Trickle-down norms

People shovel snow during the first lake effect snowfall of the season.
Editor's note:

This article originally appeared in the Winter 2018 issue of National Affairs.

American society is fragmenting. Social solidarity is withering, as evidenced by the fading influence of mediating institutions like unions, churches, and social clubs. Dwindling participation in the community organizations that used to bridge social gaps means that Americans tend more and more to interact exclusively with people who are like themselves. This “silo effect” is widespread, but the social chasm has grown particularly large between the upper-middle class and everyone else.

It is well known that, by nearly every measure, people in the upper-middle class and above have pulled ahead of the rest of America in the past three decades. Their incomes have risen far faster than those of everyone else, and they have reaped the majority of the benefits of decades of economic growth. They also tend to have more stable family lives, so the children of the upper-middle class start life with a leg up on their less-fortunate peers, perpetuating their advantage.

Both sides of the ideological spectrum recognize the problem, though they have different theories about the source. Conservatives typically blame cultural differences, while liberals point to inequality. But one of our most troubling inequalities is in culture itself.

Those in the upper-middle class are succeeding not only because of the way the labor market and educational institutions are tilted in their favor, but also because they cultivate and follow strong, positive social norms with regard to family life, work, and education. Well-educated Americans live in communities soaked in social capital, raise their children within stable marriages, go to church, volunteer for good causes, look after their health, and work hard. They are more likely to provide a stable home for their kids and be engaged, committed parents; research by UCLA sociologist Meredith Phillips has shown that high-income parents talk with their school-aged children for three hours more per week than low-income parents. Those in the upper-middle class are also active in their communities and directly support strong civic institutions; for instance, the majority of parents with a college degree volunteer at their child’s school or serve on a school committee, compared to just one in five parents with less than a high-school diploma. They also tend to be more tolerant of diversity and cosmopolitan in their outlook.

While these norms are strong in the upper-middle class, they are weaker — and in some cases weakening — among less-affluent Americans. While those in the upper-middle class exercise a high degree of control over their lives, there is an epidemic of powerlessness among poorer citizens. As J. D. Vance puts it in his bestselling memoir, Hillbilly Elegy, “There is a lack of agency here — a feeling that you have little control over your life and a willingness to blame everyone but yourself.” Drug use, too often, is a vivid symptom of this powerlessness, an external marker of internal malaise. The shocking increase in opioid use, especially among rural whites, is the starkest measure of this feeling. In 2015, 52,000 people died of drug overdoses — more, as Andrew Sullivan observes, than AIDS at its height in 1995. But, as Sullivan points out, for many in the upper-middle class, the costs are distant: “For many of us in the elite, it’s quite possible to live our daily lives and have no connection to this devastation.” The epidemic is visible on the pages of our newspapers, but not in our neighborhoods.

Those in the upper-middle class are succeeding not only because of the way the labor market and educational institutions are tilted in their favor, but also because they cultivate and follow strong, positive social norms with regard to family life, work, and education.

So, yes, culture matters. And there is indeed a growing cultural gap between the classes. But culture is not separable from economics. A lack of economic opportunity breeds despair and detachment. Insecure work leads to family instability. Engaged parenting requires time and resources that many poorer Americans lack.

Economic inequality also damages the natural processes through which positive social norms are diffused. In a healthy society, as conservative critics point out, emulation and example play an important role. We learn from each other. In particular, we learn from observing and interacting with those who are succeeding. But as we become more unequal, and the wealthy separate themselves from everyone else, the transmission mechanisms become weaker. The supply lines for social norms become stretched.

It does no good for the elite to simply preach at the poor from a safe economic distance. The sharing and shaping of social norms takes place, to borrow a phrase from the philosopher Gerald Cohen, “in the thick of daily life.” But increasingly the everyday lives of the upper-middle class are filled with people very much like themselves. The cultural changes desired by conservatives may therefore require the greater economic equality demanded by progressives.


Cultural change is typically a top-down process. Elites set and reset the cultural climate, with the majority typically following along later. While there is plenty of hand-wringing in every era about the corrosive influence of pop culture, in practice it tends to be posh culture that dominates.

