The Hand We’re Dealt

What does it take, in this era of growing inequality in the United States, for someone to move up the social ladder? How people answer that question tends to reveal how they view our society: whether they believe hard work is rewarded no matter who you are, or whether they think race, class, and economics place restraints on your chance of success.

Sociologist Dalton Conley has spent much of his career studying the forces behind social mobility. Born in 1969, he grew up in a New York City housing project where he was one of only a few white children. His parents, he says, were “liberal starving artists.” His mother was a writer, and his father was a painter whose ancestors had come over on the Mayflower. They lived on food stamps, but family friends were willing to let Conley’s parents use their address to transfer him to a progressive school in Greenwich Village, where his fellow students were more privileged. The experience gave him a lasting curiosity about inequality and race, subjects he would later write about in Being Black, Living in the Red: Race, Wealth, and Social Policy in America and Honky, a lyrical memoir about his childhood.

The search to understand unequal outcomes has led Conley to other subjects, including some that sociology has traditionally avoided, such as the roles played by fetal development and genetic inheritance. He recently completed a PhD in biology to help him examine whether certain genetic markers — DNA sequences — help determine success or failure in life. He even has a tattoo on his left forearm depicting a marker associated with learning from one’s mistakes. (He got it after his divorce.)

A former dean of social sciences at New York University, Conley still teaches at NYU. He is also an adjunct professor at Ichan School of Medicine at Mount Sinai, a research associate at the National Bureau of Economic Research, and the dean of arts and sciences at the University of the People, a tuition-free institution. His other books include The Starting Gate: Birth Weight and Life Chances (with Kate W. Strully and Neil G. Bennett), The Pecking Order: A Bold New Look at How Family and Society Determine Who We Become, and Elsewhere, USA: How We Got from the Company Man, Family Dinners, and the Affluent Society to the Home Office, BlackBerry Moms, and Economic Anxiety. He is affiliated with the progressive think tank Center for American Progress and served as a senior advisor to the United Nations Millennium Project. In 2005 he became the first sociologist to win the National Science Foundation’s Alan T. Waterman Award, given annually to one young researcher in science, mathematics, or engineering.

I met Conley in 2004 at a conference called Inequality Matters. He was presenting findings from The Pecking Order, and I was working for a nonprofit that aimed to reform philanthropy. We’ve stayed in touch ever since. Even so, I was a little hesitant to interview him. He’s used to being the one asking the questions, and he always seems to be mining for data.

We held this free-ranging conversation in his home, a converted warehouse space in Manhattan that he shares with his daughter and son and their many pets: two cats, a yippy Yorkshire terrier, three guinea pigs, a lizard, a rabbit, a fish, and a bullfrog (thought to be expired until he was discovered thawing in the pond on the back deck during my visit). Believe it or not, this menagerie is smaller than previous ones kept by the family.

Conley is precise about word choices and careful to avoid making generalizations. (He began our conversation with a caveat: “You have to put ‘on average’ after everything I say.”) Still, he comes across as easygoing and is quick to laugh at his own shortcomings. Being the father of two teenagers undoubtedly helps keep him humble.

 

Conrad: What are the questions that concern you?

Conley: I’m interested in who gets ahead in this country from generation to generation and why. My father loved handicapping the horses, so I spent much of my childhood sitting at Aqueduct Racetrack in Queens, New York, sifting through used betting tickets, reading the Daily Racing Form, and learning how to predict winners. Basically that’s what I do today, only with people. I want to know: What are the social mechanisms by which upward or downward mobility happens? How does what we inherit from our parents and forebears affect our opportunities in life? As my dad did with the fillies, I examine the data and try to figure out who’s going to win the race.

Conrad: So what does it mean to “win” in life?

Conley: I look at a variety of outcomes, but I try to stick to those that can be clearly measured. So I don’t look at whether people are happy. I don’t look at their sense of spiritual well-being. Those are important, but you face an almost insurmountable challenge when you attempt to measure them. Take happiness, for example: the Scandinavians are said to be the happiest people on earth, yet Scandinavian countries also have relatively high suicide rates.

Instead I turn to data that don’t leave a lot of room for interpretation: I look at how much infants weigh at birth. I look at how long people live. I look at how far people go in school. I look at net worth — how much money people have. I also consider genetic information. These are not subjective measures. They’re not opinions or feelings.

Conrad: So what have you found in regard to social mobility? If someone has achieved a higher socioeconomic status than his or her parents, which factors proved most significant?

