|SEARCH SITE Advanced Search|
How Welfare Reform Worked
Though successful beyond anyones wildest dreams, welfare reform left key problems unsolved.
Welfare reform celebrates its tenth anniversary this year, and celebrates seems the right word. As most readers know, Temporary Assistance for Needy Families (TANF) ended the much-despised Depression-era federal entitlement to cash benefits for needy single mothers, replacing it with short-term, work-oriented programs designed and run by individual states. Its success has surprised just about everyone, supporters and naysayers alike.
So it seems a good time to remember the dramamake that melodramathat the bill unleashed in 1996. Cries from Democrats of anti-family, anti-child, mean-spirited, echoed through the Capitol, as did warnings of impending Third Worldstyle poverty: children begging for money, children begging for food, eight- and nine-year-old prostitutes, as New Jersey senator Frank Lautenberg put it. They are coming for the children, Congressman John Lewis of Georgia wailedcoming for the poor, coming for the sick, the elderly and disabled. Congressman William Clay of Missouri demanded, Whats next? Castration? Senator Ted Kennedy called it legislative child abuse, Senator Chris Dodd, unconscionable, Senator Daniel Patrick Moynihanin what may well be the lowest point of an otherwise miraculous careersomething approaching an Apocalypse.
Other Washington bigwigs took up the cry. Marion Wright Edelman of the Childrens Defense Fund called the bill national child abandonment and likened it to the burning of Vietnamese villages. Immediately after President Clinton signed the bill, some of his top appointees quit in protest, including Edelmans husband, Peter, who let loose with an article in The Atlantic Monthly titled, The Worst Thing Bill Clinton Has Done. No less appalled, the Chicago Tribune seconded Congresswoman Carol Moseley Brauns branding the bill an abomination. And while in 2004 the New York Times lauded the legislation as one of the acclaimed successes of the past decade, the editors seem to have forgotten that they were irately against it before they were for it, pronouncing it draconian and a sad day for poor children.
Its worth recalling the outcry at this anniversary moment, not in order to have a gotcha-fest, pleasurable as such an exercise can be. The truth is that many of welfare reforms promoters were not spot-on in their predictions, either, and their expectations require some Monday-morning quarterbacking, too. But the apocalyptic scaremongering of reform opponents on the one hand, and the relative benignity of the bills consequences on the other, prompt the obvious question: How is it that so many intelligent, well-intentioned people, including many experts who made up the late twentieth centurys Best and Brightest, were so mistakenmistaken not just in the way a weatherman who overestimates the strength of a snowstorm is mistaken, but fundamentally, intrinsically, and epistemologically wrong?
Before examining why so many people were wrong, lets look at exactly how they were wrongan easy task, given the Everest of data on welfare reforms aftermath. TANF did not include a federal jobs program for the poorthough many wanted it tobut it has ended up being a WPA for social scientists, who have been busily crunching just about every number that happened to wander anywhere near a welfare recipient for the past ten years.
The most striking outcome has been the staggering decline in the welfare rolls, so large it has left even reform enthusiasts agog. At their peak in 1994the rolls began to shrink before 1996, because many states had already instituted experimental reform programsthere were 5.1 million families on Aid to Families with Dependent Children, the old program. Almost immediately, the numbers went into freefall, and by 2004 they were down by 60 percent, to fewer than 2 million. A lot of reform opponentsthe unreformed, so to speaktried to chalk this up to the booming economy of the later 1990s. But according to former congressional staffer Ron Haskins, author of a history of the reform due out this fall, that doesnt make sense: in the 41 years between 1953 and 1994, he points out, the welfare rolls had declined only five times, and only once (between 1977 and 1979) for two years in a row. Compare that with the present case, when the rolls continued their fall even after a recession began in 2001, and when 2004 marked the tenth continuous year of decline.
Caseload declines are all well and good, but what caused opponentsand many proponents as wellto lose sleep was what would happen to women and their children once they left the dole. There were four chief concerns: First, would welfare leavers find jobs? Second, would they sink even deeper into poverty? Third, would their children be harmed? And fourth, would the states take advantage of the wide flexibility the bill gave them on implementation to join what many anticipated would be a race to the bottom?
