“Marriage promotion” is a destructive cargo cult
I think I’ll basically be repeating what Matt Yglesias said yesterday, but maybe I can put things more plainly.
“Marriage promotion” as a means of address social problems at the lower end of the socioeconomic ladder is a bad idea. It’s not a neutral idea, or a nice idea that probably won’t work. It’s inexcusably obtuse and may be outright destructive. It is quite literally a cargo cult.
A cargo cult is a particularly colorful way of mistaking cause for effect. Airplanes do not actually come to remote Pacific Islands because of rituals performed by soldiers at airports. But absent other information, to someone with no knowledge of the larger world, it might well look that way. So when the soldiers leave and the airplanes full of valuable stuff no longer come, it’s forgivable in its way that some islanders populated the abandoned tarmacs with wooden facsimile airplanes and tried to reenact the odd dances that used to precede the arrival of wonderful machines. It is forgivable, but it didn’t work. The actual causes of cargo service to remote Pacific Islands lay in hustle of industries vast oceans away and in the logistics of a bloody war, all of which were invisible to local spectators. Soldiers’ dances on the tarmac were an effect of the same causes, not an independent source of action. That is not to say those dances were irrelevant to the great bounty from the skies. An organized airport is part of the mechanism through which the deeper causes of cargo service have their effect, so something like those dances would always be correlated with cargo service. But even a perfectly equipped and organized airport will not cause airplanes sua sponte to deliver valuable goods to islanders. A mock facsimile even less so.
The case for marriage promotion begins with some perfectly real correlations. Across a variety of measures — household income, self-reported life satisfaction, childrearing outcomes — married couples seem to do better than pairs of singles (and much better than single parents), particularly in populations towards the lower end of the socioeconomic ladder. So it is natural to imagine that, if somehow poor people could be persuaded to marry more, they too would enjoy those improvements in household income, life satisfaction, and childrearing. Let them eat wedding cake!
But neither wedding cake nor the marriages they celebrate cause observed “marriage premia” any more than dances on tarmacs caused airplanes to land on Melanesian islands. In fact, for the most part, the evidence we have suggests that marriage is an effect of other things that facilitate good social outcomes rather than a cause on its own. In particular, for poor women, the availability of suitable mates is a binding constraint on marriage behavior. People in actually observed marriages do well because they are the lucky ones to find scarce good mates, not because marriage would be a good thing for everyone else too. Marrying badly, that is marriage followed by subsequent divorce, increases the poverty rate among poor women compared to never marrying at all. Married biological parents who stay together may be good for child rearing, but kids of mothers who marry anyone other than their biological father do no better than children of mothers who never marry at all. As McLanahan and Sigle-Rushton put it (from the abstract):
[U]nmarried mothers and their partners are vastly different from married parents when it comes to age, education, health status and behaviour, employment, and wage rates. These differences translate into important differences in earnings capacities, which, in turn, translate into differences in poverty. Even assuming the same family structure and labour supply, our estimates suggest that much of the difference in poverty outcomes by family structure can be attributed to factors other than marital status. Our results also suggest that full employment is essential to lifting poor families — married or otherwise — out of poverty.
Let’s stop with the litany of citations for a minute and just think like humans. Marriage is a big deal. The stylized fact that the great preponderance of grown-ups with kids who seem economically and socially successful are married is known to everybody, rich and poor, black and white. Yes, the traditional family is not uncontested. There are, in our culture, valorizations of single-parenthood as statements of feminist independence, valorizations of male liberty and libertinism, aspirational valorizations of nontraditional families by until-recently-excluded gay people, etc. But, despite the outsized role played by Kurt on Glee, these alternative visions are numerically marginal, and probably especially marginal among the poor. Single motherhood is the alternative family structure that matters from a social welfare (rather than culture-war) perspective. The problem marriage promotion could solve, if it could solve any problem at all, would be to increase the well-being of the people who currently become single mothers and of their children.
But why do single women choose to become single mothers? It does not, in any numerically significant way, seem to have much to do with purposeful rebellion against traditional family norms. No, marriage of poor women seems constrained by the availability of promising mates. And why might that be?
