taking precaution there. that doesn't mean and budgetary variety and less of a structural variety. >> budgetary meaning -- >> if you said you have a met care and that doesn't mean they can't raise medicare premiums or do something else to improve the system. >> let me criticize the siltation of the breed. i have written that turned down and never followed up on it. what he assumes is something you know perfectly. there is a parameter of changes we have. you go for what you know perfectly.

>> i can add on both of those, just take that as another example and what is uncertainty. that matters a lot. you can get different results. you need to have that on the bigger issue of uncertainty and ambiguity, one of the things we haven't talked about at all is what the view of the nation should be regarding uncertainty. when you move away from the standard and expected framework or even within that we have to have the government be risk-averse. once you move into areas where there is not ambiguity and this shows up in climate change policy, we have to take a 0%3pvstand.

my technical work goes in analysis of social policy. you have to take a stance of what the right answer is. that's as well. >> i'm interested in that point. there may be circumstances where the government is in the national interest to respond to uncertainty by taking action now. >> i have a paper that came up with the journal on the effect of infrastructure. the big uncertainty is what's the effect of government spending. i cranked out because i'm a technical economist. i crank out the solutions.

just based on the criteria the government is going to use. >> we are talking about this in the context of the forecasts and how they use uncertainty. i want to make an observation and it may be growing, but it seems to me that the private sector and goldman sax of the world will not treat -- they treat it more like andrñuñ less like the ideal economic viewpoint would be.át there is a question for all of you given that the private sector is not asking or producing these estimates of uncertainty that we would like to come in the economic models.

>> if uncertainty was more on wall street that's hard to see the market for options. >> to think they were delivered in the same way the cvos are. it's uncertain. >> i think when goldman sachs predicts what the get is going to do, the clientele may be as interested as congress even though there is a great deal of uncertainty about what the interest is. >> thank you very much. what impact they would have if we focused on the inequality, close that gap considerably,

what impact would that have on uncertainty for the future especially as it relates to social security," medicaid. thank you. if we made unequality disappear. >> it's not going to disappear but it's going in the wrong direction. >> we have greater inequality and they are more likely to use the taxes and equity is largely improvements that will be more valuable and willing to spend more to get it. that also means if you have an economy where you can have lower tax rates because you don't need to affect policies with redistribution and gives you more flexibility in terms of the playing.

it does mean that if revenue needs go up, you have more scope for increasing tax rates than you would if you were already using a lot of high margin attacks to do redistribution. it would make life simpler. that's probably most of them. >> the last question, the guy in the blue shirt. they have a proposal that they have a tax increase that is triggered if uncertainty and inequality gets worse if we are not uncertain. >> each comes from social security administration. the question for the panel on this that relates to what you were saying about the prospect of elected officials of inflicting pain that may not be necessary. they mentioned this. the intention is yes. understanding experience and

worrying about risks for the future. we would like to save more than may be necessary. we may expect to be necessary. i would ask you all to look back and observe on what you experience. i think we have a lot of theoriy this this room all of us. what is your experience and observation? we have a lot of people representing the people with a range of possibilities. do they tend to look and they are suggesting at the out liar situation where they have that or they tend to be more hopeful than say given a range while it could well turn out to be better than the best expectation so we can defer action as a result. therefore giving a range of possibilities doesn't really take us in the direction of our politics that you would like. >> i'm not sure i'm an expert on any other politicians and others

here who were much better able to comment on that. what if any institutional changes could be made to make it more likely to succeed with the resources aside. it's an interesting question. there various potential answers. >> i am not an expert, but inflictingnbacñ future pain is politically much use and again social security is the obvious example. secondly worldwide we are seeing serious events filled up. i'm just back from peru and they are studying how they can have

commodity issues. we actually have experts coming later. >> too humble here. >> the one point that i might put is that the behavior policy makers and politicians is not necessarily fixed. i hate for an economist to use cultural norms, but there is a norm to ignore uncertainty. that is changeable. given now and going back we had a heated exchange and wall street shows the market is

always right. it could be the forecasters were wrong. and these things are changeable. i spend a lot of time in london and i find somewhat more to the expression with the uk than in this country. this is changeable. >> with that i will dismiss the panel and invite david cayman who is an assistant professor of law. he's a veteran of the obama administration and the national economic council. we asked him to think if you are you are uncertain, how can you

instead of ignoring that how can you build into government policy into legislation, ways to anticipate and adjust to it. by bill hogland and representative cooper who will not say they are notb? %>p1cr experts. >> first thanks to the center for organizing the conference and i look forward to the discussion with representative bill & want to thank them in advance. i want to specifically thank gene. part of the my interest in the topics and trying to build these mechanisms came from the many hours and frankly very late nights i spent trying to dream up trigger mechanisms while working for gene. i wanted to thank you for inspiration even if it came at the expense of sleep.