Conley: Only two measurable socioeconomic aspects of the parents really matter in predicting who succeeds: the parents’ education, which is the most important, and the family’s wealth, which is the second most important. By “wealth” I don’t mean how much the parents make a year. I mean net worth, including savings, property, and other financial resources.

To come to this conclusion, I used a survey called the Panel Study of Income Dynamics. I like to call it “America’s family tree.” It started with five thousand U.S. families in 1968 and has followed members of those families ever since, even after the kids grew up and moved out or a couple divorced. Because those kids have grown up to start their own families, the survey has grown dramatically in size. It now contains information on three, and in some cases four, generations. We can look at siblings and see how they have fared differently. We can compare cousins. We can document the rise and fall of people’s fortunes through recessions and booms, divorces and remarriages. The survey really is a national treasure. Unfortunately, because of budget cutbacks in the 1990s, it now happens only every other year. Keeping that study going is one of the most important efforts in my field. It’s the social-science equivalent of a 1,200-year-old sequoia: if you cut it down, it’ll take lifetimes to replace it.

Anyway, in the 1990s the survey started asking not just about income and employment but also about wealth and savings. So I looked at kids who had been living with their parents in 1984 and then followed them as they grew up and formed their own households, tracking their educational and economic outcomes. What I found is that the parents’ incomes and what jobs they had didn’t matter. Even race didn’t matter. There were racial differences, but only because there also were differences in wealth and education between races.

Conrad: Why do you think that parents’ education has such a pronounced effect? Does it give children a head start in school? Does it correlate to greater stability at home?

Conley: To be honest, I don’t really know. It could be that highly educated parents provide their offspring with a more cognitively stimulating home environment. There was a famous, if somewhat flawed, study documenting how the number of words spoken to young children varied by social class. The researchers didn’t measure the size of the parents’ vocabulary but the literal number of words they spoke to the child. Between families with highly educated parents and those with less-educated parents, there was a gap of millions of words by the time kids hit school age. But, again, it could also be that education is merely a proxy for other factors, such as attitudes toward schooling or social connections. It could even be that parents with more education pass on “smarter” genes. The answer is probably that education is working through a whole host of channels.

Conrad: A moment ago you made a distinction between income and wealth. Could you clarify that?

Conley: It’s an important distinction. Most past studies of social mobility looked at occupation or income or both. My contribution, standing on the shoulders of giants, was to examine wealth or net worth. You can think of income as a stream: if you stop going to work or get cut from the welfare rolls, the stream ceases to flow. Wealth is more of a pond, and it’s the stuff of which mobility is made. Economists have shown that somewhere between 25 and 75 percent of our lifetime wealth accumulation is a result of what we receive from prior generations. This doesn’t mean just inheritance, which usually comes late in life, but gifts along the way, such as your parents’ financing your college education or cosigning the loan for your first mortgage. Wealth is also where the legacy of racial inequality shows up starkly: the median African American family has just 8 percent as much wealth as the median white family — a gap that isn’t explainable by income differences. A white family with an income of thirty thousand dollars a year will likely have money in savings, whereas a black family making thirty thousand a year is likely to be in debt. And this isn’t because white people save more. The data show that saving rates are the same for blacks and whites.

Part of the reason there’s this enormous gap is that whites have had more wealth to pass on generation after generation. It’s also a result of discriminatory policies such as “redlining,” in which banks refused to give people loans to buy homes in predominantly black neighborhoods.

Homeownership represents the largest part of a family’s nest egg, and rates today are 73 percent for whites and 43 percent for blacks. Homeownership also creates wealth in a number of ways: You get a mortgage deduction, meaning you pay less income tax than if you were renting. Owner-occupied communities tend to have better public services — everything from garbage pickup to policing to public schools. And, of course, you can borrow from the equity in your home to start a business or send your kid to college.

In fact, when you compare black kids and white kids from families with the same income and wealth, the rates of college graduation are the same; the rates of employment are the same; the number of welfare recipients is the same. This is why I say that the issue is not race, per se. Race is just closely associated with significant wealth differences.

Conrad: In the U.S. we cling to the story of the self-made millionaire, and so we downplay the significance of how much wealth we’re born into.

