>> they decided to transfer almost to good samaritan hospital where the facilities were better for delicate blaine sur brain surgery. mrs. kennedy was in the ambulance with him.laine brain surgery. mrs. kennedy was in the ambulance with him. the suspect was grabbed by the two kennedy men and led by police back through the ball room and hotel. some of the officers had to protect him from the crowd, there were several kennedy supporters, bystanders, who were close to hysteria and there was concern for the suspect's safety. >> watch real america, sunday, at 4:00 p.m. eastern on american history tv on c-span3. coming up next, a conversation with the social security administration's chief actuary on the future solvency of social security and how economic and population trends will affect the program. from the national academy of

social insurance, this is about 90 minutes. . >> i think we're ready to get started. on behalf of the board of directors of whom we have two representatives who will be introduced to you momentarily, the extraordinarily dedicated staff of the academy, many of whom are here, and our members who are the life blood of the academy, i'd like to welcome all of you to this annual briefing that we conduct with a key group of panelists on social security finances. i'm the chief executive of the academy. i just came back along with josh from arrelington cemetery wheree commemorated the 50th anniversary of the robert kennedy campaign. knowing the age group in this room, how many of you have heard of robert f. kennedy?

i feel much better. better than and i get who? i worked for him as you may know.thanes that and i get who? i worked for him as you may know. i was on his senate staff and today was a very difficult day. yet very up lifting. what does that have to do with this? in 1968, his campaign was an 85 day campaign. he gave one speech on social security. in michigan i believe it was. and he called for an across the board increase. at the time the monthly benefit was $140. this is in april of 1968. he wanted an across the board increase that would raise to

$165. no wone what we would have toda, what is the long range impact. and more importantly, who's going to pay for it? and i don't know how he would have answered the first question. but i know how he would have answered the second question. we all are gonna pay for it. because as he would say, it's -- social security is all of us for each of us. and i'd like you to just keep that theme in mind, because you're going to get into the weeds today. a lot of technicalities which are critical to understanding the program. but let us never forget that for millions of beneficiaries, this is a lifeblood. and we have to keep that in mind. it's a lifeblood. and it may be for future generations it's even more of a lifeline, i should say, for future generations. so with that in mind, let me turn it over to the academy's vice president of policy to introduce our panel. c-span is filming but not airing

it live. that will mean if you do ask questions, identify yourself and think twice before asking that question, because it will be preserved forever on c-span. ben, let me turn it over to you. >> welcome, everyone. thank you for coming to this briefing from the national academy of social insurance. what is the outlook for -- [ inaudible ] i'm vice president for policy at the academy. the academy is a nonprofit, nonpartisan organization made up of experts on social insurance. its mission is to advance solutions and challenges facing the nation by increasing public understanding of how social insurance contributes to economic security. before we begin, i wanted to call attention to -- bill already thanked our staff, which i would echo, but in addition i wanted to call attention to a brief, which elliot sker is in the back, that's available, many of you picked it up on the way in.

today's briefing is at first glance about numbers. the financial outlook for social security. projected revenue, projected spending, trust-fund reserves, et cetera, over the next 75 years. i wanted to highlight two other numbers which are relevant today, one bill already touched upon. the first is that today is the 74th anniversary of d-day when millions of -- when hundreds of thousands of americans stormed the beaches of normandy. 53,000 americans died that day and many more were injured. so i think that's something we should keep in mind today. and as bill mentioned also, it's the 50th anniversary of the assassination of robert kennedy. both of these events bring to mind the sacrifices that americans have made for our country and for the common good. and i think it's important to remember that social security is

a program that is embedded in american culture. it's about far more than numbers. it's not simply -- social security is not just about the finances of the program or how it impacts the federal budget, or even about the dollar amounts that beneficiaries receive to help make ends meet. it anchors the middle class, it helps millions of american families protect themselves against impoverishment when tragedy strikes, whether it's disability, inability to work in old age or the death of a breadwinner. it unites us also by strengthening the common wheel. it's something all of us pay into and all of us receive benefits when we need them. it's a shared system, and it strengthens american civil society. that said, social security does face financial challenges, and it's important that we talk about those. that's why we're all here today. the national academy of social insurance is a place where people from the left, right, and center can come together to talk about our social insurance programs in a civil discourse, and also to -- and social security is probably the most

important of those programs. so with that, i'd like to introduce our speakers, steve goss, stephen goss, actuary of the social security administration. karen glenn, deputy chief actuary of the social security administration. henry aaron, bruce and virginia mccory, senior fellow with the brookings institution. kilolo kijakazi from the urban institute. and doug holtz-eakin, president of the american action forum. their bios are in the packets. so if you want more information on our speakers, please look there. with that i'd like to introduce steve to begin talking about the program's finances. >> great. thank you very much, ben. and karen and i are going to do as we described, sort of a tag team run you through some of the material. we have a number of slides. ben will yell at us if we're taking too much time. so let me just run through the first couple slides, karen will run through some. we'll go back and forth, to keep it interesting.

because it's been said that some actuaries are not as interesting. so tag team will be better. why are we here today because of social security, but in particular the date of yesterday the annual trustees report came out? every year, starting 1941, there's always been one. why are we doing it? because the congress demands it. we do what they say. it's required that every near an annual trustees report comes out, april 1-ish or so, when we get it completed. that speaks to three basic things. what are the operations of the trust in the last year, what are they expected to be in the next five years and what the actuarial status of the program? the status, really and truly, it's to inform policymakers that under current law, how are we doing financially, and if we look like we're going to be having shortfalls now or in the future, to what degree are we

expecting shortfalls, because that tells policymakers what the job is ahead of them, when they have to make changes and how big a change they're going to have to make. it's up to them to decide what the change is. the history of this is that whenever we've reached trust fund reserve depletion, congress has always stepped up, always acted on that. for this year's trustees report, we have three primary changes we would like to just toss out to you. we have a lot more detail on some slides. the three primary changes that will leap out the most is in the area of the disability insurance trust fund. the disability insurance trust fund has been much paid attention to for a long time now. a lot will recall back in the bipartisan budget act of 2015 we had a re-allocation of tax rates that extended the reserve depletion date from 2016 out to 2022. in the next report, we bought one more year of solvency because of the good experience

in social security, then after that in the 2017, we went five more years to 2028. in this year's report, we're going to 2032 for the expected reserve depletion date for social security. this is pretty good news, and why. basically because the experience under the d.i. program has been remarkable. we have slides with more detail. but applications to social security disability have been dropping dramatically. the number of people newly starting to get disability of those who are insured and not already receiving have been dropping dramatically beyond our expectations. so disability insurance, all good news. the number two item we point out here is a slightly different story, which is on the revenue side. on the revenue side, we have a couple of changes, the most significant of which is the social security program, the