This is not, in general, a bad thing. Educated, successful people tend to produce and promote valuable, rather than destructive, social norms. These then spread downward, influencing broader values and behavior. While the idea of “trickle-down” has been largely debunked in economics, it seems to hold true for cultural change. It is hard to think of many important cultural shifts that have moved up the social ladder, starting among the disadvantaged and then influencing successive classes above before finally reaching the elite. Of course, social change is messy and unpredictable, and it would be silly to claim that it always and everywhere flows in the same direction. Premarital childbearing, for example, was once largely confined to the poor, but is now common among middle-class Americans. (It should be said, though, that it shows no signs of growth among the upper-middle class; if anything, the opposite is true.)

Trends nearly always travel downward. Social norms and attitudes regarding gender roles, education, parenting, health, and same-sex relationships, have been altered first at the top and then in the broader population. This cultural trickle-down can even be seen in baby names, as Steve Levitt and Stephen Dubner showed in Freakonomics. Names that start off being popular among the wealthy become mainstream within a few decades. And this process is not the result of distant celebrity. “It isn’t famous people who drive the name game,” they write. “It is the family just a few blocks over, the one with the bigger house and newer car.”

Of course, the name the neighbors give their new baby daughter is a pretty trivial matter. But norms regarding the way they raise her, including how much they talk to or read to her, are important. Providing a strong, stable, loving family environment is even more important, so a culture of planned, committed co-parenting, usually in the shape of marriage, matters too. A cultural norm of striving for a good education is similarly positive and powerful. An openness to new people and fresh ideas equips us for the modern world, and so norms of tolerance and respect for differences are also useful. In all of these cases, the American upper-middle class is setting a good example. If over time these norms become common ones, then all of society will benefit.

If this argument sounds elitist, that’s because it is. But by and large, elite influence on culture is benign, including from a liberal perspective. Gender roles are a case in point. Few Americans today agree that “[i]t is much better for everyone involved if the man is the achiever outside the home and the woman takes care of the home and family,” according to the General Social Survey. But it was the most educated who led that shift in attitudes. By 1986, only 30% of those with a college degree supported the traditional, gender-based division of labor; the general public caught up almost 30 years later, in 2014.

The same basic pattern applies widely, for good and for ill. Just under half of college-educated Americans supported marijuana legalization in 1978. That level of popular support would only be approached by the whole nation in 2010. Just under three in four college graduates were accepting of premarital sex in 1986; by 2014 the general population had become similarly accepting. The same basic pattern can be found across a range of social attitudes: Where the educated elite lead, the nation follows.

Consider another example: tobacco use. In recent decades, attitudes toward smoking have turned decisively in a negative direction, especially among the upper-middle class. Lighting up a cigarette in an upper-middle-class setting would once have been expected; now it is almost unthinkable. It is important to note that smoking rates dropped first among more-affluent and better-educated Americans. Over time, these attitudes have percolated through the rest of society.


Social norms are particularly important when it comes to the conduct of family life. This is an area where government intervention is necessarily limited; as Bill Clinton said, “[G]overnments don’t raise children; parents do.” At the risk of stating the obvious, the way children are raised has an enormous effect on their life prospects. But social norms around family formation and parenting are diverging along class lines. More-educated Americans manage their fertility well, not through abstinence but through effective use of contraception. They typically wait until they are in a stable relationship — usually formalized in marriage — before having children. So their children are almost always born to a couple who have decided to raise a family together. Outside the upper-middle class, the norms influencing family formation are markedly different. Most women with less than a college education have their first child outside marriage. A young woman without a college degree is five times more likely than a college graduate to be a cohabiting or single mother.

Robert Putnam’s Our Kids is peppered with stories of individuals who see their plans derailed by a pregnancy that was not anticipated or planned for. Darleen, for example, got pregnant two months into a relationship with Joe, who was her boss at Pizza Hut. “It didn’t mean to happen,” she said. “It just did. It was planned and kind of not planned.” In another chapter, David from Putnam’s hometown of Port Clinton, Ohio, becomes a father at 18 years old. David’s report was much like Darleen’s: “It wasn’t planned,” he said. “It just kind of happened.” By some estimates, nearly one in 10 young women who drop out of community college do so because of unintended motherhood. As Isabel Sawhill puts it in her book Generation Unbound, “[F]amily formation is a new fault line in the American class structure.” The family instability experienced by children in less-affluent homes undermines their life chances still further, while those raised in the stable structures of the upper-middle class do well. The family gap is an opportunity gap.

