- |Forum Futures
- |Ford Policy Forum
- |Forum on Higher Education Finance
BOSWORTH: Now we’re going to turn back to the other end of the income distribution and talk about poverty and the distribution of income. We’re going to have Gary Burtless and Ron Haskins lead that discussion.
HASKINS: There are so many things to say about poverty, so I’ve tried to narrow down quite a bit, and I’m going to talk for maybe ten minutes and then Gary’s going to talk about inequality and opportunity for ten minutes or so, and then we’re going to open it up for discussion.
We have done two things in the history of American social policy that have actually had a clear impact on poverty. One of them was an incredibly sophisticated strategy, which you can see on exhibit 1. The name of the strategy is: give them money. That’s what we did with the elderly.
We used to have huge poverty rates. If you go back even further than this exhibit the poverty rates among the elderly were well above fifty percent. But then we started Social Security and under Johnson we increased Social Security, and I remember one year the Ways and Means Committee got in a battle between Republicans and Democrats about who could increase Social Security the most, and so they kept bidding each other up. So we have increased Social Security many times. And there are very good studies–including one by Gruber and MIT–that show that almost the entire reduction in poverty among the elderly is due to Social Security. So as long as we have Social Security and we at least have some kind of inflation adjustment in there, we’re going to have low rates of poverty among the elderly. Those rates are going to be less than half as much as the rate among children and therein, in my view at least, lays a problem. Why should we spend tons of money on the elderly especially?
And let me add that the money is on automatic pilot. The Congress does not vote on Social Security and Medicare, it just keeps on going. Congress doesn’t even have a budget for this; the CBO estimates how much they’re going to cost and that’s how much money they have. And yet children struggle for money and, for the first time in decades, we’re spending less on children.
So this strategy of giving senior citizens money does reduce the poverty rate. It’s now under ten percent. But among children it’s over twenty percent. So the strategy works, but it’s somewhat questionable.
Now the second strategy has been somewhat controversial. In 1996 we passed a welfare reform bill, and we did two things. First, we had very, very, very strong requirements that mothers had to go to work. There was a lot of thought among conservatives that the states would not really implement the provisions in the federal statue, but they did and they even went beyond it and they threw mothers off the roles if they didn’t meet the work requirements that the states established. You can see in exhibit 2 that the unemployment population ratio for never married mothers–the group that’s most likely to be in poverty, least educated, least job ready– had a forty percent increase in employment over a four-year period around welfare reform. I don’t think there’s anything like this in the Bureau of Labor Statistics records. It was a huge increase in work.
Black children are especially likely to be in female-headed families, about seventy percent of black kids are born outside marriage. Half of the rest go through a divorce, so black kids are especially likely to live in female-headed families. You can see here in exhibit 3 that the poverty rate among kids in female-headed families declined.
Poverty among black kids in female-headed families reached their lowest level ever within three years of welfare reform, and went down even more. Then it went back up because low income moms were tied to the market, which they have not been before. When recessions come, their employment falls off and the poverty goes up. But even after the 2001 recession, and the most serious recession we’ve had since the Great Depression, 2007 to 2009, poverty rates among these kids are still lower than they were. And the work rates among the mothers are higher than they were before welfare reform.
The second aspect of welfare reform was work support. Congress was trying to change Means-tested programs so that you didn’t lose everything if you took a job. And they did it in two ways. First, they changed the rules for food stamps to make it easier to get food stamps if you were working and still had low income. Second, they changed programs that supplemented income. The child tax credit has a refundable part. A social worker can look a mom in the eye and say, get a job, if it pays eight bucks an hour, take it, you’ll be much better off because you’re going to get the child tax credit, we’re going to help you with your day care and so forth.
A lot of Democrats said the conservative first part of the strategy was very mean to do–to really make people work and throw them off the rolls if they didn’t. The more liberal part of welfare reform was to supplement their income with these various work support benefits. But those two things together drove the poverty rate down and even today, after two recessions, it’s even lower. So these two strategies have been very successful.
By the way, the official poverty measures are extremely misleading because they do not include these wonderful benefits that I just told you about. It does not include the income tax credit; it does not include food stamps and so forth. A recent study at Columbia by Irv Garfinkel and Jane Waldfogel and some others shows that the federal safety net programs reduce poverty among the whole population by about fifty percent.