Charles Murray recently wrote a wonderful, terrible, book called “Coming Apart“. The book is wonderful, because it identifies and very sharply observes the core social problem of our time, the Great Segregation (sorry Tyler), or more accurately, the Great Secession of the rich from the rest, and especially from the poor. The book is terrible, because it then analyzes the problems of the poor as though they come from nowhere, as though phenomena Murray characterizes as declines in industriousness, religiosity, and devotion to marriage among the poor have nothing to do with the evacuation of the rich into dream enclaves. There are obvious connections that Murray doesn’t make because, I think, he simply doesn’t wish to make them. Let’s make some. We were talking about marriage.
Murray does a wonderful job of describing the homogamy of our socioeconomic elites. The people who, at marriageable age, seem poised to succeed economically and socially, tend to marry one another. Johnnie doesn’t marry the girl next door, who might have been a plumber’s daughter while Daddy was a bank manager. Johnnie doesn’t marry anyone at all he met in high school, but holds out for someone who got into the same sort of selective college he got into. The children of the rich marry children of the rich, with notable allowances made for children of the nonrich who have accumulated credentials that signal a high likelihood of present or future affluence. Of course, love knows no boundaries.
As a matter of simple arithmetic, increasing homogamy among the elite and successful implies a reduced probability that a person who cannot lay claim to that benighted group will be able to “marry up”, as it were. Once upon a time, in the halcyon days that Murray contrasts to the present, the courting would not have been so crass. There were many fewer markers of social class and future affluence. The best and brightest were not so institutionally, geographically, and culturally segregated from the rest. (That is, within the community of white Americans. For black Americans, all of this is old hat.) The risk of “mismarrying”, for a male, was not so great, as he would be the primary breadwinner anyway, and her family, while perhaps poorer than his own, was unlikely to be in desperate straits. Men could choose whom they liked, in a personal, sexual, and romantic sense without great cost. Women from poor-ish backgrounds had a decent chance at landing a solid breadwinner, if not the next President. Very much like an insurance pool, a large and mixed pool of potential spouses renders marriage on average a pretty good deal for everyone. Really bad future husbands existed then as now, and then as now women were wise to do all they could to avoid marrying them. But the quality of a marriage is never revealed until well after you are in it. In a middle-class society, it was reasonable for a woman to guess that a nice guy she could fall in love with would be able to be a good husband and father too.
Flash-forward to the present. We now live in a socially and economically stratified society. By the time we marry, we can ascertain with reasonable confidence what kind of job, income, neighborhood, and friends a potential mate is likely to come with. The stakes are much higher than they used to be. Our lifestyle norms are based on two-earner households, so men as well as women need to think hard about the earning prospects of potential mates. Increasing economic dispersion — inequality — means that it is quite possible that a potential mate’s family faces circumstances vastly more difficult than ones own, if one is near the top of the distribution. It is unfashionable to say this in individualistic America, but it is as true now as it was for Romeo and Juliette that a marriage binds not only two people, but two families. If you have a good marriage, you will love your spouse. If you love your spouse and then her uninsured mother is diagnosed with cancer, those medical bills will to some perhaps large degree become your liability. More prosaically, if the inlaws can’t keep the heat on, do you wash your hands of it and let them shiver through the winter? In a very unequal society, the costs and risks of “marrying down” are large.
As with an insurance pool, too much knowledge can poison the marriage pool, and reduce aggregate welfare by preventing distributive arrangements that everyone would rationally prefer in the absence of information, but which become the subject of conflict when information is known in advance. Because the stakes are now very high and the information very solid, good marriage prospects (in a crass socioeconomic sense) hold out for other good marriage prospects. The pool that’s left over, once all the people capable of signaling their membership in the socioeconomic elite have been “creamed” away, may often be, objectively, a bad one. Marriage has a fat lower tail. When you marry, you risk physical abuse, you risk appropriation of your wealth and income, you risk mistreatment of the children you hope someday to have, you risk the Sartre-ish hell of being bound eternally to someone whose company is intolerable. More commonly, you risk forming a household that is unable to get along reasonably in an economic sense, causing conflicts and crises and miseries even among well-intentioned and decent people. It is quite rational to demand a lot of evidence that a potential mate sits well above the fat left tail, but the ex ante uncertainty is always high. When the right-hand side of the desirability distribution is truncated away, marriage may simply be a bad risk.
If you are at all libertarian, what the behavior of the poor tells you is that it is a bad risk. After all, marriage is not subject to a Bryan-Caplan-esque critique of politics, where people make bad choices in the voting booth that they would not make in the supermarket because they don’t own the costs of a bad vote. The consequences of a decision to marry or not to marry or who to marry are internalized very deeply by the people who make them. Humans, rich and poor, have strong incentives to try to make those choices well. Both common sense, social science, and revealed preference suggest that marriage rates among the poor have declined because the value of the contingent claim upon the future represented by the words “I do” has also declined within the affected population.
