David Harris: Reflections from the human services side of Health and Human Services: Evidence, challenges, and public perceptions

October 20, 2010 0:46:00
Kaltura Video

David Harris shares reflections from his six months as Deputy Assistant Secretary for Human Services Policy. October, 2010.


>> Welcome all of you here today on behalf of the Ford School and our Center for Public Policy in Diverse Societies.  I'd also like to extend a very warm welcome to our speaker, David Harris.  David is the Deputy Assistant Secretary for Human Services Policy at the U.S. Department of Health and Human Services, and he leads the Office of Human Services Policy in the Office of the Assistant Secretary for Planning and Evaluation.  He's actually called that unit something akin to a "think tank" that analyzes federal policy issues.  So that's a way to think about what that unit does.  He's on leave from Cornell University where he was Deputy Provost, Vice Provost for Social Sciences and Professor of Sociology.  As Deputy Provost, he focused on a number of key provost office priorities including academic planning, admissions and financial aid and diversity.  As Vice Provost for Social Sciences he was responsible for leading the development and implementation of university-wide efforts to enhance the social sciences and for providing a social science perspective on Cornell's policies and priorities.  He has extremely broad research interests in social stratification, race and ethnicity, social identity, and other areas related to public policy.  And his work has applied theories from sociology, economics, and psychology to empirical studies of racial and ethnic disparities in socioeconomic status, the fluidity of race, and racial and nonracial determinants of residential mobility.  I know that for many of you, he really does not need any introduction.  His academic career started right here at Michigan where he started as an assistant professor of sociology and taught on the faculty here until 2003.  He was actively involved in the early days with the National Poverty Center, and he was responsible for one of the NPC's edited volumes, a book, that is called, "The Colors of Poverty:  Why Racial and Ethnic Disparities Persist," which he coedited with our own Ann Lin.  We couldn't be more pleased to welcome David back to Ann Arbor, but before I call him up to the podium, I did want to say that today's event is the second in our annual Distinguished Lecture Series hosted by the Diversity Center this year.  We have a number of additional events planned for the fall, and I encourage you to take a look at our website, which we've actually just recently revamped and updated, to learn about those activities.  Our next session, actually, is very soon.  It's coming up on Friday, this week.  It will be at four o'clock where we'll host a book talk and a signing by Dante Chinni, the creator of "Patchwork Nation," which is an innovative effort to understand voting patterns across a wide range of community demographic characteristics.  E.J. Dionne in reviewing the book calls it, quote, A wonderful book that takes us beyond the red and the blue while also explaining social, regional, and economic roots of partisan and ideology [sic] color wars.  So I hope that many of you can join us on Friday.  But with that, I am delighted to turn the floor over to David Harris.  
[ Applause ]
>> Thanks, Sue.  
[ Applause ]
Thanks, folks.  It really is a great -- you always say this, but this time I really mean it -- it really is a great pleasure to be here today at the University of Michigan.  As Susan said, it's where I got my academic start after graduate school.  Started first working with Sheldon and then on the faculty in sociology and pop studies and at Survey Research Center, and actually [inaudible] a little bit here at the Ford School, taught a course -- a graduate course on race, class, gender inequality my last year here at Michigan.  So it was pretty exciting when I got a letter from Susan back a couple months ago asking if I'd come give a talk here at the University of Michigan at the Ford School.  And I thought, well, it would be great to go to Ann Arbor.  I haven't been to Ann Arbor much since I left, but then the question was, well, what are we going to talk about?  And in my role as a government official I thought, what am I going to talk about?  And so I decided that I used this time to talk about three issues -- sort of bundling things into three categories, and they're three categories -- three things that I spend a lot of my time thinking about, a lot of my time working on in Washington, and they're three areas where I think academics has something to say.  We're both -- those who are here in master's program, PhD program, I think through your research, through your education, things you'll be doing soon, but also the academics, researchers, I think have something to say.  And there's plenty of opportunities for the public to comment on these things but I thought this would be another venue, another opportunity for that to happen.  As you can see up here -- I'm getting feedback here, echoes of some sort -- as you can see up here, we've got -- oh, and this is what happens when you go PowerPoint and try and go a little bit fancy.  These didn't all say "No. 1" by the way.  You think we can't make decisions in government, we can't prioritize in government, right?  Everything's No. 1.  I actually had prioritized, but it got undone.  There was No. 1, No. 2, and No. 3.  I'll see what else happens as I go through.  So the first No. 1 is evidence I'm going to talk about.  As you probably know, there's been a huge push in this administration on evidence based policy, and there are an awful lot of question about what exactly that means, how you implement it.  I'm going to talk about that some.  The second is challenges.  What I'm going to talk about here is what I've long believed is one of the critical issues to successful antipoverty policy, and that's men, and in particular, it's issues around incarceration.  And so I'm going to talk some about this.  And third, public perceptions.  So a little bit different from what we usually worry about as academics.  We're about doing the research and what's the answer and what should the policy be.  I want to take it back and step and maybe influence somewhat by my time as a senior administrator at Cornell and think about the public perceptions, think about the message and other things that are critical to the policy, as critical as having the right policy.  So before I get into talking about those three things, let me just give you a little bit of an overview on HHS, Human Services at HHS and ASPE, the area where I'm involved.  