And there are at least two studies using different data sets that showed that in 2009, when we had a huge increase in unemployment, poverty did not increase at all. And it’s because of the safety net. So the safety net is pretty good. The war on poverty did very substantially reduce poverty — it cut it in half. Now all these programs were not passed during the war on poverty but clearly Johnson got us on the trail. Subsequent Congresses added to it. We increased spending until the last couple of years, almost every year. We now spend about a trillion dollars a year, between the federal government and the states, on these programs. So we’re doing a lot.
Now I want to say something about a perspective that I favor. Individual effort really makes a big difference here. If people follow simple rules of what you ought to do to thrive in a capitalist economy, such as graduate from high school, get a job and work full time, and wait until you’re married and at least 21 before you start having babies, you dramatically decrease the probability of being in poverty and dramatically increase the probability that you will be middle class.
As you can see in exhibit 5, if you don’t follow any of the norms, the probability of being in poverty is seventy-seven percent, and the probability you’ll be in the middle class is four percent. If you follow all three, the probability you’ll be in poverty is two percent, and the probability you’ll be in the middle class is seventy-two percent.
This doesn’t absolutely mean that these three things are what cause poverty. These things are associated with all sorts of other things, but we all stress to our kids the importance of rules like this, and people who violate these rules start out with a big problem. For example, if you look at exhibit 6, the probability of being in poverty is four or five times higher for kids in single parent families than married coupled families.
That means that if the percentage of our kids who are in female-headed families is going up, poverty will be going up too; even if our social policies were successful poverty still might be going up. And that’s exactly what has happened. More and more and more of our kids are in single-parent families. You can see in exhibit 7 the non-marital birth rates for blacks and whites and Hispanics — about seventy percent for blacks, around fifty percent for Hispanics. It’s lower for whites but the rate of increase is higher for whites and has been for about the last decade, so eventually maybe they’ll catch up.
So more and more and more kids are living in the situation where they’re more likely to be in poverty. Not only that, but there is almost universal agreement in the scholarly world that the best working environment for children is a married couple family. So you get two big impacts here. One, they’re more likely to be living in poverty, which no one thinks is good for children, and their rearing environment is worse if they live in a female-headed family.
Okay, let me just say one more thing and I’ll say it very quickly. There’s a lot of talk now in Washington, especially among Republicans about poverty and opportunity and maybe thinking about some policies. There are two things they are thinking about, one of which I think is quite interesting and may actually have some impacts, and the other one is also interesting, but it’s going to cost too much money so I’m not sure it would work. The first one is a continuing tradition that we started in welfare reform, and that is that the federal government does way too much and we ought to send these programs back to the state and local level. I think Rubio in the next couple of months will introduce legislation. I know for sure that Paul Ryan will because I’ve seen some early versions of it, and they will propose returning a lot of the power and authority and the money to the states. They say they’re not going to cut the money; in fact, they’re going to give them as much as we spent last year and then they’ll increase it every year. It will be a big fight about how much to increase it, but the states would be able to fashion their own programs. They would do the same thing with Pell grants, but there are all kinds of issues here. If I had to predict, it won’t happen, but there’ll be an interesting debate about it.
And the second thing politicians are looking at is the programs that support work like the tax credits. These would be retained by the federal government and expanded and improved in some ways. The goal is to create two separate systems — one system for people who are not expected to work, like the disabled and other people. They would be the responsibility of the states, and the states would have all the safety net money and have to figure out what to do. They could have stronger work requirements. They could do, roughly speaking, anything they wanted to. And then the Federal government would be responsible for the work system that subsidizes the income and wages of low income families.
So there’s a lot going on and our programs are way more successful than they’re given credit for. They’re bi-partisan in a sense because they’re based on strong work requirements but also on real supports. We spend a lot of money on supports for low income families that work. We’ll discuss all this some more in a few minutes when Gary gets through.
BURTLESS: Barry asked me to talk about inequality, so that’s what I’m going to do. Let’s start with a chart that has launched ten thousand political debates, one hundred demonstrations from the Occupy movement and one political slogan — “we are the ninety-nine percent”.
These are numbers that a French economist Thomas Pikkety and Berkeley economist Emmanuel Saez compiled from the nation’s income tax records. It’s the celebrated chart that shows what percentage of income in the United States is received by the top one percent of income recipients. And you’ll notice that the share in 2012 was only a whisker short of what it was in 1928 in the peak of the roaring twenties. Now you’ll also notice in that slide that there is an asterisk and quotes around the phrase income share. One thing to bear in mind when people show you this chart is that it’s not the spendable incomes of American families, it’s their pre-tax cash market incomes. It’s the market income that you report on your 1040 form. This is not household spendable income. So what does it leave out?