So lets consider concern number one: Did women who left the rolls actually go to work? The answer is: more than almost anyone had predicted. According to one Urban Institute study, 63 percent of leavers were working in the peak year of 1999. True, some studies showed numbers only in the high fifties, but even these findings were much better than expected.
Nevertheless, a lot of skeptics still werent biting. It was the luck of a boom economy, they said; just wait until the job market sours. Well, the recession came in 2001, and though it was no picnic, it wasonce againnothing like what had been feared. As of 2002, 57 percent of leavers continued to punch a time clock. That, the critics warned, was only because the first recipients to leave welfare were likely to be the most competent. Just wait until were dealing with the most dysfunctional, those who have the most barriers to employmentfrom limited education or work experience to English-language deficiency or mental disability.
But even there the news was encouraging. The Urban Institute kept a close eye on the caseload composition in welfare reforms early years and found that the proportion of highly disadvantaged women was no greater in 1999 than in 1997. A 2003 study by June ONeill and Anne Hill found a large increase in the employment of some of these women: for example, in 1992 only 31 percent of young single mothers who were high school dropouts were employed; by 2000, 50 percent had jobs. And none of this takes into account the women who under the previous regime might have gone on welfare but, after TANF, with its time limits and hassles, never did. The percentage of employed single mothers rose, in the years following reform, from 45 percent in 1990 to 62 percent in 2005nearing the employment rate of their married counterparts.
What about concern number twothat welfare mothers would sink deeper into poverty? Shortly before TANF passed, the Urban Institute released a report, solicited by a wavering Clinton administration, warning that welfare reform could impoverish an additional 2 million people. Reform Jeremiahs waved the report around as scientific proof of their worst fears. Even if some welfare mothers did find jobs, they argued, they would merely be stocking shelves at Duane Reade or making hotel beds, the proverbial dead-end jobs that would leave them worse off than on the dole.
Though a lot of women did take low-paying service jobs, the unreformed got this one wrong, too. For one thing, they failed to consider the Earned Income Tax Credit, whose expansion in 1993 meant a 40 percent boost in annual earnings for a minimum-wage worker with two kids. Most leavers, though, were doing better than minimum wage. In 2002, the same Urban Institute that had predicted TANF disaster found that the median hourly wage for working former recipients was around $8 an hour. Moreover, ONeill and Hill discovered that, just as with most other people, the longer recipients were in the job market, the more they earned; four years off welfare, only 4 percent of working single mothersand only 8 percent of high school dropouts who were single motherswere earning minimum wage or less.
As a result, most welfare leavers had more money than when they were on welfare. The poverty rate for single women with children fell from 42 percent in 1996 to 34 percent in 2002; before 1996, it had never in recorded history been below 40 percent. This was the first boom ever where poverty declined faster for that group than for married-couple families. Nor did leavers disdain their dead-end jobs. Studies consistently found that ex-recipients who went on to become waitresses, grill cooks, and security guards took pride in being salarywomen.
Still, its fair to say that while post-reform America did not look like Calcutta, it was no low-wage workers paradise, either, especially as the economy weakened in late 2001. Exwelfare mothers were still poorer than single mothers overall. Some who worked had less income than on welfare. Many were not working full-time, and an estimated 40 percent of those who left the welfare rolls returned later on. In 1999, close to 10 percent of leavers were disconnectedneither working nor on welfare nor supported by a working spouse. By the recession year of 2002, that number had risen to almost 14 percent. From the beginning, studies from the Childrens Defense Fund and the Center for Budget and Policy Priorities warned of an increase in the number of families in deep poverty, and a steady stream of rumors claimed that soup kitchens and homeless shelters had crowds of ex-recipients clamoring at their doors.
But at least some of these warnings turned out to have been yet more crying wolf. Those who returned to the dole tended soon to find other means of support, getting a new job, signing up for disability or unemployment insurance, or turning to employed partners. As for deepening poverty, experts are often unsure what to make of official estimates of the income of the poorest of the poor, since they may have other sources of support than reported income. So they try to see if the income numbers conform to other measures. There was no evidence that single mothers were moving in with relatives, as you might expect if money were that tight. Harvard researchers Christopher Jencks and Scott Winship, neither of them avid reform supporters, found that, despite a big drop in the number of families using food stamps, worry among single mothers and kids about being able to afford the next meal declined between 1995 and 2000, and though such insecurity increased in the early 2000s, it never rose to pre-reform levels. Moreover, the lowest earners were buying morespending money that, according to official numbers, they didnt have.
