 So, I'm in the interesting position of actually introducing myself. I was sending an email just a few minutes ago saying that the New America Foundation didn't have anyone on hand to introduce the panel, and so they asked if I would be willing to do it. And I said, of course, because I will have a nice extensive discussion of all the many virtues of the next panelist sitting over here. Jacob Hacker. Yeah, how are you, Jacob? But I'm really pleased to be here, and the New America Foundation, I guess, can be forgiven for not having someone to introduce me because they've done so much for me over the years. We had the launch event for the Economic Security Index here roughly a year ago back in the summer of 2010, and then we also had a follow-up event looking at public opinion during the Great Recession, which was also a great event. And so today we're going to be doing a second installment of the Economic Security Index project. I want to just mention a couple sponsors for this event. In addition to the Rockefeller Foundation that has funded this work and the New America Foundation that has helped us launch it here, I'm very grateful to the fact that the Foundation created an independent advisory council to advise our work. One of whose members, Larry Michele, is with us today. The other members of the advisory group are Henry Aaron of the Brookings Institution, Gary Burtless of the Brookings Institution, Hank Farber of the, no, Princeton University, Elisha Munnell of Boston College, and Robert Solow, who needs no affiliation since he was the Nobel Prize winner in economics in 1987. I want to thank EJ, who actually also is here for the initial event. And EJ has been a friend and a provider of wonderful blurbs for things I've written for a long time, since I was a research fellow at the Brookings Institution alongside him. I had an alcove, well actually it was called an office, but it had no door that was right across from EJ and we, I'll just say that I did a lot less work on my dissertation and drank a lot more red wine than I would have otherwise, but it was definitely worth it in terms of my long-term intellectual development. I also want to thank my team, some of whom are listed here. In addition, and we have Austin is here today, right in the front row, and Stuart Craig back there, and in addition Greg Huber, Mark Schlesinger, and Philip Reem were partners in this work and weren't able to be here today. I'll stop with the thanks because I want to try to get through the results pretty quickly so that we can have a good discussion. We have a really fine panel. I won't actually introduce a panel. I'll let EJ do that, but I'm really pleased to have all these folks here today. And whenever I'm under time constraints, I think of an evaluation I received when I was a new assistant professor at Yale. It began very promisingly. It said Professor Hacker, if I had just 15 minutes to live, I'd want to spend it in your class because that way it would seem like an hour. So I forgive my, by the way, I have to forgive, Austin has to, I have to apologize to Austin for having repeated that joke and to EJ as well, but he still laughs. So we know that even before the economic downturn that many Americans were concerned about their economic security, this was a poll that the Rockefeller Foundation did in 2007. It was part of a worker security initiative. They began and which funded the Economic Security Index project. And it also was important because when we were doing our own polling work two years later, we built on some of the questions in this survey. So we were able to look at changes in people's perceptions of economic security between 2007 and 2009. And I'm going to refer to some of those polling results and what follows. But the measure of economic security I'm going to discuss is actually based not on public opinion polls but on economic data that looks at changes in America's economic circumstances from year to year. So the first question and the obvious one is what are we talking about when we talk about economic security? And basically the definition that we came up with as we were developing the index was very straightforward. That there were two basic components to economic security. One was exposure to risk and the other was the degree of protection that one had against economic loss. So in other words, economic security is basically whether people are vulnerable to major risks to their economic well-being. Now there's clearly as I just noted a psychological component to economic security which is why we did our public opinion survey. But in designing the economic security index following on the basic approach to looking at say poverty or unemployment or median incomes, we wanted to have a measure that was based on economic data that showed how people's economic circumstances changed over time. Now that dynamic aspect of economic security is the hardest to capture for two reasons. One is that it's very hard as I'll note in a moment to find good data that looks at how people's economic circumstances change from year to year or over longer periods. The second reason why I think this definition raises questions or issues is that clearly economic security is more or different than income volatility. It really concerns risks of downward economic loss. And that's really the extent that we're incorporating a core psychological element into our definition of economic security. It's that we're putting primacy and understanding economic security on people's aversion to loss as behavioral economists have called this phenomenon. It is people are highly loss averse. They fear losing what they have. And even Americans feel this way. Back in 2005, during one of the battleground surveys that was done in the run up to the 2006 midterm election asked people whether they'd prefer to have stable secure incomes or the opportunity to make more money. These are Americans after all and this is in the midst of the 2000s expansion. You would have thought I think that most Americans would say the opportunity to make more money. Surprisingly by a two to one margin they said they preferred income security. And I think this is just a measure of how strong this loss aversion preference is. So in thinking about economic security, we put heavy emphasis on the downward risk of economic loss. That still raises a lot of questions about how one would actually measure economic loss. Are we looking at wealth or income or some other measure of economic well-being? We decided after much consideration and looking at what the available sources to focus on a special measure of income. And this actually is not so far from the measure of income that is now being used by the Census Bureau in developing its poverty line. We call it available household income. It's basically income from all sources including government benefits less out of pocket medical costs and the cost of servicing one's financial debts. When one has negative financial net worth. So we say that if this measure of available income again household income from all sources minus medical spending minus debt service that when this changes drops from one year to the next by 25% or more that this household is counted as this individual is counted as insecure. We adjust of course for inflation. We also adjust for household size. And finally we make a special exception if you will for two groups. You are not counted as insecure under this definition even if you experience a 25% loss if you are either entering retirement which turns out to be a trivial exclusion. And second of all or if second you have enough financial wealth to be able to buffer to be able to basically pay the difference between your former income and your current income until you recover back to your prior income level. So in other words if you have enough precautionary savings that you that you're able to basically draw down your savings so that you don't experience a loss in your available resources then you're not counted as insecure. And I'll get to how we actually measure those things in a moment. So that means that our measure was really based on three key elements of economic security. First the year-to-year income fluctuations and in particular year-to-year income drops. Your year-to-year changes in out-of-pocket medical spending and then whether or not you have an adequate financial safety net. And just to give you a sense of what this means I'll say two quick things because they'll become salient when we start to look at the actual results. First if you let's say have $50,000 in available household income it would mean your income would have to drop to $37,500 or lower for you to be counted as insecure if you didn't have unless you had enough savings to pay for that difference until a typical person like you recovered. But one important thing to note is this is every year so let's say it's from year to year so it's an annual measure so it's from say 2009 to 2010. Then you would have to experience if you were going to be counted as insecure the next year you would also have to experience another drop in income of 25% or greater from your new base income to be counted as insecure. So you might think that's a very stringent test and indeed it is but if you think about it what this is really is in some sense is a measure of income risks that families are facing right. The share of the population or of a defined group say the age that are experienced this loss is basically a measure of the chance that someone like this will experience a drop of this and that in that sense when it goes up it indicates that more Americans are insecure and I'll come back to this in the conclusion. Why do we choose these these three elements? Well I've already noted that this is very close to how the Census Bureau is now conceptualizing poverty that is to say that medical care is a constraint on available income and debt service is a constraint on available income and at the same time I've mentioned that people who have large financial wealth holdings are able to basically self-insure against large income losses. But let me just say a word about why income is the metric that we use. It's not only that it's one that's well measured it also