I know -- if you don't mind holding questions until the end, I think that will be best, unless there's some question of clarification, which is unlikely to happen in a talk like this.  There's no formulas or any such things. HHS.  So HHS, I didn't know all of this about HHS six months ago.  HHS, of course, I knew is a principal agency here for protecting the health of all Americans and providing essential services for those in need.  There are upwards of 65,000 employees in the Department of Health and Human Services.  They aren't all in Washington, they're all over the place.  We have places in Rockville.  We have regional offices around the country.  They're spread across 11 operating divisions.  Operating divisions are those divisions at HHS that run programs for the most part.  And there's 18 staff divisions.  So staff divisions range everything from HR, budget, legislative affairs, a whole range of things fall under the staff divisions.  The operating divisions are the vast majority of the employees.  There's 300 programs run out of HHS and a large number of grants programs run out of there as well.  As you can see here, the annual budget, $850 billion.  HHS is a huge part of the federal government.  About 80 percent of that has to do with Medicare and Medicaid.  HHS has primary responsibility for implementing health reform.  So the big game, the big story, the big issue at HHS is health reform.  And it's been really a treat as someone who doesn't do health reform, as I said, knows very little about health, I've been learning a lot since I got there, and my area's not -- I don't work in the health area at HHS, but it's really been fascinating to be in the meetings for the last six months, sort of behind the scenes of health reform, which is a critically important change in American policy, but also incredibly complex.  And there's just a lot of things that have to be hammered out.  It's been very interesting to see how we're trying to make this happen.  We will get it done but it's quite complex.  The other, of course, is there's this other part of HHS which is the human services, and there I think it's fair to say, maybe some of my colleagues in other departments will quibble, but I think HHS has primary responsibility for antipoverty programs.  A lot of poverty -- antipoverty programs you think about are run out of HHS.  We work, of course, with folks in HUD and ED and other places, but HHS is a critical player with respect to antipoverty programs.  So that's HHS.  The vast majority's health, but there's this other part of HHS which is human services.  So where is that?  Human services at HHS is overwhelmingly in ACF, administration for Children and Families.  ACF runs a bunch of these programs you know, it's antipoverty.  So TANF, which used to be welfare.  You've got child support enforcement, child welfare, foster kids, just an area that I hadn't really come into contact with that much before I got to HHS; somehow on my work it hadn't really come up very much.  Just a really important and really difficult area, and a lot of kids in very, very situations, and a lot of really thorny policy issues which is a scenario I don't know how much is going on at the Ford School.  You guys talk about it, but it's really an area that I think could use a lot more attention from academics and from others, the area in child welfare.  Disaster response and recovery; big surprise to me.  I actually spent a fair amount of my time late spring and early summer on the oil spill related issues.  I had no idea I'd be doing things like, that but HHS has responsibility for the public health, recovery and response with respect to disasters, and the part of that that deals with human services is actually in ACF, and our role as a liaison with ACF and the rest of HHS, got heavily involve in aspects, the human aspects, not shellfish and all that stiff, the other parts of the oil spill.  LIHEAP which is around assistance to people for heating and cooling expenses; Head Start; and of course, there's the Fatherhood and Marriage Initiative.  So those are the big things that happen around human services and across HHS, and they're sitting in ACF.  So what's ASPE?  Well, ASPE -- I only knew about ASPE -- the big thing I knew about ASPE before I got this call and was asked to come join ASPE was, of course, they fund the poverty centers.  Sheldon knows that.  A lot of you probably know that.  You think ASPE, word association, ASPE Poverty Centers.  But there's a lot more to ASPE than the poverty centers.  It's an important piece but there's a lot more.  ASPE is one of these staff divisions.  We sit in the office of the secretary.  We're principal advisors to the secretary on policy development.  We're involved in a lot of things -- policy coordination.  So we're heavily involved in activities coordinating across HHS, and often, a lot of our work is between HHS in other departments.  So I spend an awful lot of time with the people from HUD, ED, Justice, and other places.  Legislative development, working on leg proposals that come up, often, but not always from the operating divisions to trying to figure out which of those we want to move forward.  Strategic planning for HHS as well as operating divisions.  And there's a whole other piece of our portfolio, as Susan said, it's sort of the think tank part.  We are a piece that tries to fund projects that are thinking both a few months ahead and several -- couple of years ago.  We're funding some of our -- well, some of our work is sort of quickly trying to understand answers to things. So for example, on poverty guidelines, there's a poverty rate, which we all know, but there's also something called the poverty guidelines which is related, and the poverty guidelines actually affect eligibility for a number of programs around the country, school lunches amongst other programs.  We had a whole bunch of things we had to deal with very quickly this year because for the first time since we've been doing this, we had a deflationary period, and that was something that was not anticipated in the legislation.  Had to quickly figure out if we allow these rates to go down, how many people will lose their access into what programs.  That's a very quick -- but we also have a longer run things around research.  It's four offices.  DALTCP, which is Disability Aging and Long-Term Care Policy, health policy, human services policy, and science and data policy.  So I'm the Deputy Assistant Secretary for Human Services Policy, this part of it.  There's about 30 people in my office including Kim Clum, as some of you may know, who's a recent alum of Ford School in anthropology here.  We have the poverty centers, there's a national poverty center here, of course ;there's three other poverty centers, area poverty centers.  There's the National Center for Family and Marriage Research.  