learned stories, but there are points about tierney, sun ship and daughter ship, that nature and nurture and politics, even about democracy. a" --ight on "q and children of monsters, which looks at the lives of the children of 20 dictators, including stalin, mussolini, mao zedong, and saddam hussein. somewas able to talk to knowledgeable people. i could not talk to family members, which is usually the case in the preparation for this book. there are only so many around and so many willing to say what they know or develop their feelings or experiences at all. i was digging around for any scrap i possibly could because these sons and daughters, most of them -- some of them are famous and of art and. some of them become dictators. but most of them are footnotes and asides and you have to dig

to find out about them. at 8:00 on c-span's "q and a." >> now a look at increasing retirement savings. the form examined social security, individual retirement accounts, and refundable tax credits. this is about one hour. mr. weller: good morning and welcome to the center for american progress. i'm christian weller. i'm a senior fellow here. carmel martin was supposed to do the opening remarks but she is sick. with each passing generation, america's retirement crisis is growing. in 1983, almost a third of working households would learn to live on much less during retirement.

by 2013 come of that number grew to more than half. this trend is getting worse, not better. young families are having even more trouble than previous generations to save for retirement. while this problem is for all middle income families, there are some families suffering more than others from the retirement crisis and the expected shortfall in spending. low income households are more likely to experience short retirement. this is true for low income households who don't work for an employer that offers retirement benefits. what can we do to tackle this problem? governments already offer substantial tax benefits to help people save for retirement. it is to incentivize people to save for retirement. income tax encourages workers in their states to save for retirement. even though the federal and state governments are spending a lot of money for growing tax revenue to help people save for retirement, we face a growing retirement crisis. part of the problem is the

dichotomy of more government spending and less retirement savings is maybe not enough tax incentives. they are often inefficient. high income earners get the most help from the existing incentives, multiple times more as we have posted on our website. we have to do more to help families tackle the retirement crisis and think about ways of how we can reform the tax code and improve incentives to help those who need the most help. possible solutions include expanding refundable credits, designing incentives and offering more ways to let people save for retirement outside of the employee-employer relationship. we have our panel that will tell

us what the data says and what is feasible and how we talk about the solutions. the first panel will be moderated by professor teresa ghilarducci at the new school university in new york who will leave the engaging discussion on the current crisis. the second panel will be moderated by the director of policy for the poverty and prosperity program to allow households to save. we will be joined by the chief economist of the treasury department. she will offer keynote comments. let me welcome the first panel moderated by teresa. thank you. [applause]

ms. ghilarducci: really wonderful to have you three here with me. let me start with keying off what christian set is we have a retirement crisis. many people have focused on this idea that government may not be able to pay for the elderly who will be of retirement age. but, a bigger problem, another way to think about this crisis is these folks will not have enough money when they get to retirement age. so, in your remarks, i would like you to address what are the facts to help us all understand how big the crisis might be? the second issue is off what christian said is we spend a lot of money in the federal tax code and also in the state level that

is invisible, indirect. it is in the form of taxa competitors, money we don't collect because people get a deferral when they contribute to their qualified tax retirement account and the contributions. this indirect subsidy -- how large is it? and what kinds of reforms? not what might happen, but what policy would best help the people we just talked about to get more savings adequacy? lily, do you want to start? ms. batchelder: thank you so much for having me. this is a wonderful organization. i'm glad to be here today. as both christian and you

mentioned, we are facing an ongoing retirement crisis, retirement security and people having enough to spend that they can maintain their levels relative to prior retirement. particularly, low income people can have a very basic level of income. this is a problem concentrated on lower and middle income households, households of color. part of the solution is to increase the minimum benefit or social security which is now below the risk of poverty line. another huge piece of the puzzle is the focus today which is tax incentives for retirement. it costs about $2 trillion over 10 years. teresa's work suggest that it is 20% more if you include the state level because the state piggybacks on the federal level. i wanted to focus on three issues related to this. the first is the fairness of the current incentives in terms of people who are participating in

employer plans. the second is how to gain access. the last point is to improve the effectiveness of the plan themselves to improve retirement security. on the first subject, the author of the research shows our evidence is that incentives do not increase how much people save out of their own money. they do increase how much savings they have for retirement which may sound a little paradoxical. one way to think about it is they take him $50,000 of income. they would say $5,000 without any saving incentives. you give me $1000 saving incentives. what the evidence is suggesting is that $1000 will go to increasing my 401(k) to some degree, probably most or all of it. i will not actually increase how much i'm saving.