first, i wanted to distinguish this discussion from the discussion that came before. alan's paper addresses the very important issue of how much we should be saving for the future given the uncertainty that you face. the issue is somewhat different. irrespective of whether we choose more or less because of uncertainty, we make decisions that will affect our future. we have little choice in that matter. when major legislation is enacted today or on future dates, it will have effects going forward. that legislation will continue as time goes on and each as the world changes. in ways at the time that legislation was designed. unless it adapts to the circumstances, we will have a situation where what i call policy drifts. as we drift off source that was

intended where there was no correction from the legislation itself from policy. this has areas from cyclical to social security. they focus on the legislation that can't be used to address this problem and building flexibility into legislation so that it can update in ways nn2)to ises even if we end up in a situation where the world is different than what we expected. i emphasize automatic adjustment mechanisms as a tool and something emphasized by the panel. i think that is probably among the most attractive ways, but i don't think it will be appropriate to all circumstances and will discuss other tools we have at our disposal for trying to address this. first, let me lay out the basic problem. the process of legislation is

often characterized by inertia. not always. there will be moments where large legislation gets done and bargains get struck as it is followed by periods where little legislation is actually done. so the figure i have up here displays this visual in terms of the process we may be looking at. it is quite a simple figure. the idea is that tell us gets set at an earli circumstances change in some ways to begin to drift off of the path intended and policy may not be performing as it was intended to. there may be thresholds below they will not correct the drift. they only correct if you drift off from the intended path. there a number of ways to explain why this pattern may develop and includes the fact that congress with can only focus on so many issues at once

and include the difficulty of negotiating across two houses and the president. especially an increasingly partisan environment and the exact explanation congress tends to move in fits and starts and doesn't always respond to new information. this has real effects on policy and the country and i want to illustrate that using two policy areas. one coming from the recent obsession and the other from the social security policy that has been discussed already. first the great recession. the great recession may be remembered in fact for the large legislation in response to the economy's tail spin. it may seem an odd choice for illustrating what i call policy because congress seemed in fact to respond. the major legislation was designed at a time of great uncertainty about the state of the economy.

while it was expected to suffer the recession and the severity had not become fully apparent. as of january 2009 for instance, the budget office expected the unemployment rate to be about three percentage points above the full unemployment rate in 2009 and 2010. as it turns out, the actual rate was something like four to six percentage points higher than the full employment rate assuming policy remained the same as of the beginning of 2009. i should be clear the fact that the initial projections were incorrect and too optimistic and became evident. the rate shot up to higher levels than have been expected. what was the congressional sonsestionressional response to this information? beyond what was earlier enacted and 2010 saw some largely as a result of the congress extending measures of cliff-like

expectations. this is a lesson on which i later focus. probably not the best way but one way of getting congress to focus and update areas. the point is that circumstances had changed and policy had drifted away from the intended course to the extent there is such a thing without such response from congress. >> now taking the response of social security. as the panel discussed, the problem is not just about the short-term economic trends but about the long-term and we face uncertainty there as well. when it comes to social security, we have seemed to have a general commitment for self sustaining social security system. they agree on that. generally. in 1983 and in the face of the

solvency problem there was a deal to maintain solvency in the system for at least 75 years.jjqtn but the deal was made in the face of uncertainty. as it turned out, they produce a system that will not be as solvent as they expected. they pointed around 1933 and was shown and the trust fund never becomes insolvent and that it occurs as early as 2028. we have drifted off the course set in 1983 but there is no response in the form of social security since that date. so now the question comes, what do we do in the face of uncertainty and sometimes