oesidi and the hospital insurance program are affecting our taxable payroll, which derives from the earnings of everybody in the country. and if the earnings of everybody in the country do not grow as fast as we expected before, we'll have less that are subject to the tax and we'll get less payroll tax. lo and behold, we are suggesting some of that this year for a couple of reasons. one is, we had revisions by the department of commerce, their bureau of economic analysis. those are the ones who cook up gross domestic product numbers and national income numbers, all of that. they came up with revisions to their 2016 numbers that we had used, the initial ones in the 2017 trustees report. the revision showed a smaller amount of employee compensation relative to the size of gross domestic product. the ratio of compensation to product they revised down for 2016. lo and behold the numbers they have about into mid 2017 are also lower than we had been expecting and projecting in the prior report. gdp is pretty much been realized

for what we were projecting, but the share of the gdp going to employee compensation, earnings, and taxable earnings is reduced. so we have less tax income coming in. and that is a negative. the other factor on that is that as many have noted and i think we have a slide showing this in more detail later, that what we call labor productivity, the amount of output per hour of work by u.s. workers has not been growing at the rate we were expecting. and as has been done in past year, the trustees have decided we will accept some small portion of that shortfall as being a permanent loss to our future economic prospects, which lowers the level of gross domestic product, meaning slower growth rate in gdp over the next ten years. both contributed negative to the revenue we have for all programs. that caused our reserve

depletion date for the oesi program to change from the very beginning of 2035 in last year's report to the end of 2034 in this year's report. it's two or three months' change, happens to flip a year, it was only a portion of a year. the third thing we want to mention really is, and karen and i have talked about this a lot, what do we say about the status of the oesi and d.i. combined program as a whole? we look at the deficit and last year's report was at 2.83% of payroll. think of that in terms of the 12.4% of payroll which is the tax that people pay on their earnings. so 2.83% is the shortfall for the next 75 years. for this year's report, it's higher at 2.84% of payroll. however, is that a good thing or a bad thing? in last year's report just on

the basis of moving up one year to having our valuation period be one year later, we were estimating it would be up to 2.88% of payroll. it's not gone up that much. so relative to our expectation of last year, all other things, assumptions, methods and realized data, have on balance been better than we were estimating in last year's report. on that, the difference between income to the program and expenses to the program for most of the future period of the next 75 years, is actually a little bit better, which contributes towards our overall deficit, not having risen as much as expected. with that, let me pass the baton to karen to cover some of our fun graphs. >> great. thank you, steve. so this graph right here probably looks familiar to those of you who have been to this briefing before. what it shows is the trust fund ratio which is really the asset

reserves at the beginning of the year over the cost expected for that next year. so the higher, the better, obviously. the pink line shows the oesi by itself. the black line shows the combined oesdi trust funds and the blue line shows d.i. couple things to note on this graph. once the reserves reach zero, that will be at the bottom line, you can see the d.i. from last year to this year extended out those four years, like steve was mentioning before. oasi and oasdi are very similar to what we projected last year. you can also see the tax rate re-allocation pretty clearly on this graph. back in 1994 and again in 2016, we made changes to the tax rate, or more specifically, congress made changes to the tax rate. we didn't do that ourselves.

and you can see that that really affects the trajectory of the d.i. fund. >> very, very quickly, on these next two or three slides, it gives a little more detail that you might want to reflect on later about what is going on with the disability insurance program. you'll see on slide 5 here, these are applications coming in applying for disability benefit for the social security program. that's not just disabled workers, widows and children, but almost all disabled workers. you can see the applications were over two million at the peak of the worst part of the recession back in 2010. and they have dropped very steadily all the way through 2017. and you can see what we projected these applications to be doing in prior trustees reports. we've expected it to be dropping after the peak, but to turn back up around, up to a more stable, expected level. that has not happened. and the really interesting part of this, there's more to come.

because in the year 2018, we see that applications have continued to decline, which we did not assume when the assumptions were set about the end of the last calendar year. >> you can also see in these graphs that we're projecting in the long run the applications will end up in about the same place. that's a trustees assumption to assume it will go back to a more normal level. >> more normal, yeah, what we have believed is more normal. >> what we expect is a more normal level. >> and whether we should be modifying that is another question. on the next graph, this is very, very similar, the disability incidence rates, the number of people newly starting to receive disability benefits as a share of those insured and not already receiving a benefit. this looks very much like the other graph. we see on the expectations in the last several trustees reports have not been realized. the rates keep dropping. the one thing that's a little bit different on this and a nuance item is for the year

2017, you can see the incidence rate went up a little bit. you might say, how could that be when the applications just keep going down year after year after year? this is a little bit of good news. many of you have probably heard about the administrative law judge backlog of cases awaiting a hearing. for the first time, the social security administration has started to make progress, we're starting to get the backlog down. we had a surge in the number of dispositions made by the eljs, and with that, many of those people were in fact allowed benefits and that gives a little bit of a surge in the number of new allowances which raised our incidence rate for that particular year. but remember part of the increase there is not because a natural increase of disability incidents, but catching up on cases that were waiting determination. and one other item, the culmination of all these things put together, what does this mean for costs? >> there was a straight line on

the number of people in receipt of disabled worker benefits under the program. along came the recession, you can see in 2008, 2009, and 2010. but by the end of the period they're a little bit better. in between, they fluctuate a little bit. so why does that happen? the big thing in the very near term is we're expecting lower payroll tax revenue, and we'll get into more detail on that later on. after the near term, we're expecting demographic effects to really kick in. >> this next figure is sort of the annual shortfall of income versus cost in each rear. in last year's trustees report

versus this year's. so the red line is this year's report. so what you can see is things are a little worse in the very near term, but by the end of the period, they are a little bit better. and in between, they sort of fluctuate a little bit. and the big thing in the very near term is we're expecting lower payroll tax revenue. and we'll get into a little bit more detail on that later on. after the near term, we're expecting demographic effects to really kick in. we have some changes to our assumed fertility rates in the near term. and in the longer term, mortality. there's been bad mortality experience recently. higher death rates than we've expected. which is bad for all of us, but that's actually good for the social security program, unfortunately. the sooner people die, the less

benefits they get. so sort of a bad news/good news story. >> i think that covers it well. and one thing to note is towards the end of the 75 year period, you can see the trustee's report has a smaller negative balance so actually by that time, i think largely bauer of the mo l mortality effects, we're in better shape towards the end of the 75 year balances. >> so real quickly, we will not go through all these numbers. if you want to delve into more details, please feel free to let us know. you can see at the bottom of the page, we're saying the net change in our actuarial balance is a minus 0.02% of payroll. that is a slight worsening from last year. most of that is due to the first line on the page which is the change in the valuation period

because we're picking up every year our valuation period is 75 years long. the last year that is new for this valuation is a year where we're kind of at a balance. so we're picking up that one year, it makes things look a little worse all by itself. so that is a minus 0.06. >> which explains almost all of our net reduction. >> so factors pushing things in the other direction, we have a bunch of demographic effects going on here. interesting one is the fertile y i ity rate. you may have seen that the 2017 news on birth rates much lower than people have been expecting. i think it was about a 1.75 children per woman expected in her lifetime. our long term assumption is two children per women. so that is pretty low. it is the lowest it's been since, what, 1976.

