>> behavioral effect and should we get rid of this? for many european countries we have even after distribution. anks. >> thanks, alan. and this particular talk -- a little bit more academic solid

try to level set a little how they congress really think about this issue and the bridge and so forth. i will be essentially channeling out our block career and what i'm going to talk about. alan auerbach -- that ebv perfect objective moderator for the session for that purpose. but also in thing assay is called for i out his phone number toward the end. just kidding. so federal debt. in particular we heard about the $34 trillion the federal debt is in you. the 30 fort wayne dollars debt is not really that -- alum referred to in japan and other places where this trust funds with the government will come to the undead. what we really care about is the debt held by the public. that's ultimately what the government owes and veteran 27.5 children. it's that world war ii levels right now but, of course, one big difference is that it is

increasing as far as the eye can see. that's not the current projection. i will post my slides for people so i will touch a few things on each slide and in particular we have an exploding level of debt. cbo -- we both agree at the table also shows the sensitivity of that debt. some men shall bit about that this morning. time to talk but right now but we have done calculations like how high can the debt go before capital markets cannot possibly rationalize keep on contributing capital? even, ultimately when you see a spiraling condition like this it really comes down winter capital figure markets to get a? when you start believing that congress will eventually do something about it? even the most optimistic expectations we calculate debt

is about 190%, 200% debt-gdp ratio is really kind of that upper maximum. so the debt problem also has been massed a little bit. we think about programs like social security -- masks. when intro rate, social security, it's going to be through the population growth and productivity growth. that will give the internal rate of return. that's hot tax bases grow over time, population growth and productivity growth. that's what sometimes called labor a justice40 really aggressive couple decades ago. some of that with females into the workforce. they give us a one-time level shift but the stock would be permanent growth effect going forward. were forecasting the debt is likely going to go down quite a bit. keep in mind we are in the good days and away. we have a capital glut glut, a

high amount of capital per person were because of baby boomers going into retirement. that's not going to persist but that will help mask some of the problems going forward. and so, by the way a most as to what about japan? that's that come up a coup. as alan pointed this out, japan and the united states really are not comparable for lots of different reasons. our household saving rate is very different in the united states relative to japan. in particular even if you do some adjustments, measure capital gains for example, as part of savings, and japan's one a few and dozens of sites where housing prices actually lose value through the earth movement. even if you make those adjustments we still say much less in the united states. in terms of national city they still actually have a higher saving rate than we do when you count for household and subtract public just saving. let me get now a little bit more

into the weeds a bit. i'll try to do this without equation but just pure simple graphics. what is debt really measure? this is a question that alan -- larry kudlow and others have tried to address. not surprising, students of marty stoles dying to try to get people to think about this issue as well. let's consider, is that really measuring much. in particular let's consider a really simple to period model, i should've called these periods because each year could represent maybe a couple decades and so forth. in particular we have generation one here who is working year is year one and they retire in here or period two.

they are followed by an overlapping generation who when they're retired followed by a next generation who are working and then they retire in the following year and so on. so suppose we do the following can we do a tax cut where treasury gives one dollar to generations, generation one, index by the working here. and we don't pay for it from generation one. what we do is we simply tax future generations two, three, four, and so on just for the interest payments. so we just roll over that one dollar constantly making interest payments. suppose we start and simple world, a borrowing rate of government, 10% growth rate is zero, and just to keep it simple, generalizes and what you have is simply you can see ultimate generation one is well-off. they get a dollar that frees up in terms of year two, $1.10

because a freeze of money they can then save for your two. interest rates 10% and within finance of that simply by rolling the dollar over and hitting future generations with a ten cents per generation interest payment. okay. so far so good. treasury debt of one dollar. construct the exact same cash flows without using government debt. in particular, what we do is we simply make a promise to generation one working age in year one that when you retire in year two we're going to give you a transfer of $1.10. how are you could pay for that? by taxing the next generations, pay-as-you-go, $1.10. and by the way, were going to give them the same promise where we're going to say when you turn