It leaves out Social Security. It leaves out unemployment benefits. It leaves out Medicare. It leaves out Medicaid. It leaves out food stamps. It leaves out the entire United States tax system, which includes the earned income credit that Ron just mentioned. It ignores employer benefits for health care and other things. In short, almost everything that society has done to ameliorate inequality is left out of this picture. And people still claim to be appalled that inequality as measured with this income statistic is as high in 2012 as it was in 1928.
So what did we have in 2012 that we did not have in 1928? We didn’t have Social Security. We didn’t have unemployment benefits. We didn’t have Medicare. We didn’t have Medicaid. We didn’t have food stamps. We didn’t have employer health plans for their employees. To pretend that this is showing the actual distribution of inequality of wellbeing in the United States is quite a stretch because it only includes the market cash income: wage income, salary income, self-employment income, interest dividends, and rent payments if you receive rent payments from somebody else. But it doesn’t include all of the things that society has done to try to reduce inequality and poverty.
The chart is certainly telling us something very important about inequality, but it’s telling us less than what we need to know to evaluate what is happening to the share of spendable income that Americans are dividing. One of the items that’s left out, as I just mentioned, is non-cash benefits from government or employer welfare plans, notably health insurance. In exhibit 9 you can see some Congressional budget office estimates of the importance of these items in the income of rich, of middle income and of low income families in 1980 and in 2010. These items, which aren’t included when you’re just talking about market cash incomes, now represent a quarter of the incomes of people who are in the bottom fifth of the American income distribution. None of them count in the standard statistics on poverty. None of them count in the income share of the top one percent.
If you look at the top one percent on this picture, you’ll notice that all of these items added together only constitute one percent of their incomes, and that was true in 1980 and 2010. These income items are much, much more important to people the further down the distribution you go, and they’ve become more important over time, and they’re not counted in standard measures of the distribution of well-being in our country.
Now consider the numbers in exhibit 10 that the Columbia University social scientist, Jane Waldfogel, and Irv Garfinkel and their two colleagues put together. Contrary to what the loudest complainers tell you, middle and low income families in the United States continue to see improvements in their living standards. If you take a ten-year perspective, incomes in middle income families and lower income families have indeed improved on an after tax basis.
These numbers are comprehensive income measures put together by the Congressional Budget Office. They include the value of the health benefits that our employers give us, that elderly people receive through the Medicare program, that poor people receive under Medicaid. They include food stamps. They include the subtractions from your income because you have to pay taxes. And you can see, in the ten years after 2000 that the top one percent of households saw their incomes fall, whereas in the middle of the income distribution, and especially at the bottom, incomes improved. So the facts aren’t quite as they’re sometimes advertised, but that’s because people are often using measures that don’t include all of the sources of income that people receive in order to pay for their consumption.
That last exhibit was a little bit misleading — why? Because I was comparing the peak of an economic expansion in 2000 with 2010, which was the first year after the end of a very, very bad recession. It turns out that our system doesn’t protect the after tax incomes of the very well to do very well at all. They don’t get unemployment benefits from their losses; they don’t get more Social Security or any other kind of benefits because they’ve run into hard times. The tax system does protect them some, but it doesn’t protect them nearly as much as it protects the incomes of people in the middle of the distribution.
Now if we take a longer term viewpoint, if we go all the way back to the end of the 1970s, exhibit 11 looks a lot more like what you’re used to hearing about. This is after tax income measured in a very comprehensive way, and you can see that the incomes at the top have doubled, whereas in the middle class they’ve only risen by forty, forty-five percent, and the bottom increased about fifty percent.
This is the trend that the Occupy Wall Street movement was about. This is what bothers people, that over several business cycles, over the last forty years or so, there has been a dramatic improvement in the relative well-being of the very well off even if we include the effects of redistributive programs like the income tax system, food stamps and everything else. Yes those programs have sheltered people at the bottom so that they have not suffered, but the long-term trend has been toward big gains in market income at the top, which haven’t been fully compensated by changes in the income tax system or government transfer programs. Exhibit 12 traces what happened to incomes of the people in the exact middle of the distribution from 2007-2010, just before and after the recession.