turns out that large income losses are among the most salient risks to individuals. So we did the survey in 2009 and in March we asked people to basically rank from one to seven one being a highly anticipatable easy to deal with risk and seven being a risk it's very hard to anticipate and prepare for and noticeably the ones that are very high on the scale are not retiring poor which people seem to think you can plan against or housing affordability but losing health insurance a job being disabled or having big medical costs and those are all core components of the economic security index as we developed it. We also as I said have the standard 25% and while this also it has a nice intuitive feel we also were pleased to find when we did our survey that it fit very closely with people's own self-assessed capacity to deal with large drops in income we asked people how long they could go before without their current income before hardship sets in unfortunately we designed the survey before we knew that we were interested in doing a 25% standard but one thing what we can say is if you look at this that three months is probably about the median 64% said four months or less and that means that for a typical household or individual a 25 threshold is probably a pretty pretty close to the amount of income loss that is three months of income that would induce hardship now I just want to emphasize one thing before moving on is that our measure of whether you have adequate financial wealth is based on financial wealth not housing wealth and our main reason for not including housing wealth or debt in our measure is that housing is a pretty difficult asset to liquidate when you face major income losses certainly in recent years it's been a difficult asset in general but our idea is that when someone loses a job or is disabled their housing is not likely to be a very big source of insecurity of security they need to have a place to live after all when the economy is down or when they're out of work it may be difficult for them to use their home as a source of cash and in any case we did some analyses to look at how sensitive our measure is to our treatment of housing and we found that it basically lowers the index a bit prior to recent years but of course in recent years with the housing market down it's much less of a buffer against economic insecurity so how do we calculate economic security well in the last report that I presented last year we used the survey of income and program participation the survey of income program participation is a large survey panel survey done by the census bureau a panel survey is a survey that follows individuals over time most surveys are not panel surveys most surveys are one shot random samples of the population this is a panel survey basically interviews someone one year or one month and then interviews them the next year or the next month over time for some period of time the longest panel survey the panel study of income dynamics has done this for about 40 years the survey of income and program participation uses these many panels from the early 1980s on where it surveys people for a period of three to five years two to four years Austin's nodding two to four years and in our initial analysis we use the survey of income and program participation the sip but we decided against it for this report and actually moved to another approach that better captures recent the developments in recent years so I want to explain that report and its motives so basically now rather than using the survey of income and program participation we went to the sort of mother load for income and and demographic data namely the census bureau's march current population survey this is a survey that is used for all of the major reports about health insurance and poverty that you read every year it is not technically a panel survey it is basically a survey of geographic residences but it turns out that you can actually that reaches different people every year but it turns out that because of the structure of the survey of the survey you can actually use it as a panel survey for two year for one year to the next by looking at people who have not moved from one year to the next in March of each year so we're basically linking data from March to March and the the advantages of this over the sip are several fold first of all the sip as I mentioned is not a continuous panel survey it's a series of short-term panels and there was no panel survey in the field during the onset of the Great Recession so basically we have no data on income loss from the sip from roughly 2007 to 2000 through 2009 second of all the March CPS is not only this gold standard that's used for other analyses but also a very large survey that with proper weighting can be used for state analyses and so being able to do state by state comparison was a big advantage third and this is getting kind of technical but what we discovered with the survey of income and program participation is that it has a very high attrition rate and a relatively high attrition rate and as a result we were concerned that toward the end of each of these panels three or four-year panels you were actually getting biased results because a lot of people were dropping out of the survey that's not a problem with the March CPS basically we have in each case a two-year panel from one year to the next and the attrition rate and the like does not change much over time so it's not that's not an issue the main drawback of this approach is that the March CPS only allows us to look at geographic units rather than individuals which means that we cannot look at people who have moved so Austin actually has done some analyses using the survey of income and program participation to see how much of a bias this might introduce and it is the case that people who move experience higher income instability than those who do not but they are relatively small portion of the overall sample and in fact migration has been declining in the United States it's contrary to many people's belief over this period from the 1980s until last few years it falls over this period and we our own estimate suggests that this has only a very small bias and remember that since people who move have higher volatility it's a bias that basically reduces the level of insecurity that we find okay so we have I'll show you the results in a moment but that's those are the motives for the shift and we're very confident that our our new approach which we'll be using going forward is is both in advance over the previous approach and will provide a much more detailed data not just for the most recent period as I'll show you but also we can actually go back before the mid 1980s using this data to look at income trends from basically the late 1960s on and that's one of the next steps we're going to be taking okay we at the same time have a constraint that we had before there's not good medical expenditure data in either the CPS or in our previous survey the consumer sorry the survey of income and program participation so we basically developed an imputation method that is very advanced and quite good at predicting out-of-pocket medical spending and not just in one year but from one year to the next and I'm happy to talk about that for the aficionados more and then finally we bring in medical we bring in wealth data from the survey of income and program participation and that our our our benchmarking has shown that that again is a very reliable way of predicting wealth levels remember we're not focusing on wealth changes this is just the wealth level so basically whether you have debt or whether you have enough financial wealth to to be counted as secure so our so we're less concerned so the so the imputation procedure does not have to predict changes in wealth from one year to the next which is a good thing because it turns out that there really aren't very good sources on wealth dynamics finally I mentioned that the the amount of wealth that you need is based on how long it takes for you to recover you might be thinking well how would you know that given that the CPS is just a two-year panel and for most people they don't just immediately bounce back obviously if you drop to 25 percent in the second year we don't know what happens to you after that so what we did is we used a panel study of income dynamics which as I mentioned is available for 40 years to estimate the long-term recovery of income just to illustrate those results because I think this is actually one of the cooler things that we're able to do you can get a sense of of what the dynamics of income loss are in this figure basically we're looking at three different sizes of drops so in the first year of course in the in the year before the drop you have 100 percent of your pre-dropped income and then you drop down so this is a small drop 25 to 35 percent small by our standards here you see a 35 to 50 drop and then here's the really big drop half income or more and you can see that recovery is quite slow though you recover in roughly the same period of time to be clear these are just illustrative these are just the observed patterns in the panel study of income dynamics there are a bunch of statistical reasons why we don't use these exact numbers we use a basically a model a hazard model that allows us to predict how long it takes for people to recover and then from that calculate how much of their income they would need just to be clear you'll see in a moment that the debt the exclusion of people based on wealth has a fairly small effect in general and it and therefore you know changing these these estimates modestly would have very little effect okay lots of buildup let's actually see what's happened to economic security now I don't want to keep you in suspense I'm sure that given given the previous report given what we know about the economy that you probably know where this is going here are the the downturns you can see that the most recent downturn was was quite prolonged relative to both the 1991 and 2001 recessions so you can see here both that it's a it's a very cyclical trend a very cyclical pattern with a downturns producing an up up tick in insecurity and that quite