With respect to coordination of across -- to my surprise, the first couple of months I was in the role, I probably spent half my time on homelessness.  I had no idea I'd be doing that.  But about half my time probably, because in June, the federal government rolled out the first ever federal strategic plan to end homelessness that had five and ten year goals on ending various forms of homelessness.  Worked closely with HUD and others on that.  So the research projects and the leg proposals, which have been quite interesting trying to figure out.  That's why we we're waiting with bated breath for two weeks from now to see what we think will happen -- what will happen with Congress and what that implies for the work we're trying to do.  So that's the setup.  That's the background.  That's the context in which this is occurring.  Let me talk about the three things.  So the first one's evidence based policy.  So this long quote here is from former O and B director Peter Orszag.  Peter Orszag was director of O and B from the beginning of the Obama Administration until July I believe it was, this previous summer.  This is a quote -- the blog post that Peter Orszag put on the White House blog about six months after President Obama took office.  And the keys here are -- basically what he's saying is, for a long time the federal government has funded programs and it hasn't been clear that those programs are effective.  It's not been clear that there was evidence for those programs, but we kept doing it.  And we can't do that anymore.  Basically what we have to do is we have to think about evidence. We have to -- we have to put our money into supporting generation -- whoops -- people generating evidence, doing research to generate evidence, but we also have to make hard decisions.  And we have to ask, is there evidence for this program?  And if the answer is no, we have to think, maybe we shouldn't be doing it.  And if we do a program and we can't show that it's effective, we have to think maybe we should stop doing. Maybe we should do something else. So that's evidence based policy.  And two of the key places where it's mentioned in this blog post -- this isn't the whole thing -- are teen pregnancy and home visiting, which are two areas that with both sitting squarely in HHS, and we end up getting involved in.  So evidence based policy.  Now, I think it's -- well, especially for a bunch of academics who spend their lives thinking about evidence and research, hopefully, it's difficult for you folks to think, we shouldn't have evidence based policy, right.  I mean, if you knew that one group of policies worked and another group didn't, you would generally want to support the policies that, you know, we thought would have an impact.  But when you move to evidence based policy, there are an awful lot of questions that come up.  And what we've been doing in sort of this, if you will, this think tank piece of HHS and working with others across the federal government is trying to anticipate some of those questions and trying to answer some of those questions.  And, again, this is an area where I think a lot of folks in this room, a lot of folks on this campus and other campuses have a lot to offer about how we ought to think about this.  One of the first questions is, well, what counts as evidence?  This experimental design, of course, right. What if we don't have experimental design?  Can anything that isn't experimental design count as evidence?  So quasi-experimental design, is that okay or not?  What if -- what are the implications of saying it has to be experimental design?  A long of things in human services, it's just not feasible to think that you'd have a random assignment across some things before you knew that they worked.  So there's one huge question a lot of methodologists and others [inaudible] chime in, evaluation people -- what counts as evidence?  But then you get to some other issues.  We can tell if a program's effective, but it's pretty clear that more -- I'd say just as if not more important than asking the question, is the program effective, is asking the question, why is the program effective.  Because what we're interested in is not saying, wait, that was great, that worked.  We're interested in taking this and trying to replicate this, trying to scale it up.  If you don't know why it worked, it's awful hard to figure out how to replicate it and how to scale it up.  It's very difficult, but it's something we've got to get better at doing is trying to figure out why things work.  And that's actually one of the reasons I think that -- and I don't think -- I don't think there's disagreement among people I'm talking to in Washington about this, but the notion of evidence based policy, it kind of needs to be accompanied by something called theory based policy.  So you need to have evidence but you also need to have a theory about how things work and why they work that way to allow you to later replicate and scale up.  But then you get into another issue, which is -- I like to say is, you know, evidence anywhere is not evidence everywhere, all right.  So evidence anywhere is not evidence everywhere.  So if something worked in a particular place in a particular economic context, that doesn't mean that's evidence for every subpopulation in every part of the country in every economic context.  So you have to think long and hard about what kind of modifications will be required to make this work in a different setting for a different population.  And I'll give you an example at the end where things didn't go so well.  And lastly, this issue of fidelity.  This is an issue that the evaluation people talk about an awful lot.  Fidelity.  Fidelity basically means, if you had a program that -- the program was run, this demo, you got the results you wanted, you need to have fidelity to that model.  Which means you do everything that was done in that model.  If you don't, you have deviations from this, you don't have fidelity.  But the question is:  What exactly does fidelity mean?  As I said earlier, there's a question about what are the key elements that you need to have, but there's also a question, if -- fidelity, for example, if you had a population, you had materials for them to read as part of this program, right, and those materials were in English, and then you go to south Texas, right, fidelity actually would mean changing them to Spanish probably, or having an English and a Spanish option.  You have to think about the context when you think about fidelity.  If you're -- if you're -- if you're a stick in the mud about fidelity and you say no, no, no.  These are exactly the programs we used -- materials used, it won't work.  So a couple examples:  The Healthy Marriage Initiative.  