You can see that this recession was just a terrible disaster for middle income people, measured solely by the market income that they received. They sustained huge losses, almost ten percent of their incomes were lost, wage income, salaries and so on. However, if we take into account market income plus government transfer programs — such as unemployment benefits, increases in Social Security and so forth — some of the loss is compensated for. These are middle income families. They’re not getting food stamps, they’re not getting Medicaid, but nonetheless those unemployment benefits and so on did replace some of the income losses that middle income people sustained in the great recession.
That last line over to the right of exhibit 12 shows what happened to the after tax incomes, including transfer benefits that people received. They fell a little less than one percent. So I just want to underline something that Ron said. The programs that we had to protect people’s incomes worked the way they were supposed to, both the automatic programs that gave people unemployment benefits and Social Security and so forth if they lost their jobs or became disabled. Furthermore, all of the programs that the public loves to hate in the stimulus program protected the incomes of middle income families from calamitous losses that they were suffering in wages, interest dividend payments and so on. Yet, for reasons that only a political scientist could explain to me, the public thinks that the stimulus programs were a big failure. Market incomes for middle class families fell by almost ten percent. But on an after tax, after transfer basis, incomes fell a little less than one percent because of all of the things that are automatically built into the social safety net, and the add-on things that Congress put into the law at the beginning of the great recession. So if there are any political scientists present maybe you can explain to me why the public thinks that these programs, which were so successful in protecting their incomes, are regarded as such a failure by typical voters.
HASKINS: Questions, comments? Yes?
SPEAKER: I understand how you measure the effect on poverty for females who have children before twenty-one. Can you explain to me how you measure poverty for males who have children before they’re twenty-one? How do you identify them? And how do you make the causal connection, since many of them deny that they’ve had children.
HASKINS: First of all, we have a policy that addresses this issue of males not acknowledging their children and it has been immensely successful. We have, for example, a program where we ask males voluntarily in the hospital, who are not married to the mother, to sign on the dotted line that yes they’re the father. And then they wind up paying child support and we have the toughest child support system you could imagine. So it’s a little misleading to say that you have all these males running around who have kids that they don’t acknowledge. They do acknowledge it and we have other programs where we actually go out and look for them.
The way the data on males is collected is based on regular Census Bureau data. They do interviews with a sample of the population, including single males. Now single males are probably underrepresented because they are sometimes hard to find, so there’s some truth in what you’re saying that the —
SPEAKER: They’re in jail.
HASKINS: Single males are underrepresented, however they’re still in there. I don’t know what the percentage would be but it would be well above half, I’m quite certain of that. So I say that this does apply to males and to females. Females obviously wind up with the children in around ninety percent of the cases, and that makes their life tougher. We now have a lot of good data in the last ten years because of the “fragile family” study that followed a sample of people who have babies outside marriage. We know a lot more about these families than we used to. The mother goes on to another relationship, the father goes on to another relationship, and the mother may have children with another man. Think of the complexity in these households. In fact, the Department of Health and Human Services is now supporting research under the theme, complexity of poor families. Think of yourself as a child in these families. You’ve got maybe a sibling, but then you have half siblings, your father is outside the family, and your half sibling’s father is outside the family. Both of the fathers, we now know, are trying to see the kids. That tails off over a period of years, but they’re still trying to see the kids. They’re often having problems with the mother. The second father doesn’t want the first father around — so there’s all kinds of drama in these households. This is not a good circumstance under which to rear children.
SPEAKER: Looking at your first slide about the poverty rates by age groups going back to the fifties. Has the definition of poverty changed over that time?
HASKINS: For the official poverty measure the answer is essentially no. It takes into account inflation but otherwise it has stayed the same. That’s one of the flaws in the measure. Years ago, 1991 I think, the National Academy of Science did a big study. A couple of people at Brookings like Becky Blank were on that Commission and they made recommendations about a much improved poverty measure. It improved in an enormous number of ways, but for one it took into account the kind of income that both Gary and I have included in our charts, which is ignored under the official poverty measure. The second thing was it took into account changes in the average family living circumstances, so it represents something like a relative poverty measure, relative to the other people in the society. It’s called the supplemental poverty measure. Becky Blank went to the Census Bureau and eventually became the acting secretary. While she was there she declared that we’re going to collect all the data we need and publish the supplemental poverty measure every year. Now we have this special measure, which is the measure that the group at Columbia used. It gives you a much more accurate picture.