And that takes us to concern number threethe kids. Children were the unreformeds most lethal weapon: the image of kids starving in the streets, sleeping on grates, begging from strangers, and neglected and abused by desperate mothers, was enough to make the most robust reformer queasy. But the predicted Dickensian purgatory also turned out to be wrong. There may have been an increase in the number of children in foster care, but child abuse and neglect numbers are, depending on what measures you use, either unchanged or down.
More striking was what happened to rates of child poverty. They not only went down; by 2001, they hit all-time lows for black children. And though the numbers drifted up again during the recession, they were still lower than they had been pre-reform. On other measures, the young kids of exwelfare moms are no worse off than under the old regime. Though some studies find lower achievement and more problem behavior among adolescents, the big picture doesnt show teen children in more trouble post-reform. After 1996, juvenile violence and teen pregnancy continued to go down, as they had since the early nineties.
As for the anti-reformers final concernthe states race to the bottomthat dog didnt bark, either. True, the enemies of reform might point at the 20-odd states that introduced a family cap, which sought to stem illegitimacy by denying any increase in benefits to women who had another child while on welfare, and whose efficacy remains uncertain. But theres little question that the unreformed were wrong here as wellfor the fourth time. The states were, if anything, nicer than the feds. No state barred cash benefits to teen mothers, though TANF permitted them to do so. Forty-seven states made it easier than the old system for leavers to keep some of their cash benefits when they first went to work. Many states, including New York, did away with TANFs five-year time limit for all intents and purposes by using state dollars to pick up the tab for those still on the dole at the time and deemed unable to work.
As caseloads declined, the states moved the federal money they would have spent on welfare benefits into work supporttransportation, child care, and the like. In fact, under the states management, welfare has morphed into an unprecedentedly generous work-support program. The real proof that the states were not the scoundrels that opponents had warned they would be came as Congress debated reauthorization after TANF expired in 2002. Reformers argued for even stricter federally mandated work requirements, while those who once warned that the states would engage in a race to the bottom demanded more state control.
This, then, is where we find ourselves today, ten years after reform: a record number of poor single mothers off the dole and the majority of them gainfully employed; less poverty among single mothers, especially black single mothers, as well as their kids; children adjusting well enough; and state governments taking care of their own. The situation is so far from what experts predicted that, as New York University political scientist Lawrence Mead has put it, it brings to mind the Sovietologists at the fall of the Soviet Union.
So how do reform opponents explain what went wrong when things went right? Some, like the team at the Childrens Defense Fund, just continue to rejigger the numbers in dogged pursuit of evidence that the situation is much worse than anyone will admit. Others have been self-reflective. In 2004, Andrew Cherlin, one of the nations top family scholars, gave a revealing speech to the annual meeting of the American Sociological Association, in which in light of his own studys benign findings, he tried to reassess his own earlier opposition to reform. How should progressives respond to what we know so far about welfare reform and children? he asked. By progressives, I mean leftist/liberal and feminist scholars and observers, who, based on past experience, probably constitute 80 to 90 percent of this audience. His answer suggests some serious soul-searching. [T]here may be something to the idea that long-term dependency on public assistance is detrimental, he conceded, though he had always rejected this idea out of hand prior to 1996. Poor mothers derive a basic dignity from work. In fact, he continued, as a result of what I have seen, I now think the term Ωdead-end job≈ is a label that often doesnt fit the perceptions of low-income workers; and I will not use it again.
It was a thoughtful admission, yes, but only a start. Progressivesa term that, given the present context, seems due for retirementneed to go much deeper than that. It is not simply that they were wrong in making a few isolated predictions; after all, there were bound to be legitimate questions about how to transform welfare and about what would happen afterward. Its that those predictions rested on a scaffold of moldy assumptions not just about poverty but about what kind of country we live in, how human beings decide to live their lives, and what role government plays in those decisionsin other words, about our politics in the deepest sense. At times, Jason DeParle writes in his fine book American Dream, as he describes the debate in Washington in 1996, it seemed the very idea of America was on trial.