So this has been out there publicly.  The Healthy Marriage Initiative is something that started in the previous administration.  It's based on some work that primarily came out of Oklahoma that showed that looked like we could positively affect relationships, positively affect marriage -- strength of marriages and so forth.  And we implemented this, I believe it was nine sites across the country.  It was primarily -- not exclusively -- primarily tested, developed on more middle class populations, on primarily white populations in Oklahoma.  Some of this work also came from college students, work with college students, psych labs and that kind of stuff.  Well, one of the places it was implemented was in Baltimore, and the group in Baltimore was overwhelmingly black, overwhelmingly very poor, and, in fact, unlike some of the other groups who were cohabiting or in marriages, this is a group where a lot of these folks were barely in a relationship.  They had a kid together perhaps, but they were barely in a relationship.  Well, not only didn't we get the same positive results, what was troubling about it was that we actually got a significant and a substantial increase in rates of serious domestic violence.  And we're not [inaudible] I just had a briefing on this and said, "Well, what is serious domestic" -- being a good survey person, right, "What is serious domestic violence?  What does that mean?"  Punched, used -- assaulted with a weapon, sent to the hospital.  We're talking serious, you know, things that aren't ambiguous.  We had something like a 60 percent increase in the rate of ser -- serious domestic violence among the treatment group as opposed to the control group.  So why did this happen?  Well, we're not -- frankly, we don't 100 percent know.  We have some good indications about it, but my sense is that part of it has to do with this issue of evidence anywhere isn't evidence everywhere, all right.  Having a program that worked really well for a more advantaged, more white population in a different part of the country might not work for a population where we tried it in Baltimore.  And in this case, the consequences were pretty significant and pretty disturbing.  We learned a lot from it basically.  Home visiting.  Home visiting and teen pregnancy are extremely active right now.  There's whole questions about what's evidence.  And so you can go online and see, we've had reviews done by MD I think -- we had reviews done by Mathematica out of all the research we could find on home visiting, all the research we could find of teen pregnancy, and then identified -- in teen pregnancy I think it was 28 programs.  28 programs that we believe have evidence.  And this is a grant competition where you can apply -- states can apply and say, what should these 28 programs are they going to use.  But there's some really important questions that are going to have to be answered as we were reviewing these applications, right.  If you say, we're going to use this in Phoenix," right, this is part of Arizona, so we're going to use this in Phoenix, well, the question is:  How are they going to modify it so it works in Phoenix?  Who has the expertise to review the application?  What do we do with it?  So there's going to be a lot of important questions.  Obviously, don't want to go back to a world that isn't an evidence based policy, but there's a lot of difficulties.  Whoops, sorry.  That's the telephone.  Second -- and it's still No. 2, so I did prioritize.  So for a second imagine you're in a poverty class, right.  So let me give you some highlights from inequality courses.  Poverty rating, right, this is a 2009, came out a couple months ago -- or a month or so ago.  Poverty rate in the U.S. is 14.2 percent overall; it's higher among kids; it's about a fifth of American kids are living under the poverty line.  And you can see that there are racial differences, which is N, and Lin and I wrote this book -- this is why we have the book.  You can see whites and Asians on one end of this continuum, you know, just around ten percent plus or minus poverty rates.  Blacks, Latinos on the other end, 25 percent or so poverty rate.  And that's troubling but I always like to put on here what the poverty line is, because then you realize, this is more troubling than you think it is at first; because the poverty line, family of three -- that's mom and two kids is what I put up here -- $17,285 a year.  And my favorite exercise when I was teaching with undergrads is I would have my students break up into small groups and I'd say, "Okay, you're a mom and two kids.  What are the necessities?  What are the luxuries?  Let's price out the necessities, and let's figure out what it is.  Now, what's that for a year?  And then figure out how we get back to $17,285."  And it's only when you do something like that that you really realize how low $17,285 is for a mom and two kids.  So first thing we know, we've got a lot of poverty in the country.  The poverty rate actually went up.  It didn't go up as much as people were worried it would, shouldn't be surprised it went up given the economy, but we have poverty and we have disparities by race and ethnicity, and, in fact, the number of people who were under some livable amount is probably higher given to where the poverty line is.  Second, test score gaps.  You can see them by race, you can see them by class, they're large.  And in the book I did with Ann Lin, I thought the most interest -- one of the most interesting if not the most interesting graph in the entire book is in the chapter on education.  I'm sure you all have it by your nightstands.  If you'll turn there tonight and look at it, you'll see there's this chart, and basically there's three curves.  There's one, two, three.  The gap between them's pretty constant across the whole space, right.  What you have on the y axis, I think it's verbal ability, what you have on the x axis is age from -- starts at 36 months and it goes to age 13, okay.  Now, what are these groups?  The groups are higher income whites, lower income whites, higher income blacks, lower income blacks; so there's two things that are really interesting about this:  One, there's only three curves but there's four groups.  That's because the higher income blacks are sitting on top of the lower income whites for pretty much the whole space.  So you really see three.  If you look, you can see four, but first glance -- first time you look at it, you just see three.  That's the first thing that's interesting. Tells you there's a race/class interaction.  The second thing that's interesting, and I think this is really fascinating, the gap at age three is the same as the gap at age 13.  So that tells you schools are part of the story, but a huge chunk of the black/white test score gap we're worried about is before kids ever enter the classroom.  