age two at period three were going to give you the benefit of $1.10. how were going to get that is tax that next generation after them the same amount. pay-as-you-go. notice the first generation gets this windfall of one dollar, or year two, $1.10. subsequent generations all lose ten cents in present value. how did you lose it? it's because when we have been give $1.10 to the previous generation, they could have, in fact, had that money and invested themselves and give up the interest rate of ten cents. in present value its identical cash flows, everything is exactly the same between these two approaches. but what is called treasury debt and the other one is called what we economists say implicit debt. if you got this exercise i will tell you right now probably the hardest concept of economic

special capitol hill is trying to explain to playmakers the meaning of implicit debt. but that's all we're doing here. in particular, we have one dollar of explicit debt backed by the full faith and credit versus one dollar implicit debt. yes, there are some legal distinction between these two, although some implicit debt like so security is indexed to inflation in a way and even wage growth the weight of explicit debt may not be. but either way our job as policymakers, not policymakers, but as scorers and analysis is to look at law as is. and economically why do we care about that? i blew through this a little earlier, but ultimately why we care about debt is because a person order today mention the crowding out effect, in particular if that is being used

primarily to find immediate consumption, then that means we are giving at someone's deficit. debt competing with private capital for the same cash flow from household to international capital flows, and, therefore, you have a reduction in capital because each household who buys that use that as saving from their perspective the fact that if they want immediate consumption fell for it is irrelevant. think about intergenerational fairness, it's exact same cash flows, one explicit, , the other implicit but no difference between the two. if you think about issues of clothes, i blew through this, but something idea of -- we talk about the earlier today. what happens eventually is either congress has to do something really, really big, or if congress doesn't do anything, kind of reminds me of several

years ago when a japanese investors what the wharton school, doing executive education for them and they are like, as well as chinese investors, they were like are you going to pay us back? we lent you all this money. i going to pay us back? i said yes, we will pay you back every dime that we promised you. there's no question. we are not going to default on that. we will make sure to print enough money to pay back every single dollar and time that we promised. the point is that when you think about what is inflation? inflation is ultimate attacks, a tax on net long-duration old labor has to be closure somewhere in the model. he may say this is all good theoretical exercise. economists have noticed especially if you got -- grilled into this, alan auerbach made this point they musters a goat what they call the labeling issues that it is just labeling but how big a point is this

economically? it's a big point. in particular, the current, explicit that is about $27 trillion. however, if you look at the implicit debt, that is about, only for so scared a medicare in particular that you simply ask the question, how much have we given in the form of transfers in the social security and medicare people in the past as people see how much i we projecting to give them? versus how much have the fit into the system, alden properly and present value. the difference is about $62.7 trillion according to our calculation here. want to be very clear. those of you in the weeds we call this a closed loop liability. this is the open group. this is not accounting for shortfalls the future generations that are not currently debt. that never gets even bigger than

this. so as a measurement, federal explicit debt is not a great measure. implicit debt does a better job. it's hard to explain of course, but nonetheless it is simply more comprehensive and better. you may say, or correspond i should say to what ultimately we would think we would care about in terms of economic effect, in terms of intergenerational fairness, things like that. let me give you an example. let's look at the social security 2100 -- sources could act of 2100, and in particular i'm going to look at the original version of it. it was a pretty clean act. things have changed, do like temporary benefit changes, things like that. let's put that aside. let's look at the most original version of just a couple years ago just to make it crystal

clear. in particular, the social security 2100 act both the salsa could act and penn wharton budget model agree that this basically solves the 75 year problem we estimate is not quite fully solvent in 75 year shortfall. we have different projection assumptions and so forth, but nonetheless we projected mostly solved the shortfall that exist. i have a lengthy study and so forth you can look at if you want more detail on that. as pointed out earlier, almost everybody look at what's called scheduled benefits. in particular even the score keeps her supposed to look at current law as written, we don't do that for so scared medicare. in particular, because if we look at little current law that would be slashed benefits by as much as a quarter in less than a decade and we're not really, no, we can talk about that. by definition so security would