inability of congress to respond to new information? part of the answer should be to make it more robust and the legislation can adapt and without congress necessarily having to take action. i call this automatic adjustment triggers and the form is indexing. it's a trigger set to go off and updates and legislation appropriately for those conditions. first let's go to counter cyclical tell us. in the face of uncertainty, it can be made to more automatically respond to changes in economic conditions in order to support demand and withdraw when not. they strengthen what are the automatic stabilizers. while the idea is old, it is worth revisiting and ex-panting in at least two ways. first, when the government does do significant discretionary stimulus with congress stepping in in a period of weakness to

adjust demand, it can potentially build automatic adjustments into that legislation. legislation. it is sensible. many of the largest provisions could have been made on economic s. second on a permanent basis they have automobile stabilizers. only unemployment is triggered off by when the unemployment rate crosses stresh holds. yet the unemployment rate does a good job of indicating when they are interior recession. the gray period here shows periods of recession and a spike indicator intricates the beginning of a

recession. on other indeces we can come up with we can adjust spending on a state by state basis and adjust tax credits and so on. we would have better performance. now returning to social security, in social security we can index both taxes and engets either to the long-term social security projection or to individual indeces that are relevant and solvency. they can have the agreed upon goal with the social security system and assuming that remains the goal the system can update to be consistent with that. now, one point that peter diamond meat is based on indeces other than long-term solvency. i would note that there is a and

we might have to build in different protections for the actuary and doing those projections. i would note that when it comes to doing the actual social security plans, they do them based on the> u(áv 75-year projection ands qu including the diamond saving social security plan. this would suggest that he would do target the 75-year solvency. >> with that said, that would be an improvement. if we did it through indexes and parameters, that would make sense as well. that's the revenu%ék÷side as peter diamond suggests. the goal of such an automatic neck nichl would not be solvency. they will get around to it later and focus changes on later generations.

the idea here is to spread the adjustment appropriately and share risk across generations. i also don't mean to suggest that this is a way we will get to where we are today. right now congress and the president needs to agree on a compromise. the point is to build in these mechanisms into compromise legislation to preserve the deal it has made and spread risks across generations rather than waiting for adjustments and concentrating them on certain generations. medicare of course is also an area of major uncertainty. our health cost trajectory is uncertain and the question is how do we react to that? let us say how automatic adjustments can be used here. they can be used to adjust one could set a base line and whatever is the desired trajectory and adjust financing

to offset changes relevant to that trajectory. we do this when it comes to medicare premiums. irrespective of health care costs, they are offset about 10% of medicare spending. the same is not currently true of payroll taxes or general revenues including income taxes that finance medicare. as this graph shows, if health care costs followed income growth payroll taxes would finance about 1/4 of medicare over 75 years. if they grow faster than income annually consistent with the base line, they fall to just over 10%. they could be adjusted for the trajectory. for the payment side to figure

out the limits of them. that's because we have less information about appropriate policy of responses to the different cost trajectories and we don't know how to reduce cost in effective ways 20 or 30 years from now and whatever the responses are they can't i]d]ç triggered via formula. >> there other tools beyond the automatic one it has been historically used in this way. congress can delegate to agency with better title adapt policy and in medicare this is part of a logic around the reforms in the affordable care act that it is undpleer they get off the ground given congressional opposition and aspects of its

own structure. this delegation makes sense for helping to address uncertainty on the spending side of the equation. medicare. in my paper i discussed other tools as well in addition delegation that can be used to address policy whether in medicare or other areas. this includings alarm bell triggers and mechanisms that try to force congress to address an area even as circumstances change so instead of having policy adapt, the idea is to force their hand to be focused on an area and compromised if circumstances change to a certain degree. each of these mechanisms comes with trade offs. ä< circumstances and along these lines, automatic adjustments seem to be superior where they can be developed. that is the first place i turn, but others may give greater weight to the other factors identified as being relevant.