and one of the things we look at all the time is birth expectation surveys. we ask women how many children they expect to have in their lifetime. and that is consistently been around 2.2. so a little bit of a mystery why things are so low right now. so lower birth rates in general are worse for the social security program. it is not a good thing. so recent data and some slight changes to our assumption for fertility in the near term cause that minus 0.08. mortally as i mentioned, death rates are worse than expected which is good for the program. a 0.05 there. steve, do you want to go into the economics? >> i think they are probably already mentioned. think about how our ultimate level of gdp is lower. it has a small effect here. we have a lower interest rates,

we're taking longer within a ten year period to phase up to our ultimate interest rate which is the ultimate rate has not changed. and starting values as some of the starting values and near term economic assumptions, other ones you already mentioned, had small positive effects. >> and then last point 0.05, other new data, methods, improvements, we're always working on improving our methods, incorporating all the new data we can. so this year that was a little bit of a positive for the system. >> the next slide is an illustration on the labor productivity. you can see if you look very carefully, in the last few years since about 2009, in this recession, we've been having labor productivity, that is the increase in the amount of goods and services produced per hour of work by our economy has been way below the 1.7% handle rate

of increase that has been pretty much historically been the case. in fact it has been higher than that. you can see from 1949 to 2016, it lnaveraged almost 2%. the last almost decade it has been well below that. and one thing that we have done in this year's report is to accept some of that shortfall as being a permanent loss in the level of output per hour for american workers in the future. and you can see that everyone in sort of another sort of framework on slide 11 here, the lower potential gdp already mentioned, you can see back in 2010, we were projecting the level of gross domestic product under a full employment situation where there are enough jobs for people who are really wanting to work, and that was the black line. you can see we have year by year taken a little bit off of that expectation as we've had this continuing period of labor

productivity growth falling well short of expectations. there a possibility at some point that will turn around. we'll have massive investment in the workforce and we'll have labor productivity go up much faster for a while. but we have simply not seen that as yet. but we only incrementally have been accepting some of the drops. >> as already mentioned a couple of times, mortality experience has been a lot worse than we've expected. the black line here is the actual death rates. you can see that since about 2009, they have been fairly level. generally we expect mortality to improve over time. it really hasn't been since 2009. another thing to point out, this is all ages. we do look at things by specific age groups, but this is sort of a summary of what we've done. the colored line, you can see in every report we've been expecting mortality to continue to improve.

it hasn't happened in recent years. and i talked to my folks in the office and they have preliminary data for 2017, it looks like that black line will actually go up a little bit next year. so still not looking good. >> continuing the trend. >> this is another figure that probably looks familiar to a lot of you. it shows the difference between the cost of the program which is the blue line and the income which is the red line. the blue line you can see drops down at the time of reserve depletion in 2034 to equal the income line. so the blue is really covering the red at that point. that is because once reserves deplete, we can't pay out any more than we're taking in. so those lines will become equal at that time. the dashed blue line is really

the cost of all scheduled benefits. so after reserve depletion, we will not be able to pay all of those. we will be able to pay whabout % in 2034 going down to about 74% by 2092. so another takeaway from this graph is the difference between the dashed line and the solid line is really the funding shortfall that we have to figure out a way to fill. >> one tiny aspect on this, trying to throw in good news here and there, the 79%, the spoints of reserpoint of reserv depletion, that is up from 77 cents last year. and at the end of the period, the 74 is up from 73 cents. so it does vary from year to year, but we'll take any little positive we can. >> right. here is another graph that looks very similar. instead of looking at a percent

of payroll, this one looks at a percent of gdp. so the focus of this graph is really is the program stus takenable take sustainable in the long run. if the cost in income lines were wildly diverging, we'd say no. but it is clear from this that the gap is remaining relatively level over time. there are ways to fix the problem. >> so once the baby boomers all retire and are followed in their into the steps by the lower birth indicarate generations, w of stabilize ergo it stays about flat. just need some adjustments. >> and one more graph that again looks very similar, this is ratio of beneficiaries to workers in the population. you can see it has got the exact same picture as the cost of the program. so what this really says is that

the cost of 10essocial security almost solely driven by the demographics. >> and to that point, this next chart shows you really how it is drivendemographics. this is the ratio of number of people in our population 65 and older to the number who are between 20 to 64. often referred tos at prime working ages. and this has almost exactly the same shape. and we wanted to illustrate the reason why the ratio is going up, it is demographic. and specifically it is really more than anything else birth rates. you can see the black line is what we're actually projecting under current estimates, but you can see the blue and wrred line are what if back after the end of the baby boom period we had stayed at 3.3 or even 3.0

children per woman. 3.3 is what it averaged during the baby boom period. if we stayed at those levels, it only very gently has been rising because of increased longevity. the big hop we have is simply because of the drop in the birth rates. and that is reflective in the cost of our program, cost of the medicare program, cost of lots of things including pvgc probably. uncertainly illustrations, we throw this in because want to make sure that you know we do not just simply make the sort of central intermediate assumptions. we have at least actually several ways of showing variation in what the estimates might be because we know we're unlikely to nail it and have it exactly the right numbers all the time. the central numbers are what we have on the dashed lines for our actual intermediate projection, but then we show our high cost and low cost where we vary all the principal assumptions toward

better or worse for the financing of social security, but in the solely lines the 95% confidence interval for the annual cost rates for the program under our methodology. and we'll give you a preview that we've been working for quite a while now, it is indicated in the trustee's report from time to time that we think this range is probably not as great as it should be. there are some technical things like parameter uncertainty, i'm sure everybody reads about that every night before going bed, that we should be able to expand the range. and i think there will be a study coming out in the not too distant future. probably in trouble for mentioning it, but we're hoping to get there. and this other chart we wanted to show you is something that a number of us have looked at and been a little concerned about. we know what we pay attention to for actual status as demanded by