never have problems under current law because we would just cut benefits at the time occluding on 90 old people and so forth at that point. we really look at scheduled benefits. however, here's the thing that's pretty interesting but the 2100 act. it actually did decrease explicit debt. how? because this is all relative scheduled benefits that in particular when the shortfalls happen in so security they show up as though if it's a general revenue shortfalls, as though it's future debt that is required to make these payments. but what we show, implicit debt goes up by more than dollar per dollar than the decrease in explicit debt. and how is this happening? because part of that act actually removes salsas to benefits taxation and that loses some revenue come quite a bit of overtime in the book benefit taxes are with a constructed,

index inflation can more people pay the overtime. tickets are bit of that and the shortfall increases going forward. goes up by more than explicit goes down. as result of that we find it contracts the macroeconomy even relative to peer deficit financing of scheduled benefits. by the way, without almost every reform to social security that basically gross economy relatively scheduled benefits come basis one -- it was generous in terms of giving benefits to older people. it puts it explicit picky mice okay maybe we'll get the baseline wrong -- getting. we are not measuring debt correctly because after all were doing implicit transfers. i am saying even a policy change

is i getting it right. in particular if you look from a treasure perspective, looks like this is lowering future treasury debt both of the economy and is not even the right foreign policy change. other than that i say it's great. but keep in mind explicit debt really reflects a historic compass that a dishonest metric but it reflects history. it was created during a time were lots of government spending was more brick-and-mortar. i would say it's pretty samuelson diamond r bock approaching before modern computing methods. let me go into more. how can we fix this? this is where the history of the common bottle that is often used

to analyze this policy. it's a popular growth model, a brilliant man so much comes from frank ramsey, died at an young age unfortunate but he was a growing person into particular is basically as a model that relates capital and labor to output. with this representation of households, this model does not treat debt. in particular there is a zero cost to debt, regardless of the difference between r and g. just no relationship between the two. debt is kosice in the model. here's a little intuition that robber barrow years later gave us for this -- robert. in particular, i have a typo i will surely to you in a second. let's go back to our entitlement example of you. women with the same thing of

government debt. in particular now what it added here grandparents, parents and you. suppose now that generation one that's working age, that to grandparents, and you, you are working, you start generation three, kind of your working age, and what we said to grandparents? we said we are going to give you a dollar and ten in transit. how will we pay for? generation two was a $1.10, and so on and so forth. you make example as before. but suppose the grandparents were going to leave money to the parents. i haven't told you how much capital they have. they will these two dollars to the parents and the government comes along and gives them $1.20. the grandparents leave two dollars to the pressure the government comes long since we will give you $1.10 from parents. they gave the money right back.

that's going to follow all the way down the line. when i say the net minus ten cents, the ask around this at the particular the operation, fiscal operations of these transfers are fully neutralized within the family. if this sounds particularly extremist to you, robert barrow was one who showed this in 1974 that led to hundreds of papers on this idea about family transfer offset some things like this and so forth, general conclusion is probably not. any and maybe closer to 13 cen the dollar come something like that. but what it did show here was, which i think it's a real insight is that we think about the ramsey model, it's really a family, a dynasty here. ultimately in that model there's no role for debt because and it doesn't matter what r versus g

is in the framework simpered because there is always going to be money given, given right back across the generations. so also by the way that model since it has like this long representation family here, you can only have one of them. you cannot have any heterogeneity. so for interest and progressive taxes and all that type of stuff you can have that progress is also good benefits. you can have this in this framework simpered because the most patient dynasty, the most patient mafioso found will accumulate all the capital it is no role for anything anybody else. how do we fix this? here's the reduced form attempt to fix this is what i call, reduced form in particular there simply limited in scope. if have a rule based on say

empirical that simply says for every one dollar of new debt let's reduce capital accumulation by icct says based on empirical relationship. the problem is this actually does encompass suppose we leave the assessment of crowding out. it actually doesn't cover simplistic debt. it only covers the explicit debt. an example i just gave you of the social security 2100 act or the pay-as-you-go transfers, that's going to still show up as a debt reduction that's therefore crucial as a capital increase, positive macro, even though we know implicit debt went up by more than explicit that went down. so it's going to miss that. also some issues on how we can keep the 50 cents and so forth. and that is an average marginal effect. it will is good in the particular range that it was computed and in particular the