Conley: In sociology we talk about ascription versus achievement. It’s similar to nature versus nurture in biology. Ascription means what is determined by conditions of birth and is therefore out of our control. For example, your physical appearance. Achievement is your performance — anything over which you do have some control. Achievement-based rewards are considered more legitimate in our society. We say we’re a society of equal opportunity and that the conditions of your birth don’t really affect you as much as they do in, say, post-feudal Europe, but that’s an idealized image. The truth is that there are plenty of ascriptive factors that affect how Americans fare in life.

We typically measure the extent to which your family background matters by comparing how alike you and your parents are, or you and your siblings. When there are significant differences, that’s generally seen as an indication of more achievement and less ascription, but not always. For example, you and your sister, assuming you grew up together in the same household, may have had equality of family wealth, but you could still be more attractive than your sister. Economist Daniel Hamermesh and others have found that attractive people enjoy a premium in the labor market — and, of course, in the marriage market, which is another important, if less often studied, factor in economic attainment.

Within black families it appears that skin tone is a good predictor of which siblings will do better. This seems paradoxical, since I just said that race matters only because of its association with family wealth, but there is sorting going on within families that may not show up in a large-scale analysis across families. In white families an overweight sibling will tend to do worse economically than his or her thinner brothers and sisters.

Another factor is sexual orientation. I found that LGBT [lesbian, gay, bisexual, and transgender] individuals tend to have more social mobility than their heterosexual peers — sometimes upward, and other times downward. LGBT people from wealthy families tend to fare worse because they often don’t get full access to the family’s resources and are kept at a distance by their kin. But those from disadvantaged backgrounds typically show better outcomes than their siblings. A working-class lesbian from a dying Rust Belt town might move to a big city with a more robust labor market. A gay black kid from the ghetto might distance himself from street life and leave home as soon as he is able.

Only two measurable socioeconomic aspects of the parents really matter in predicting who succeeds: the parents’ education, which is the most important, and the family’s wealth, which is the second most important. By “wealth” I don’t mean how much the parents make a year. I mean net worth, including savings, property, and other financial resources.

Conrad: Speaking of the ghetto, in your memoir Honky you write about growing up poor in a neighborhood where you were one of only a few white children. Were the issues of race and class as stark to you then as they are today?

Conley: Definitely not. My awareness of race took a long time to sink in. I remember my sister got a “real” Barbie while her friends all got dark-skinned Barbies. And in first grade I was the only one who didn’t get hit by the teacher — an African American teacher, I should add. Also I didn’t get the same treatment by the police as my neighbors did. But before I started school, I didn’t really understand race. Once I wanted a little sister so badly that I stole a baby girl by grabbing her stroller in the park of our housing complex. I didn’t get very far, of course. Her parents turned out to be the leaders of the local black-separatist movement! [Laughs.] But at the time I saw no reason why people with completely different skin tones couldn’t be from the same family. Only later was I able to make real sense of the race and class dynamics that were rife throughout my childhood.

A friend of mine grew up in the reverse situation. He was the only African American kid in a rural white community. He says that in preschool another kid asked him why his skin color wasn’t coming off when he washed his hands. My friend then attempted to scrub the brown off his own hands. It was the first time he’d realized that skin tone was important.

In the U.S. we get hung up on skin color because of our history of slavery, but it’s not inherent to the concept of race. In Japan, for example, a minority group called the Burakumin experience much the same bigotry that African Americans do in the U.S. The Burakumin, who make up about 3 percent of Japan’s population, are descended from people made homeless by the Japanese feudal wars during the shogun era. They have been socially isolated since the fourteenth century.

Even though the Burakumin look identical to the majority of Japanese — seven hundred years is nothing in evolutionary time — they experience discrimination. There are firms that racist Japanese parents can hire to make sure that their potential son- or daughter-in-law does not have Burakumin blood. As with African Americans and white Americans, there’s a life-expectancy gap, a test-score gap, a health gap, and an economic gap between the Burakumin and the Japanese majority. When individuals from both groups immigrate to the U.S., however, all those differences go away. They are all just Japanese.

Conrad: Are there examples of other “races” with no physical indicators?

Conley: In Nazi Germany Jews were marked with a Star of David because they weren’t physically different. In Rwanda the Tutsis and Hutus are said to have different physical features, but ultimately the distinction relies on whether one’s ancestors were a nomadic people or a farming people. In Brazil there is a continuum of many shades, rather than the two simple categories of “black” and “white.”

Conrad: In The Starting Gate you write about the impact of birth weight on educational attainment. Why do you think lower birth weight correlates to less likelihood of graduating from high school and college?