It's there at age three, okay.  So that's test score gap, right.  So we've got some other challenges from the human services side in America.  And the last one I'll share with you here is infant mortality rate, which I always tell my students I think is the most disturbing indicator in -- in America. One of the most disturbing indicators around individual quality of life.  It's impossible, I would argue -- I submit it's impossible to blame the victim in this case, right.  You can't do it.  You can blame people for being unemployed, you can blame them for not getting an education, you can't blame them for not living to see their first birthday.  That's what this is.  And this tells you that in America it's -- you have -- overall it's 6.8 -- 6.8 -- I'm blanking -- 100,000, right? I think it's 6.8 out of 100,000 live births don't make it to their first birthday.  That's similar to Hungary, Poland, and Slovakia, and it's not as good as a whole bunch of countries we want to be as good or better than or you expect given the socioeconomic status -- Japan, Denmark, England, Cuba, right.  They all have numbers that are as good -- or those numbers are all better than ours.  But the story gets more disturbing when you look with subgroups.  You see that for whites and Mexicans --  and here this is this paradox that's been talked about a lot -- you see the number's 5.8 to 5.6; Asians 4.9; but here's African Americans again, 13.7.  So what that basically tells you is, if you're born to a black mother -- because we don't -- that's how we handle interracial couples by the way, it's just race of mom -- so if you're born to a black mom, you're -- you're -- you're only half as likely as kids born to a white mom to live to see your first birthday, okay.  So what do we do about this in the federal government?  We have a lot of programs.  A lot of programs that are trying to deal with this.  Those programs are overwhelmingly focused on children and primary caregivers.  And those primary caregivers are basically -- overwhelmingly, that's women, okay.  So we've got programs as I mentioned some of them before -- we've got TANF, we've got Head Start, we've got [inaudible] nutrition, we've got WIC and SNAP -- SNAP used to be food stamps, as you probably all know -- race to the Top, Promise Neighborhoods, coming out of Education.  We've got EITC, public housing, Medicaid, CHIP.  We've got health reform on the way but for now Medicaid and CHIP, these are almost all -- on most of these you can get it.  It's possible if you're a single man, but it's very difficult.  But what about men?  What about prime age men?  What about men who are, say, 25 to 35 who live in poor communities, who -- the people Al knows, right -- Al Young [inaudible] college -- people like Al Young study, right, people who didn't graduate high school, they have weak work histories, they've been incarcerated.  What about those guys, all right?  What do we have for them?  What's the story with them?  Let me give you a few facts about these men.  First, let's think about earnings and employment.  So if you look at individuals who are nonresident parents with a child support order -- these are overwhelmingly guys, okay -- nonresident with a child support order, 32 percent of them, almost a third, have zero reported earnings, okay; another 19 percent have less than 10,000 a year reported earnings; and 14 percent have between 10 and 20,000 which tells you about two-thirds of the guys out there with child support orders, they make less than $20,000 a year.  How about education?  Now look at those who are working -- so we've already cut out the bottom folks who aren't working at all -- look at those who are working nonresident parents, again, largely men, with incomes less than 200 percent of the poverty line.  What's their skill set with respect to education look like?  20 percent don't have a high school diploma or a GED; another 50 percent have no more than a high school diploma or GED; so 70 percent of these guys don't have an associate degree, bachelor's degree, graduate degree any of those things we think are critical in the new economy -- 70 percent of those guys.  But the other part, the one that I think we don't talk about enough, these -- the first two, education, employment, we talk about those a ton, we have for decades.  But there's this other piece that we don't talk about very much and that's incarceration.  What I like to say is that, trying to understand the intercity without thing about incarceration and life course is like trying to understand affluent areas without thinking about college transitions.  You can't do.  Incarceration's all over the place in these communities.  It explains a ton about these guy ,about their families, and about their communities.  What do the numbers look like?  Well, this is '07 data.  If you look at men between the age of 18 and 55, two and a half percent are incarcerated currently.  This is only current, not lifetime.  You've got big race disparities again, and it's the same kind of things that you usually see.  You usually see whites and Asians on one end of it continuum and blacks and Latinos on the other.  You can see it here.  You've got 1.4 percent and .6 percent respectively of whites and Asians incarcerated men; you've got 6 -- 8 percent of blacks, and 2.8 percent of Latinos.  But that's all men.  I like to hope that I have a pretty low probability, okay.  So what if we narrow it down, some of these guys who we're thinking of in some of these communities.  Let's take blacks age 26 to 35 with less than a high school education; 28 percent of them are currently incarcerated -- not lifetime, that's currently incarcerated, okay.  So that leads to a statistic that Bruce Western generated at the bottom, this quote:  "Among children born since 1990, four percent of whites and 25 percent of blacks will witness their father being sent to prison by their 14th birthday."  So that's the link to families.  And for me this became -- really hit home.  I had this great -- this interesting opportunity at Cornell to give a talk in a maximum security prison on the book, "The Colors of Poverty."  And I said, I'm busy, right, and so someone asked me something interesting and I thought, yeah, I'll do that.  And then on the way there I thought, why did I agree to do this?  Why did I agree to spend my time talking to these guys?  Why didn't I spend my time like mentoring at an elementary school or something?  What am I doing?  Why am I going to hang out with these guys?  And it was an incredible experience, especially because the book, if you've looked at it, it starts with a story about a hypothetical character, and I talked about that character.  