better way to do is with a more structural, i will talk more about that a little bit later here. so how do we generate a six-year? we know how to fix. alan r bock. that's how we fix that. [inaudible] >> oh, no. [laughing] but you can. you can download it for free. and i give you my photocopy. note in the market. in particular, we can computed these models in particular the lifecycle framework, hence the title of the talk you. it simply saying we will explicit a model, model how the public, explicitly model the business sector, government and there's now been does the papers written based on this type of framework. a with the first to take this

implement and make a big and do a confrontational -- computational. at the time thought to be impossible but they, in fact, did it. the whole story behind that. in particular it's not what your model. the point is what makes this model i think the real way to set think about fiscal policy is simply being explicit about things. we use this as a budget welcome part of our larger platform. a time to go into a lot of detail. i can have to provide slot slides but the point is we start off with michael's stimulation, 60 some different at the beach, grow, get married come have kids, so forth. ultimately feeds into overlapping generation model. one reason why we do this is we one realistic demographics of lots of other things. but g model also to michelle in

terms of computation. that reduction is something we have been solving more and more. let me do some examples of how this will treat a few things differently. we posted some analysis, what we call options for policy or federal budget reform. in particular we never made the recommendation which they much like seaview innocent as it we're not opinionated way comes to policy. we have posted a brief that basically looks at three different options for trying to reduce the federal debt, and the idea behind is that each one would take at least not kill the economy. so optional and is basically high income taxes and tax of corporations, see how far we could go with that one we don't call it the liberal option but effectively that's a people think about.

option two was how you get rid of, talked earlier about the adjustment. also another one. get rid of the employer-sponsored health adjustment all these get rid of any access are wrote to the median health plan out that that's a big -- so forth. so what happens in this plan, what we show is indeed the economy grows and so forth. if you were to compare this partly because of capital increases the crowding in when death comes out so forth, suppose call bundle three. this is what we don't call it the modified but essentially has a grab bag of different things like a value-added tax, a carbon tax in the rid of some of the deductions and so forth. also slowing benefit growth and so on. notice, sorry, hopefully you can

see this. it's the total debt over time between the two plans actually goes down more on their bundled 310 of bundle to make. so total explicit federal debt goes down more in bundle three but notice the macro gains are actually smaller with bundle to than they are with bundled two. the intuition behind that is simply that explicit debt is not measuring the change in implicit debt. implicit debt goes down more still in bundle two, suffered because of the reduction paid to go. bundle two accommodates things like flat benefit and so forth. we can see even with larger reforms come visit with some of the stuff really shows up. the real advantage of the lifecycle framework is, i was a better microphones to macro and back again.

it captures things that economists care about. for example, the payroll tax is not like any other tax. others talk about this with her in particular the payroll tax is not distorting as the same 12.5 or 15.3% tax levied on individuals, simple statutory tax that goes than pays for public goods, roads and so forth. the payroll tax is what's distorting some good because you are in wages to pay the payroll tax. if you didn't earn those wages you like it's also scaredy benefits. fact in the rules were r equals g and is no redistribution and there was no differences in lifetime, how long people live and things like that, really simple welcome the payroll tax would be non-distorting the reason is it's just giving you a return that you otherwise would've gotten capital markets. we want to account for that at the lifecycle framework automatically gives you that

type of micro foundation. also, count the fact that lots of times the interactions on the conventional site, how we often show things, is on conventions that sometimes these interactions will be quite negative. i am changing the social security retirement age and medicare retirement age. you can't compass things as independent. because they're going to have effects across each other. some increasing the corporate tax rate and the minimum book tax rate, those are not independent. they have strong interactions with each other. so often the interaction effect on conventional basis is negative, but would you think about in a macro, you the macro framework as well, i can often be positive. when change distance could affect the tax base and on social security as well. also the lifecycle model can also explain something that is not well understood.