Conley: Research in the medical field shows that birth weight casts a long shadow on life expectancy. For example, babies born shortly after the Dutch “hunger winter” during World War II were lighter than average and suffered higher rates of cardiovascular disease fifty-odd years later.

My goal was to see if those effects extended to socioeconomic attainment as well. By comparing siblings from the same mother, I found that being lighter at birth lowers the chances that a child will graduate from high school on time. Because I compared siblings, I could rule out a whole host of competing factors that would have been present if I’d compared kids from different families. Comparing twins is even better, since both siblings experienced the same prenatal environment.

We don’t know all the mechanisms, but we think that when a fetus is deprived nutritionally in the womb, it shifts resources from neural development to the immediate task of staying alive. Maternal stress, too, diverts fetuses’ resources from what will keep them healthy in the long term to what will keep them alive in the short term. And stress can also trigger premature birth. So there are a number of things going on even in sibling studies.

How does our biological stock interact with social circumstances? There’s an old adage that a gene for aggression might land you in the boardroom if you’re born into the upper class, or it might land you in prison if you’re from the ghetto.

Conrad: Is this why you decided to study genetics: to better grapple with the cards we are dealt at birth?

Conley: There are two elephants in the room that confound sociologists’ statistical models of cause and effect. One of them is culture, which is notoriously difficult to define and measure. The other is our genetic stock, which can affect health, cognitive ability, and so forth. For a long time in the social sciences we just fenced off genetics and said it was taboo, since it was assumed that any genetic findings might appear to legitimize inequality, blaming poverty on a person’s DNA, for example. And, at the very worst, it might lead to eugenics. Now we have a torrent of genetic data because of the falling cost and increasing ease of genotyping — measuring someone’s genetic makeup. We can now actually open up the DNA “black box” and shine some light into it. Step one will be to integrate genes into models of social inheritance, allowing us to look at people who have the same genetic marker so that we can rule out its effects and more confidently draw conclusions about the influence of social factors. For example, if we’re looking at longevity, and we find that a rich person lives five years longer on average than a poor person with the same genotype, that’s a robust finding. Nobody can suggest that underlying genetic factors were driving both the poverty and the decreased longevity. By ruling out genotype, we can better document the effect of social forces.

Step two will be to ask: How does our biological stock interact with social circumstances? There’s an old adage that a gene for aggression might land you in the boardroom if you’re born into the upper class, or it might land you in prison if you’re from the ghetto.

Here’s another example of gene-environment interaction: A number of studies have tried to tackle the question of how very young children are affected when their primary caregiver (usually the mother) goes to work full time. Some studies found that kids developed more behavioral problems when sent to preschool; others found that preschool better prepared children for kindergarten. Then a more in-depth study discovered that kids with less-educated parents did better when they went to preschool, whereas kids with highly educated parents did worse. So the effect of the parent working depended on the parent’s education.

And how can we explain that some kids are resilient whereas others are irreparably damaged by early-childhood stress or neglect? Some behavioral scientists think that genotypes may be the determining factor for which kids are “orchids” — that is, really sensitive to their environment — and which kids are “dandelions,” or less affected by their environment.

Conrad: Might there be unintended consequences of opening the black box of genetics?

Conley: Of course. We could see a trend toward so-called designer babies, as parents prescreen embryos for physical and behavioral characteristics. In fact, I think this definitely will happen. But that’s all the more reason for social scientists to be at the table — so that reductionist understandings of the roles that genes play don’t carry the day.

Conrad: How do sociologists conduct experiments?

Conley: In all social sciences except psychology (which performs lab-based studies), we have two types of experiments: natural and field. A field experiment is one in which the researcher randomly divides the subjects into groups and relies on the law of large numbers to ensure statistically significant results. A natural experiment is one in which we look at a situation that already divides people randomly into two groups. For example, the sex of your first child is random. It creates a natural experiment in which we can look at how the sex of a first child affects the parents. We’ve learned that, if an unmarried couple has a boy, the dad is much more likely to stick around, marry the mother, and financially support the child. I have a PhD student who’s studying this in the developing world, and she’s finding that if the first child born to a married couple is a daughter, domestic violence increases, the mother is more likely to have to work outside the home, and she’s more likely to have additional kids in an attempt to produce a male heir.