the congress is what is the actual status of the trust funds under current law. there is another way of looking at some of the number which is is the budget's growing convention and we want to make clear for those who pay attention to the budget scoring and look at these big increases in the debt held by the public that a lot of those projections are based on a specific budget scoring convention which is that if the reserves deplete, that in effect presumes that all of the benefits will keep getting paid which of course the law does not allow and how will they keep getting paid? by having money somehow transferred over from the general fund of the treasury which will be borrowed from the public in order to keep paying the full benefits. the problem with that while that is a possibility, that would require a change in law. so we just suggest anybody who shows that kind of a projection of the debt held by the public under the budget scoring convention, it should be qualified as saying assuming the law will change in order to be

able to make this happen. >> all right. so we have shown you quite a few slides focusing on the funding shortfalls for social security. so how do we long-term? bottom line, we've got to fill that gap, like i showed you on a few slides. congress can make certain choices addressing that gap. number one, they could raise scheduled revenue after 2033 by about one-third or they could reduce scheduled benefits after 2033 by about one-fourth or some combination of the two. one other thing people have suggested is that maybe we invest more of our trust fund reserves and get more interest income. it's a possibility and i know $2.9 trillion sounds like a lot of money in reserves, but it's really not in the context of our program. we could get a little bit more

income by doing that, but it really will not be a huge help. >> in our final slide before passing the baton is all the places you should look for more information on our web page, including for the new trustees' report we have a massive amount of information up there explaining a lot of the detail and a lot of extra numbers more than the report itself. >> henry? >> okay. i'd like to start by referring to the last slide that steve referred to, the reference to their site, the actuary site on the web. it's a gold mine. anybody who is serious about social security and doesn't dip into that is simply not availing him- or herself of the available information. now, they've emphasized, karen and steve, the changes and -- from last year and the detail

elements of the projections. i'd like to step back a bit from that and suggest that the major important take away from this report is that there is very, very little news in it. it is negligibly changed from last year or the year before or the year before that. the reports have consistently painted the same picture for the nation. there's adequate money to pay benefits for now and the next 15, 18 years. at that point or somewhere around there plus or minus a year there is a financial problem, we won't have enough money to pay all scheduled benefits. and something needs to be done at that point. that's the take away from this report.

one can get to some degree into details and i want to mention just two of them for purposes of underscoring what i think is an important point about the methods that they use in doing their projections. i want to focus on two important assumptions that go into the long run estimates. the first is the one that they emphasized at the beginning, the improvement in the prospects for disability claims. there are fewer of them than there were in the past and fewer than were expected for this year in the past. the trend is down. they have not fully incorporated that craig into their long run assumptions. in fact, it influences their numbers only over the next four or five years and then everything goes back to the same baseline that existed before. i will come back to why that's

important, why it opens them to criticism and why i think it's the right thing to do. the second assumption that's key is the fertility assumption. this year their projection is based on an assumed fertility rate considerably above that reported this year. that income has been trending down. that's a very important assumption in estimating long run costs. they have not adjusted their numbers yet. as they have not adjusted the long run assumptions with respect to disability. now, many people look at year to year changes in the values of specific parameters that go into their estimates and they say, oh, you are off the trend, things have changed, and expect and want the actuaries to change

their projections based on the latest news. consistently they don't do that. rather they wait for trends to become well established until there is a logical basis, an informational basis for explaining why trends have shifted and then they move gradually in the new direction. that means that -- it helps explain, in fact, the observation i made at the outset, the similarity of projections from year to year. ask yourself how useful these projections would be if they jumped up and down year to year by large amounts. those jumps would discredit the projections themselves, they would be responding to events that very often indeed much of the time reverse themselves because there are deviations from long-term trends that are

offset and returned to base values. so i think in looking at the projections and in looking at the observations that you will hear made about the projections, keep in mind that this is a very damped adjustment mechanism that is used in making these projections. it's exactly in my view the right thing to do, as long as they are responsive and they have been, to evidence about long run trends. you may hear some disagreement among us about whether particular values should be different from where they are, i'm not going to defend everyone of them, but i think the general approach is exactly the right one and you should keep that in mind. now, if this is the bottom -- if the bottom line is there is not a lot of news in this report, then this brings us, i think, directly to the critical question which one of their slides referred to. we've got a long run financing

problem, how are we going to fix it and when are we going to fix it? well, i think i would have few dissents in this room if i said not likely this year, indeed not likely probably for the rest of the current administration. we might have an opportunity in the next administration, whether the next president is a democrat or on the off chance it was a republican other than the incumbent. but i want to suggest that the consensus among close followers of social security is that congress is going to punt this thing as far down the road as it can and the reason it's going to do so is whether you are a republican or a democrat, taking the steps that are necessary to close the financing gap are politically -- are likely -- are

politically very dangerous in your constituency. if you try to raise taxes as a republican you are in deep trouble. if you try and cut benefits as a democrat you are in deep trouble. and looking ahead, many people would argue that a likely solution to this problem in the long run is going to include both. waiting a very long time to take the necessary steps is in my view exceedingly dangerous and this is the key point i'd like to leave you with. let's imagine we have come to 2033 or actually a couple of years later when the trust fund will have dropped below 100% and steve or his successor, maybe karen, will report that the system is no longer in close actuarial balance and red flags get raised.

at that point one could solve the problem, the financing problem, entirely by raising taxes, technically. no problem. you can raise taxes quickly, it would result in a small drop in take-home pay for workers, but it's not a major event. we move taxes around from year to year quite a bit anyway. at that point you cannot solve the problem by cutting benefits quickly. you can't solve the problem because it would involve the estimate of cutting benefits by a quarter for everybody now on the rolls at that time and everybody now coming on to the rolls in the next -- following that date. it is impossible to conceive that members of congress, of either party, would vote for a 25% cut in benefits for every

retiree in america. all right. well, you say, well, they don't do so much of that. they don't cut benefits for people now on the rolls. well, then you would have to cut benefits more for people coming on to the rolls fresh. you are going to tell 64 year olds, 65 year olds, you were planning to retire with a certain amount and you're going to get 35%, 40% less in social security. not likely. so that then leads to what i think is the scenario that should scare both conservatives and progressives, and that is that congress in that situation enacts sizable benefit cuts that don't take effect immediately, sustains benefits for those already on the rolls, doesn't raise taxes and borrows funds in order to pay for a very extended

transition. that means the higher debt scenario that steve pointed out would require a change in law to make cbo's projections correct, could become a reality. it's bad economic medicine, it's bad pension policy medicine and it's something in my view that argues strongly for trying to move up the date at which we address the projected long-term deficit in social security as early in the 2020s as it is possible to do. >> so i want to thank the national academy for the chance to be here today and to join this group of esteemed panelists. i especially want to acknowledge the work of steve and karen, not just this year, but as a sustained contribution to public service that i think is too infrequently recognized and for

which they are not thanked enough. my job is very simple, my job is to explain that henry is 100% right and far too calm and that's what i'm going to do. you know, as he pointed out, there really isn't a lot of news in this report for those who have been following the evolution of the system. the kinds of things that i can say today i said in 2003 when i was the cbo director. a lot of this was entirely foreseeable, it was the demography and the structure of the program. the numbers do move around bit to bit. a lot of care is taken to correctly characterize the outlook for it and, you know, what we have found out is that as we expected, you've got a problem, a mismatch on a sustained basis between scheduled benefits and the revenue dedicated to the system. and that is the problem. it needs to be dealt with and i think it needs to be dealt with very quickly for a different reason than henry laid out.