there's a quite striking rise in the most recent downturn I think there are two things to note here first the first is just how sustained economic insecurity has been most during this recent downturn for each of the last three years every year 20 percent more than 20 percent of Americans have experienced a 25 percent or greater drop in available household income without an adequate financial safety net the other thing to note is that this is a long-term trend as well and this is true even if you take out the most recent years it it looks as if with each with each downturn the recovery is back to a slightly higher normal plateau than the previous one I'll return to this theme and thinking about what where we might be going it should also be said that we've done some analyses with the panel study of income dynamics and we'll be doing analyses with this this current population survey to look at previous years and there's no nothing that we've done or other scholars have done suggests that the mid 1980s was somehow a low point in economic security a high point in economic insecurity in fact most of these measures of income volatility show a rise over the the 1970s and early 1980s as well so this was not a an artificially low point so based on our measure we can I can show you what I mentioned earlier that the different the contribution of the different elements of the index so the most important factor given how we constructed the index is these year to year declines of 25 percent of income or more and you can see that that basically follows the pattern of the rest of the index if you add in medical costs you can see a slight increase in the overall trend now that slight increase as I'll show you in a moment is is a lot smaller than the increase in out-of-pocket medical costs over this period at the median out-of-pocket medical costs increased about 36 percent over this period which is a very large increase but the point is that to count is insecure in our measure that your out-of-pocket medical costs have to push you over this 25 percent loss threshold they have to change from one year to the next so a lot of that increase and in out-of-pocket medical cost is probably represented by people have stably high spending rather than fluctuations in spending now add in debt you can see that debt really doesn't do much for the index until you get to the most recent years and I'll show you why and although you already know the answer and then finally if we take into account wealth you can see that it generally brings the index back down to somewhere between the income and medical cost measure the yellow line and the income only but it gets a little bit closer over time to the income and medical cost measure which is suggestive that the wealth buffers in particular are less of important and offsetting income loss in the most recent years so the financial wealth holdings the buffers are crucial for looking at this yellow line here this sorry this this white line here and you can see now on this figure why it is that the last few years have really shown a separation of the income and medical costs and the income metal costs and debt lines at the median household financial wealth that is uh non-housing wealth and liabilities added up your net financial wealth has been zero so there's been no and there's only a slight as you can see in the mid 90s there's a slight move above zero but it's very small at the 25th percentile you can see that it's been a that debt has increased and at the 10th percentile of debt hold of wealth holdings it's plummeted that's about a ten thousand dollar household increase in debt between 2000 net debt between 2004 and 2010 on medical care just to substantiate the story i told you there has been a pretty big increase for the non-elderly um and i'll show you'll see in a moment why i have to have these axes so high 80 for the age you can see they spend massive massively more there are two things to note here one is that the median and among higher income citizens especially in the last few years there has been a fairly dramatic increase this may not look that large but it's basically a doubling of the share of americans uh 18 to 64 who spend more than 10 percent of their income uh on medical costs out of pocket but um but at the high at the lowest income levels among both the age to um have been have been assisted by the medicare prescription drug benefit and expansion of coverage for dual-eligible and among the non-elderly who've been assisted by the expansion of medicaid and s-chip um medical out-of-pocket expenditures as a share of income have actually stabilized i want to make one very important point here there are two ways to interpret this one is that we've done a great job protecting low-income americans against rising medical costs the other is that we might not be doing that great of a job because we don't know whether or not this uh stable spending as a share of income is a reflection of better policies protecting these citizens it surely partly is or whether or not it reflects and to what degree it reflects the fact that people are putting off medical care uh or um or or um going or receiving care that's not paid for in the form of charity care and that's one of the problems with our measure is that medical out-of-pocket expenditures which is i think the best possible measure but not the only possible measure uh doesn't allow you to distinguish between spending uh for those uh low spending for those reasons low spending could either reflect lack of care or low spending could reflect um good insurance okay so the last slide i want to show you before concluding is um an illustration of the demographic differences we find we go through um a whole host of them race education um age um and education uh race education age household type i want to look show you age because i think it first it helps us see the role of medical costs and second of all um it's it's telling i think given um the brunt who's born the brunt of this recession so first i'll put up the results um sadly i have to put them up one by one uh for with the esi without medical expenditures so this is each age group here and then by period 1985 to 95 97 to 2007 and then the the great recession period you can see a pretty clear pattern going up across each of these groups note though that for older americans these two periods do not look particularly distinct for younger americans they look much more distinct that is the long-term trend towards increased insecurity seems much more dramatic for younger americans when you add in medical costs what you see is that for older americans they look much less secure and this is what we found with the census bureau data as well with poverty that medical costs are a much bigger bite uh older americans incomes um and so in this respect the other thing i would note is that despite a lot of talk about how the only the young who's born the brunt of this downturn um it actually looks as if there's been a very big increase um in income instability for older americans uh as well um and uh this may reflect in part the fact that older americans are much more likely to be in the labor force than they were a generation ago okay so um but i'd say we're talking about almost 23 around 23 percent of 18 to 34 year olds experiencing a quarter or more drop in income from one year to the next um in the most recent downturn every year it's a pretty dramatic number so this is not the end of the fund reports that we'll be providing over the next few years uh and we're also going to be providing uh based on this new cps approach state by state results soon uh we we have um commissioned we're in the process of commissioning a new survey um and as you've seen we the first survey has produced a lot of interesting findings about people's views of economic security um one thing that i will mention um that i think is quite crucial as we think about where we're going is that um is two things actually first of all we we we just we're at the oecd organization for economic cooperation development austin um and philip where they are with me and we presented some cross-national results um i won't be giving away too much and saying that the united states looks less secure at least in the run-up to this recent downturn than do other countries um the um it will be interesting to look at those findings going forward as the europe um crisis um has effects on on european economies and but at least on our first glance the higher levels of inequality in united states are matched by higher levels of of insecurity and income instability as well though the one thing i wanted to mention was this role of government policy remember this is taking place against the backdrop of a fairly substantial expansion of unemployment insurance and very aggressive efforts to try to um help the economy starting in 2009 um and in thinking about going forward i think there are two contrasting trends and for those who want to um look at the results more fully you should go to this website that on the one hand um our measure is a measure of income loss of income instability so if if families or house uh or individuals lose a job and they're basically stuck at a low income they're not treated as unstable right and insecure um so it's really very sensitive to increases say in the unemployment rate or increases in the fluctuation of health care costs not so much by the the the baseline level of unemployment or health care spending that said it does look like it captures the recession pretty well however those levels um which are above 20 in recent years are against the backdrop of this major expansion of of of government aid and it's not clear to me which of the two effects that we should see going forward will dominate on the one hand um if we continue to see very slow growth but the unemployment rate kind of creeps back down and we don't have a lot of new people entering the category of the job list we could should expect this rate to go down on the other hand the expiring of a lot of these uh measures um will will could very well lead to um an increase at least among some populations of insecurity think of the two million people who are facing the cutoff