I like natural experiments: just poring over the data and trying to see something that other people haven’t. The problem is that the questions you can ask with them are limited. You’re like the drunk who is looking for his keys under the lamppost: The police officer asks, “Why are you looking for your keys here?” And the drunk says, “Because this is where the light is.”

Conrad: You’ve often been critical of sociology and sociologists. What do you have against the field?

Conley: How does the Groucho Marx joke go? “I wouldn’t belong to any club that would accept me as a member.” Of course I feel I can make derogatory comments about sociologists because I’m one of them, but if someone else makes fun of them, my hackles go up. [Laughs.]

I think for a long time sociologists didn’t take seriously enough the distinction between causation and correlation. The fact that two things occur together doesn’t mean that one causes the other. A third factor might cause both. Sociologists also haven’t been great at seeing that the causal arrow, if it does exist, can go in both directions. For example, a recent study claimed that when parents helped kids with homework, it led to worse performance in school. When I looked at the researchers’ data, however, it was clear to me that the causal arrow was most likely going the other way: kids who were already struggling wound up receiving more help from their parents.

I had a kind of “Saul on the road to Damascus” moment shortly after graduate school when I read a book by sociologist Susan Mayer called What Money Can’t Buy: Family Income and Children’s Life Chances. Back then, sociologists generally believed that, because low parental income correlated with fewer opportunities, poorer health, higher risks, and so forth, the way to improve poor kids’ chances was to increase their parents’ income. Mayer found that we were overestimating the impact of income, and she showed that other factors must be involved.

Mayer’s questioning the dogma back in 1997 made me realize that we had to be more thoughtful about our approach to these questions. Since then, I’ve been on a mission to establish cause and effect with much more rigor. That’s partially why I wrote a textbook for introductory sociology classes. I wanted the discipline to become more scientific and less focused on its founders, Émile Durkheim, Karl Marx, and Max Weber. Physicists don’t keep referring to Newton’s findings when they write papers today. Most psychologists are not talking about Freud.

I’m not saying there’s no place for the founders. I occasionally return to them to generate new questions. When Marx published Das Kapital in 1867, the world was very different. He was talking about alienation of labor in the context of assembly-line factory work. So what does that alienation look like in the digital age? There’s an interesting question.

Conrad: Don’t you address that in your book Elsewhere, USA?

Conley: Yes, I attempt to update Marx’s notion to the twenty-first century. He talked about how, in those days, a craftsman knew how to make a chair from start to finish. He was master over the object. But in factory capitalism, all each worker knew how to do was hammer in this one nail or screw this leg into that base. He or she didn’t actually know how to make a chair. Therefore the objects workers made became masters over them. In today’s digital economy, many of us don’t even produce an actual physical product. We’re working with pure information. So we feel doubly alienated: we have no deep understanding of products we consume, and many of us aren’t even sure what we’re producing.

Many of us are trapped in this “elsewhere” of rising inequality, longer work hours, and ubiquitous technology that allows the cycle to accelerate. Sacrosanct boundaries like work versus leisure or home versus office or consumption versus investment are eroding. But, short of dropping out and building your own log cabin, it’s hard to escape this. We have trouble even cutting back on how often we check our e-mail. Yesterday I was at a committee meeting on technology and teaching at NYU, and people were arguing about which technological bells and whistles to employ to help engage the students, given how much competition there is for their attention. They are “elsewhere” even when they’re in our classes. I think the single best thing we could do is to put our classrooms in Faraday cages — copper meshes that prevent any wireless signals from passing through. Then students could learn for an hour without being distracted. I could probably use one, too.

Conrad: I find it upsetting that it’s come to that.

Conley: Technology is always viewed as a boon at first; then the drawbacks become apparent. In the sixties the creators of the Corporation for Public Broadcasting thought that TV would be the great equalizer. Here we had this medium that could reach poor households where kids were not getting enough cognitive stimulation. Twenty years later the question had become: How do we get poor kids to watch less TV? Now the conversation has started about the “digital divide,” and people are pushing to get poor kids online. Soon we’ll have the opposite problem: How do we get them off the Internet? In the near future I expect only the rich will have the privilege of being offline.

Conrad: Getting back to your critique of sociology: How can sociologists ask better questions?

Conley: Good questions are informed by logical thinking and understanding the difference between causation and correlation, but they are also developed by our becoming “import-export dealers” in the marketplace of ideas.