i'll and up with $45,000 of consumption and $6,000 in my 401k rather than $5,000, but not changing my consumption. this is a really important fact. it means these incentives are not spurring people to save more. it is like the government depositing into your 401(k) and possibly a bit of it into your checking account as well. i think we need to focus a lot more on the fairness of the incentives and whose savings we should be supplementing the most. my answer would be lower to middle income households rather than the wealthy. in fact, the current retirement incentive is very disproportionate to the wealthiest households. wealthiest households. so cbo and jct again have estimated about two-thirds go to the top quintile, two-thirds of the value of these incentives whereas 7% go to the bottom 40%. and if you look at just the top 1%, they get 50% more of the benefits than the whole middle

quintile. >> wow. >> so one way and probably the main way to address this unfairness issue would be to shift more towards refundable tax credits. >> yes. >> and you could completely replace the current system of deferral and deductions and taxing the money when it comes out -- often many years later -- or taxing the money when it goes in and then not taxing the earnings at all. bill has proposed totally replacing the current system. you could also shift more in this direction by cutting back on the tax incentives for the wealthy, whether that's through reducing the contribution limits, something like the president's proposal to limit the value of lots of defunctions and exclusions including those for retirement savings to 28%, a cap on the amount of tax-preferred savings that you can have as an aggregate balance and then using some of that to make the savers credit

refundable, potentially expand it. and you might even want to use some of that revenue for deficit reduction or other priorities given that $2 trillion is really an awful lot of money. an additional benefit i just want to mention of a refundable credit is that it would enable you to structure the tax incentive as a match that's directly deposited into the account, and there's some empirical evidence that this would increase the responsiveness due to framing effects. >> you explain that just a little bit? what's a framing effect? >> sure, sure. this is just the fact that let's say i get a refund equal to 25% of my savings. so i save a dollar, and i get 25 cents back. i'm really putting in 75 cents out of my own pocket to end up with a dollar in my account. but if instead at the match you can have the same result and have it as a 33% match, i put in

75% and then the government in this case would match 33% of my savings -- >> yeah. >> and most people don't do that math in their head, so 33% sounds bigger than 25%, and they respond more. and then also just the fact that it would go into the account in some reform ideas would eliminate some of the friction. >> uh-huh. >> so the second thing i wanted to talk about was expanding access, and right now about a third of people don't have access at all to a retirement plan. about half don't participate. these numbers are much higher if you're low-wage, if you're part time, if you're working for a small business. so about 60% of the bottom quartile don't even have access. and the fact of the matter is people really don't save for retirement if they're not covered by an employer plan. very few people directly contribute to an ira. so even if you made the tax incentives more progressive, you'd be failing to cover or increase retirement savings

among a lot of people who are not covered or participating in employer plans. and i think that's a very important aspect of retirement security that we need to focus more on. one way to expand coverage would be through the automatic ira proposals that some folks in the room were pioneers of. i see david johns out here. the president has proposed -- his proposal would require businesses to have more than ten employees to offer a payroll deduction ira and would provide tax credits if they do so including larger ones if they are auto-enrolled or creating a qualified plan. while this idea was originally a bipartisan idea, it has really been stymied by the fact that it has included a mandate and that brings up images of health care reform in the minds of a lot of republicans on the hill. so another approach is john's proposal which is substantially to increase the credits for employers offering plans but not

necessarily apply mandates or penalties. and in the meantime, i think another really important effort is to get states to be pioneers of this. so if we're not going to see these legislative proposals enacted in the very near future, there are states that are wanting to create auto-ira plans. and the department of labor's actually working on guidance clarifying that they are allowed to under erisa. and then the final thing i want to touch on is plan design. this can have a tremendousfect on coverage -- effect on coverage. a lot of people know that if you have auto-enrollment, auto-escalation plans, you see a hot more participation, a lot more savings x. people's behavior, as i think john will explain, is a lot more influenced by how easy it is to save and the structure of savings than it is by these savings incentives. we've seen a big shift towards auto-enrollment since the pension protection act, but there's more that we can do to spur that.

and even more ambitiously, i think we could think about calibrating some of our incentives or trying to get employers to calibrate their defaults towards adjusting the auto-enrollment defaults based on income. so given social security is on net progressive and lower-wage workers get a higher share of their income replaced, in an ideal world you actually would probably have different savings rates among different income groups. and you could go in that direction. and then one other thing i think we need to be thinking about long term is how the savings is invested. so, you know, one important development has been due in part to the increase in auto-enrollment. we've seen a lot more investments in target date funds which automatically adjust your portfolio over time to be less risky as you approach retirement. i think it would be great if we could begin to think about defaulting people, and this would potentially require

legislation or guidance into products that gradually annuitize and/or purchase long-term care insurance over time. because if a goal is to insure a certain level of retirement income and address health care shocks in retirement, you actually need less money saved to do that if you are partially annuitized or purchase long-term care insurance. and then the final important initiative that i have to mention with respect to how funds are invested is the department of labor's conflict of interest rule. >> yes. >> we've been working on it. and this is addressing the very large problem of people are losing about $17 billion per year due to receiving advice from their financial advisers that is not in their best interest because of conflicts of interest that these financial advisers are receiving. and the proposal which is, still needs to be finalized would require financial advisers who receive these conflicted payments to provide advice in