So the official poverty measure has not changed essentially, but the supplemental poverty measure really makes a big difference. By the way, my prediction is we’re not going to change the official measure. Let me tell you why; it’s real simple. I was in Congress and worked on the Ways and Means Committee for fourteen years. A lot of our formulas for distributing money among the states are based on the official poverty measure. If you had a better poverty measure, it would change the way that money is distributed. The surest way to guarantee a fight in Congress is to change the way states are treated under any formula — you’re asking for warfare. So I don’t think it’s going to happen in our lifetime.
SPEAKER: Given what you’ve just shared with us, is the implication that it’s actually working pretty well and we don’t really need to worry about inequality? Why is it that people have this negative perception and begin to occupy Wall Street? And what are the implications for policies around financial aid and higher education? We’re talking about poverty and I don’t know whether these kids ever go to college.
BURTLESS: Generally speaking, when you hear people refer to numbers about adverse trends and middle class living standards, or income standards of the poor, they are using an indicator of people’s well-being that is not very comprehensive and excludes most of the things that society has added to the market distribution of income over the last fifty years. They think it is a comprehensive measure of how much people have in spendable income. And that is incorrect. The relative trends are not nearly as unfavorable as is reflected in this picture. Nonetheless, I did emphasize that if you take a long enough perspective, since the end of the 1970s, changes in market income have indeed had a profound effect on how much prosperity has been obtained by different parts of the income distribution.
Schools that get fat donations from their contributors — Harvard got $150 million dollars yesterday — may see gifts that size a little bit more regularly given income trends like this. On the other hand, if you’re a college or university that does not have a very good endowment, does not have three billion dollar hedge fund managers among your alums, you might be a lot more worried about market income trends; because you have to finance your operations with people who are coming and who are paying the bills or whose parents are paying the bills. As a result, there might be a growth in the inequality of institutions that parallels this inequality in the more general public unless affluent donors give huge gifts to some of the poorer institutions too. That’s what I would worry about.
SPEAKER: Do we need to worry about the minimum wage?
HASKINS: Let me add three things to this discussion directly relevant to your question. A big part of the problem is the number of jobs that we have. That has declined. Our unemployment rate would be much higher if the employment to population ratio had not changed over the last ten or twenty years. It’s going up for females, but it’s going down consistently for males and people draw out. They leave the economy so they’re not in the denominator and so our numbers really cover up how serious the problem is. A lot of low income people just cannot find jobs. Part of the reason that the 1990s we were so successful was because there were so many jobs. And the recovery has not generated jobs like any past recovery. It will be something like 2018 before we get back to where we were. That’s something to be very worried about, especially if we want people to earn their own money.
The second thing — I just want to add to what Gary told you because it’s astounding to me. Over ninety percent of the take from the federal income tax system comes from the top twenty percent. We have one of the most progressive tax systems in the world, and by contrast the bottom forty percent pays negative taxes on average. We send them money through their income tax, child tax credit and so forth. And those numbers constantly improve, in other words, the rich pay more and the poor pay less. This, by the way, is the origin of Romney’s famous comment about the forty-seven percent saying that they don’t contribute.
Now, the third thing is that college is the best anti-poverty, middle class program imaginable. We have really, really good numbers on this. A University of Michigan panel study of income dynamics has been following families since the sixties — so we can take parents and their children and we can compare their income during the prime earning years. Here’s what they showed. If kids had parents that were in the bottom twenty percent, the probability that they would be in the bottom was forty-five percent. But if they get a college degree, the probability is sixteen percent. Forty-five percent to sixteen percent. This is the most effective anti-poverty program known to man or God.
By contrast, if you look at the top twenty percent the probability that they make it to the top, regardless of whether they go to college or not, is about five percent. If they get a college degree, it’s almost twenty percent. This kind of impact on income is just spectacular. So we have a lot of programs where we try to get kids ready for college, unfortunately those programs fail. We have not figured out how to help low income kids make it to college and do well in college. A lot more make it to colleges, but it’s very low compared to kids from families who have more money, and their drop-out rates are even higher; often that leaves them with a bill because they borrowed money to do it. So the idea that we ought to get them all to go to college is flawed. But mostly it’s flawed because they’re not ready for college, and the programs that we designed at our universities and even community colleges to make up for what they lost are a bust too. We have all kinds of evaluations that show that. There are a few that are somewhat successful. But if you’re a college administrator and you want to do your bit for poverty and for opportunity in the United States, figure out how to get the K-12 system to educate those kids better and then help them more when they get to college.