Conservatives used to blame the liberal elite for leading the masses away from marriage and traditional families. By normalizing divorce, embracing sexual freedom, and celebrating women’s independence, the upper-middle class tore up the social scripts that guided flourishing family life. Fine for them, the critique went, since they have the resources to manage without strong social norms, but those who are less well-equipped will struggle without them. Now the criticism is that the elite are living traditional lives themselves but remaining steadfastly agnostic about how others conduct their lives, refusing, as Charles Murray put it, to “preach what they practice.”

Murray is right that there is an elite reluctance to preach about the virtues of a particular lifestyle, even their own. They are tolerant of diversity, pluralism, and individual choice, and they do not want to be judgmental. But even if the elite became less squeamish about moralizing and more willing to act on Murray’s injunction, it is not clear what that would look like, and even less clear how their preaching or persuasive efforts would be received, especially in the current populist climate. Most Americans do not seem to be in the mood for a lecture from the upper-middle class.

The healthier, quieter evolution of social norms is the one that takes place through the power of example and emulation within our existing communities and social networks. Demonstration is more powerful than exhortation. Norms that evolve first among the most educated, and are seen to work, can spread more widely; at least, that’s what history suggests.

There are signs today of this kind of cultural trickle-down, for instance, in the increase in the proportion of poorer parents who are now reading books to their children; a once-elite preoccupation is now becoming a norm. Teen-pregnancy rates have also dropped sharply in recent years, down by 42% since 2007. The decline has been especially sharp among black and Hispanic girls, narrowing the race gap. The social norm has clearly shifted.

It is reasonable to hope that elite norms concerning family formation could similarly spread. The problem is that the most powerful means of norm diffusion — emulation — requires exposure and connection. But the class fractures in our society have put a great distance between us.


Social norms trickle down through three main mechanisms: regulation, persuasion, and emulation. Each of these is threatened by the division and fragmentation of American society.

The first way elite norms are transmitted downward is through regulation, via rewards, rules, and penalties that are explicitly designed to change behavior in the short term and, ideally, habits in the long term. Here, simple self-interest is the primary motivator. Perhaps the most obvious example is the “sin tax” on morally illicit or socially unproductive activities. Cigarette taxes are designed explicitly to drive down smoking rates. They work, too: Big cigarette-tax increases between 1991 and 2005 resulted in substantial declines in youth smoking rates. Regulations can also incentivize desired behaviors; the Earned Income Tax Credit boosts work rates.

The healthier, quieter evolution of social norms is the one that takes place through the power of example and emulation within our existing communities and social networks.

You do not have to be Gary Becker to see how the direct manipulation of costs by the state alters behavior and, ultimately, social norms. If you subsidize it, they will come. If you tax it, they will leave. But there are more subtle ways to guide behavior too. As the “nudge” revolution in policy thinking demonstrates, more nuanced social interventions can be designed as acts of “libertarian paternalism.” Here, the goal is not to change norms, at least not immediately, but simply to change outcomes given the predictably irrational nature of much human behavior. Perhaps the best-documented example of this nudging is auto-enrollment into company pension plans, which dramatically increases participation rates. If people “default” into it, they will likely stay.

The second channel through which social norms can be communicated, especially from the top, is through persuasion. Here, those who wish to shape behavior rely in part on the capacity of the citizenry to respond reasonably to argument and evidence, and in part on a shift in attitudes. Again, health provides perhaps the most obvious example. Public-health campaigns focus on the risks of smoking, unprotected sex, or poor diet. But leaders can use their “bully pulpit” to promote broader social change, too, on issues as varied as drunk driving, engaged fatherhood, teen pregnancy, and so on. Iowa’s “Avoid the Stork” campaign, for instance, aimed to boost the use of effective contraception. Isabel Sawhill calculates that state governments could see at least $5 of savings from each $1 spent on well-crafted campaigns of this sort.