of unemployment benefits if unemployment insurance is not extended um I think what we can say safely now since forecasting the future is is always difficult uh since it hasn't happened yet what we can safely say now is that America's economic security is under challenge as as more than at any point in the last quarter century so since uh Jacob couldn't introduce himself fully I just want to say that he's absolutely right he spent nine months outside my office right outside my office and I have been learning from Jacob ever since in fact I saw Kurt Gans is in the audience so he actually will remember this from G McCarthy's campaign there were little buttons that said for McCarthy before New Hampshire as he after he did well in the Hampshire primary I was four hacker before he was famous and because he spent those nine months outside my office I've claimed credit for all the brilliant things he has done ever since personal credit the other point I would like to make is that the social justice needs warmhearted friends but it also needs hardheaded of friends and while I can testify that Jacob is warmhearted as you could see from those wonderful charts I love the way things are moved around moved around although my conclusion is uh we will entirely get rid of economic insecurity when people lose all their income by the time this is over and then there's no place to drop from but it needs hardheaded friends and I think what Jacob and his colleagues have done here is exceptionally useful because unemployment is an important measure but it's an imperfect measure even uh income growth is an imperfect measure partly because it depends on how broadly shared the income growth is and I think economic insecurity is a genuinely helpful measure of the problems we face and if Jacob and his colleagues keep working hard eventually of the progress we've made so keep at us Jacob this is an exceptional panel and I will introduce them in the order in which they have uh agreed to speak the Larry's biography that I have here says Larry is a nationally recognized economist that's true but not recognized enough for my taste he deserves even more than he already has he's president of the economic policy institute he's been president there since 2002 he first joined ebi as a research director he's done a lot of different kinds of work but one of the one of the many great things he's done is he's co-authored the 11 editions of the state of working america which former labor secretary Bob Reich said and I agree with him uh remains unrivaled as the most trusted source for a comprehensive understanding of how working americans and their families are faring uh in uh the economy um he is um he was uh Larry had a fellowship at the U.S. Department of Labor oh no no I want you to be nationally recognized even more he was a badly member at Cornell and I like this fact in his bio his PhD in economics is from the University of Wisconsin which is for many generations has provided our nation uh with learned progressives so thank you to the University of Wisconsin um John Cohn is senior editor uh of at the new republic uh his uh blog is essential reading his writing on healthcare uh is essential I always told people if they wanted to understand everything that was happening on the healthcare bill just read Jonathan Cohn and Ezra Klein uh and uh I um I was extremely well informed as a result of that I can also he is um um he is a senior fellow at Demos he's a contributing editor at the american prospect um and you must have been about 18 years old the first time you edited me and I can say um when I wrote a piece for the prospect and I can say that it's not surprising that Jonathan is such a good writer uh because he is a superb editor he's received lots of awards for um his writing and in addition to doing the best coverage of the healthcare law he also wrote one of the best books about uh the healthcare system and matiglasias is slate magazine's business and economics um correspondent he is also essential reading before joining the magazine um he worked for think progress the atlantic tpm media and uh the american prospect he is the author of heads in the sand um published in 2008 and his next book um comes out and I can't wait to read it in march of 2012 its title is the rent is too damn high um so we all look forward to that um I'm going to each of our respondents will um speak for about uh five minutes I'll go sit down uh during that I will ask a few questions uh but I want to bring the audience in as quickly as possible and uh I will do that uh so first uh larry michelle thank you ej and uh thank you uh jacob for inviting me and being on this panel with uh these great young pundits um so I don't know what should I do first um you know this is great work I look forward to incorporating it in the next state of working America uh they'll be I'm sure it'll be at least one table and one graph uh reflecting these data um it is a little surprising it didn't increase more in the downturn uh one of the things you don't one of my um disappointments about the presentation is that it's hard to get the actual numbers from the uh the book itself but luckily there's this there's a website where I can go get the actual uh data but uh you know as jacob pointed out in the last report and again this there's been a uh a secular increase in economic insecurity over time and if you look at the data for 2006 and 2007 by the graph and how much it went up by 2008 9 and 10 it doesn't really jump up as much as my priors would suggest it would and perhaps that's you know we have such a great welfare state but perhaps it's something else I don't know um the reason why I think it's surprising is that um I frequently emphasize how extensive the impact of unemployment is uh 9 unemployment does not mean 91 percent of the people of the workforce are okay you know nine percent uh is the unemployment rate there's an underemployment rate of around 17 percent if you capture how many people are not only unemployed or underemployed on average each month but at some point during the year which is much higher it's roughly double that so there's roughly a third of the workforce that's going to be unemployed or underemployed at some point during the year if you look in polls respondents tell you that uh in last June 42 percent of people said they were unemployed or a member of their family was unemployed because we're looking at family income here which is shared right so that's 42 percent there's another 38 percent which overlaps which they are someone in their family experienced hours reduction a wage reduction or a major benefit reduction right so there's a lot of people really getting hammered so to speak so in some ways it's um uh surprising to see so little if you look at I looked up the numbers on long-term unemployment compared to 2007 uh in 2011 there's roughly five million more people each month who have been unemployed at least six months and there's roughly four million three point eight to four million each month that have been unemployed more than a year so that's a big increase over that period of time and it's not all one year change like this but um so in some ways we've experienced something really awful and the index um uh you know has gone up but it's not going up um uh remarkably it has not gone up as as much as I might have expected uh and I'm going to just make a few data nerdy points now uh one is in the description of the increase in insecurity um it's not so great to just look at the end points uh you know if you want to estimate a trend you would want to estimate a trend with unemployment rates or something like that not you know not the 1986 versus now when we're in a trough usually economists compare trough to trough peak to peak things like that I mean I'm sure it's going up and it's going up a lot remarkably you know within each business cycle which is useful to say uh again I'd love to have the actual numbers in the tables and or graphs I find the pick the selection of years to be uh odd they happen to be like the first 10 years the second 10 years and what's left over but um you know we economists like to look at you know cyclical peaks or business cycles so the um you know the years you have 1986 to 1995 I go like what's that you know uh you know it's the first 10 years of the the data you have I get it but uh anyway then the last thing I'll say is um there's a lot of discussion of disparity of economic insecurity and I think that's really uh important uh but uh I think it's important not to focus on the levels as much who's got the highest but how they've changed so um for instance um it's less interesting I mean it's interesting that college graduates have less uh insecurity than high school grads but it's equally or more interesting to me that the level of insecurity increased about as much for every educational group table of figure seven especially if you throw out the the two brackets which only have around 10 of the workforce less less than high school and uh more than college those are kind of outliers but everybody else uh and you can actually see I think the sum college people actually increase their insecurity increase more than uh high school and it's also useful to know that um if you look at people with a college degree in the recession their economic insecurity is um you know higher than what high school grads insecurity was before the recession and much higher than what high school grads had in your first 10 years so uh you know this is really an equal opportunity uh mess uh we're in uh and people of all educational groups uh are have experienced um a lot of damage and thank you very much congratulations on this great report the rocket filler institute uh should respond by starting an initiative to protect workers in the downturn and to explore worker economic insecurity that's a joke uh it's a very very inside joke sounds like a funding application no no no no they closed down their program for their initiative on worker economic insecurity earlier this year in the midst of the worst um downturn in the 80 years um uh why also want to start by thanking Jacob for doing this report and actually for his work over the years you know when I