i am the problem. that's usually true, but in this case i am the problem because i am the trailing edge of the baby boom generation, i'm the trailing edge of that demographic shift. i'm 60 years old. i'm an affluent person who can afford my retirement more than most americans can and if you grandfather me and you do not reform social security for me, you have grandfathered the problem. and we are perilously close to doing that. so i think the amount of -- the urgency should be heightened even above what henry is saying for that reason, as a matter of social security policy. i believe as we sit here one of the ironies about the social security system is that this is one of the most effective and sustained social insurance programs we have ever seen, social insurance is designed to remove financial risks from the lives of americans, particularly those who are not particularly

affluent and aging. if we leave it in its current position every year that passes it becomes a source of financial risk. people do not know what their benefits will be come 2034, 33 or 35, whatever that date may be as congress proceeds to not take any action on one side or the other. that's incredibly ironic and disappointing to me. we don't want a system that's supposed to remove financial risks become the source of the risk. we should move quickly to fix this. we should also move quickly to fix this for at least two other reasons. one is this challenge comes in the context of much larger budgetary challenges. anyone who looks at, for example, the congressional budget office baseline projections knows that. there is an enormous amount of red ink elsewhere in federal finances. the fed care program runs an annual deficit of $350 billion a year right now. it doesn't have a dedicated source of finance like the payroll tax, has a combination

of some payroll taxes, some premiums and a whole lot of reliance on general revenue. we have a whole lot of the federal budget looking at that general revenue and this isn't enough. to pretend we can borrow our way out of that problem is to court economic disaster in my view. it takes place in a larger context, i think that's a reason to move. the other reason to do it is it's pretty easy to fix. you know, if you get on the metro you can fix it a couple different ways every day. there are payroll taxes coming in, benefits going out, benefit formulas and you can imagine different variations on all sides that will get the system into balance quicker, slower, but can get there. that's not true for some of these other problems. medicare is going to be incredibly hard to address, much, much harder. so in my view a congress should take an easy problem, get a little warm up exercise so as to take on a really hard problem and there is no way it's going to get around taking on both

over the next ten years in my view. so the bottom line is really simple. i think we should fix this now. i genuinely believe we should have fixed it before now, but my life is one of repeated disappointment and if you do policy work in d.c. that's what your life is, sorry. but i just would applaud the academy for holding this every year and i want to thank the trustees and the actuary and staff for what they do to illuminate the problem. now the issue is to fix it. >> let me just step in for a second. put in a plug for the national academy of social insurance who recognized steve goss with its first bob ball memorial award, now, what, 12 or 15 years ago. we recognized that steve has been an absolutely invaluable resource on both sides of the aisle without fear or favor,

whoever has come has gotten straight answers and good information and has done as much as or more than anybody else to educate the american public on social security. so thank you. >> henry, that's so nice of you to say, but i would just add to that the academy has done an even better job since that year. every year they come up with better and better people for the ball award. and i think there's at least one such person here on this panel. >> thank you, henry. >> my job here today is to talk about the role of social security in people's lives. social security is a retirement benefit that is critical to the economic security of the elderly, but it is also an insurance program that is needed at every stage of life. social security retirement benefits reduce poverty for the elderly from about 41% to about 9%. however, the program also provides insurance protection for young workers and their

families. if a young worker dies leaving behind a spouse and two young children, they would receive survivors benefits valued at about $700,000. if a young worker with a spouse and two young children became disabled, the family would receive about $725,000 in disability and retirement benefits. many, if not most families, would not be able to afford this kind of protection from disability insurance or life insurance in the private market. social security is critical for children, too. over 3 million children under the age of 18 receive benefits as dependents of workers who died, became disabled or retired. this is greater than the number of children who received temporary assistance for needy families in 2015.

social security is responsible for lifting over a million children out of poverty. for women social security has been a valuable source of income as a spouse, divorced spouse or widow of a worker and for women who have been in the labor market for a long time and for those who are entering the labor market in increasing numbers the program is vital to their own work histories. women are more likely to spend time out of the labor market, usually caring for family members. social security benefit -- the social security benefit formula helps to compensate for this by first counting the 35 highest earning years rather than typically 40 years, thereby eliminating five years of zero or lower earnings from the benefit calculation. and second, because women

continue to face a gender wage gap, they are helped by the benefit formula which is progressive. so lower wage earners receive a higher -- receive a benefit that represents a higher -- higher share of their pre retirement earnings than higher wage workers receive. this benefit formula also helps to compensate for labor market disparities faced by people of color. research documents that racial discrimination still plays a role in determining who gets hired, compared to white workers, workers of color are more likely to be unemployed, remain unemployed longer and to work part-time even though they want full-time positions. the benefit formula which eliminates five zero or low earning years helps to

compensate for years of unemployment and part-time work. researchers have also documented that occupational segregation that results in people of color being disproportionately represented in occupations with lower wages and access to benefits such as employer sponsored retirement savings plans, even after controlling for education. social security is the only retirement benefit that many workers of color have, and the progressive benefit formula helps to compensate for lower wages received. social security disability insurance is particularly important for workers of color because they are disproportionately represented in sectors with the highest rates of illness and injury. and survivors benefits are important to african-american families because they have a lower life expectancy. while social security already plays an important role in

reducing poverty and in improving economic security, the benefits are modest. the average benefit in 2016 was $1,250 per month. although the benefit formula helps to reduce the disparity in benefits across workers, the actual dollar amount received by low wage workers can leave them and their families in poverty even after a long work life. so benefit improvements are needed. there is a need for an effective minimum benefit that provides a floor at at least the poverty level. there are several proposals including the basic minimum benefit that was offered by the bipartisan policy center's commission on retirement security and personal savings. there is also a need to improve