that is if you look empirically as a possible weak relationship between debt and interest rates. cbo study on this but most historic debt, , alan pointed ts out this morning, is counter cyclical in nature as a result, what happens when you actually have a negative shock in economy? supply goes up but the demand usually goes up as well. in particular, so we don't expect from a countercyclical increase in debt that you will necessarily have a reduction in a capital or increase in interest rates because supply and demand are changing at the same time. with a realistically calibrated overlapping generations we can almost perfectly replicate that. this by the way goes to another

thing alan auerbach and larry kudlow pledges go, generate a data set which you then perform reduce estimates and see if you're replicating what you seen elsewhere. let's talk about how is government debt, suppose government debt is immediately consumed versus how if it is infested and so forth. here's one thing i was simply say about this, is that it's true that government investment, if you take that debt and you invest in something that is long duration, public roads and things like that, you can actually get a positive return, things like that. however, when you do it realistically, realistically calibrated framework, that high marginal product of capital for government, government capital, actually falls very quickly. there's no way saying it. you can't just look at the marginal benefits. you have to look at it

comprehensively. let me give you a little bit of an example here, and that is in particular let me dive into this next topic and i will go wrap up in about 15 minutes. that is, what are conventional to spatial tables measuring? this one is, i will not be dogmatic but all will say to decide on this one. but let me give you a couple examples, something just to think through. this is bundle two blown up so you can see things what's going on here. change in social security benefits, flattening the benefit, the retirement age and so forth. you could look at conventional distribution analysis like, for example, suppose we said this is going to begin in 2025. what is the impact on different households? one way of doing that is you could just line up households by their incomes and say how much

are they going to win or lose from this from taxes, from benefit changes, all that type of stuff. by the way since this is phased in over 20 years slowly and so forth, you may want to repeat this exercise over time. but the thing that's weird. if you look at those, that particular plan, thus no particular reason the highest incomes, how much are going to lose from this in terms of actual dollar amounts should actually be higher than a bit of a lower income. that the art and conventional analysis. what's happening with conventional distributional analysis is it ignores lifecycle effects. in particular, why are people in, who would be in the 99th percentile? typically not retirees. they are people who are higher income while whether work. so even if you sort incomes,

include all sorts of incomes can wage income, solstice could income, capital income and so forth, you get a pure lifecycle affect year at the particular looks like the highest income are actually not being hit, even though in dollar amounts they certainly are if you were to look at it that way. now by the way the other policy bundles come to get something that looks more intuitive. that is if it's just tax changes, looks of the more intuitive that the region paid to those taxes and so forth in a conventional type of way. even if this embellishes, , goes too far the opposite extreme. because what you really care about typically is asking people how much we value this reform before the know exactly what their income shock is. here's the problem with distributional analysis. it treats everybody, whether

you're a twitter five-year-old making 75,000 feet or our house for household making the same amount, you know, a family of four, they are all treater applet in the the same income bucket. yet clearly those two families are very different. they also ignore how progressive tax is including solstice. also provide insurance value against risk it's hard to ensure privately. i have a couple papers years ago where i show you can actually grow the economy by going to a flat tax system by getting rid of solstice could and so forth. throughout the entire transition path from a macro perspective it looks awesome. however, from a well for perspective people are worse off simply because of insurance value of those programs that they provide. so you can't, economists have never thought macro alone was a

fix. how do we fix? asked alan again. in particular this up with the lifecycle framework also provides. at the particular it doesn't simply line of people based on distribution analysis that we catch it was called equivalent variation. in particular basically asking the question, how much would you value this policy change? in particular, suppose we didn't do the policy change. how much money would have to give you to make you exactly indifferent to policy change? for various technical reasons, it incorporates things like risk sharing value of the policy, macro effect on the lifecycle affect and so forth. it can lead to very different answers. think about the inflation reduction act, which we said at the time, this was before the treasury the competition, electric vehicles and so forth, is we would reduce the debt a little bit, deficits by $260 billion over ten years. it actually grow the economy a