Sometimes, public messaging goes beyond the presentation of simple facts, and moves into feelings, including shame. In 2013, the Michael Bloomberg administration in New York City launched a controversial public advertising campaign to reduce teenage parenthood. Advertisements around the city dramatized the truncated life chances of children born to teenagers; in one, a tear-stained toddler stared out, declaring, “I’m twice as likely not to graduate high school because you had me as a teen.” Critics of the campaign pointed out that it was using shame as a tool to change behavior, but many public-health campaigns, especially the most hard-hitting, combine a presentation of facts with an implicit message that we should feel some shame if we simply ignore them.

Reality TV may be especially powerful in shaping norms. Economists Melissa Kearney and Phillip Levine found that the MTV shows 16 and Pregnant and Teen Mom, which documented in detail the tough lives of real teenage mothers, lowered rates of teen pregnancy by almost 6% nationally. Popular TV shows in Brazil lowered fertility rates by showcasing families with fewer children. The effects were strongest among women closest in age to the stars of the most popular shows; among women aged 35 to 44, the fertility rate dropped by 11%. One notable aspect of the Brazilian case is the TV production company explicitly set out to use their shows to achieve the policy goal of reducing the size of families.

The third means through which norms are transmitted is emulation. Most of us are strongly influenced by what our friends, neighbors, and colleagues think and do. All around us are people who may act as role models, or as salutary warnings. It is human nature to look to others for guidance on behavior and norms to fit in better in a particular society; when in Rome, it’s natural to want to do as the Romans do. This impulse to emulate may be particularly strong when we compare ourselves to those who are most successful, a phenomenon Adam Smith highlighted in The Theory of Moral Sentiments. Taken too far, admiration for the elite can be enervating. But the natural desire for what Smith calls “fellow-feeling,” and the consequent desire to copy elites is part of the human condition:

It is from our disposition to admire, and consequently to imitate, the rich and the great, that they are enabled to set, or to lead what is called the fashion. Their dress is the fashionable dress; the language of their conversation, the fashionable style; their air and deportment, the fashionable behaviour. Even their vices and follies are fashionable; and the greater part of men are proud to imitate and resemble them in the very qualities which dishonour and degrade them.

Of course, it is one thing to show that people do tend to emulate those above them in the social hierarchy; it is quite another to suggest that they should do so. But emulating successful people is not in principle a bad idea. As a distinguished group of thinkers noted in Aeon magazine in an essay in defense of hierarchy, “We can acknowledge that individuals differ from one another in embodying excellence of various sorts, and these various forms of human excellence elicit from us a special kind of positive regard that philosophers call ‘appraisal respect.’”

There is a delicate balance here between elitism and equality. The idea that positive social norms trickle down through the emulative deference, or “appraisal respect,” of the non-elite for the elite is an uncomfortable one for liberals. But at the same time, this process can only work in societies with some degree of solidarity, which is dwindling today as America fragments along class lines.

The three channels through which social norms trickle down — regulation, persuasion, and emulation — often overlap, of course. Tobacco use has been reduced through high taxes and tighter regulations, but also through high-profile public-health campaigns and a dramatic shift from glamorizing to stigmatizing smoking — both on the screen and among friends and family — which created a political climate amenable to tougher laws. Similarly, the greater exposure of sympathetic gay characters on mainstream TV reflected the growing acceptance of same-sex relationships, but also helped to create an environment in which more people felt able to open up about their sexuality, which in turn led to more acceptance and ultimately to legal reforms. When combined this way, the three complementary channels can be a powerful force for change.


The idea that social norms evolve first at the top and then trickle down through these three mechanisms is an uncomfortable one for liberals, especially when it comes to the mechanisms of persuasion or emulation. There is an inescapably elitist feel to the idea that people on the top rungs might be leading the way in terms of cultural reformation, even when the evidence is strong. Liberals are afraid of being seen as judgmental of those who are less fortunate, and this is a legitimate and worthwhile fear. Indeed, the process of norm trickle-down only works when those who are successful have a mindset not of superiority but solidarity. This is about show, not tell.