survivors benefits. currently the surviving -- the surviving spouse receives 100% of the deceased worker's benefit or 100% of their own benefit, whichever is higher. this can result in a substantial reduction for the surviving spouse in their income. one proposal is to provide the survivor spouse with 75% of what the couple would have received and cap this so that the increase is focused on lower income survivors. a third proposal is to establish is caregiver credit to compensate for time out of the labor market to care for family members. for example, workers could receive a credit of the median wage for up to five years during which they were caring for a family member. and a fourth proposal is to reestablish the benefits for

students up to age 22 if they have a deceased, disabled or retired parent and they are pursuing post secondary education. this benefit was terminated in 1981 as a cost cutting measure and under current law unmarried children can receive benefits through age 18 or 19 if they are students in elementary or secondary school. so benefit adequacy is necessary, but so is restoring the solvency of the program and to restore solvency and pay for benefit improvements the following proposals have been offered: one is to gradually raise the maximum amount of wages subjected to the social security payroll tax. currently only $128,000 -- $128,400 of annual income is

taxed. this is regressive since lower wage workers pay tax on all of their wages while higher wage workers pay tax on a fraction of their wages. historically the maximum tax has covered about 90% of maximum wages but this has designed to 83% today. raising the maximum taxable wage to capture about 90% of national wages again would substantially increase the income to the trust fund and would not effect lower wage workers. second proposal is to cap the spouse benefit. currently the spouse may receive benefits based on his or her own work history or 50% of their spouse's work history, whichever is higher. >> but given the growing share of women in the workforce

earning benefits through their own work history a reform that has been proposed for years is to cap the spouse benefit. a final proposal is to gradually increase the payroll tax for all workers and employers. the national academy of social insurance estimated that an increase from 6.2 to 7.2 over a period and that would be nor workers and for employers each, over a period of 20 years would require someone earning $50,000 a year to contribute about 50 cents more per week. so social security is essential at all ages, but benefits are modest and leave some families in poverty even after a lifetime of work. proposed changes can be made to both increase economic security

and the equity of the program as well as restoring program solven solvency. >> thank you. thank you, everyone. i would like to open it up for questions but before so doing i would like to take moderator's privilege and ask a question myself to our panel. 2018 is the first year we have had to tap reserves. we are projected to have to tap reserves towards the end of the year to pay scheduled benefits. what is the significance of that? perhaps we start with one of the actuaries and move on to our other panelists. >> i guess one little note on that. anytime we expend any money on any benefits or paying karen and my salaries that results in tapping into the trust fund reserves. we have to -- we have to redeem reserves -- redeem investments at any time. the real question is we are moving in 2018 because almost exclusively of the drop in the

tax revenues coming in which were -- indicated a little bit of a surprise here based on these things towards the idea that we are going to have to redeem a little bit more in reserves than we will be investing. so the total nominal dollar amount of our reserves will begin dropping partway through 2018 as opposed to 2022. that's significant. i guess from a budget scoring perspective that can be argued to have some particular significance, but from the point of view of the ability to pay the social security benefits, assuming that as always our investments are always invested as required by law, interest bearing securities backed by the full faith and credit of the u.s. government. those will be available so from that point of view the ability to pay benefits really does depend upon the reserve depletion date. it doesn't really have any significance in that regard but certainly it does in other regards which i'm sure henry or doug might want to opine on. >> i will just say for me i

don't think there is a lot of significance. there's operational significance, but i look at the outlook for the program today, the same way i looked at it a year ago and, you know, the problem was understood to be there and it hasn't changed. >> what steve said, what doug said. >> okay. thank you. >> one more question, if i may, about the disability program. a lot is going on in the disability program in terms of its finances. one of the -- you said that the applications and claimants to incidence rates were going down but another item in the report was that the initial benefits are lower and i understood that, henry, you asked this question on my behalf earlier today. >> well, since they answered it for me i will let them answer it for you. >> okay. thai. so why are initial benefits down and does it have to do with the backlog? >> so it's really interesting. it's more about the mix of the new awarded benefits.

let's see. the see if i've got this straight. so while the backlog was growing, there were fewer awards at the alj level, the administrative law judge level, more at the initial determination level. one thing we discovered this year is that awards at the alj level are actually about, what is it, 10% to 12% lower than those at the initial level. this was news to us. something we just discovered as we were digging into it. so while the backlog was growing, there were more of those, so benefits were actually elevated a little bit at that time. now as the backlog is finally starting to go down, we're seeing sort of a more normal level of benefits and they've come down because we've got 10% to 12% difference, a changing case mix, average benefits are

down about a half a percent. is that right? >> the key point here is that it doesn't have anything to do with what alj's are doing. >> that's right. >> it has to do with the characteristics of the workers who are receiving final decisions from aljs, on average they have lower earnings and hence get lower benefits. but it's nothing that the aljs are doing. >> no cause for concern. it's really no news. >> and we do have some speculation as to why that's the case, if you want to -- >> yeah, it is really interesting that people who get awarded a little bit after their initial application at the administrative law judge hearing tend to have lower -- as henry loaded -- on average their earnings lists reese are weaker. we suspect part of that might be because people who are awarded very quickly, people who have very obvious disabilities in many cases maybe disproportionately and we need to check this may

disproportionately be cases where they have had sort of a sudden elevation of their impairment or had an accident, therefore, they might have had a very, very good earnings history right up to the point where they became disabled and have a higher benefit level. people who get awarded at the administrative law judge level are more likely to have a more complex case, more likely to get into terms of vocational considerations. these are people who have probably had an impairment that has gradually been getting worse over time and might have compromised their earnings ability some years before becoming disabled. >> questions from the audience. tj. please identify yourself. >> i'm tj shut cliff at the arc of the united states. thank you so much for today's panels. for steve goss and karen glen i was curious if you've taken a look at what's driving the lower di incident rate as well as the lower application levels that we've been seeing over the last few years, if there are any variables that you've ruled out and if you have any hypothesis for what is driving this.