Consider the assumptions of these progressive pessimists, as well call them, about what motivated American opposition to welfare. Their answer appears to be: greed laced with a hefty shot of malice. They repeatedly compared what they viewed as the paltry social welfare spending of the United States with that of Western Europe, whose large welfare states supposedly reflected the citizenrys generosity. Americans, they said, were either indifferent orsome on the extreme Left arguedhostile toward the less fortunate. Americans want to punish or scapegoat the poor. They enjoy the politics of meanness, a phrase coined by thenTimes columnist Anna Quindlen and repeated persistently by the liberal commentariat. This view led to the expectation that states would race to the bottom. After his intemperate Atlantic article, Peter Edelman warned that ending the federal entitlement meant that states could do anything they wantand what they wanted, implicitly, was to let the poor starve.
Progressive pessimists were blind to the promise of reform partly because they believed that the American discomfort with welfare was really a mask for racism. The term progressive may imply forward thinking, but in many ways the pessimists are still living in George Wallaces America. Welfare recipients were, and are, disproportionately black: African Americans totaled about 37 percent of the welfare rolls in 1996, though they were only about 12 percent of the population. If Americans didnt like welfare, pessimists reasoned, it was because they didnt like black people. Wags referred to welfare reform as racism in drag and to workfare as slavefare. A perfect example of the Lefts assumption that racist motives prompted welfare reform is a 1999 book called Why Americans Hate Welfare, by Yale political scientist Martin Gilens. Far from being mean-spirited, Americans are actually a fairly generous people, Gilens argued. If they hate welfare, its not because they are tight; its because they dont want to help black folks. With the help of a subliminally racist media, they look at black welfare recipients as lazy freeloaders, just as they did in the Jim Crow South. That Gilens wrote his book just as the Americans who hated welfare were helping to push black child poverty rates to their lowest levels in history seems not to have caused him a moments cognitive dissonance.
Progressive pessimists were especially gloomy when it came to the American economy. You cant blame them for not foreseeing the economic exuberance of the second half of the 1990s; few did. Less defensible was their deep-rooted assumption that poverty was structural, as permanent a part of the American scene as the Appalachian Mountains. Economic growth might be useful to the Mercedes-driving rich but not to the poor, they asserted. No matter how strong the economy, there still wouldnt be enough jobs, employers wouldnt hire welfare mothers, and those who did would offer them only the most demeaning and temporary work.
From this vantage point, any talk about personal responsibility was only more evidence of racially tinged victim-blaming. Nothing distinguishes progressive pessimists from their pro-reform counterparts more than their attitudes toward self-sufficiency. For the pessimists, poverty in America was so severe, and dead-end jobs so demoralizing, that they almost invariably shattered the individual will (and, incidentally, led to such social pathology as child abuse). Thats why, if manufacturing jobs moved out of the ghetto, the pessimists explained, the poor could not be expected to move where there were more opportunitiesthough throughout the eighties and nineties, millions of poor immigrants were doing just that.
For reformers, on the other hand, there was much to learn from immigrant optimism, as its sometimes called. People are capable of far more resourcefulness and resilience than welfare recipients had been given a chance to show, they contended. This understanding was one reason why Bill Clinton rejected the advice of most of his advisors and signed welfare reform in August 1996. Ive always known poor folks, he told Jason DeParle in an interview quoted in American Dream. Ive just never thought they were helpless. DeParles own research reached the same conclusion.
Social science tended to reinforce the dark assumptions of progressive pessimists. After all, as Andrew Cherlins speech to his audience of sociologists suggests, the pessimists dominate the research community. The handful of pro-reformers doing poverty research didnt get much respect, until the Republicans took over Congress in 1994. According to a 2000 article on the politics of welfare reform by Diana Zuckerman, a onetime Democratic congressional staffer, House and Senate Democrats were working under very liberal leadership of many congressional committees, who were used to getting their numbers from groups like the Childrens Defense Fund and the Center for Budget and Policy Priorities. Staffers tended to dismiss the very different findings of more conservative experts as ideological, an irony that would be laugh-worthy if its consequences hadnt been so dire.
But bias is not the most important reason that social-science research supported the pessimists and failed to predict accurately what would happen after welfare reform. In poverty research, labor economistscredited with having the most sophisticated measurementsare key players. But labor economists have a limited framework for analyzing human motivation and almost never deal with actual poor people, a combination that leaves them vulnerable to the cause-correlation confusion that so bedevils social science. Worse, what they cant measureand what therefore never entered into progressive thinkingis the key role that values, or culture, plays in making people into productive citizens who go to school, do their homework, plan their lives, and work to support themselves and their families.