There is also a strong connection between a central concern of many of those on the left — economic inequality — and the valuable elements of top-down cultural change. Remember that Levitt and Dubner described the most important cultural influencers as “the family just a few blocks over.” But today they are likely to be more than a few blocks over. They are on the other side of the city, in a different neighborhood altogether. Just as water can only trickle through layers of rock that are close together, so the transmission of social norms, and of changes in those norms, requires a degree of proximity, and therefore of social equality.

The divide that has opened up in recent decades has led to increased distance between social classes, and especially between the upper-middle class and everyone else. This distance can be measured in dollars. Since the 1980s, households in the top fifth of the distribution have seen a $4 trillion combined increase in their market income, while the bottom 80% shared a $3 trillion rise. On almost every economic variable — wealth, wages, income — the top fifth has pulled away from the bottom 80%. (It is true that the top 1% has pulled away most dramatically, but it is not true that this is the whole story.)

The economic distancing of the upper-middle class has cultural consequences, a trend that has been described by scholars on both sides of the political spectrum, notably Murray in Coming Apart and Putnam in Our Kids. One of the most vivid symptoms of class separation is neighborhood segregation. Wealthier neighborhoods have grown far richer, better educated, and more culturally homogenous than ever before. At the same time, cultural elites who once lived in a broader set of communities have become concentrated in a handful of what Murray labels “Super Zips.” Protected behind the invisible walls of zoning restrictions and high house prices (kindly subsidized by the state through the mortgage-interest deduction), the upper-middle class increasingly lives in a cocoon of its own making.

One of the central ideas of American community life is that it can produce communal sympathy across class lines. But now, along virtually every axis of interest — class, religiosity, television preferences, political ideology — communities have become more homogenous in the last few decades. Neighborhoods are more economically segregated, with affluent families more likely to have only affluent neighbors. Selective colleges are dominated by upper-middle-class students. Churches are segregated by race, class, and political orientation. Social media has added another polarizing dimension to community life, both online and offline.

When people live in communities where almost everyone looks, thinks, and lives alike, communal sympathy is likely to be replaced by tribalism. This fragmentation and tribalization of society carries a number of risks, not least of which is a loss of mutual respect across class lines and a concomitant weakening of the mechanisms of downward norm diffusion. This vicious cycle reinforces the class divide and the callousness and resentment it fosters, fraying the fabric of American society.

There are implications here for both public policy and our personal conduct. In the policy arena, greater redistribution to blunt economic inequality will help, but only if accompanied by a lowering of the class barriers erected by exclusionary zoning regulations, lack of affordable housing, unfair college-admissions policies, the secession of school districts, and so on. Here, the challenge will be to persuade those who are prospering to give a little ground and look beyond their immediate self-interest. It will take concerted and intentional action to create more economically integrated communities and institutions, especially against the tide of rising inequality.

But creating stronger ties across class lines is a personal challenge, too. It is too easy, perhaps especially for liberals, to subcontract the business of making a more equal society to distant policymakers through the occasional vote. For one thing, the policy changes listed above won’t happen by themselves, especially since many of them depend on decisions made very close to home. We have to show up at community meetings after work to support policies that promote greater class integration at a very local level; we have to turn up, for example, at the housing-committee meeting to call for less-restrictive zoning laws, or at the school-board meeting to support greater school integration.

We need to ask ourselves, when we volunteer our time or donate our money, is it to help people in our own, narrowly drawn community, or to help those in greater need? How much money does our school PTA pass on to a school with a less-affluent population? What kind of organizations do we support? Where do we look for a new summer intern for our business? Whom do we choose to mentor? Here, there is a challenge to conservatives. They will likely welcome the emphasis on taking personal responsibility, including the responsibility to help others, but too often conservatives are most enthusiastic about helping people like themselves.


The upper-middle class is sustaining and cultivating many norms and values that most conservatives welcome. But one of the main reasons these norms are not, as yet, being shared more broadly is the inequality that results, at least in part, from policies that have fueled the economic divide.

Many of the norms of family life being cultivated and sustained by affluent, educated Americans would, if shared more broadly, promote greater opportunity. But inequality has damaged the mechanisms of norm diffusion, and therefore the prospects of more upward mobility. Conservatives who rail against cultural decline and liberals who rail against economic inequality may have more in common than they think.