Krugman fundamentally misunderstands Sweden

Paul Krugman is profoundly inspired by Sweden. He has stated that the ideal society he dreams of is Sweden around 1980. Since Krugman is working to transform the United States in the image of another society, we would expect him to put a lot of effort in understanding his utopia. Unfortunately this does not seem to be the case. 

In a responce to Ross Douthats thoughtful column, Krugman writes “In Sweden, more than half of children are born out of wedlock — but they don’t seem to suffer much as a result, perhaps because the welfare state is so strong. Maybe we’ll go that way too. So?”

This is highly misleading. In secular Sweden, family traditions differ from those of the United States. Cohabitation (“samboförhållande”) is formally recognized and treated by the law as virtually identical to marriage. Swedish couples typically cohabitate, get children and only then get marry. Statistics Sweden explains:

“Living together without being married has long been common and majority of the children born in Sweden are born out of wedlock, but usually cohabiting, parents. Cohabitation can in many respects
equated with being married, and young adults has been widely accepting of couples with children remaining unmarried. Despite this, most couples choose to married eventually. Of the couples that are followed in this report and still lived together at the end of 2010, 73 percent married, while 27 percent were still cohabitating….About 10 percent of couples did not live together when the child was born, but most of these couples have lived together before or after birth. Approximately 3 percent of all couples never lived together and had a child outside of a relationship.”

So only ten percent of children in Sweden are born to couples who are not either married or co-habituating at the time of birth. Even that exaggerates, since many couples start cohabitation after birth. Obviously what we are interested in about is the child having two parents, not if they are in a christian marriage or secular Swedish cohabitation. Only 3 percent(!) of children in Sweden are born to single mothers. Swedes can afford to be so politically liberal ideologically because they are so socially conservative in their private behavior.

In fairness, when you read that half of Hispanics children are born out of wedlock, that too includes cohabitation. Accounting for later separation or divorce, according to the Census Bureau 18.7 percent of Swedish households with children are single-parent households (this share is lower among ethnic Swedes). Among Hispanics in the United States by contrast 37 percent of households with children are single-parent households. 

The New York Times itself writes: ”The share of Latino children living in single-parent families soared six percentage points to 38 percent since 2000, a larger increase than among blacks or whites.”
 Krugman thinks that children growing up in single parent households in Sweden (like me) “don’t seem to suffer much as a result, perhaps because the welfare state is so strong.” This too is incorrect. Just as in America, Swedish single parent households and their children have far higher prevalence of social  problems than intact families. This in-depth study of absolute poverty in Sweden finds

“The risk for poverty is more than three times as high among children of single parents compared with children of cohabiting parents, 28.2 and 9.1 percent in 2009.”

Prominent left of center economist Anders Björklund and co-authors directly compare outcomes for children living in non-intact families in the United States and Sweden: 

“In this paper we compare the relationships between family structure and children’s outcomes in terms of educational attainment and earnings using data from Sweden and the United States. Comparing the United States and Sweden is interesting because both family structure and public policy environments in the two countries differ significantly. Family structure could potentially have a less negative effect in Sweden than in the United States because of the extensive social safety net provided by that country. We find, however, the associations between family structure and children’s outcomes to be  remarkably similar in the United States and Sweden even though the policy and social environments differ between the two countries; living in a non-intact family is negatively related to child outcomes.”

In both Sweden and the United States, children of single-mothers earn less and are less likely to go to college. We don’t know for sure to what extent this is caused by single-parenthood itself or by confounding effects. In both countries single-parenthood is correlated with less social capital and other problems that both cause single-parenthood itself and other undesirable outcomes. This does not change the fact that population groups characterized by single parenthood have worse outcomes even in welfare states. 

Krugman thinks that because of the Swedish welfare state, it doesn’t matter if families arebroken, they and their kids do fine anyway. As shown above this is incorrect, Krugman is relying on utopian theories about how Sweden works rather than empirical analysis about how it actually works. 

The main problem is not that liberals like Krugman have childish fantasies about Sweden. It is that they are  proposing radical social engineering of the United States based on their superficial understanding of Sweden. Krugman’s response to Ross Douthats about the U.S moving towards non-intact families through the twin forces of demographic transformation and social breakdown was thus “Maybe we’ll go that way too. So?” works.  

Let me spell out what Ross Douthat is trying to say Paul. After decades of intense competition, liberals have finally and permanently triumphed against Conservatives. However you didn’t do it it by convincing existing Americans to adopt a welfare state. If the composition of the electorate was what it was a couple of decades ago, Romney would have won in a landslide. Liberals triumphed through the twin forces of demographics transformation and social breakdown. 

The defining characteristic of Sweden in 1980 was not that it had high marginal taxes; it was that it was a homogenous society with intact families, high trust and stratospheric levels of social capital. Because of the way through which the American left triumphed, rather than realizing your dream of recreating Sweden in 1980 you are going to end up with a more unequal, more socially dysfunctional and less cohesive America than we have today, let alone Sweden in 1980. 


National Review’s research-wonk Reihan Salam commented on this blog-post, concluding:

Suffice it to say, this isn’t the last word on the matter. There is near-invincible confidence in the potential of the Swedish social model to mitigate the consequences of family breakdown in some quarters, and no doubt some observers will conclude that it is the Swedish welfare state itself that holds families together.

We can shed some light on this matter through my usual method of using Americans who report Swedish ancestry as a control group. Using the American Community Survey 2006-2010, I estimate the percentage of unmarried households among households with children. 

Among Americans with Swedish ancestry born in the United States, the share is 18.1 percent. This is similar to the Swedish numbers, below the U.S national average and below the average for non-Hispanic whites. Let me emphasize that these are Americans with Swedish ancestry born in the United States. They hence live under the American system and are uninfluenced by the Swedish welfare state. The fact that Swedish-Americans have similar outcomes to families in Sweden indicates that culture and social capital are more important explanations for Swedish family stability than economic policy.

Comments are closed.