-wedlock birth rates are not even correlated, but Kurtz assumes that such correlation exists, then concludes causation from it. In fact, the data show that cohabitation and out-of-wedlock birth rates began rising in Scandinavia long before the enactment of homosexual partnership laws. This trend is apparently partially a result of laws governing heterosexual unmarried couples. In many Scandinavian countries, cohabitating heterosexual couples have most of the same rights as married couples, which obviously reduces incentives for a cohabitating couple to marry.

As a study by William Eskridge, Darren Spedale and Hans Ytterberg in the journal Issues in Legal Scholarship concluded:

The long-term trend in Scandinavia has been lower marriage rates, higher divorce rates, and higher rates of nonmarital births. This has been a trend lasting at least two generations -- long predating registered partnership laws adopted in 1989 (Denmark) and 1994 (Sweden). The trend has mainly been cultural and social. To the extent that law has made a difference, one would expect the liberalization of alternatives (cohabitation) and exit (no-fault divorce) to be the key legal developments contributing to these changes in marriage and divorce rates. In both Denmark and Sweden, these legal changes occurred between 1969 and 1980, and the data in Tables D-1 and S-1 reveal a close correlation between these particular legal changes and lower marriage rates and higher divorce rates. Less dramatic legal changes, such as state support for working women with children, have also contributed to rising births of non-marital children.

Can Kurtz convincingly show that registered partnerships accelerated the pace of change in marriage that had been going on for most of the twentieth century, and that had been dramatic since 1970? Because registered partnership laws have been in effect for almost 15 years in Denmark and almost 9 years in Sweden, these countries might offer laboratories for testing his hypothesis. If state-recognized same-sex partnerships “contributed” to the decline of marriage and the rise of illegitimacy, even if indirectly by reinforcing an expanded-choice norm, we would expect to see (ceteris paribus) something more than falling marriage rates, rising divorce rates, and rising non-marital birth rates in Denmark after 1989 and in Sweden after 1994; those were the trends before 1989 and 1994. Rather, we should expect to see marriage rates falling faster, divorce rates accelerating upward, and a surge in non-marital birth rates. The data reveal no such trend. Not only do the registered partnership laws in Denmark and Sweden not correlate to super-normal plunges in marriage rates and super-elevated divorce rates, but some of the trends move in the other direction. The 1990s see no stake through the heart of marriage--indeed, the institution shows renewed signs of life in the new millennium. It is Kurtz's hypothesis that dies, according to the data.

Kurtz's argument -- which O'Reilly embraces -- is that when legal rights are accorded to gays, through some unspecified process heterosexual couples begin bearing children out of wedlock and refuse to marry when they otherwise would. There are many government policies that alter the incentives to get and stay married, and thus have direct and unsurprising effects on marriage; for instance, the availability of no-fault divorce leads to more divorces, and laws providing the same rights for cohabitating heterosexual couples as married couples lead to fewer marriages, as couples make private commitments to each other without seeking the imprimatur of the state. But neither Kurtz nor O'Reilly has provided a plausible justification for their belief that extending rights to gay couples harms heterosexual marriage.

From the June 1 edition of Fox News' The O'Reilly Factor:

O'REILLY: Are you familiar with the Swedish study on marriage, ma'am?

DR. NADA STOTLAND (vice president, American Psychiatric Association): No -- which study are you referring to?

O'REILLY: OK. They did a sociological study in Sweden that said that marriage between men and women declined drastically since gay marriage was legalized there. And now up to, I don't know, 60, 65 percent of all Swedes are not married. So the institution of marriage, basically, in that country collapsed, because there was no tradition to it.

And you know, I was wondering before the -- your organization made your decision to come out and say this is a good thing, whether you took a look at that study.

STOTLAND: Well, I just got back from Stockholm a couple of weeks ago, and I didn't hear anything about loads of people not being married. Though we have a lot of trouble --

O'REILLY: These stats are irrefutable. They're government statistics.

STOTLAND: Well, we have a lot of stats about what's good for people's mental health. And that's what we go by.

O'REILLY: All right. But I think you may be making a mistake, and that's why I'm bringing it up. I'm not badgering.

If the institution of marriage collapses in a country of about six million, OK, maybe it wasn't because they legalized gay marriage. Maybe there were other factors. But there is no question the institution of marriage in Sweden has collapsed.

I don't think that's a good thing, and I think if it happens in America, it will lead to chaos. Would you agree with that?

STOTLAND: We have a 50 percent divorce rate in this country. Is marriage collapsing in this country without same-sex marriage?

O'REILLY: Well, you would -- you would then have to apply that would be -- apply to the gay marriage, too. They'd come in at the same rate.

STOTLAND: But are you saying that we're going -- that we have a crisis in heterosexual marriage now?

O'REILLY: I think we have some problems. I think we have some problems.

STOTLAND: Yes, well, relationships have problems. But it's good for people to be able to form them. Very important.

O'REILLY: All right. Dr. Satel, do you see my point about Sweden? Maybe I'm not -- maybe I'm wrong here, but I don't know if it's a good -- I think -- I think Dr. Stotland may be correct in the sense that it probably is a good thing for people to commit to each other, whether they're gay or straight. It doesn't matter.

But I'm coming at it from a societal point of view. I don't think it's a good thing if marriage collapses. And it has in Sweden, and some people say it's because the traditional marriage is thrown out the window. Go ahead.

SATEL: Yes, I understand. I know that study and I know Stanley Kurtz has worked on it. And it actually is somewhat controversial. I haven't looked at it in depth myself.