was initially working on health care reform when I was a little past 18 um uh I was flailing about for a while trying to figure out how to put all these pieces together and actually Jacob played no small role with his work both statistical and on the history of the American welfare system and and giving me an analytical framework that I could then tell a story about how American health care had changed not just in the last 10 20 years but in the last 100 years so I'm very grateful for that I'm grateful for this report uh which I think will uh I hope will get a lot of attention um Larry is a lot nerdier than I am and and I mean that in the best sense possible as a policy wonk who aspires to nerddom um so I'm not going to try to be so nerdy I'm going to instead uh come at this as someone who uh thinks like a journalist and as a journalist you know I'm always primarily concerned with the WGS factor which is the uh who gives a um something uh and after all I mean it's not as if you you need a a a lovely produced and foundation backed extensively peer reviewed report to tell us that you know there's a lot of people struggling right now um it's hard when you struggle it's not a good thing and we should do something about it um but uh there are reports like this actually have an important piece in our political ecosystem I I think sometimes those of us who live read and breathe these numbers sometimes forget that but the fact is a report like this you know it will get headlines it will generate uh news attention and um as somebody who makes a point every morning of turning on my I live in Ann Arbor, Michigan so I turn on the Detroit local radio station every morning and I always listen to the one line I'm going to get on the am radio about some news development in the business cycle and you know things like this they make you know you'll hear that one line and it'll get repeated and on the nightly news it'll be a segment and this will be this will drive coverage reports like this drive coverage they they play a very important role um and it's important that these things be rigorous and I think you're to be complimented on trying to do something that wasn't you know I've seen a lot of these reports not all of them have such distinguished uh committees looking over them and not all of them go to such lengths to be uh rigorous about the data um and that's not to say it's a perfect uh uh statistical measure you've produced here I think anybody who has spent any time with these measures will will will agree that there is no perfect measure of economic uh security um Larry has raised some interesting points why is this not bigger why are we not seeing larger numbers here in the middle of this vast economic downturn um you mentioned another one during your presentation uh if you're trying to capture the impact that medical expenses has on incomes um and your measure is a delta it's a change from year to year you're not you're going to miss a lot of people with chronic disease and again anybody who knows healthcare will tell you that's where a lot of the big spending is that's where a lot of the big uh hit is so you know this is not a perfect measure but it's a measure and I think we look to these measures we put them all together and it builds a case that uh builds a case for media coverage it builds a case for political action um and you know sometimes we find interesting things we look at these numbers I was uh very struck by figure two let me find it here in the report maybe I wasn't the only one which had the regional differences of uh which parts of the country had uh the larger changes and uh as uh where is it it's very beginning page five thank you page five and we see that the north central northeast uh had uh were slightly more economically secured by this index than the south and the west now being from michigan we don't often get to say that we're you know in the vanguard of economic security so this obviously played to my regional pride um but that's a very interesting phenomenon I would be interesting to know is this a reflection of the economy the nature of the economy is a reflection of uh government programs union density or is it again because you're measuring the change here as opposed to the static picture is this simply the fact that hey we in michigan we've been enjoying a recession for 20 years we're just excited the rest of you all caught up to us um uh on the question of uh the picture you showed before on uh medical expenses by income and by age group you could read that as as you said in one of two ways one story you could tell is that well here we have the parts of the population where we've intervened most aggressively in the last 10 20 years low income groups and we see that they are more protected it appears against medical expenses medical expenses is that because we've done a better job or is that because they're still having such a hard time they're just uh postponing health care or is it both um that's not something you can answer with this report but you know this is an interesting avenue for somebody else to explore and I hope I hope we will um but I think the third point and this was the one that really struck me and it was the first graph you showed and it's uh it appears twice in your report is figure one and then again I think figure six figure three sorry which is the long-term upward trend in uh economic uh insecurity and if you look at the numbers and if I'm reading this graph right what's really remarkable here is that the figure for 2006 the percentage is actually higher or roughly the same height as it is at the peak of the previous recession so this is we were already had economic insecurity equal to the previous recession before the new recession even began that's a pretty remarkable uh statistic now again I'm I'm I'm not nerdy enough to sit here and tell you whether that statistic is absolutely correct whether it's missing something but you know this story of rising economic insecurity as a long-term trend and not a short-term trend that certainly rings true to me it's consistent with everything else I've seen and I think it's very important as we enter this political season obviously the focus is going to be on the economy how do we get jobs how do we rescue people from this downturn but there's a longer story here and when we talk about the the other 99 percent we talk about long-term strategies for trying to bring the federal budget under control with the combinations of taxes and expenditures I think it's important remember that something's been going on for a very long time in this country something that's been eroding at the basic fundamental idea of economic security that once upon a time most people in this country enjoyed and uh simply getting creating more jobs and getting the economy on sound footing isn't necessarily going to take care of that problem so there you go and turn over to my and Matt here all right um well I'm so glad to be here this afternoon and I want to thank Jacob for for having me um you know it's it's always really like one of the key people I point to when I try to persuade people that like ideas really do make a difference in the long term in the political realm and that you know things aren't aren't just about elections and power and I think we've seen that over and over again in in his career um but with with with that in mind I I did want to share a couple of points of perhaps uh puzzlement on my part as to what it is where we're really seeing in the in the long-term trend here um that you know a lot of the time when we look at the economy we're looking at very large sort of macro aggregates did household income go up or did it go down and of course it's a big country hundreds of millions of people you know very many households so those broad aggregates you know mask an enormous variety in individual stories that you're seeing so in any given time some people are doing much worse than the average and others are perhaps doing better um so you know one reason you might see just an an increase in the number of people suffering very large losses is that if the overall growth rate slows down then you know that's going to mean that more people get sort of pushed below that 25 percent threshold um but I'm not sure we would really want to describe that as an increase in insecurity I mean it is an increase in insecurity but if the increase in insecurity is simply being driven by a slowdown in the in the overall rate of income growth that to me is a different story from a story if just hard luck cases are getting even less lucky than they used to be uh it is different policy implications uh among other things is this really a question of do we need to strengthen our buffers against severe deprivation or is it a question of we need to look at you know why is it that that overall wages for people who are doing okay have tended to slow down um because as as he noted in his presentation this has actually been a period of time in which we we seem to have acted at least on the medical front to do a lot to bolster the sort of unluckiest cases to expand Medicaid to expand s-chip during the recession we've expanded unemployment insurance um so you know we're doing things at those kind of severe bad outcomes um the other thing that's just amazing looking at at this chart is the um the the financial wealth uh one that that pulled up on on the screen you have here page 13 figure five um you know it's it's really striking that the median household has managed to accumulate absolutely no uh non-housing wealth on net um basically ever in the course of this time period that you know if you look at the the survey results that we started this with people claim to be extremely concerned about volatility and security in this particular way that they say you know that they they care more about the the prospect of losing out than about the opportunity to get ahead um you know the natural thing you would expect households to do if that's really their priority is to start stocking up uh substantial financial buffers if you're very loss averse person you know that's that's what you're going to do um in practice we we have not seen that uh americans have a very low household