the best interests of their clients. so i'll sop -- stop there. i've taken up too much time. >> no, no. you talked about lack of coverage of most private sector workers. you've given us some ideas about how to tweak the design of people who have it. you hope, you're hoping that auto-iras cover the people that don't have an employer with tweak bl designs, and you talked about the big fault lines in the way people invest now. most of the time asking experts for advice is the one big fault line. and i'm interested later on after we hear from john's framing about what about the odd design we have in this country which is that we allow lump sum withdrawals? we're seeing already this week what's happening in the u.k. now that they've just given the sort of pension freedom to take out lump sums. it's not good. you know, whether or not you think, you know, you're not in politics anymore, so whether or not you think we should stop

lump sum withdrawals or withdrawals at all before retirement. but, so just keep -- i'd love to hear your voice on this. >> okay. [laughter] >> john, how would you frame the problem as not being a people problem, not sort of a budget problem, and then how would you frame and have people, you know, point the way forward to fixing it with the money we already have on the table? what did you say, 120 -- >> two trillion over ten. >> and you add another 20% from the state. >> yep, yep. >> and so every year that's 100 and -- 200. >> over 200 billion. >> a little less the first year, and a little more the tenth year. >> if you show the gun in the first act, it's going to be shot in the second. so the gun is $200 billion a year, all right? can we use that better? >> thank you. first of all, thank you so much to the organizers and to teresa for moderating this panel. it's a great pleasure to be

here. so i think that you phrased it, is this a budgetary problem or a people problem -- >> yeah. >> and i think the easier way to see this as a deeper people problem is i think if you substantially expanded the budget that we were putting forth on retirement incentives the way they are organized today, you still wouldn't really go very far towards fixing the problem. and i think that that's because these tax incentives -- and really the research suggests most forms of incentives whether they're company matches or the savers credit or the tax deduction, they just aren't that effective at increasing savings. and that's for three reasons. so, first, and i think this will not come as a great surprise to any of you who have talked to people that are not in this room, most americans do not pay a great deal of attention to the fine details of savings policy. [laughter] for instance, if i think you've doubled or have the existing tax incentive for retirement savings, i just don't think most people would notice. i think they would literally do

nothing. now, of course, there will be some people who respond, and so that leads me to the second problem with tax incentives which is that it does not appear that they actually increase the amount that people are saving. so lily mentioned this, so just to give you a sense of why this might be true, you know, say your firm comes out, and they're going to give you a match rate, and you say this is fantastic, i'm going to put some more money in the retirement account, that's only the first step, right? you have to not just deposit the money in the retirement account to start with, but you have to actually spend less on whatever else you were going to do that year, and that turns out to be much, much harder. so what we see is that people do put a lot of money in retirement accounts when they're offered these incentives, but essentially if each dollar you put in this -- for each dollar you put in this retirement account, you either save less in some other form, or your credit carding debt goes up, and it just gets offset in a way that might even make the problem worse.

for instance, you're borrowing on a credit card that's got a much higher interest rate than your savings account will return. now, as lily said, this essentially reframes the government tax incentive more of a lump sum deposit that the government is making in alternative people's savings accounts, and i think that you want to then evaluate that on the basis of fairness, as lily said, but also on the basis of efficiency. what is it that we are trying to do by augmenting people's savings. and i think in addition to distributional concerns, the motivation for government support of retirement savings is that we think that some people on their own will not be able to save enough for retirement. and so what i think that clearly points to is that weç shouldñrç directing these subsidies which are effectively lump sums towards those people who are least prepared or least able to save for retirement. and by structuring it as a match, we're doing the opposite. first, we're giving the largest

lump sum to people who are already saving the most, and second, e people who are already saving, even if they're savingless, the people who are saving -- they're understanding the problem, they're thinking aboutñ]oki] retirement in a wayt the people who have the deepest problems are the ones who aren't thinking about the problem at all. and they haven't started saving not because they thought about it really hard and decided it was not best for them to save, these are people who just haven't thought about the problem at all. those are the people we need to help the most. how do we help those people? as lily mentioned, there's increasing evidence that defaults or nudges or you can call them what you want, but these other forms of increasing -- encouraging participation are much, much more powerful. and for each of the three ways in which tax incentives are ineffective, a default or an auto- escalation is effective. so first, just take a default, for instance, it exactly works on the people that aren't paying attention x. so now you're