Human beings tend to do pretty much what they are expected to do. When the culture expects self-sufficiency, people will try to achieve it. When the culture sends mixed messages about self-sufficiency, as it did during the old welfare regimeparticularly to the minority poorsome will not try to become self-sufficient. Experts couldnt calculate that simple dynamic, since fuzzy notions like values, norms, and culture are not easily quantifiable. The economists assumed that people were mere rational calculators, maximizers of self-interest, who would naturally try to be financially independent if the optimal incentives were in place. Moreover, used to drawing conclusions from small, experimental programs, they could not model the cultural effects of the large-scale policy shift that occurred in 1996.
The architects of TANF certainly relied on economic incentives, including the definitive oneno benefits after five years. But the bills sticks and carrots added up to something greater than the sum of their parts. In a variety of waysending the federal welfare entitlement, forgoing the customary escape hatch of training programs in lieu of jobsit was an unambiguous expression of a social consensus about work and self-reliance. The idea was not to get people to respond to particular inducements. It was much bigger than that: to try to get them to change their orientation toward life.
The impact on the ground was unmistakable. Sharon Hays, the sociologist author of Flat Broke with Children: Women in the Age of Welfare Reform and no friend of TANF, describes a bureaucracy completely transformed by the new law. With the help of an influx of federal money and firm leadership, local offices spiffed up their waiting rooms, hung inspirational banners proclaiming perseverance, success, and determination, and became job-search and -support centers. Like caseworkers, almost all welfare recipients noticed the public enthusiasm for reform signaled by all the new programs, the influx of federal dollars, and the surge of media coverage that occurred in the first years after implementation, Hays writes. In other words, through TANF, the culture spoke, and poor women listened.
All of this might seem to lead to the conclusion that welfare reform has been a triumph for conservative thinking. That would be overstating things. TANF was never simply about ending welfare dependency. As part of a larger bill called the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), it was designed to improve the lives of the formerly dependent more broadly by nudging them toward middle-class life.
The Left always thought of moving up as a matter of money, not behavior: if people earned a middle-class income, they believed, middle-class conduct and aspiration would be sure to follow. Conservatives tend to see it the other way around: middle-class mores are necessary for economic success. If people adopt bourgeois habits and ambitions, they will work hard, save, and plan, and eventually have the money to make a down payment on a house or pay parochial school tuition. In the case of PRWORA, supporters took this idea a bridge too far. They imagined the work ethic as the engine that would carry all other virtues in its train. Jobs would bring discipline to the lives of poor single mothers and transform them and their children. Work would turn them into bourgeois strivers.
And you do hear stories that seem to support that theory. Take Jewel, one of the three protagonists of Jason DeParles American Dream. After failing repeatedly, she finally got her GED after the book came out, and is now studying for a nursing degree, even while she holds down a full-time job. She is still with her boyfriend of ten years, and he, in turn, has kept straight in the six years since he was released from prison, working during most of that time. Though they havent married, they are raising their son together, pooling their money, and behaving in most respects like a married couplehelping, as DeParle told me, to stabilize and encourage each other.
But taken as a whole, youd have to conclude that welfare reform has not been the extreme makeover that supporters had sought. And the reason is that it has barely touched the single-mother problem. Reform optimists predicted that by heightening womens self-respect and belief in their future, work would make them more marriage-minded. Women, realizing welfare wont support them, may begin to make better choices: demanding more from the men in their lives, delaying childbirth, teaming up with breadwinners, journalist Mickey Kaus theorized. Reformers also hoped that work requirements would act as a deterrent: girls seeing their mothers and older sisters juggling a low-paying job, an apartment, and children, all without a husbands help, would shun such a life.
Moreover, PRWORA had provisions to attack the single-parent problem in a variety of ways, from cracking down on deadbeat dads to abstinence education, experimental marriage-education programs, and bonuses to states that brought down their out-of-wedlock birthrate. The act was supposed to have a clear message: children should be born to married parents who are responsible for raising them.