savings rate compared to what you see in many foreign countries um you know and that includes countries where the median income is a lot lower than it is in the united states nonetheless people are are setting more aside for for various kinds of reasons obviously we we had particularly during the most recent expansionary period there was a lot of encouragement both official and sort of unofficial encouragement for people to think of um their home their own or occupied housing is a great savings vehicle even though it's in fact um not a savings vehicle it's a it's a net liability if um the price of your house goes down as everybody has has discovered um so you know i mean simply in a in a qualitative sense this strikes me as the the thing where there's the most obvious focus there's a there's a disconnect between what it is people are actually doing with their money and with their financial decision making and what they say their priorities are and i suspect that that disconnect isn't simply that suddenly everyone became you know very stupid and doesn't know that you can save money um it's rather that you know we've had a various different things going on that have encouraged people to make certain kinds of choices that um obviously have not panned out well and that you know we need to think about facilitating encouraging ways for people to actually put aside money in a in a meaningful way or to return to the kind of uh defined benefit programs that used to provide more automaticity to to those things that um there was a big concept of a quote-unquote ownership society but that what we had in practice is obviously a debt society and that that leaves people much much more vulnerable to to income loss and we saw in the chart that you know uh the wealth cushion does not undo these effects but it but it makes a big difference i mean if people are able to save in a realistic way that can help them with a lot of these problems and that's what saving is for but we don't have nearly enough of it um so i guess that's what i've got thank you i think first i want to give jacob uh chance to reply uh to some of these comments larry's peak to peak trough to trough i looked at some of your dates and i was thinking well those look like important years for the boston redsox 86 uh came to mind but you can explain that's probably not the uh rationale but if you could deal with some of larry's points i too was struck as jonathan was by um how well the mid-midwest and the northeast were doing in these numbers um is that really the bend down so long it looks like up to me phenomenon or what's going on um there and um and matt's point about uh savings and the fact that we were a debt society and the difficulty that in order to hold up their standard of living people have had to resort to borrowing because they couldn't increase their incomes anymore um if you could deal with those three questions and then we can move on from there and any other comments you want to make in response to well i want to keep it brief because i really would like to open it up so let me take up the savings question first um all i can say is that when we've looked across countries we've found a similar relationship that is countries with more unstable income seem to have lower savings rates um so either this is a limits of individual rationality it could be something about um the um the coincidence of instability and other factors that depress savings rate but but it seems as if people who are experiencing and it's certainly at the individual level that's the case as well that the groups that are subject to the highest levels insecurity have the lowest savings rate but there are there's lots of obvious confounding factors so i think it's worth it's really worth examining more um austin has done a little bit of analysis of retirement savings and volatility and if i fair to sum it up as saying there isn't doesn't seem volatility doesn't seem to have much effect one way or the other on retirement savings so which is itself puzzling in both could be puzzling in both directions um what i will say is that i think your other point about the so our savings our financial wealth measure does not include holdings and defined contribution plans and so there has been a big shift over time from defined benefit plans which don't show up on balance sheets and defined contribution plans that do show up on household balance sheets these probably do reflect defined contribution but not defined benefit that's they correct they they include not we exclude both crucial both yes well defined benefit will will not be there obviously um it and so we exclude both but we also include both withdrawals from both the retirement so but the point is only that that shift there's a good there's a question there about whether or not households that are holding retirement assets in defined contribution plans whether they they see them in part as a substitute for other forms of savings you can after all borrow against a 401k plan you get it as a lump sum when you change jobs and lots of people spend that lump sum rather than saving it i would also say that um house there is this big elephant in the room which is house price values we're going up during this period and um so the real test for the theory right is going forward but my sense is that you need that what we've there what what's needed to encourage savings is both cultural and in terms of policy and that clearly it wasn't in place in the united states over this period i would just caution to say that i don't think with a lot of these large-scale dislocations that savings alone is only a partial you know small but important part of the the response you know i when we did the last report we did not have actual data for 2000 for the post 2007 period so we projected forward and the number we got was essentially the same as what we're ending up with here however it's worth noting that the levels were in general slightly lower in the survey of income and program participation and so in that sense we were surprised in that we would have expected it to be slightly higher i would note part of the answer which we're going to be exploring is that it could be public policy part of it is this important point about the nature of the the index which it captures instability um and in income so we'll certainly pick up the initial increase in unemployment but it depending on how that's timed and how it's shared within households it might be smaller than you would think um lastly i would just say and um and so this is something that we'd like to explore more um lastly i would say is that i think we we've very we worked very hard in general to make sure that the measure was i wouldn't say it's conservative as possible but that it was really capturing um very clear um um and and easily identified losses of this sort i think that there's certainly a lot more that goes along with unemployment and long-term unemployment than what we're picking up with this measure and anytime that you are also being very careful with data what you tend to do is in my experience is you tend to depress trends over time so that that it's possible that we're one to be even less you know less vigilant with regard to misclassifying people is insecure that you might get a slightly different trend so i will um i will only mention one quick thing um that john um brought up and then one quick thing that matt brought up really quickly um so um on we did a series of regressions when we were doing the projections and it's worth noting that overall growth did not seem very kindly correlated with our measure we also did a bunch of regressions that had time trends in them and controlled for unemployment and needless to say that the upward trend is statistically significant even when you control for these other factors why do we start in 1986 um there's two two reasons one our prior report based on when the sip begins had to start in 85 so we thought there was symmetry there and that we would eventually go back to earlier years it's actually very hard to do the earlier years with the march cps um and we can't we can't do the medical spending and the debt side of things for earlier years but there's another reason which is in 85 is one of those data void years in the sense in the march cps you actually can't get 85 so um we can't get 95 as well and we just interpolate between those 94 and 96 so there was sort of a another justification for starting there there was no other reason than that and I guess that's not a reason that um would would seem any less arbitrary than the ones you offered but that is the reason so um so then uh uh john just asked about um what would you know a very good question from a medical care expert which would I think implicit is sort of what would be the right measure of medical care risk and I think it would ultimately have to be some measure that combined actual spending and what spending you know in some sense should have been um and one way you could do that was look at similarly situated people right who are similar but you'd have to get met you'd have to get health status variables to really do that so the problem is really you know spending is highly variable even among people who are diagnosed with the same conditions but I think that would be the way to go you would know that someone is massively over under spending if they have a condition that is very costly and report a condition that's very costly but they report very little in the way of medical spending thank you very much um I want to get to the audience right away so I'm going to ask a question which I want to have in the backgrounds as for you as you answer the audience question my bias both as a newspaper columnist and as a Brookings person is what good is a report if it doesn't have policy implications so I'd like each of you as you're answering other questions to be thinking about what one might what sense what might one might make of this in policy terms and here in New America they give you some very helpful hints on moderating up here it says remind the audience we are filming and everything is on the record remind questioners to wait for the microphone that good gentleman back there