Promoting marriage among this population is not merely ineffective. It is at best ineffective. If the marriage-promoters persuade people to marry despite circumstances that render it likely they will marry poorly, the do-gooders will have done outright harm. Pacific Islanders no doubt bore some cost to build their wooden planes, lashed to a mistaken theory of causality. But lives were not destroyed. Overcoming peoples’ well-founded misgivings about the quality of potential mates with moral exhortations and clipboards of superficial social science might well destroy lives. It would create plenty of success stories for marriage promoters, sure, because even bad bets turn out well now and again. But it would create more tragedies than successes, tragedies that very likely would be blamed on personal deficiencies of the unhappy couple while the successes would be victories for marriage itself in some insane ideological version of the fundamental attribution error.
Fortunately, people aren’t stupid, so marriage promotion is more likely to be ineffective than devastating. But why go there at all? There is some evidence, for example, that where prevailing social norms prohibit premarital fun stuff and push towards early marriage, people do marry earlier and they marry poorly. Social norms matter, and even smart people are sometimes guided by them to do stupid things. Let’s not reinforce foolish norms.
None of this is to say marriage is bad! On the contrary, despite my lefty hippie enthusiasm for transgressive goat sex and stuff, I think in the context of the actually existing society, the prevalence of durable marriages is a reflection of social health. Marriage is part of how we organize a good life when a good life is on offer, just like airports with people guiding planes on the tarmac are part of how Pacific Islanders might organize trade for valuable cargo. But before the odd dances on the tarmac must come the production of goods and services for trade, or at least some kind of arrangement with the people in faraway places who control the airplanes. Before you get to smiling families, you have to create the material circumstances that render marriage on average a good deal. For poor women in particular, it very often is no longer a good deal.
But what about the children? One variant of marriage-centric social theory refrains from pushing marriage so hard, and simply asks that people delay childrearing until the marriage comes. (See e.g. Reihan Salam for some discussion.) If a woman is likely to find a good spouse at a reasonable age, then it might make sense to suggest she delay childbearing until the happy couple is stable and married, since kids reared by married biological parents seem to do better than other kids. Even that is subject to a causality concern: Perhaps childrearing is best performed by the kind of mother capable of finding a good mate, and at a time some unobservable factor renders her both ready to raise a child well and likely to take a husband. This would create a spurious correlation between the presence of biological fathers and good kid outcomes. We can’t rule that out, sure. But we have no reason to think it’s so, and lots of common sense reasons to think a biological father in a stable marriage improves outcomes by contributing to better parenting. So, I’d agree that women likely to find great marriage partners should by all means delay children until they have actually found one.
But women likely to find great marriage partners already do exactly that. Single motherhood is not a frequent occurrence among women who expect to marry happily and soon. The relevant question is whether we should discourage from having children women who reasonably expect they may not find a good spouse at all, at least not while they are in their youth. That is to say, should we tell women who have been segregated into the bad marriage market, who on average have lowish incomes and unruly neighbors and live near bad schools, that motherhood is just not for them, probably ever? We could bring back norms of shame surrounding single motherhood, or create other kinds of incentives to reduce the nonadoption birth rate of people statistically likely to raise difficult kids. It is possible.
I think it would be monstrous. I believe that, as a society, we should commit ourselves to creating circumstances in which the fundamentally human experience of parenthood is available to all, not barred from those we’ve left behind on our way to good schools and walkable neighborhoods. Women unlikely to marry who wish to have children by all means should. The shame is ours, not theirs. It belongs to those of us who call ourselves “elite”, who are so proud of our “achievements” that we walk away without a care from the majority of our fellow citizens and fellow humans, from people who in other circumstances, even in the not so distant past, would have been our friends and coworkers, lovers and spouses. It’s on us to join together what we have put asunder.
- 23-Jan-2014, 12:55 p.m. EEST: “invisible to local spectators” Thanks Noumenon!
- 24-Jan-2014, 2:25 a.m. EEST: “caused airplanes
landingto land on Melanesian islands”, fixed misspelling of Reihan Salam’s name.
- 24-Jan-2014, 11:35 a.m. EEST: “as though phenomena