primarily going to affect the 85% of people who aren't paying attention as opposed to the 15% of people who are. second, it turns out that defaults when they increase contributions to retirement accounts, they actually do increase true savings. so it's, there are a couple of different reasons why that might be the case, but in contrast to the tax incentives which i think is the increase in savings and response to that tends to be a rationed, well-thought-out thing, when people get defaulted into contributing more into their retirement account, and sometimes money is just kiss appearing from their -- disappearing from their paycheck. they end up spending less, which is what exactly what you need te more. and third, they're targeted to the people who often have the least savings and who need the most help in accumulating a nest egg for retirement. and in this way i actually think that defaults, you know, you mentioned the political problems with mandates, i think defaults

might actually be better than mandates because it is going to help target these, this government intervention in a way that's most effective. so, for instance, with a mandate there may be some people who really in some years they shouldn't be saving. maybe they're earning much less this year than they otherwise would, you know, lily just came out of government service, so she, you know, is taking a hit for the sake of the country, and maybe it wasn't worth it so much for her to save while she was earning at the government, but now that she's back at nyu and earning a princely sum for teaching there, it makes sense to sock more money away. that's a silly example, but there are people who go through bad times and good times, and forcing everyone to save the same amount no matter what's going on, i think, is not right. if you have a default, it encourages people to save, but if you really need the money, it's not forcing anyone. so i think that these types of nudges should be put at the forefront of policy efforts to try to help the retirement

savings crisis, and i think that even though the normal thing we think about with tax incentives coming from tax reform might not be as effective, tax policy can still be very effective in encouraging firms to take up these types of very effective policies or end counseling -- encouraging firms to give workers access to these retirement accounts at all. for instance, being able to have direct deposit of savings directly added to your paycheck instead of having to save up and write a check at the end of the year turns out to make an enormous difference in how easily people are able to save. and so i think just in the minute or two left, kind of two other things that, you know, while we're thinking about the tax system is trying to encourage better retirement savings behavior more generally, kind of two other, i think, really big deal things, we've talked about getting the money into the account, we only talked a little bit about how the money is invested in the account. there's enormous amounts of

evidence that people pay extraordinarily high fees relative to what they need to pay. it's been getting better over time. i think the conflict of interest rule is a good step further in that direction, but a lot more encouragement of people to invest in index funds, target date funds, things that are relatively commoditized and low-fee. and then the final thing is it turns out to be way too easy in the united states for people to take money out of their retirement accounts. you mentioned the lump sum withdrawals at retirement. i think the bigger problem is when they take out before retirement, right? because at least you've gotten the money to 65, you've gotten some part of the way there. the evidence suggests that in the u.s. for every dollar that's invested in a retirement savings account, nearly 50 cents is coming out in that same year. and some of that is coming out due to real-life hardship withdrawals that we don't want to shut down, but a lot of it's coming out in ways that i think are not great. so, for instance, people take

out loans from their 401(k), and then if you switch firms, you automatically have to repay the loan, in which case almost everybody defaults, and that money is just lost from the retirement system. for instance, the work of bridget shown really small changes could increase the amount of savings that we manage to get to the retirement age by 25 or 30%. and so you kind of get people contributing more up front, you get people maybe 25 or 30% from paying lower fees, another 25 or 30% from not taking out the money early, and i think you take some really large strides towards solving this problem. >> so i wanted to ask you this over coffee or in front of a hundred people. >> either way. [laughter] >> so we have $200 billion a year on the table or $2 trillion over ten years. add the states in, you have a little bit higher. you say that we are spending that money incorrectly because we qualify plans with bad designs, and you would want

designs with lots of tweaks, but you would not mandate that the money go in be and stay there. -- go in and stay there. and i've been intrigued by this idea that that would disrupt people like lily or somebody else who over their life cycle year to year have consumer needs and that the government would be forcing people to save too much, and they couldn't optimally structure their spending. this came up in the recession, and i was asked by a member of congress, can't people take money off the to -- out of their 401(k)s now because they need it in an emergency. and the response is, well, if you're going to call et a retirement -- it a retirement tax credit, that's what it's for. would you want people to be defaulted into social security and then use their fica money when they need it in those times they need the money the most? i mean, how far are you going to go with this default rather than

mandate idea? >> so i think that it's very important to have a base floor below which people cannot -- >> take out, okay. >> and i think social security -- >> okay. >> -- is effective at providing that. i think it could be more effective, but i think, you know, again, not to say that we could not improve it, but i think allowing, you know, we are trying to provide an efficient system to get people to save for retirement, but i think we are also trying to avoid a problem where we have elderly who are -- >> so just to follow up on that, would you mandate 2% on top of social security? because we know that social security isn't providing enough. >> so i would not mandate anything on top of social security. i think that's the role for this flexibility. i think people should be encouraged to save a lot on top of social security, but i think -- >> you would -- >> yeah. i think that you really need