Perhaps without PRWORA, things would have been worse. After rising steeply for three decades, the increase in the rate of out-of-wedlock births did slow around 1994. Robert Rector of the Heritage Foundation points out that if the pre-reform trend line had continued, 42 percent of all babies would be born fatherless today; instead, the number is about 35 percent. Something different happened in the 1990s that didnt happen in Europe, where illegitimacy rates continued their inexorable climb, he observes. Moreover, the marriage rate among black Americans was substantially higher in 1998 than the trend line from 1960 to 1990 would have predicted. The number of poor couples who were cohabiting at the time of their babies birth also rose. Still, a slowdown in the spread of illegitimacy and more shacking up among poor men and womena notoriously fickle domestic arrangementare not what optimists had in mind.
Ending welfare dependency, then, is not likely to turn poor mothers and fathers into child-centered soccer parents. Though researchers havent found that reform has had any adverse effect on children, they havent found much positive impact, either. Jason DeParles book and others that look at the lives of poor families after reform give a pretty good idea why. Just because a woman has to be at a job at 8 each morning doesnt mean that her childs father has become a paragon. For that matter, just because shes following the dictates of an alarm clock doesnt mean that she is envisioning a better life for her children.
In Adrien Nicole LeBlancs Random Family, the boyfriend of one working exwelfare mother minds her children while he bags crack cocaine in the kitchen. All three of the children of one of DeParles protagonists, Angie, have quit high school. Her eldest son describes his aunts boyfriend, a drug dealer, as his role model. As a teenager, her daughter began spending time with a family friend who happens to be a prostitute. When DeParle questions Angie about why she doesnt try to discourage the friendship, the mother of three huffs, Im not supposed to let my kids visit her cause thats her chosen profession? . . . You dont judge people about stuff like that! Today Angies daughter is 22, with no husband, two children, and a job as a checkout clerk. Its better than being on welfare, sure, but not much for the history books.
What reformers especially didnt factor into their marriage visions was that PRWORA, like so much of American social welfare policy, had little to say to the prospective husbands. While experts sometimes overstate the shortage of marriageable men, it is true that employment levels for men with a high school education have been declining since the 1980s. The situation is especially dire for blacks. At the same time that poor black single mothers were setting off to work in record numbers, the employment rates of their would-be husbands barely budged, despite the boom economy of the late 1990s and despite their rising education levels. (These numbers dont even take into account the 21 percent of non-college-educated young black men who are incarcerated.) By 2000, employment rates for black men were around 25 percent lower than for whites and Hispanics.
Put all of this togetherlow-income single mothers, idle (and sometimes imprisoned) men, children with no vision of a way out (and no help from their schools on that score)and you get a group only slightly different from the welfare families we had before: the struggling, working-single-mother family. So far, it seems better than its predecessor. Life in many inner-city neighborhoods has improved noticeably from its nadir in the late 1980s and early 1990s. Paul Jargowsky and Isabel Sawhill recently published a paper called The Decline of the Underclass, showing a sharp decline in census underclass tractsareas defined by their thick concentration of underclass behaviors, including welfare dependency, dropping out of school, and crime. If neighborhoods are improving, part of the reason, though its impossible to calculate how much, is welfare reform. But at least as big a reason is that many of the most dangerous people have been sent away, so crime has gone down.
So what do policymakers do next? First, TANF, which Congress has just reauthorized for five years onlyafter several years of dickering over child care and work rulesneeds to become a permanent fixture of American policy. The Left will always use reauthorization debates as a stage to push for more funding for day care and a higher minimum wagepolicies that do nothing to change the fragile cultural landscape in which the single-mother family struggles. To make that transformation, more men have to become attached to work and to marriage, rather than to crime or prison. However, unlike with women who were dependent on welfare checks, policymakers have fewer levers to change the culture of unskilled, uneducated men.
As for the revival of wedlock, aside from as yet unproven marriage education programs that will reach only a small number of people anyway, social policy can play only a limited role. Whats needed instead is a blitz of marriage talk from the churches, from the black leadership, from the political class, from celebrities like Bill Cosby, from radio and television spotsfrom anyone and everyone who might get the ear of the welfare population. This campaign will not lead poor mothers and their boyfriends to surge into wedding chapels; most never planned to raise children together, and they have pretty dismal long-term prospects. Rather, the marriage blitz needs to speak to a new generation of young people still able to rethink the possible outlines of their lives.
Policy has given the culture a nudge, but now the culture has to change itself.