i want to offer a few words of caution. first with triggers and expirations. these are meant to force congress's hand to adapt policy under circumstances by creating a set of circumstances that are so distasteful that congress just has to act or at least that's the theory. a colleague of mine calls these the shot in the foot mechanisms. the sequester being the leading chchl. they come with a severe risk. mainly they may go off and congress may fail to respond. the drift may be worse than it was and so they can end up being counterproductive along these lines. >> second the goal should not be confused with reducing year to year volatility in the

government. in the face of that they may be attempting to make it unresponsive to changing circumstances and thus say reduced year to year changes in the fiscal balance. that can be done through the targets like a driver's licensed budget. this is less responsive than what we have today. they can diversify risks and they can respect adjust if growth exceeds certain levels while adapting. i should also say that i at least am optimistic given the fact that a number of these have been used and be expanded but

they are in terms of uncertainty. >> iment to note what i should have said before you really only a taste of them from the website. we have written a paper which we post ed that posted that expresses it. i'm pleased to be joined by three people who have practiced thissard andancy.

spent 33 years in the federal government and 25 of them in the senate. most of that time on budget issues. we don't hold cigna against him. he has been director not once, but twice. gene had something to do with smm every every time he has been in washington. jim cooper is a democratic congressman from tennessee something he has done for all of eight of 32 years and not afraid to say what he thinks even if

it's not popular. i feel like there is a question hanging and maybe i can start with you. >> you are getting me in further trouble, but it is the profession of c students. that's on a good day. i am strongly in favor of alan's bias towards savings. that is very unpopular and almost a hall mark to be able to

delay gratification. what have you done for me lately? >> i started my career and in charge of making estimates for the food stamp or snap program. i twped these models and i had the estimates going forward and called over to the chairman's office to explain. i wanted to be pressing with the metrics and i have the standard that my point is the estimate and what it looks like. he looked at me and said young man, we don't appropriate in changes. that always led me to the conclusions that at the end of the day it's a number.

are they incapable of doing it? alan's paper and that discussion, i think we really have to seriously about it in the last 25 to 30 years and in terms of what happened if fiscal policy and on one hand i think much of the premises of what we did was precautionary and you can project not only that you might have a downturn, but you might need a stimulus of what

that does.ñ a bias has been one of the most harmful things. i believe that we have great central bankers and we are seeing an effort by bernanke and fiscal policy has failed in europe and trying not to fail in japan. while we are the least ugly kid on the block, we under perform. i do think we have to really take a fresh look. i think that one has to take

more seriously the do no unnecessary harm mandate. they have been a terrible thing and where they do protect the low income programs. how terrible it could have been to cut food stamps and found out only a few years later that there was not even a case that you needed to do that.h3r9çhdññ

you have people who are not well off. that has to be built in. >> let me turn to the paper. jim cooper when do these legislative mechanisms trigger expiration dates and alarm bells and work best or are useless. >> they help design legislation from itself. that is a big job. i would suggest that david's excellent paper is the first part of a trilogy.

congress is responsive and prompt to the private sector. it's only public goods like policy and social security and medicare where we are slow and you can divide those because many firms benefit. it turns out we are responsible to med tear and really only the public good. the taxpayer is the only person in america who is not allowed to pay. special interest up and downwash wash pay us for performance every day. they pack contributions and speaking opportunities and other things. why is the taxpayer there? partly constitutional.

the yet that everybody makes the same. that's nice. but it's a handicap because you get what you pay for. i wish that he had put in his paper behavioral economics. the type one brain have determined so much activities. i'm not implying that my colleagues are reptiles but we have to speak to the audience and people want a fick 86. people can say more drink or

feel better, but we have to go with the of the demagoguery. we don't even have to show up for work. the trend is such that the most popular way to run is to promise you will never visit. we extend all of those and we should do the triggers and have some and maybe we can do more. we are a well-oiled machine.

even in this latest when it comes to meeting. in don't we get our demagoguery as well as the folks here. >> whether or not it is your folks in the sense of where he has it right and wrong? where they earn the public interest and whether they become like this sequester. >> let me raise a little question about the whole uncertainty framework. then i want to go on the cable to discuss that goes to what may be more purnt. as david said, we have finders of triggers that never saw the light of day.

is it about uncertainty or is it morem democracy and our democratic system building in ways to respond without having to force ourselves to go so repeatedly through the system thatsq1 we have trouble responding that. is different than the uncertainty. it's more intelligent people recognizing like a boss might recognize they travel a lot. that might be a bottleneck that

is difficulties of people that are coming and doing something. the stroppingest case is on the cyclical issues. i think that the reason i said that, david just gave the unemployment unemployment. over 5% in the first quarter. about 7%. i don't know when that was adjusted. the question is was that the reason? my memory is you couldn't get a dollar more and get 60 votes in the senate. it was much more the lit cal constraint. i think people would have taken more if they thought they could.