thank you. >> so, yes, we have looked quite a bit at it. it is still sort of a mystery to us. we've looked at things like is there geographic variation, is there a rural urban divide, are there differences by age groups. the answer is really no. it's down all across the board. so we don't know. if you guys have any great ideas, please let us know. >> we know much of the drop since the peak in 2010 which is just post the start of the recession, the drop has been steady from then. we were expecting some drop and we got some drop. but the degree of the decline, our applications were a little over 2 million in 2010, they are between 1.4 and 1.5 million now well which is below the level we had in 2007 at the peak of the last economic cycle. especially given the nature of our population age distribution now we would expect quite a bit

more disability applications than we're getting. so as karen said, a bit of a mystery. we're looking into it, but there's nothing that has really leapt out. is there something just sort of in general in our society? where people are viewing themselves as more able? that would be wonderful. but, again, as karen said, cards and letters we are always looking forward to to any thoughts you might be able to offer us to help us understand this. >> other questions? sn>> so if we wait until 2034 - >> sir, could you identify yourself, please. >> it's not important because my position has nothing to do with anything, i'm not with any institution. i work as a physician in the va. if in 2034 we go only with tax increase do i understand correctly it's 2% on the individual, 2% on the employer? that fixes it? >> well, that would really be up

to the congress to decide what they want to do. >> what is needed. >> what is needed. >> right now 12% combined contribution and you say by one-third does that mean 4% more solvent? >> right. well, we would need -- we would suggest a one-third increase in revenue, which over that remainder of the 75-year period would be about 4% of payroll. they could be a little bit smaller than that initially in 2034 and have it rise up because our deficits do -- >> it's in the ballpark. >> it's in the ballpark. >> 2% for employee, 2% for employer. >> that would be a solution. >> the second question, i understand that benefits are adjusted for inflation and indirectly the collection is also adjusted for inflation, but sometimes there is a provision that can be adjusted by weight inflation which is if it is lower than inflation.

is that correct? and how would that affect this? >> okay. what you are a 're saying, we h provision here, right, where if our trust fund level gets to be less -- i believe it's less than 20% of annual outgo, there is a provision in the law that we may have to adjust our cost of living adjustment, which as you said is based on the cpiw for urban wage earners and clerical workers now. if we go below 20% trust fund ratio in a year and the average wage growth is less than the increase in prices then we use the lesser of the two. otherwise we always use the cpi. >> so would that affect materially -- would such a choice affect this materially? >> well, we do not expect to be living in the range of between 0 and 20% trust fund ratio very

much at the time. historically all of the policymakers and policy wonks like all of us have said that for this program even though we are not an advanced funded system and because we can't borrow at least under current law, we have to keep this reserve level at at least one year's worth of outgo. that's the target to be at least there. we are at almost three years worth of outgo in the trust fund now. so to get and stay below 20%, between 20% and 0% for any period of time would be tricky. part of the reason for that is because recessions still happen and we have estimated over the years a normal recession, not a great but, but a normal rescission must result in dropping the trust fund ratio by 20%. the trustees and congress has targeted to keep our trust fund ratio well above the 20. >> if we can go back to the first of those questions. i'm terrified of the idea of waiting until 2034, i think that's a very bad idea.

someone asked me one time if we wait what does it take to hold current retirees harmless. i did a back of the envelope calculation where you have to immediately raise taxes enough to have no shortfall at all and i'm going to pass my homework over to the trustees to get the math checked, but i got about 6 percentage points for the payroll tax increase. 4 seems optimistic to me. >> he said one-third. >> i heard that, too. when i did the math i got a far more dramatic increase. i think the spirit of it is neither 4 or 6 might be right and honestly if you are a betting person bet on their number. that's a big increase in payroll taxes overnight because we waited. i don't think we want to do that. >> there is another perspective, however, which is also an argument for not waiting and i may have my arithmetic wrong and i would invoke steve or you,

doug. i think if one would reverse the recent tax cut one would raise revenues over the 75-year period. when i say reverse it i'm assuming the individual cuts are permanent not phased out. if one reversed that one would go most of the way if not all of the way in terms of financial flows to closing the long-term gap in social security. >> and, again -- >> it just won't happen. as long as you have the current electorate and the current structure of people. >> we don't know what will happen until candidates for the presidency identify the issues that they want to bring before the american public and to which they want to devote their administration. i think it was a long shot, it turned out to be a very dangerous one, but a long shot that got us major healthcare reform in the obama administration.

i for one didn't believe it would happen and i was surprised. i think a lot of us will be surprised if a farsighted candidate skillfully articulates the importance of social security to the nation's future, the modesty of the steps that could be taken as i think doug and i both agreed in the near term if we act reasonably quickly to deal with it, and we could get action. whether it would -- i'm not suggesting it would involve reversing this particular tax cut. what i am suggesting is that it is a measure of the magnitude of the steps that would suffice. we had an easy enough time cutting the taxes. thinking in those metrics rather than a major chop out of earnings in 2034 is a more

constructive way to think about it. >> if i could, just to clarify the arithmetic on this thought experiment of if we were to just simply raise the payroll tax rate, in 2035 we indicate in the report that our shortfall is 3.4% of taxable payroll which means at that time in that year -- pardon? >> okay. go ahead. >> at that time in that year 3.4% of payroll additional would be what we would need. by the time we get out to 2095 it's 4.45% of payroll. so that would mean the 12.4 which is half and half for employers and employees would have to be raised by about 4% through that period, but it could be as little as 1.7% each for employers and employees in 2034, rising up to a little bit over 2.2% for employers and employees each by the time we get to 2095. not that anybody is advocating that as a solution. >> and the other thing just in terms of understanding how different people think about it and how i often don't sound the

same as others, looking over 75 years you can do things relatively gradually and get the system itself into balance. my concern is that the rest of the federal government has melted down in year 30 and we haven't taken care of an extraordinarily large budget problem that exists outside of social security. so i worry about those things which counsel far too much patience. in the big picture we don't have that kind of room. >> let me remind you of the chart that steve showed us which is that the magnitude of the deficit gap outside social security is smaller than that suggested by the cbo projections of debt since those projections include the counterfactual or counterlegal assumption that all scheduled social security benefits will be paid by borrowing, the gap between that and revenues. >> ken? >> ken buffon actuary at buffon

foundation. i have a question for steve goss, it's more related to methodology than to the results of the latest trustees report. would you please explain exactly how the trust fund assets are valued in preparing the financial projections. are the securities taken at their principal value? and if so, what is the effective discount rate as applied to the perspective cash flows from interest income and maturity proceeds? and then if that's so, would a different result be obtained by values directly the expected cash flows from interest income and maturity proceeds at appropriate discounted rate? and then are the results of the trustees report for the value of the trust fund cash flows affected in any way by the prevailing interest rate environment? as for example whether the yield on ten-year treasuries is 2% or 4%.