flexibility because many time toes either people need money -- many times either people need money for current expenditures, if you have mandates, also you often end up -- even if it's just a small fraction of people. if you impose enormous hardship on 1% of the population because you haven't structured the mandate in the right way, that can offset almost everything else you're doing that's good with the rest of the system. so i think, again, the real advantage, you know, i gave one example, but the real advantage of having a system where it's kind of a strong hi-suggested -- strongly-suggested default but not one we're going to absolutely force you to take is that you get the main effect of a mandate for 90% of the people. it's just that for that last little bit that really need the flexibility, you're allowing it. >> it's a self-mandate. okay, thanks. thank you. bill, you've always helped me

envision that pile of money on the table, you know, by -- with your work on tax expenditure and calling out that we're really spending lots of money on retirement security. how would you spend it better? is this a people problem or a budget problem? >> no. i guess i would think of to it as both. but, first, let me say thanks to c. a.p. and to you for inviting me. on the one hand, it's very gratifying to hear stuff that i've been saying for 25 years now coming out as conventional wisdom from people as sharp as lily and john. on the other hand, i have to go third, and they've taken most of the sensible, thoughtful points. [laughter] so i'll try to, i'll try to weigh in on different aspects of this. the way i think about it is that, broadly speaking, there's four ways that government can influence retirement accumulation. the first one is mandates. and it goes without saying, but

i'm going to say it anyway, that social security is at core of the retirement system. it should be there, it should be a mandate for the reasons john mentioned, and that's sort of the building point from any additional discussion of retirement policy. the second way in addition to mandates is incentives, of course, tax incentives. tax expenditures, whatever you want to call them. and there's a wide variety of them with a somewhat checkered history as lily and john have described. the third category is information, education. the government can provide people with information; the social security statement that used to go around is an example of some of the stuff that cfpb is doing is trying to get information out to people on various aspects of saving. and the fourth approach is, which i'll refer to as nudges or defaults, what richard saylor

calls choice architecture which has shown to be a powerful influenceym on saving behavior. historically these four ways have developed asñr substitutes( auto-enrollment came along because firms didn't want toç y the cost of a matching incentive and soç on. and, you know, tax incentives came along because social security wasn't meantç to prove all of the retirement needs of people just to provide a base. so historically they've largely come as substitutes for each other, but i think we should think of them as complements in the policy world. and the example just to put an example in your head, let's suppose we want to increase saving, retirement saving by low and middle income households. well, what do you do? the fist thing you do is you default them into a 401(k), or if they don't have a 401(k), you

default them into an auto-ira. the second thing you do is you reform the savers credit so that it provides a refundable contribution to the account. and the third thing you do is you provide them with the information be or education so they know what they're doing with the funds. and you further design the account along the way to that contributions escalate, they're invested in diversified, low-cost funds. i'll talk about what happens when it comes out in terms of annuity in just a second. but it's the combination of the policies that would work, i think, much more effectively than one or other. so you get the person into the account with the nudge or the default, rather, you get them contributing more with automatic escalation. you make their saving more rewarding by providing a matching contribution, and you equip them to figure out what

they should do with it through education and information. so i think that's sort of a framework for where we should go with policy. that suggests, coming back to tax expenditures, that incentives can be part of the solution. and i want to emphasize both the "can be" and the "part." the can be is if they are designed well, if they are going to people for whom the contributions represent net saving, then they will raise national saving. if they're going to people for whom the contributions represent saving they would have done anyway or asset shifting or whatever, then, in essence, they're a waste of money. and one way to think about this touching on the earlier conversations is there's an easy way to take advantage of tax incentives and a hard way to take advantage of tax incentives. the hard way is to reduce your current standard of living in order to get the benefits of the

tax deduction. that represents lower consumption, as lily mentioned, and it represents higher saving. people for very obvious reasons choose the easy way if they can. the easy way is shifting yours assets or saving you would have done anyway and not having to reduce your standard of living. you know, people are very excited about tax subsidies for saving b, but nobody is excited about reducing their standard of living. and if you look at the ads that the brokerage industry offers people, they don't say, you know, cut your standard of living and take advantage of this tax incentive. they emphasize shifting opportunities. and for natural reasons. i mean, if you want people to participate, you want them to participate in the easy way. anyway, so the incentives can be part of the solution, but they can only be part of the solution. if everyone were a

perfectly-informed neoclassical consumer, maybe they could be the whole solution. if they were designed right. but given the imperfections that people have in terms of not being foresighted, not being able to implement plans even if they get plans, we need more than just incentives. incentives cannot be the whole solution. so incentives can be part of the solution, is what i would like to emphasize there. let me mention a couple of other things that i think are important. lily touched on the -- i think it was lily -- touched on the state-level activities. my colleague, david john, who's here has been doing a lot of work on that. i think that's a crucial avenue given the gridlock at the federal level. and as lily mentioned, the federal government can help