And the first person in the audience raises -- this is a maximum security prison -- he said, "That's me."  And he talked about why it was him and what had happened in his life that led to him being there, and it was almost like -- it was incredible.  They all started to talk about this sort of stuff.  But I realized on my way there and especially on my way back, the reason I was there was not for these guys.  It was for their kids.  Because I started to realize, all these guys, almost all these guys, are connected to children.  They're affecting these kids' sense of what they ought to be doing in many, many ways.  So you can see this is the why they matter.  They marr -- they matter because of income but they also matter because of other stuff, social capital culture; they matter because of expectations and aspirations; they mod -- they matter because they model what success is for their kids and their communities.  They're the phrase "old heads" that you've heard perhaps, and thought neighborhood stability.  They can be the old heads who create stability, or they can be the old heads who get in the way of stability.  So these men are absolutely critical.  And as I mentioned already, Al, there's lots of folks who have done fantastic work, urban ethnography.  One of the ones I like to assign I think it's really wonderful it's "The Story of John Turner" by Elijah Anderson.  If you haven't read it, I strongly encourage you to.  It's a really great way and a relatively short piece to get a sense of the importance of young men in inner city communities and the difficulties of addressing their challenges.  And also, I think you get from that, but I certainly get from other things, quite pessimistic or concerned about our ability to address the problems of poverty by just focusing on kids or just focusing on kids and women.  We've got to think about these guys and they're big challenges.  Here's something I did once in a -- after a meeting and it's on my wall -- not on the wall.  It's on a piece of paper that's taped on my wall.  I did not, you know, mess up government property.  And so this is actually one of the things I think about is, you know, I was in some meeting and I saw some presentation and it seemed very compelling about young guys, and I started to think about the following:  If you look at the statistics, we talk about reentry programs a lot, incarceration, right.  You can almost talk about pre-entry programs.  Why would I saw that?  I'm not saying all these people are destined, but if you look at the data, more than half these kids in a lot of neighborhoods, the boys are going to jail.  We know they're going to jail if you look at the numbers.  We don't know which ones, but we know something's going to happen.  So what's happening here is, we've got boys born in extreme poverty, right, and there's a path.  And there are programs.  My worry is that a lot of the programs out there, they're not off ramps, they're rest stops.  So I worry that you go along, maybe you were in and out of the system or maybe someone provides some help before you got in the system, and it slowed you down and you went to a rest stop, right.  But after not too long, you're back on this path.  Maybe you had another program.  This is what I felt like when I was hearing these guys talk about different programs they'd been in, but, you know, you're headed down here perhaps; not finishing high school, not going on to college, not developing the skills you need to support your family and so forth.  You got to figure out an off ramp.  We need good outcomes.  We need an off ramp.  And one of the things I've spent a lot of time thinking about HHS, working with folks in my area and working with others, is trying to figure out what are the off ramps.  Do we know -- and that's evidence based policy comes back to -- do we know what the off ramps are, right.  If we had all the money and all the political capital and so forth to do it, what are the off ramps.  'Cause, again, I don't think -- unless we can figure out the off ramps for these men, I think -- I'm not too positive about our ability to make huge progress with respect to urban poverty.  So what are we doing about that?  One is awareness.  So one is just talking a lot about incarceration, trying to make it clear to everyone that you know, in more affluent communities, we don't have incarceration.  I hard -- I don't think I know -- I do not know anyone who's currently incarcerated, all right.  And a lot of people with my socioeconomic status would be like that.  You don't know anybody's who's incarcerated.  And as a result, I think we think about it too little when we do our research, and we think about it too little when we do our policy.  It's the opposite of the Chicago sort of Urban Poverty Affect on people like Bill Wilson and the work that was going on in the '80s.  It was all in big city and so forth and we thought about urban poverty a lot.  We don't think about this enough.  And so one is just awareness.  A second, though, is to think about existing programs.  So it's actually the case right now in -- I think it's about nine states that -- and I don't remember -- I don't know if Michigan's one of them.  New York is not.  I don't think Michigan is but I'm not sure -- that if you are incarcerated, that's treated as voluntary unemployment with respect to child support, okay.  Incarceration equals voluntary unemployment.  Which means that arrearages pile up while you're incarcerated.  So you get out of jail, now you have a prison record which makes it harder to get a job, and we're going to garnish your wages because you're way behind on child support because you weren't paying it while you were in prison and it racked up because we thought, that's voluntary unemployment.  That's nothing different than saying, I just don't want to work.  That's one of the things we're thinking about.  Does that make sense and what impact is that having on individuals, families, and communities.  There's a bunch of other programs we're thinking about from the lens of incarceration.  And the other is new programs.  Okay, if recognize this in the federal government, what should we have out there to help folks who either are incarcerated or seem like they're on the wrong path, they're on the way there?  Lastly, public perceptions, okay.  This is the best I can do.  When I was at Cornell I used to always joke with the dean of the vet school that it just wasn't fair, because we'd give talks to the board of trustees and he'd put up pictures of like kittens and so forth, and, aww, and then I'd come up with data.  Well, let me show you this bar graph about poverty.  So the best I can do is kids, all right.  Kids are better than puppies, right?  I [inaudible] anybody know who these kids are?  Anybody recognize these kids?  Yeah.