Take, for example, the link between abortion and crime that economist Steven D. Levitt discovered. Levitt and his coauthor, John Donohue, found that states that had legalized abortion earlier in the 1970s saw crime drop sooner in the 1990s than states that had legalized abortion later. Their hypothesis was that legalized abortion reduced the number of unwanted children, who are more likely to grow up to commit crimes. But how did he even think to look for the crime-abortion link? My understanding is that it came from his reading about how Romanian dictator Nicolae Ceauşescu had outlawed abortion in his country, for a short period, and two decades later there had been a spike in crime rates. By reading widely and making connections across domains, we can learn to ask good questions.

There are many ways to take a simple question and make it into a deeper inquiry about human social life. In conceiving of studies, sociologists are well served to connect a very precise empirical question to a larger theoretical question. In a recent study I did, I looked at the long-term effects of service in Vietnam on veterans’ family life. I asked: Did Vietnam veterans suffer higher divorce rates?

Now, I could have just compared those who served and those who didn’t, but, as we know, the privileged often got out of serving. So any control group of people who didn’t serve would have been skewed toward those with more money or power. My solution was to divide people into two groups using their draft numbers, which are by definition randomly assigned. This allowed us to assess the effects of military service with minimal bias from social status or other confounding factors. What my coauthor, Jennifer Heerwig, and I found was that serving in Vietnam actually reduced veterans’ likelihood of divorce. We hypothesized that being in the military — going through basic training, living in a platoon, and what have you — taught people to give up some individual autonomy, not unlike the way being married does. Of course whether having fewer marriages end in divorce should be society’s goal or not is a separate question.

You could interpret that study as a narrow exercise in asking whether military service leads to higher divorce rates, or you could use it to ask a more interesting, deeper question: Is an experience of socialization in one institutional context — namely, the military — transferable to socialization in a different context, and why or why not? The deeper question often gets answered — or, at least, explored — through a series of studies that build upon each other.

Conrad: So you weren’t assigning a negative or positive value to divorce?

Conley: That’s right. This brings to mind the negative-income-tax experiment from the sixties and seventies. Back then, some Democrats, along with President Richard Nixon, wanted to change the tax system so that everyone would have had a guaranteed minimum income. The only difference between the Democrats’ plan and Nixon’s was that Nixon wanted to impose work requirements. But, either way, everybody would have gotten a check.

The idea was killed in Congress, but an experiment was still conducted to study the effects of a negative income tax. The U.S. Office of Economic Opportunity gave a guaranteed minimum income to some families on welfare and compared the results to a control group. The researchers found that the families with the guaranteed income were better off economically in the short run, but their unemployment spells lasted longer, and their divorce rates increased.

The progressive Left said this was great: People weren’t having to take terrible jobs; they could wait until less-demeaning, better-paying work came along. Meanwhile the increased divorce rate meant that women were no longer dependent on men for economic survival and could walk away from bad marriages. But, from a political perspective, it’s hard to sell a policy that will create more unemployment and divorce. So we’ve not heard of the negative income tax since.

I think we should be following the kids — now adults in their forties — whose families took part in that experiment and asking if the ones whose families received the guaranteed income are better or worse off than the control. Maybe the fact that their parents could stay at home more when they were young had positive ramifications for the children. Or maybe growing up with divorced parents hurt them economically. I briefly looked into trying to find the Social Security numbers of those participants, but a lot of the records are lost. The early seventies was not yet the information age we live in today.

Conrad: Was your desire to do that study based on your being a divorced parent?

Conley: Sure. Just as with my race and class research, I was motivated in part by my personal life. Maybe, through analysis of the negative-income-tax data, I could have shown that divorce has fewer negative effects on children than we might think.

Divorce does affect them in many ways, of course. In my research on sibling differences for The Pecking Order I found what I call the “Cinderella effect”: When the oldest child of a divorced couple is a daughter, she tends to be particularly disadvantaged, since she often picks up the slack of the missing parent and makes sacrifices to that end. One might assume that learning how to handle adult responsibilities at a young age would enhance a girl’s chance of success later on, but what we found was that, more often than not, eldest daughters got stuck in caretaking roles and did not leave the parental home to seek higher education or job opportunities as readily as their younger siblings did.

Conrad: Do you think the idea of a guaranteed minimum income has any hope of making a comeback?

Conley: I never say never. Society is always evolving. In the future a guaranteed minimum income might be the only viable solution to extreme inequality.

