Gary Lee’s The Limits of Marriage is highly logical, wonderfully informative, and enviably written.  But is his functionalist view of the decline of marriage true?  Superficially, yes.  Fundamentally, no.

He correctly argues that most single moms would not be better off if they simply married the fathers of their children.  Why not?  To put it a little more harshly than Lee: Because the fathers of the children of single moms are rarely “father material.”  On average, they have low attachment to legal employment.  On average, they have little commitment to monogamy.  And on average, they commit a lot of crime and serve a lot of time.  Marrying such men is the height of folly.

Still, do you know what else is the height of folly?

1. Dating such men.

2. Falling in love with such men.

3. Having unprotected sex with such men.

Since most single moms have done all three, Lee’s view that they’re “acting rationally, doing the best they can under difficult circumstances” is highly implausible.

This is especially clear in view of the youth of most single moms.  If you’re 45 years old, with one last chance to have a baby, perhaps you should take whatever guy you can get.  Even if he abandons you, you’ll still  have a baby.

Contrarily, if you’re 20 years old and have yet to find a responsible man with whom to have children, the prudent course is to spurn irresponsible suitors and wait.  You’re young; maybe someone better will come along!  Biologically, you’ve can afford to delay at least a decade.  During that decade, moreover, you have many practical ways to raise your social status.  Get extra education.  Acquire more skills.  Search for a better job.  Relocate.  These don’t just provide greater financial security; by the power of homophily, they improve your pool of suitors.

Lee is admittedly skeptical of such self-help:

The declining marriage rate is just one of the problems afflicting poor, less-educated women that may adversely affect their children.   In McLanahan’s (2004) analysis, these women continue to have children because there are few “opportunity costs” for them – having children may limit their opportunities for further education, good jobs, higher incomes, economic security, and possibly marriage, but they are unlikely to attain these goals even if they don’t have children out of wedlock.  Having children isn’t a barrier to achieving things that can’t be achieved without children.

On reflection, though, it’s hard to understand his skepticism.  “Further education” and “higher incomes” are low bars.  Does Lee really think that single moms couldn’t plausibly get an AA degree if they had delayed motherhood?  Couldn’t plausibly have found a job that pays 25% more?  Sure, if “good job”=”investment banker” and “economic security”=”tenured professor,” then delaying motherhood rarely leads to “good jobs” or “economic security.”  Yet by normal standards, delaying motherhood remains an excellent catalyst for self-help.  Indeed, even if motherhood-delayers were unable to get better jobs, at least they could work full-time for ten years and build up a nest egg using whatever money they would have spent on their kids.

Lee could object that I’m ignoring his demographic evidence: While one poor woman can improve her situation by delaying motherhood, poor women in general cannot.  Demographics are especially unfavorable for poor black women:

[I]f every unmarried black man in the age range appropriate for marriage for a 25-year-old black woman were to get married, 32 percent of these women would still be without husbands.  (Yes, they could theoretically marry white men, of whom there is a slight numerical surplus, but interracial marriages are far more likely to involve black men and white women than the reverse.  See Wang [2012].)

On inspection, however, these results heavily depend on Lee’s stipulation that women must marry men close to their age.  All else equal, this is an understandable mate preference.  But all else is not equal.  If men your own age are too immature to be good fathers – or simply too unproven to evaluate – common sense tells you to give moderately older men a chance.

Not only are men who are five or ten years older more mature on average; they have a long track record you can use to predict their long-run behavior.  A 30-year-old man who has been consistently legally employed for ten years straight is highly unlikely to suddenly become a violent criminal.  Unlike Lee’s interracial marriage scenario, moreover, many single employed older men are already interested in marrying a moderately younger woman.  Of course, if the woman is lower-class, the older man she marries is also likely to be lower-class.  Still, there’s more than a marginal difference between marrying a mature man who consistently works in low-skilled jobs, and marrying a teenager who spends half his time in jail.

The last big problem with Lee’s story is that he hastily accepts the mainstream view that real wages have sharply fallen for low-skilled Americans.  Once you take CPI bias seriously, however, this is highly debatable.  The best data says that even poor Americans who work full-time are markedly richer than they were in the 60s, when single motherhood was rare indeed.

I suspect that Lee would reply, “How then would you explain the large increase in single motherhood?”  He might even quote himself:

Both non-marriage and single parenthood have increased much more rapidly among poorer people.  Any explanation of the “retreat from marriage” in America over the past half-century must focus on the categories of the population whose marital behavior has actually changed.  We know who is not marrying today: poor, less-educated people, particularly but not exclusively if they are members of minority groups.  What is it that is keeping them from marrying?  Are they, as Waite and Gallagher (2000) assume in their concluding chapter, unaware of the benefits that marriage would confer?  Do they devalue attachments to others, and value a more hedonistic lifestyle that can best be achieved without the encumbrance of a spouse?  Are they avoiding responsibility?  Or are they perhaps acting rationally, doing the best they can under difficult circumstances?

Here is a rough sketch of my preferred story:

1. All humans are somewhat impulsive, but the degree of impulsiveness varies.

2. On average, impulsiveness causes poverty.  The greater the impulsiveness, the greater the poverty.  So the very poor tend to be highly impulsive.

3. In traditional societies, however, social pressure and stigma against impulsive behavior sharply reduce their incidence.  Teens like unprotected sex, but fear social suicide.

4. In the 1960s, social pressure and stigma against single motherhood started to deteriorate for largely cultural – not economic – reasons.  (While the expansion of the welfare state was one notable economic factor, it wasn’t decisive).

5. The decline in social pressure and stigma have gradually snowballed ever since.  (Welfare reform probably turned back the clock, but only marginally).

6. People with low impulsiveness continued to avoid single motherhood despite declining traditionalism, hence their continued rarity among the middle and upper classes.

7. People with high impulsiveness, however, have increasingly acted on their impulses ever since.


I say my story explains everything that Lee’s story explains – and more.  Like Lee, I can explain why single motherhood rose so much for the poor, and so little for the rest of society.  Like Lee, I can explain why single moms rarely escape poverty by marrying the fathers of their children.  Unlike Lee, however, I can accept the obvious fact that having unprotected sex with irresponsible young men is a great way to mess up your life.  Unlike Lee, I can acknowledge the many ways that even poor women in the First World can improve their lives with planning and hard work.  Unlike Lee, I can explain why absent fathers usually move on to new women, instead of doing their best to provide for the kids they already have.  And unlike Lee, I don’t have to claim that the American poor were materially better off in the 60s.

Still, shouldn’t Lee get credit for not appealing to exogenous “cultural shifts”?  I think not.  Yes, economists like the idea that economic forces determine everything.  Frankly, though, only severe hindsight bias allows economists to believe such extraordinary claims.  Example: Religion has declined as wealth has increased.  Yet this was hardly a foreseeable result of economic growth.  Why couldn’t we have a rich society where traditional religion remains culturally focal – where we consume our wealth with opulent cathedrals, pilgrimage vacations, and hundreds of channels of religious programming?  The same goes, I say, for the decline of traditionalism more generally.  The world of The Jetsons – which combined futuristic technology with the ethos of the 1950s – didn’t happen.  Yet economically speaking, it totally could have.