enable state governments to do more by clarifying the regulations. i want to talk about it, we talked about automatic enrollment. everybody knows about automatic escalation, automatic investment. and, teresa, you raised this issue about lump sums versus taking the money as an annuity. david and i went to a few other people, mark and lena, to come up with a proposal for automatic annuityization which is actually a really hard issue to think about. if you automatically enroll someone in a 401(k) and you get the contribution level wrong, it's no big deal. they just change the level. if you stick somebody in the wrong annuity, that's a mess because they're stuck there for the rest of their life, and annuity needs varian i substantially. -- vary substantially. some people have a lot of kids that they want to give inheritances to, some people have a lot of health needs, so they want to keep their money flexible for that.

some people might think they're going to live very long and so don't want to annuitize, etc. so annuity needs vary a lot. what david and i and others came up with was the idea of test driving annuities. the idea that people would automatically be enrolled, but it'd be on a temporary basis. they would get a two-year period when they wanted to start taking money out of their 401(k), and that would sort of get over the hump and sort of the big -- small issue with annuities which is, basically, try this at your thanksgiving table. tell your relatives, you give me $100,000, and i'll give you $500 a month. they'll look at you like you are nuts. why would i ever want to do that? but people don't see it that way. so the idea is by getting people to experience that rather than trying to anticipate it

beforehand, just have people try it for two years. if you don't like it, you can take your money out in a lump sum. so there are ways to think about automatically annuitizing people without forcing them immediately into a lifetime commitment. let's see, two other points. one is -- and john has written about this. i'm surprised he didn't mention this. we need to start thinking about ways of divorcing retirement saving from the employment system. employers, it's sort of an accident of history that the dc system ended up in, being employer-based. the db system you can understand why employers set up db structures to manage their work force, etc. a dc program really doesn't have those features, but because the db system was in firms, the dc

system came in firms. you can do all the matching, all the incentives you want in a 401(k) that is not, that would not have to be employer-based. i won't say more than that because john has written about that recently. but i think that's an important way to, important thing to think about especially the latest hot topic is the gig economy and workers who are not really -- workers, people who are functionally workers but for legal reasons are contractors. they don't actually have employers, they're all independent contractors. that's sort of the kind of the tip of the iceberg here. but more generally, we need to think about there's no reason why your retirement security should depend on whether your firm feels like offering a retirement plan. last point, we keep throwing

around this $200 billion a year number on tax expenditures. some people would argue the number's not really that big because it's -- >> [inaudible] >> deferred expenditure receipt, not a complete subsidy. regardless, the point i want to make is we should take 1% of that amount on an annual basis and use it to fund research to understand what works in saving and what wouldn't work. the amount that we could -- we have learned a lot in recent years, but we could learn a lot more at very low expense relative to the stakes involved. so besides the various policy initiatives aimed at people, we should also aim some of them at researchers who are also people but interested in the research

issues. >> and the consensus on the panel, the researchers -- that's a great idea. [laughter] before we go to questions and answers, can i just do a quick elevator speech? you've got less than 30 seconds. is a refundable tax credit a good idea to spend those hundreds of billions of dollars on the table? and how far does it move the needle? how far does it help soft the problem? -- solve the problem? lily, do you want to go first? >> yeah. i think it's a good way to shift the incentives in terms of improving the fairness of the system, and i think some of the credit should go not just to employees to supplement their savings, but to make sure that more, a lot more employers are offering plans. >> interesting. >> so that you really increase the participation rate above the roughly 50% right now. >> increase the participation rate of employers. that's really interesting. that's -- >> they're offering plans and then get people to participate in their plans because people don't generally save outside of

the employer context. >> right. we're in a short elevator ride. >> i think, you know, tax incentives, the key question is compared to what? if it's compared to the current system, i think it's a small step in the right direction. >> right. >> but i think there are many other better things that you can do with that money if you take a step back and -- >> so you would link the refundable tax credit to changes in design? >> yeah. i would use a lot of that money to rethink the design of -- >> i'll come back to what i said earlier. incentives can be part of the solution. this would be if you -- refundable credits aimed at lower and middle income households would be helpful because almost all of them face marginal income tax rates of 15% or less and, hence, the deduction system doesn't do much to help them. i don't think it'll deal with the overall cost issues at the high end, but it could be part of the solution. >> okay.