SPEAKER: I’m interested in the three measures that were mentioned that are highly correlated to not being in poverty. Is there any optimism that there might be some approach by Congress that really gets at those three measures?
HASKINS: Barry invited Gary because Gary’s the optimist here. So Gary, answer the question.
BURTLESS: I’m not optimistic. I just did a careful analysis with a lady from the University of Wisconsin— Maria Cancian. She’s going to be the new Assistant Secretary of HHS for these very programs. She and I looked at the census data going back to 1970, so forty years of data through 2010, and we looked at family composition changes. We found that marriage just plummets. The family form that has increased the most is a single mom with children and this is just completely wrong. I don’t think we can overcome it, especially because politicians are very reluctant to talk about marriage. They may give some broad speech or something, but every time I talk about marriage I get all kinds of nasty mail, ‘who are you to judge?’ and that sort of thing. But if you look at it from a kid’s perspective the decline in marriage is the worst thing that’s happened to the probabilities of their well-being as an adult.
SPEAKER: Do you think President Obama may have a unique advantage being African-American to do things that other politicians would be reticent to put forward? Do you see any leadership coming forth?
HASKINS: He has done a little bit. He’s talked some about marriage, but not more than Bush — Bush talked about it a lot. But before Bush, I don’t think anybody talked about marriage more than Obama. So he’s been okay. But I wish he would do a lot more, especially for young black males. Everything I’ve talked to you about, quadruple it for young black males. The work rate among young black males under age 24 is like forty-five percent. Women will actually say that they’ll have his baby, but they won’t marry him. They don’t want to be stuck with this guy because he doesn’t work, and if he does have income he doesn’t necessarily share it and bring it home. Some of the guys are violent. I mean, there are a lot of problems with these guys. So I’m not optimistic in general and I’m even less optimistic about minority families.
BURTLESS: There are, however, some countries that have even worse records of stable relationships between men and women but have much lower poverty rates. If the society is prepared to give enough supports to single parents so that they can raise their children at least in lower middle class comfort, then you can have a society with a low poverty rate. There are societies that have even higher rates of marital dissolution and lack of marriages forming than the United States but don’t have child poverty rates at the level for single parent families that we do.
HASKINS: Let me add to that though. I’m going to be cautious here because I don’t want to necessarily challenge what you say. In Europe what Gary says is true, but those families that are not married tend to stay together, especially if they have children, at much higher rates than in the United States. So that compensates a lot. If we could keep these parents together working as a team, then I think there would be a real chance.
SPEAKER: And there’s positive reinforcement and the importance of education. They’re very supportive of education and insistent that their children get educated even if they’re not married.
HASKINS: I agree with that. I’ve even conducted some of these studies myself. That is exactly what they say, but, do they get married? Do they try to get in a neighborhood where they can have a better education? Their decisions do not reflect the level of commitment that their words do.
SPEAKER: You can be committed to your child and be a single poor mother, but you may not have the education to decide to live in an efficiency apartment on the other side of town so my kid can go to a better school. You also may lose your social network if you move to the other side of town. Maybe your mom, who’s 65, is taking care of your kids so you can go work at Wendy’s. So leaving your social network negatively impacts what I would consider to be the right decision for the kids.
HASKINS: Yes, I certainly agree with that.
SPEAKER: I have a question for you about a new trend that I’m a part of, which is I’m a single mother by choice. I don’t have the stress of a negative male in my child’s life. And I have a master’s degree. I have friends with Ph.D.s as well who are doing the same thing. And we will never be in that bucket. Is that being skewed out of the data or are we lumped in?
HASKINS: I’m not aware of any studies that compare mothers like you with the typical single mother who might not have been married, has a conflictual relationship with the father and so forth — I just don’t know. I do know this: we do have studies of mothers and fathers who are married and have high conflict and that’s equally damaging to children. So there are some families that probably should break up. But that still does not necessarily reflect on your situation.
When we analyzed the Census Bureau’s marriage data by income, everybody goes down except college educated women. Starting in 1980, their marriage rates stabilized; they hardly ever have babies outside marriage, and they have a high marriage rate. They marry later, but they still have high marriage rates which really have not declined since 1980.
BURTLESS: And the duration of their marriages is longer.
BOSWORTH: Okay, thank you. Thanks to all for coming and safe travels home.