we have a different share of medicare and medicate based on poverty. having that share ke creasing. the federal share beinglhñ( increased is an a.m. adjustment. post people know they need to be done.nó95ñ>o6 on the conservative side, it's a way of turning off.

not that you are having fiscal problems. i think that's where the triggers are the most 3r0789ing. they play an important role and i think it's much more in the action and overcoming constraints than the uncertainty. to design the trigger with the percentages of revenue and cuts and the distribution seems to me as difficult as actually casting social security reform. if what you are trying to do is overcome the constraints i don't know that that adds much to it. on medicare, i think the delegation more often about expertise than uncertainty. it's crazy sometimes for any of us to get to the detail and the

different market baskets. they have a delegation and what we used to say is we called them mad, newtually assures disruption. they did something so eventive they would let that happenq zy or nobody could be stupid enough to let that happen. we were trying for most of the time to have a different type of trigger which now probably has a greater tas. they are not done in a pretty way, but they are liveable.

people may be more prone to not force everybody back to the table. will it be more liveable? the last point that is most topical right now is to have a single place that expands in the short run and the response you get from intelligent legislators, we don't trust that the second half will happen. i think having fiscal and economic policy is having the

mechanism that gives people confidence that they can perhaps put their foot on the accelerator and they will happen eventually. the most relevant is health care. whether you read the charm you slow down in health care that is dramatic. that has to make life tougher for members of congress. now you are doing difficult policies on health care entitlement that you are not sure how much you need to do. that goes back or you could take the bet and encourage people because you know inxjs= 2025 if you

don't, something is going to happen. that to me is the most promising area right now for helping us deal with policy. encouraging policy and behavior and the tough choices will be made. >> when are they creatible and are are they excuses? >> delegation and auto trigger. i hate to sound like a scrooge here in the holiday season, but the only there that i have come around to believing is the auto triggers. we have a lot of those. we are out there. i find that the other kinds of

approaches and the triggers thatgox you call such agreement, they choose a reason for having them. there was a thought that they wish congress would act. they did act and not the way you wanted them, but they did act. i just want to say in terms of some things james said, i think the best approach here is you said it put it into effect.@#se you ever really focusing on the

stimulus bill. what you would like to see and i'm supposed to be observing the question is whether or not we should have triggered in. i just have to raise a red flag here that i think one of the worst possible outcomes for the long term is to have some sort of automatic increase in discretionary spending. i'm not there. so some of the issues that you raised -- >> if so -- because once they are put into effect, once the discretionary spending goes up you say we turn them off. i have heard that before. and second of all, i look at your charts you said that when unemployment started up i saw a lot of peaks in unemployment that didn't lead to a recession in the chart that you put up here. i think there's so much

uncertainty out there that i would rather focus on two other kinds of triggers that congressman cooper's position on behavioral. i wish we had a trigger that said no budget no pay. i would say also if you don't pass appropriations by the beginning of fiscal year, you have an automatic continuing resolution at last year's level. wouldn't be in the mess that we have been in the last week. those are the kinds of triggers i think have an impact upon congressional decision. >> i want to mention the ones i mentioned were on the mandatory side. they do go on and off. if unemployment goes down, those are areas where as i said they would go up automatically, go down automatically. you wouldn't put congress through having to get it exactly right. we require congress to vote when things go bad.

that could be automatic. if that's what you mean. >> yes and medicaid are the two places that we have in place. we already do it to a degree expanding it further making it automatic automatic. i think would make a lot of sense. i think it is quite -- i think you could make the case because it turns off and it doesn't have a moral hazard component. i can see that being a more a plausible that could actually happen? >> let me respond and then we'll have time for a couple questions. >> on a few issues, first, when it comes to the question of triggers and stimulus and for trying to stabilize demand, so actually i think one risk identifies you trigger things on. first i note that we actually in terms of jumps in unemployment we could predict when you are entering a recession and large jump in unemployment highly associated with recession.