i'd love to open it up more questions. yes. and can -- yep. can you introduce yourself also. >> i'm david mitchell with the aspen institute. i had a question, just folks could talk a little bit more about the distinction between legitimate hardship withdrawals and kind of the unproductive leakages? maybe touch on an emergency sidecar idea that david leafson's talked about, how do we distinguish between two things, how do we keep people in retirement and let people get access to the money when they really need it? >> yeah. i'm on a bipartisan policy center commission on retirement reform. we spent three hours on this. and a lot of the provisions you can drive a remodeled kitchen through the hardship withdrawals, you know, for home repair. so thanks for that. lily, you want to start? >> yeah. i think what you're raising is a really important issue. i think as related but distinct from the retirement security

crisis. so there is also a major problem of very few households have a basic rainy day fund to deal with both shocks to their income, but also shocks to their consumption expenses. and that actually, at an aspen institute conference, one of my colleagues at nyu, jonathan murdoch, talked about his research on this, just month to month there are dramatic shifts in people's disposable incomes and what their demands upon their income are. and i think that's something that, you know, maybe is a somewhat separate topic from todayok butok is reallyç worthy careful policy attention. in terms of how the retirement savings system works, i think, i sort of want to go back to this issue of annuitizinging and lump sum. and, you know, one solution may be -- i think bill's proposal of sort of trying out annuities is intriguing and worth pursuing.

another possibility is just to gradually annuitize people over time. and this deals with several issues. one issue is that there's, of course, very large adverse selection problems in the annuity market. and so if you had employers defaulting people into products that just are target date funds sort of gradually shift more from stocks to bonds over time, you could imagine a new kind of target date fund that gradually also shifting you into annuities. you would, first because it was through an employer and defaulted, deal with a lot of the adverse selection issues. you'd also deal with the interest rate risk. they face a lot of risk that they're getting a relatively bad deal on that annuity because of what the interest rate happens to be that year which heavily impacts the annuity price. so i think that would, you know, potentially help with less leakage by also give sort of a glide path where in the interim or earlier on in the mid part of

your career you have much more of your account not annuitized, and as you approach retirement, you're having somewhat more of your account annuitized. the other thing i would mention is you mentioned the my ra program, and i think this is a really intriguing program for the issue of precautionary savings and a rainy day fund because even though it is structured as a retirement savings account, it actually works pretty well as a precautionary savings account. it's no risk, it pays a relatively high return for being no risk. it's no fee. the amount that is invested in that product is capped at 15 to ,000 and then treasury's going to have to make a decision about what product it's rolled into once it exceeds that amount. and it's on a roth ira basis which means the money that you put in yourself you can withdraw without penalty. and so that makes it a nice opportunity for people who want, you know, a basic account of up

to $15,000 to deal with these shocks, to have something that is earning a good, safe return but that they can withdraw at least their own contributions at any time for those crises. don't think that's an answer to the retirement security crisis writ large, but it's a partial answer to the other major problem that you've raised. >> so i think that, you know, i don't have kind of a silver bullet for exactly how you delineate, you know, what's a proper hardship expense, but what's striking is many of the withdrawals people make currently in the u.s. don't happen because there's some crisis and we're trying to decide whether it's actually a crisis or not, but they happen almost by accident. so the default on loans, i think, is the best example. i mentioned that briefly, but let me go through that in more detail. you can take a loan from your 401(k), and you have some period to pay it back, and that's fine. that's exactly how it should be working. now, if you had your 401(k) with an employer and you start to

roll your money over because you're switching employers or even if you just leave your employer, you are forced to immediately pay back the loan which which for many people is simply infeasible. so what happens is through things that happened in their life that have nothing to do with retirement and need not actually be bad, right, you get a better job elsewhere and now you're forced to pay back the $50,000 lobe that you took out -- loan that you took out in other in order to help buy your first home, then suddenly you have to default on that, and now you've lost forever opportunity to have that money in retirement savings, and you have to start all over. and i think, you know, while exactly the question you posed, i think, is an excellent one, think we can get a lot of mileage out of just preventing these almost accidental withdrawals. a very similar thing happens a lot of times when people roll the money over. they don't realize what they're doing, and sooner or later it's out of their retirement account. so -- >> i think you asked a great question, and i think what john

said is right, no silver bullet on this. personally, i feel like i'm the dove up here about early withdrawals. i feel like if you tell people when they're 25 or 30 you're putting this money in and you can't get it til you're 62 or 65 or 70, i feel that that's going to cause people not to want to participate. i don't have any good evidence on that. well, i have an evidence of one, my son, who's 26. and opened a 401(k) two years ago and was asking all these questions. it turns out that the idea that he'd have to lock it away for certain until he was 60 -- which seems like infinity, okay? -- was a downer. the idea of saving part of his raise, he thought, was a brilliant idea. so the save more tomorrow stuff, he's like the most enthusiastic endorser of that. but i think it's important to get the money in and rather, and even with the laxer withdrawal