Today the economy is a winner-take-all affair. We’re seeing an enormous increase in wealth at the top, with stagnant or declining fortunes from the middle on down. In the sixties and seventies the U.S. was as egalitarian as it has ever been, but by 2007 the levels of inequality were the same as in 1929. Many thought that, after the recession in 2008, we’d see what we saw between 1929 and 1931: a steep drop in inequality. After the Wall Street Crash of 1929 there was no government bailout of the financial sector, and the free market was allowed to destroy many great fortunes and level the playing field. In 2008 we saw a very different policy response. The Troubled Asset Relief Program, signed into law by George W. Bush, propped up the financial sector. We didn’t let the market reset itself, and so inequality has increased instead of declined. If these trends continue, rich people could conclude that the easiest thing to do is to make sure everyone has a minimum income. It might stop people from complaining when all those middle-class jobs we’ve lost don’t come back. Back in the sixties the negative income tax was seen as setting a minimum income, but today it might be seen as buying off the 99 percent.

Conrad: Would that be a good or a bad thing?

Conley: I don’t know. My friends in Scandinavia who lean pretty far left have seen the negative effects of a guaranteed income. They say that some people just want to live on the dole and lack motivation. I can’t confirm whether that is actually a result of their income policy — which has gotten stingier lately, anyway — or other factors. It’s just an anecdotal report. But it does jibe with the longer spells of unemployment in the negative-income-tax study.

Conrad: Are you particularly attracted to counterintuitive questions and answers?

Conley: I think a good research question is like a boxing match: no one wants it to end in a round-one knockout. You want to pose a question with two possible answers that appear evenly matched. For example, if I asked you who had fewer mental-health or disciplinary problems due to their experiences in the Army in World War II, the country recruits or the city recruits, what would you say? You might expect that recruits from the country did better because they were used to a Spartan and physically demanding lifestyle. On the other hand, people from the city are used to living in cramped quarters, and that might have been more advantageous. So there are two intuitive answers, and you hope to debunk one and confirm the other.

In fact, it turned out that the city recruits did better. Sociologist Paul Lazarsfeld used this study in the fifties and sixties to refute those who claimed that sociology was the study of the painfully obvious.

Conrad: In your 2005 op-ed in The New York Times, you raised a provocative question about a father’s rights regarding abortion.

Conley: Whenever I write an article for The New York Times, my goal is to get people to think about an old issue in a new way, either by revealing new empirical results that run counter to what most of us think or by presenting an argument that readers might not have thought about before.

With my abortion piece I meant to pit the concept of bodily autonomy against the belief that rights and responsibilities go together. The fundamental argument for abortion rights in modern society is that women should be able to decide what to do with their own bodies. But we also hold a biological father responsible for child support. So I asked whether a potential father should have some rights in reproductive decision-making if he is to be held responsible later.

Also I wanted to get people thinking about some questions that I expect will come up in the near future, such as: What happens if it becomes possible to remove an embryo and place it in an artificial womb to develop? Is it the mother’s right to destroy these growing, dividing cells or merely not to have them in her body? And what rights does the father have if the embryo can be removed safely and viably? Does that change the equation?

But I made a mistake. I basically came out and said that men should have a say in the abortion decision if they are to be held responsible for their children. This was based on my own feelings about an incident in my past, when an ex had chosen to have an abortion against my wishes. The op-ed was like an emotional e-mail that I should have let sit for twenty-four hours before sending. Rather than provoking thought and letting people come to their own conclusion, I voiced my own, highly personal, opinion. I messed up. And, though I stand by the questions I raised, I no longer agree with my conclusions.

Conrad: Any predictions for the next twenty-five years, based on your research?

Conley: Inequality in life expectancy has been accelerating. For the highly educated, life expectancy is going up and up. For less-educated people, it’s stagnant. I expect the trend to continue. It would be great to see a social movement arise whose aim is to increase life expectancy for those on the bottom.

Conrad: What are the practical applications for some of your findings?

Conley: Personally I’m interested in shaping policy. I want the research that I do to be relevant to the choices we make as a society — for example, in our government programs.

Take welfare. If we want to make it a temporary way station and not a way of life — as the 1996 welfare-reform act purported to do — then we need to eliminate the asset limitations on those who qualify for it. Right now a family has to spend almost all its savings and have next to zero home equity in order to qualify for welfare benefits. By that point people have dug themselves into a financial hole that is difficult to escape. If they could receive income assistance before they reach that point, they would have an easier time getting back on their feet. And, at the other end of the economic spectrum, the use of estate, gift, and inheritance taxes can limit the effects that wealth accumulation in one generation has on the generations that follow. These policies need not be embraced to punish those who amass wealth and inherit it, but can be targeted specifically at those instances in which the buildup of assets by some thwarts the chances for others to acquire wealth of their own.