>> if doug or orthello would like to answer that, yoert otherwise we will skr steve. >> technical question, this is an actuarial thing but it's important. first of all, one characteristic of the investments, i just mentioned earlier that the law requires any of the investments which is every dollar that comes into the trust fund is immediately invested in interest wearing securities. that does cover a fair amount of territory. there are marketable bonds and other things that could be invested in, if the trust funds invested in marketable treasuries then the fluctuations in market value would become important, however, for quite a few years now all of the investments have been made in what we call special issue bonds to the trust funds and they have interesting characteristics. there is no arbitrage because the interest rate, coupon rate assigned to any investment made by the trust funds is based on what is the actual effective market yield on all outstanding

marketable treasuries in the prior month with a remaining time to call of maturity of at least four years. so it's straight up, the initial interest rate that's assigned to any newly issued bond is right at market. however, when we have those bonds, if we hold them to maturity and then -- and then we redeem the bond at maturity, we will have every june 30 clipped the coupon, gotten the interest rate and that will be put back into the trust fund and at the point of maturity then we will redeem the par value of the bond at that point exactly as a marketable bond would be. where there is a difference is if we are required to redeem a bond before it reaches its maturity, the market value is exactly the par value. there is no market value, per se, because it's a nonmarketable security, we are just dealing with treasury on that. that's a special feature. now, is that a plus or a minus? it depends on what the interest rate environment s since the

time that bond has issued interest rates have gone up a lot, that would tend to make the market value lower. we don't suffer that lowering of market value. if interest rates have gone down a lot since it was issued then there is no change, it's always a par value. in terms of the discount rate that we use for doing our valuation and our projections, what we do there is we -- for that we look at the portfolio yield of all of the bonds that are being held in any given point in time which is the sort of weighted average of all the interest rates on all the bonds being held at that time and we use that as the annual rate of discount for our projections. now, that means then that raises the question how about if we reach reserve depletion and we reach a point where we don't have anything invested, with he simply use what we estimated at that point to be the new issue rate for the rate of discounting, which of course is exactly equal to what we in the trustees anticipate will be the effective market yield on four

year longer duration bonds in the marketable environment. >> okay. >> let me just add one note. buried in what steve said is an important fact. the assets that social security holds are better than the ones that are sold and are the best asset in the world to the public, and the reason is they carry a put at par. no other bond has that and the financial markets would price that and charge a premium. >> bill. >> my question, this is to henry, doug and kilolo this is the latest in a series of trusty reports where three trusty positions are vacant, the social security commissioner and the public trustees. so to you, is there symbolic importance to these vacancies and when these vacancies had people in these positions did they have a significant substantive effect on the report

itself? >> it's only two positions are vacant. we do have an acting commissioner who works as a full-fledged member of the board and have for these past several years. >> i think that it's important to have the positions filled in terms of just the strength of the work that's put forth. when you have all the positions filled and the report is coming out with -- it's been approved by all, i think it just makes the report stronger. in answer to your second question, i'm not aware and i've been working on this issue since the mid 90s -- i'm not aware of there being a difference in what the report is going to say based on those positions being filled

or not. >> i think the vacancies are symbolic of the larger breakdown in the capacity of the parties to get things done. and i think they are a really sad reminder of that. certainly i believe they should be filled. i think the public trustees have a very important role. steve is going to do the work, his team is going to do the work, but how you think about the meaning of the results, what it says about where the social security system fits in the larger policy objectives, you will get a very different set of opinions between an administration and their cabinet officials and the public trustees. i think their voice should be hear heard. >> i'm allen cohen with the center for american progress. i have a question for steve and karen. is it the case that there will be a point five or six years from now or four years where the riyal case of the disability trust fund, the moneys that were

going into the disability trust fund, would be reallocated back to the oasi fund, oasi fund? and if that's the case what effect does it have on the estimates that you have put forward here? >> so not in current law. in current law the riyal case was set for three years. 2016, '17, '18. it will go back to the way it was in 2019. of course, congress could always decide to reallocate again if di experience is as favorable as it has been recently, there is a possibility the depletion date to expand out past oasi and we could reallocate the other way. we don't know at this point. >> chris? >> i'm chris flynn. i have a 50-year association with the private sector and defined benefit, defined contribution.

i was puzzled in that if we were represented with this actuarial report in the private sector the first thing we would be doing is looking at the rate of return on the assets and yet that is the one solution that you dismissed out of hand. it's particularly curious in light of ken's question and your answers which implies that the fair market value of the bonds held in the trust fund, which there is no fair market value since you can't sell them, but if you could sell them henry implies that the fair market value is actually higher than the book value because of the special provision. so please tell me why you are dismissing -- >> okay. i wouldn't say i'm dismissing it. i would say that the value of our assets as sort of a percent of the annual cost is way less

than any private defined benefit plan. defined benefit plans are generally intended to be fully funded. social security is not. it's more of a pay as you go system. we do hold some reserves, right now it's at about three times annual cost. in a private plan that's more like 20, 25. so just relatively speaking, the assets themselves are not as important as they would be in a private plan. >> so is that reason to leave money on the table? >> absolutely not. absolutely not. i think traditionally congress and policymakers have believed that it's more important that the assets be safe and protected by the full faith and credit of the u.s. government rather than investing them in the market. but that is certainly an option available. you can see on our website we score several provisions for putting certain percentages of

assets in the market. it's a possibility. >> we have time for one last question. yes. >> katie smith and this question is for steve and karen. i want you both to elaborate more on why it's beneficial to project for 75 years. a lot of people talk about you could do a shorter window and with things being so uncertain with the economy and of course we could pass immigration reform, i think we might get pay leave and workforce participation but just the benefits of a longer window, 75 years. >> well, 75 years is just one of many, many durations over which we look at the prospects for the financing. in addition to the 75 years we also look at a 50 year and a 25-year period and even a ten year short-term period. pardon this, but we also look over the infinite horizon period. the 75 years is traditionally for a long time now been considered to be the basic period over which we look for a couple of reasons, i would

suggest. first of all, let's see, pretty much everybody who is a current participant in social security is likely not to live a lot more than 75 more years. so all of our -- >> what? >> well, it's likely, except for some of the very young women because they are going to live beyond us anyway, but by and large the remaining lifetime of most of our participants, even some of our youngest workers, are likely going to be covered in the next 75 years. so it's important to be able to show the prospects. the other is, again, because the hallowed halls that we are in here, when people in the congress are considering making changes and by the way we have a whole bunch of provisions we showed you and proposals identified, when people are considered -- especially -- are considering changes because of the nature of this program and because of most of the kinds of proposals that come forth, there are not immediate massive changes, oftentimes they are changes that might not start for

a while and when they do start they phase in gradually. so 75-year period is really necessary to show the full mature ramifications of a proposal. one example is the 1983 amendments which included in it an increase in the normal retirement age, which didn't even start to effect until 17 years later and it's full effect won't be completed until 2022 for people reaching 62 in that year. so the 75-year period has a couple pretty good reasons, but we do have the numbers in the report to look at shorter periods. >> okay. thank you, everyone, for coming. if you go to our website nesi.org you can find a video of this event within a few hours of now. thank you. in just a moment we will take