>> Daisy, Francisco, and Anthony.  
>> You've got it.  It's Daisy, Francisco, and Anthony from the movie "Waiting for Superman.  "Anybody seen "Waiting for Superman" other than one person's obviously seen it.  I'm not surprised because if you saw the movie, I don't think you'd forget these kids.  So I saw this movie this weekend, "Waiting for Superman," -- and that tells you I did not finish the talk on Friday by the way -- but I told this on -- I saw the movie on Saturday, and it was really powerful.  I don't know if you agree.  I strongly recommend it -- government -- a government official can't endorse a movie I suppose, but this was a good movie.  It was really compelling.  And what was interesting about it, it's basically without a spoiler -- any spoilers here -- basically what is the story is the five kids who are in situations, they're in schools, public schools, that are not meeting their needs. The parents do not believe they're meeting their needs.  And so the parents want to enroll them in charter schools which they believe to be high performing charter schools, they want to give their kid a chance.  And the movie does a very good job of making you believe that at least four of these kids, I'd say, if not all of them, they're in really bad shape if they don't get in this charter school.  I mean, they're in schools where the vast majority of kids don't graduate and -- don't graduate high school when they get -- they're in that path, and those who do graduate, for example, in California, the majority of the kids who graduate aren't qualified to attend the UFC system, they just haven't completed the classes you need to do it.  So you can see parents working really hard, and it's incredibly powerful.  And I think it -- and it ends, which was heartbreaking, the lottery.  These schools actually have lotteries.  And everyone goes and you sit there and they drop balls, you know, with numbers on them or something, somehow some random process, and the numbers tick down.  They show the pictures of the mom -- usually -- I think they're all -- well, one's a dad -- the parents and a kid sitting there, and they say a name, and, you know, X -- ten spots remaining, nine spots remaining, eight spots remaining.  Some kids get in, some don't.  I won't tell you which ones do which ones don't.  But it's really heartbreaking.  There's this moment at the end, it's just heartbreaking 'cause you know what these kids are in for.  And what I found -- one of the things I thought was so striking about this was, I thought about, again, as a privileged person, and I think we're all -- in my role, I think we're all privileged people that we can spend our time doing things like this.  -I almost never come into contact with these kids.  I don't know these kids.  They're not part of my world.  They generally don't go to my kids' school.  If they go to my kids' school, I might know one or so, but I don't know them very well.  And I don't think I'm unusual in that sense.  We've got an awful lot of segregation in this country by social class.  So we've got residential segregation by social class.  This is a paper at NBR, a working paper, and it looked, for example, used census data from 1970, '80, '90, and 2000, and said two-thirds of metro areas experienced increase in segregation of the rich from the poor between '70 and 2000.  But those people aren't randomly distributed across metro areas; two-thirds of metro areas, but 85 percent of the metro area population lives in a place that saw increase in income segregation between 1970 and 2000, okay.  And there's a lot of other work to support this assertion that we have -- that we have had a growth in residential segregation by income.  But we also have social segregation by class.  So there's some really interesting work using 2006 GSS data by Tom Dupree and his colleagues, and they looked at networks.  So they looked at two kinds of networks.  They looked at acquaintance networks.  People say, yeah, I know so and so [inaudible].  But they also looked at trust networks; who are your friends?  Who can you go to when you really need something?  And what they ended up concluding -- I have a quote here from the paper -- what they end up concluding is that levels of social segregation, network segregation by race are quite high, but they're also quite high by religious behavior, by employment status, by political ideology, and they had other indicators in there in well -- as well, these social class type indicators.  So not only do we live in places that are segregated, but our social networks are segregated.  We also have infor -- what I call informational segregation.  So I think about this -- my family moved down to DC about a week and a half ago.  Before that I was commuting back and forth and, you know, as a good modern person, I read the paper but just on my iPhone, right.  And like any of you, I don't know if you've had this experience, you know, I've had the, you know, iPhone, you're already in bed, the iPhone hits you in the head, you're reading, right.  That's how I read the paper at night, okay.  Some of you know what that -- that's like.  But then my family moved here and we started getting the "Washington Post," the print version.  And it reminded me that there's something very different about reading the print version versus the iPhone version.  And that is, you end up reading stories you're not looking for.  And you read stories you're not looking for because, oh, there's an interesting headline, there's an interesting picture.  You can sort of get that when you look on the iPhone or online, but nowhere near the magnitude in my experience of when I'm reading the physical paper, okay.  