rules than it is to insist that people not take the money out til they're 65. we already have a retirement mandate, that's social security. that is money that you cannot take out til you retire. but i have a very hard time telling somebody who's got a house and kids and loses their job, you know, and they've got, say, 50 or 100,000 socked away in their 401(k), i have a hard time with a policy that says they can't access that to keep their family in their house or something like that. so i think we all want more retirement savings, but i think the perfect can be the enemy of the good here. >> hi -- [inaudible] reporter for financial adviser magazine. this is the ivory tower panel. has any research been done on what the decline in home home

ownership has cone to retirement readiness? i think a strong argument can be made that home equity is a very important part of retirement savings. >> yeah. i'm looking at a couple to researchers -- of researchers in the audience, some from my team. tony webb here at boston college and new school and diane oakley. so it is a really important part of retirement security. comments here on home equity? >> i think, you know, home equity is the second largest source of retirement assets behind social security, so it's a very important source. i think that -- i have not done calculation, but i am guessing that the reduction, the recent reduction in homeownership hasn't had that big of an effect on retirement readiness if only because people seem to be relatively hesitant to tap home equity as a source of income in retirement even though it's a source of asset.

>> [inaudible] spending more money. >> well, you're not spending more money. i think you're just saving less. sorry, you needn't save less, but the point is if you save more in your house and then you don't ever sell your house until you die, which is, i think, how most people do it, then it's kind of not that helpful. i mean, it's -- there's a last stop, but -- >> [inaudible] >> just a second, just wait. any other, any other questions? >> first, can i weigh in on that? >> yeah, yeah. >> there is a paper by researchers at the urban institute a few years ago that showed that there's a correlation between paying off your mortgage and retiring. i think that's what you're getting at. you finished your 30 years of payment, your cash flow goes down and, therefore, the income you need to survive in retirement to maintain your living standards has gone down. so there's a correlation there. what's not clear is whether,

whether the drop in mortgage payments, the end of the mortgage payment is then allowing people to retire. it's not clear if it's causal. it might be that people who know that they're going to retire later decide to, you know, refinance their mortgage in their 50s and, hence, have mortgage payments for longer. but if you're doing retirement adequacy calculations, you know, how much income do you need, that dropoff in cash that someone, you know, when they're paying their mortgage versus having paid it off is a significant amount of money. so when people talk about, well, you need a replacement rate of 70% or whatever, that kind of calculation -- one of the reasons it's not 100 is that there's an assumption that you've finished paying off your mortgage, you're not making payroll tax contributions anymore, 401k contributions anymore, those are the types of things people try to calculate in when they figure out what replacement rate you actually

need. >> we time for one more question -- we have time for one more question. yes. yeah, hi. >> hi -- [inaudible] and a lot of the current savings incentives are not efficient, they're not effective, and many of your ideas that you've communicated i find to be very compelling. so my question is, can you comment on the level of support that these ideas have for working men and women on the hill or within the curb administration? -- the current administration? because i find your ideas very fascinating. >> that's such a good lead-in to the next panel, but any comment on the political popularity of your ideas? >> yeah. i mean, i think, first, a lot of the ideas have been proposed by this president. and i don't want to sugar coat it, i think some of these are very ambitious legislative

proposals that will probably take multiple years to get enacted. but i think there are signs of positive progress on the hill. there was a senate finance committee working group on retirement issues that indicated partisan support for the idea of refundable savers credit. they also indicated some interest in defaulting people boo lifetime income options -- into lifetime income options. there, i think, has been some interest in cutting back on some of the retirement savings incentives at the very high end. for example, there's a stretch ira proposal that limits the extent to which you can continue to reap the benefits of retirement savings once your retirement savings have been inherited by your heir. and that has, at times, gotten bipartisan votes. so i think this is sort of a long haul effort, but there is movement in a positive

direction. >> i have nothing to add. agree totally on this one. >> yeah. >> so i'll toss in two cents. i think that there's a political sweet spot in encouraging saving for low and middle income households. the sweet spot being that democrats care about it because it's low and middle income households, and the republicans care about it because it's saving. that's oversimplification, of course, but i think that if -- my sense is that both parties can support that type of initiative without any sort of stretch of their basic view of the world. and that if and when they do decide that they want to actually do something, that that would be a major candidate for action precisely because it would be consistent with both of their views of the world.

>> leslie, as i said, that's a great lead-in to next policy. thanks a lot. thank you. >> thank you. [applause]

susan: c-span welcomes to "newsmakers" this week, congressman kevin brady. he's a chairman of the key subcommittee and the vice chair of the joint economic committee, and with the new speaker elections this thursday, the day we are taping, he has announced his interest in succeeding paul ryan as the head of the ways and means committee. thank you for being with us. let me start with the new speaker of the house, paul ryan, who announced that he wanted a fresh start to the house of representatives. how long do you think his honeymoon is likely to be?