little bit of most of this is not that reduction. it's because, in fact, from a static conventional analysis that looks like it's all loss, everybody is losing because they are simply paying higher taxes, means more taxes. but for my dynamic basis, most households born future are actually better off in the soviet because just the capital accumulation. it's also less carbon, increases the strength to be selective and clear about things like greek education, who benefits from that? some people already get it. you don't want to count them twice. you'd want to treat that for individual activity. in case of carbon. think about, so have read of ths dynamic table, simply insane if, in fact, you happen to be in the bottom quintile, how much does a

splendid make you better or worse off? at a particular birth your computer to you if you're aged 40 and lows quintile, this one hurt you $700. that is a one-time payment. but if you're basically i'm bored, that is you is your 25, you are in future here, this plan even though the model helps you by about $1.9000. in the case of a social security 2100 act will be fined we find is older people like it because they are not directly hit by it. they like the crowd out of capital because it raises interest rates, but future generations from those born in future are worse off simply because the increase in implicit debt. you might say okay fine, how do we get rid of the macro stuff? that's controversial and so forth. you can still do calculations without having the macro. one way to do it is easy

calculations. you just simply fix the choices house are making whether we think about this small open economy if you wanted to, and we can now do analysis like let's suppose we raise the retirement age up to eight to 72, two months a year come fully face of in overtime. and we don't account for many macro back about the console account for lifecycle effects. that one should be way less controversial and we can see some positive and there some negatives for future generations. but when the count for lifecycle effects what happens is we get a change, almost everybody is better off through capital accumulation. partly result, existing market friction known as fiscal externalities and so forth. i realize this is a lot but let me tell you what i think is the

real challenge is in the space here. i think arguments against models are not so much arguments about all models that you can incorporate frictions, this is not being explicit. i think there's a large cost of entry into these models. and the fact of the matter is the economist are the ones to blame. we face very few rewards to making our models reusable. we don't care. in particular its very few rewards for making our models clear, reusable, so i call it -- hide your heisey write in plain sight. writer really crummy code all you can understand so when you posted as a general question of posted, it was hard for other people to then say this loss of property and use for the purposes as well. there's also free rewards are making a model that is really comprehensive as well, that includes a lot of different fiscal systems and interaction

and so forth. if you want to do serious cost-benefit analysis, i think that is what is required. it's often like somebody knows, knows how to run the model, one of the model off-the-shelf here. what we need is an integrated platform where the inputs are generated from something like a micro simulation and makes this thing really scalable and computable. you say it doesn't open source of the problem. it really does minor in economics because just to be 5000 lines with some fortran code doesn't do much right now if you can't do with the code right now. what we really need is a strong software design. in particular -- i model you can download and run on your pc. there's probably a pretty small, smallest model. what's the goal?

apple inc. enemy. which create a problem that is accessible. why not have arguments about assumptions, , about what goes into the model. but let's reduce the fixed costs that goes into the model. this requires a collaboration. my student can now come to the mess but because we try to solve these dimensional issues to make the model more flexible overtime. economist repetitions in eaters and computer such as ali to work together. the head of our technology group says great site secretary s great injury. i tell him i think michael colton, he claims it comes from someone else. i've not been able to find it but nonetheless i think that's exactly right. we need to be a software process here so that we have debate on common code also lower the cost. that's our goal at that penn

wharton budget model is was do that and we spent a lot of time really trying to quit a a plam of the people can eventually use. where nothing yet. i will end with this. sometimes people ask me, i can't try to put cbo at a business? it's like no. just the opposite. we are trying to be accomplished not a substitution in the production process. in particular i think -- absolutely critical part of your kaposi going forward. everybody loves to hate on cbo and the other groups like that. i will say one thing way worse is not having them in particular just look to europe how budget often works there, the policymakers, the politicians will tell the agency what number to come up with. it's a really problematic process of there. but they are busy and academics