The reality is that it’s going to take a million small policies to nudge us in the right direction. Of course there are some big things we could do, too. We could put funds into a college-savings account for every American child at birth so that, by the age of eighteen, each of them would have enough to attend a four-year university. Or we could grant every citizen a share in a government-run investment fund as a birthright. We should all be investors in and receive dividends from one of the most efficient economies in the world. Then maybe we could relax our obsession with work and wages and job creation.

Today the economy is a winner-take-all affair. We’re seeing an enormous increase in wealth at the top, with stagnant or declining fortunes from the middle on down. In the sixties and seventies the U.S. was as egalitarian as it has ever been, but by 2007 the levels of inequality were the same as in 1929.

Conrad: How hopeful are you that any of this will come to pass?

Conley: Not very — at least, in the short-to-medium term. I’m not sure how often policymakers listen to social scientists when they make decisions.

Then there’s an even more fundamental debate about how stable the findings in social science are. Let’s say we discover the Holy Grail: an ideal experiment with predictable outcomes wherever it’s conducted. The problem is that society adapts. Sometimes the discovery of a cause-and-effect relationship makes it go away. The best example I can think of is the stock market and the weather in Manhattan: An economist once figured out that when the weather in Manhattan was good, the stock market had a slight uptick. He could have made a lot of money on that discovery if he’d kept it to himself, but instead he published a paper about it, and Wall Street analysts immediately factored Manhattan’s weather into their models, causing the effect to dissipate. I hope that economist got tenure for his trouble. [Laughs.]

Sometimes society changes on its own, making earlier findings obsolete. In 1951 psychologist Solomon Asch brought an unsuspecting subject into a lab, showed him three lines of different lengths, and asked which one matched a line on a separate card. There were also seven phony test subjects in the room, all of whom gave the wrong answer. Asch wanted to see whether the real subject, who was asked last, would go along with the others or trust his own perception. Asch ran the test many times. Three out of four subjects gave the wrong answer at least once, disbelieving or ignoring what their own eyes were telling them. But that effect has diminished since then, suggesting that society now places greater value on resisting conformity.

Sometimes the results from a small-scale study won’t hold true at a large scale. For example, recently a thousand poor kids in New York City were given vouchers that enabled them to attend private schools, to see if they would perform better there. They didn’t, but even if they had, a thousand kids isn’t very many within the largest school system in the country. If every student in New York were likewise to be given these vouchers, there would be system-wide effects. That is, the composition of entire schools would likely change as students flocked to or from them. Meanwhile rich parents would find new ways to give their children advantages, beyond private schools that previously only they could afford. It’s not that the original experiment wasn’t valid. It’s just that the nature of the experiment changes when you increase the scale, and new effects arise.

So it’s hard to nail down a causal relationship in the social world. We’re constantly learning from social science and then invalidating its findings in the ways I’ve just described. It’s pretty frustrating, but I guess that’s progress.

Conrad: You sound a bit beleaguered.

Conley: I don’t actually feel that way. I mean, the incorporation of the knowledge and resulting dissipation of the effect is in some ways the payoff of sociology. I certainly hope that by knowing about the Cinderella effect, for instance, I can try to make sure my daughter does not suffer from it.

Conrad: When I reread your childhood memoir, Honky, it struck me how much fear you felt. For example, you described the steel front door of your apartment as your “security blanket.” Did you originally go into social science to feel safe?

Conley: I might initially have sought to understand the social world because I didn’t fit into it very well, but then I started to realize the limits of our knowledge. In every science, even chemistry and physics, each new answer raises more questions. In the social sciences more than others, though, we are a part of the system we’re observing. Plus there are ethical limits to what we can do to find answers. We can’t radically alter people’s lives just to see how they react or adapt. The Stanford prison experiment — in which college students were randomly assigned to be either prisoners or guards — was called off halfway through because some of the “guards” were abusing the “prisoners.”

There are enormous challenges. Nothing will ever be completely solved. But I don’t think that means social science should close up shop. No science offers total certainty. I believe it was the philosopher Bertrand Russell who said that, since all prior theories have been proven false, so must many of our present beliefs be false. The only certainty is that there is no certainty.