So what it means is, with the decline in circulation -- I think the "Chicago Tribune," you could have picked any paper, a lot of these went it zero because they went away in this period -- over the 16 years you saw the average daily circulation for the "Chicago Tribune" go from 601,000 down to 452,000, while the population's going in the other direction, okay.  But there's another way to think about this informational segregation, and I'll close with this chart, which I thought was really interesting from Pew Research Center.  It's the kind of thing you thought was probably true but I hadn't seen the results before I came across this.  Basically, they had a survey and they asked people a bunch of questions about themselves.  You know, what's your political affiliation; you know, are you conservative, moderate liberal; whole bunch of things; they had sexual orientation; a whole bunch of stuff.  Then they said, where do you get your news?  All right.  Do you watch or listen to various news sources.  And so all you have here is a very simple table.  It simply tells you by political party and then by political ideology what are people -- where are they getting the news.  And you can see Rush Limbaugh, right, has 63 percent of the people who watch or listen to Rush Limbaugh say they're Republican, but only 23 percent are -- 10 percent say they're Democrats, 23 Independent; and here's the distribution across the whole sample:  25, 33, 34, Republican, Democrat, Independent respectively.  On the other extreme, you've got Keith Olbermann, right.  Countdown on MSNBC.  You've got 3 percent Republican, 60 percent Democrat, and 29 percent Independent, okay.  It's pretty interesting.  Think about what you watch and so forth and see where you are on here.  And see, eh, yeah.  I guess I'm right with a bunch of people like me.  And then you can see it also over here, not by Republican, Democrat, Independent, but by conservative, moderate, liberal, okay.  You see the same basic story.  And what I think's interesting about this is -- and, of course, nowhere near the first person to comment on this -- but this is different from the old days of, depending on your age, your parents, or your grandparents, when you had, you know -- it wasn't that -- this many choices.  It was, am I going to watch Cronkite or something else or something else on the other networks.  Brinkley I guess [inaudible] one, I don't even remember, but there were a few networks, and that's where you got your news.  And there was the local paper where you got your news.  That meant that you and your neighbors had more of a shared sense of what was going on in the world.  What this is telling you is, because I've watched a lot of these and I'm sure you have as well, the story's really quite different across these different forms.  What that tells you is, that individuals are not just living in places with people who are more like them and hanging out with people who are more like them, they're getting information that's consistent with the information that people like them get and not like the information that people unlike them get.  So what?  [Inaudible] and why am I talking about this is the DAS and Human Services Policy.  Because my concern goes back to Waiting for Superman.  It goes back to the fact that I think a lot of the differences on policy are because we have different information, because we have different experiences.  It's true that once you get the information you can come to different conclusions, but I really think a lot of it is because you have different experiences.  And I'll give you an anecdote that illustrates it.  So from my time here at Michigan I became a faculty member of the Alumni Association.  The reason I became a member of the Alumni Association is because I ended up going to Camp Michigania to give a talk.  Anybody here know Camp Michigania?  Yeah.  We love Camp Michigania.  We've been eight -- I think ten years in a row now.  I was a speaker twice and now we just go as campers.  And I loved being the speaker at Camp Michigania.  My wife wasn't so keen on it, and the reason was, you give a talk -- and I would give talks about things you can imagine, inequality type things -- and the entire week people would be talking to me; at the archery range, in the cafeteria, everywhere, about, "Oh, I another thought about this."  Well, what I noticed is, these are very smart people.  They had very strong opinions on day 1 when they heard my talk.  After a week of like, "Well, you know, I was this thinking a little bit more, and I have another question about this."  Or "Have you thought about that?"  I could see them start to change.  I could see them start to say, "Hmm, well, maybe it's not all bad."  It's not like they were all of a sudden saying, "Boy, I was some lefty and I went to some right wing person."  But they could start to say, "Hmm, the world's a little more complicated than I realized it was."  Well, we don't have that opportunity very much anymore, is my argument, given the levels of segregation and information.  And what it means is, our role in human service is twofold:  Our role is to advise the secretary to generate the best policy options we can; but we've also got to figure out how we communicate that to the public.  Because if we don't find ways to communicate this to the public, it doesn't matter what great policy ideas we come up with because people are less likely to support them is my contention because they don't have the experiential knowledge that you get from a rare case of watching something like "Waiting for Superman," because we live in segregated worlds.  So that's what I had to talk to you about.  Those are three things -- they're back to three No. 1s -- three things that occupy my time at HHS Human Services, will continue to occupy my time, and I'm happy to take questions or talk about any of these or anything else that you might want to talk about.  Thanks.  
[ Applause ]