need to really get involved in these to help and bring that sophistication melissa bring that sophistication with throwaway code. let's would bring in a platform that becomes reusable. and the united states we've been contacted and is not quite yet but that's what worked on is to really make that base usable by everyone. eventually by the way were going to start grantmaking process, how use model sets, we will pay you to try to get the framework used and so forth. all right. [applause] >> kent, why don't you sit it and we can have a conversation and see if we have any questions. i want to just make it . about the first thing that kent emphasize, which is the distinction between explicit, or

the comparison between explicit and implicit debt. it may sound to some of you may sound as a carrot and a may sound like well, that something of a theoretical level, may be interesting but how does really affect policy? when i tried to discuss this with my students i i find a vy good real-world example. during the george w. bush administration there was a proposal, social security reform. it was part of the ownership society initiative, which would have allowed individuals voluntarily do have some of their payroll taxes redirected into their own retirement accounts, whether they were private or personal. there were different language used to describe but the idea was they would be put into account that people would have control in terms of investment. and an exchange those individuals who got such

accounts, get money put into their accounts from the payroll taxes, would have a reduction actuarially calculate n in the future benefits. meant to sort of offset at least in present value calculation the benefits they would get from the taxes that they were now allowed to invest. and the government in order to keep the social security system going because some of the payroll taxes in this pay-as-you-go system were now being put into these accounts, government would've had to issue additional debt to cover the shortfall. it would've been kept as pets were substantial in terms of how much the debt would've been created. and there was a lot of concern and discussion about how this

was irresponsible policy because it was created all this new debt. but if you thought about it it was really just a substitution of explicit debt for implicit debt. because what he was doing was it was saying look, if you are promised us a good benefits and you opt into the system, you are getting fewer social security benefits but you're going to get your own retirement benefit so you're essentially being held harmless, you may be better off because of something you choose to do, but the government is going to have lowered liability now in terms of the benefits are going to get but it has a higher liability to the people it is just issued bonds to. so what was the effect? the effect was there was a policy change there but in terms of the amount of government debt outstanding it was a convergent of implicit liabilities into explicit liabilities.

if you have a view of think about these things together, you could understand it as a policy change that substance in terms of investment decisions and so forth, but not in terms of the overall liability for the federal government. ..

in terms of economy and carry more weight straight on the one hand eroded over the last years and on the other hand it's very strong political commitment. there differences in the

facilities and they are somehow in this construct. in this dynamic analysis. so what you do to make your model work?

in this what they are referring to is in this framework if it isn't sustainable what i like to think so in this model you have to assume 30 some years the government will get its act together.

the value added tax in a hypothetical and a progressive income tax and it can matter i do it because it is forcing you explicit about what you are doing. as i said in the past, trying to close the locomotive bob rejection that shows the next 30 years is your 30 and magic happens here. that year we are going to keep it stabilized going forward you'll get rewards for thing that and we have options to do

it. >> the amount of explicit that by u.s. investors essentially composite that? >> this particular configuration and that going up today.

it is still what they are in, it's making to them and therefore that is bigger. it is true the shortfall the real way is that scales of something. right now giving a bottom line

supposed the rate taxes immediately and forever how much we have to raise tax revenue delete and forever just to have enough money to make our promise obligations and the number is about 33% right now so we would have all sorts of revenue and that is assuming feedback. with the new spending plan we have a tax revenue.

discretionary spending and so forth, is the problem we happened still without macro effects it would be enhanced even more. he faced enormous football in the united states. that is not working so great it's a current policy and optimistic assumption and by then, it's almost but no return so high it would not be able get away from that is not clear?

one thing that surprised me is how showed a range 30 years office 5.9% less likely three years growth. at the same time the baseline which probably can't happen in the alternative is not 5.9% lower about realistic alternatives and where we are likely to be if we stick with the baseline. >> it's hard question and you

have to have this concept. the problem is real problem there so click story. talk in the u.s. and the long-term outlook. we believe in this specification. and it is the opposite of that.

conflict that is problem. everybody says it can happen in the u.s. it definitely can't. people can change their consumption behavior and demand this so yes, you're right it would have to end. we have run out of time. thank you. q.