has it identify yourself and keep their questions concise and then it's a pep talk it says be a firm moderator don't shy don't be shy about urging a question or a presenter to get to their point or get back to topic so I'll just say that off that's only what it tells me here so that gentleman is right next to you let's start there maybe we can collect a couple at a time so we can get more people in sir all right I'm Arnold Packer we always see in journals such as the economist the benefits of flexible labor force and the problems that come about with rigid labor force policies in places like Spain and elsewhere does this suggest that there would be more happiness if our labor force was more rigid Larry why don't you take that one I think you're gonna accumulate oh yeah that's yeah please oh go ahead you go to this person right here on that same vein I haven't heard any my name is Greg Woodhead formally with the AFL CIO now retired I haven't heard any explanation about union density either in figure three the long-term trend and in the regional explanation and then that lady back there I think yeah I'm Shannon Brownlee I'm the acting director of the health policy program here at New America I have a question and a comment the comment is that that embedded in your statistics is the fact that that average wages have been stagnant in large measure because of health rising health care spending and health care costs which come out of workers wages particularly low-end workers wages the question is can you can you parse your data by zip code because I think you would see some very interesting regional differences if you were to look at it on the basis of hospital referral regions where I work my other hat is at Dartmouth and what we've seen is that health care spending is rising very very rapidly in some parts of the country and really not rising rapidly at all in other parts of the country and I wonder if that if you would see you would see that economic insecurity embedded in your statistics if you were able to pull it out by zip code there you want to start since I already invited you to yeah well I want to just on the last question I think this is a new myth about health care being the reason for stagnant wages I don't I don't believe it whatsoever I understand that there's something to it but low wage workers very few actually get employer provided health coverage and their wage trends are as bad as everybody else's you can also look at any particular period of time when health care costs are going up or down it's not necessarily the times when you're going to have poor poor wage growth so I think this is I mean there is some trade-off but to elevate it to a macro explainer of wage stagnation I think is not right but I'll leave the rest of the questions of that to other people yeah I like Arnie's question about flexibility well I mean it's it means it's flexible if you have flexible labor market meaning you can easily get screwed but no one helps you to soften the blow that's a bad problem but if you if you had flexible labor market in in Denmark but people helped you to get by then you wouldn't find this so it's not so much as flexibility as much as it is the whole system union density Greg's question I'm not sure I mean other than I'm not sure how the as a big believer in the importance of unions I'm not sure how that connects actually right unless you got laid off yeah or unless you unless you and there's not that many I don't think this has a lot to do with people who are in continued employment taking a very large-scale wage cut which can happen with a union contractor or not but I'm not so sure this is one phenomena that that relates to that who wants to go I can just say a couple quick things because we we have looked in the panel study of income dynamics of individual level correlates controlling for other factors of volatility and union union membership does seem to lower income instability the problem is the holding other factors constant is very difficult and this is sort of what the peril of a lot of the comparisons a lot of the naive comparisons of wages or benefits across union and non-union workplaces but I would guess there is an effect there and the fact that unions are a smaller share of the workforce is part is part but not a big part but part of the story I haven't looked across the regional dimension but I think that'd be worth examining in general I just want to note that one of the you say benefits or drawbacks of our measures it's completely naive to compositional changes in the workforce or the society right so so people have gotten older but then there's more minorities in the population it's a picture of American instability as it actually has changed when you start to look within demographic groups as Larry was saying you don't find you find that groups have very different levels but that there's been an increase almost across the board which I think is a notable fact and fits well with the the kind of conception of what's happened to our economy that focuses on trends that affected a fairly large share of the population not not just those at the very bottom and on Shannon's question I would only say I'm there it clear clearly is the case that total compensation has risen more than cash wages it's also the case that when you look at the congressional budget office data for example which includes tries to include employment-based health benefits that there's still very modest growth at the middle so as Larry was saying perhaps more strongly than I would it doesn't obviously undo the conclusion that there's been a slow growth in total income total compensation including health benefits we did a test just to see if this had any effect on our our index because you might think that as people that we're we're we're we're counting medical costs but we're not taking into account whether you get employment-based health insurance and that's a form of compensation and so we did check this and it doesn't if anything we're probably once again biasing toward finding greater stability because actually there's been an decline and the average likelihood of getting insurance at your place of work and so if anything it's it's essentially our ignoring it is is making us see people are slightly more secure than they would otherwise be Jonathan you have a review on the health yeah I was just gonna say I mean I I think it's both things can be true though right I mean it can be absolutely the case what Shannon I mean that I mean you're measuring uh medical costs both me you would see it mass you would presumably see some impact of higher health insurance premiums and the income figures plus there's the higher out-of-pocket costs that you've observed I mean there are some there were some large figures there I mean can I just say something there John because I think this is actually an area of research that's needed which is that when we went back and looked at some other data sources on the 70s what we found is that uninsured people had much higher out-of-pocket medical costs than insured people in recent decade and like the last 15 years or so we find generally speaking that people who have insurance are spending more especially if you're including employer provided premiums which suggests that those who are uninsured are either getting charity care or not getting care which is sort of the point that we were talking about earlier so it's something that really requires more investigation right I mean I was going to say I mean I think you know how you can break this down in the day this debt is very hard to break down I mean this we run into this every time we try to break that down and I have no idea whether it would show up in the zip code data or not um although the zip code data about hospital referrals is very real that part I know is true um but I think there's some trade off I mean whether it's the big macro explanation I imagine is important in salvage there's some trade off I think most people agree with that and the fact that you would have it could be an explanation for what's happening at upper more upper income levels I would think and not at lower income levels is plausible to me at least I mean so this lady in the front row I'm sorry I'm making you run around thank you Ramsey all in with the National Council on Aging I have two questions the first is about disaggregating and cross tabulation looking at race gender and age and if there are plans to do some analysis on that front and the second is there was a figure shared in the presentation in terms of different thresholds of hardship and how long it takes one to recover and I as I recall the figure showed up on average it was about seven years regardless of the threshold and I was just wondering what the assumptions were around that recovery for an individual so another person want to get in right in front there I love Washington think tank audiences disaggregating cross tabulation thresholds high-end crowd I'm Mitzi Wertheim with the Naval Postgraduate School I'm going to ask a slightly different question we live in a society where it's all based on continuous growth and continuous consumption how does this fit into the kinds of things you're talking about from a policy standpoint you want to start with the question about so on the disaggregation I really encourage you to take a look at the web tool you can actually do up to four cross tabs I think generally speaking you get up to three and you start to run into small end problems and we won't let you see the the data if it's fewer than I think 50 respondents who represent that group so if you try to do single mothers who are 18 to 34 and who have less than a college education and then you try to add one more on like race then it usually says you can't go there but but you can you can look very closely at these different cross tabs and if you take a look at the at the site you'll see that the the the the y-axis is very high it's about 40 percent so means which which means that I'll I'm going to explain so it means that there are some groups for which we had to put the axis up that high because they reach up to levels that are close to 30 30 to 40 percent chance of this kind of large experience of this large income loss on this your second part of your question Austin's actually here and he and he's the expert on these recovery paths that that was illustrative but what we do it what we're intuitively what we're doing is we're looking at for people who experience losses of these three different sizes that I put up there what is the typical recovery path that is what's the median recovery path that's half of people recover more quickly and half of people recover more slowly and it turns out that it's it takes a while for people to recover at four to six years is that correct so four to six years is the typical recovery path it's slightly different than those numbers I put up because those are naive just looking at the actual data whereas we do a statistical modeling that takes into account the fact that our data is censored and other we because we wanted to have a large enough end to be able to reliably estimate that we basically looked at people by drop size by family income prior to the drop and by age so it doesn't look at race so actually in that sense we're kind of probably giving an extra we're lowering the threshold for some very disadvantaged groups because they almost certainly will have a longer recovery path but as it turns out it just really most people don't have enough savings to be able to offset these kind of income declines on the second part about you know continued growth you know my view on this is that while growth has a huge role in thinking about our long-term economic well-being that this measure of income stability is to some extent a little orthogonal to growth and this is sort of speaks to Matt's question as well and because we found across countries that that rates of growth don't seem to be strongly correlated with income instability it's not the case that you and this may speak to the flexible labor markets question as well it's not the case that there seems to be a huge sort of growth instability trade-off and one of the things that I think is most interesting in these results and it is I was concerned when we started doing this work that that these downward drops were just the flip side of upward mobility that people had a big jump in one year and then fell back but when you draw when you look across countries you could sort of draw a diagonal line looking at increases and decreases right and if there's the same number of people experiencing 25 percent increases is experiencing 25 losses then the country would be right on the diagonal most of the countries are up here right so they they have more increases than decreases the United States is right on the diagonal so over this period at least it doesn't look like that there's been a big upside gain for these downside losses. Matt I can't resist throwing the big think future of growth question to you. You know I wanted to say on consumption something that's that's interesting that if you look at you know all the different variants of this stability line it's a it's very jagged you know things are very different one year to the next you can see a long-term trend I was looking recently at the longer-term trend in simply in personal consumption and that is much much much smoother you know one way or another people through most of this period are finding ways to sort of juggle you know using whatever credit cards housing wealth so on and so forth and then what you see in the recession is a huge and really unprecedented collapse in personal consumption which I think you can infer is sort of like the time when when all the tricks ran out you know there was sort of nothing more to get at and that I mean in some ways shows the the realist sharpest point of instability in people's lives when you have no more way of getting stuff you know people people have a lot of ways of compensating for income loss historically and they really just don't anymore. Which raises interesting questions about the future of growth given this experience will people go back to the if if they ever have the opportunity again to the mix of borrowing converting this or that or will we be operating a lower level of consumption for a while. Yeah there was a report out today from from the New York Fed showing much lower levels of debt than had existed even even a quarter ago and I called them up and I asked you know well is this actually are people paying their debts off or people going bankrupt and they said well you know in a couple months we'll have that. Last round of questions please sir and then way in the back my name is Robert Hurstine I just am speculating whether the increase in US population due to immigration during the period you are covering might consist heavily of people who enter the workforce at a very vulnerable level and whether that might in turn account for the increase in your blue line on income vulnerability. That's a great question. I'd be curious what kind of how you cut your data on that and then yes ma'am. My name is Lee Young. I appreciate this effort about just income stability or how to help the homelessness or poor or lower income but problem is my major concern is really the data the reliability of this data that you use to really measure the reliability of this situation and the effort you want to do is whether the nonprofit organization or government policies they really put some effort and resources to resolve the problem because a lot of times whether the money go to education or go to health human services those money resources are not really benefiting the targeted population rather than those money or even people's resources have been deprived or by those people who benefit themselves rather than benefit our society or poor income. So I wonder if you have this kind of effort or you know any kind of study try to resolve this type of problem. Could I just on that question read a sentence out of the couple of sentences out of the report which I think are partly responsive because I appreciate the fact that you pointed out that and I'm quoting here for many Americans especially the poorest are those who have the fewest financial resources the 25 percent threshold of loss may be too high about one in five Americans according to the survey conducted in conjunction with the study indicate that they would face financial hardship if they were forced to go for as little as two weeks without income. So I think it does address part of that but why don't you why don't you take we'll close taking those two questions Larry. Well I'll save Jacob some time and try to respond to the immigration one a little bit. This I'm sure this is not a complete answer but if you look at figure 10 which has Hispanics white non-Hispanic whites Hispanics non-Hispanic blacks you'll see that there's not much of an increase in economic insecurity in the first two periods among Hispanics and there was a lot of growth of immigration in that period so it seems like immigration wouldn't be driving it and in fact if you look over the entire beginning to end periods it might be that actually non-Hispanic whites have the largest increase in economic insecurity so that would be their group that probably has the least growth of immigration so it wouldn't seem like immigration is driving this. That is what I would deduce. I mean that would have been my answer Larry. I was trying to say thank you very much. I only add that there might be indirect effects of immigration that wouldn't be captured here on the labor market there might be indirect effects of immigration on the labor market but my read of the economics literature on this is that that it's probably not those are probably not that large compared with other factors that are going on here. So I would just speak for a second to your point in the second question which is to say that I think that one of the drawbacks of any index that is based on existing data like the Census Bureau data is that is comprehensive as it is it doesn't include a lot of the questions you would certainly be interested in right now about how people have actually coped with these economic losses so we did ask such questions in our own survey and I would just note as another sort of plug for the levels for what it's worth we don't have a comparative survey from earlier periods that when we calculated the annual income loss we asked people whether they had a 25 percent or greater income loss and we got 19.4 percent which is essentially exactly what we got for the income loss component of our index for the same year so I don't know what the comparison is in the past obviously but it again was remarkably close. We asked about unmet needs which I think are close to the consumption discussion earlier we asked about whether people had gone without food or medical care because of the cost or whether they'd had difficulty paying their bills and questions of that sort and it was striking that people who'd experienced various shocks as we measured them such as losing a job or having very high medical costs were much much more likely to have those kind of unmet needs and even that was true even among middle middle class families and so I think that's while well Matt is certainly right that aggregate consumption trends have been pretty stable there's some evidence that and it's still a debate is that it looks like consumption volatility has increased during this period but I think he hits it exactly right and sort of implicitly quoting that old aphorism attributed to the economist Herb Stein known as Stein's law that if something can't go on forever it won't and that you know borrowing like people were borrowing especially against homes that were overvalued sending a second earner into the workforce and the like these were strategies that had inherent limits and now we're sort of facing up to that those limits today but but I would say in closing on just to sort of thank everyone for I want to thank everyone for being here and and and say that Larry in particular I'm going to take all your comments to heart and there will be a new version of the report at some point that will that will have more justifications and nicer graphs and and the state of working America has set a high standard that we all should aspire to thank you all very very much