if you draw a line and increase in unemployment rate that has occurred and occurred in two points. you can actually whether using that index or another index there are ways of targeting periods of recession. second, looking historically while i agree that there have been some provisions that have stuck in law when you put them in place when it comes to stimulus policy at least on the spending side seems like we tend to have the opposite problem certainly in the last recession. so we had the turn off of state fiscal relief while expended once not further extended. when it came to infrastructure spending which had a slow spend out and did not have dramatic cliff but was not further extended.

it is actually not clear to me that if we had types of triggers which did increase infrastructure spending automatically during periods of economic recession that that would necessarily be overwritten. i think there is a chance they could better improve economic performance without congress not allowing them to go back down. vp+÷ so on a few other points when it comes to the entitlements i completely agree with gene that when it comes to trying to build these kinds of triggers into social security it looks a lot like negotiating a final social security deal. in fact i completely agree. xhkñ÷ i think this would be built into an actual social security deal where there is an agreement on the parameters of the deal and say we want to stick to that deal so we build in a trigger to stick to the deal. if it can happen in '83 it can happen if we reach that again. the only other point i want to make is when it comes to delegation and the entitlement programs part of it is expertise and part of it is knowledge. when it comes to delegating in medicare some of the experimentations they have going

on at the centers for medicare and medicaid services we didn't know what worked so we empower them to potentially ramp up that experience and implement them - broadly. that is an example where we have uncertainty as to what works and congress will not have to revisit the issue and delegate the ability to fully ramp up reforms if it actually does work. >> we can take two or three questions and then we will answer them and then we have to break. in the middle there and then in the back. >> spent about half my career in j÷ federal and government and fiscal institutions and working with states. one correction to gene spurling. medicaid match rate has 4 1/2 year leg between date matched

and year used in. it is pro-cyclical. that is not something i recommend. i would like to get you to measure the difference between big bang theory of stimulus. >> i was saying the same ning you were. i was trying to build in having a share based on what is happening in the economy at that moment and not on the poverty data. so we were actual in vigorous agreement. >> the difference between the big bang ask the trigger response. gave most of the state assistance through medicaid that states may not change their eligibility in which they didn't which meant states cut back on everything else including education which now has become the baseline for education for the future. this is what happens to productivity. is this a good thing? >> one in the back? a couple of rows back. zdíz and then one on the left side.

>> really a comment. that is there really is -- when you talk about the entitlement program very little uncertainty about the cause of the problem. in fact, there hasn't been uncertainty since the 1970s when the birth rate dropped. we have known about it for a while. it is sort of a decade-long certainty and actually goes forward. in fact, '83 reforms probably would have benefitted from a longer horizon because we know what was going to happen and there is very little uncertainty about that. that applies to a certain extent about medicare and the other demographic dependent programs. i'm not quite sure i understand why the triggers and automatic items really are relevant for those programs. >> there is one over here. >> thanks. i just wanted to make a cautionary point about fiscal

policy. this last down turn was very different from previous recessions. previous recessions from peak to trough lasted ten months. on average. because of that i think fiscal policy rightly got discredited as a short term stabilization mechanism. in previous recessions monetary policy was more effective because you didn't have the collapse of the financial sector. so this time was different and it paid to do fiscal policy. it was clearer. it was going to be deeper and longer and monetary policy wouldn't work. the idea of using fiscal policy in general when the unemployment rate spikes or the gdp number is

negative runs the risk that the lags are such that by the time you have actually hit the stimulus the economy no longer needs it. >[fd >> we have to get back on schedule so very brief responses and we will take a short break. >> so starting off. when it comes to social security and uncertainty you are right that part of the challenge facing social security system in our fiscal situation more generally is well known and now we have baby boomers working their way through the system. with that said there is considerable uncertainty as to exactly what the picture is going to look like going forward. even as i mentioned the optimistic scenario that the trustees put out shows the system remaining solvent on a continuing basis. that is one possible scenario. there are other scenarios, as well. the point is while we do know some points of the challenge remain considerable uncertainty about productivity growth and rate of immigration, et cetera, factors that go into effecting social security balance.

this system adjusting to some of the factors. when it comes to stimulus, marty's point is a good one, the main fact when it comes to fiscal policy that more things be automatic and not have a significant lag. one could say and you may end up with federal policy and in the situations where you need a significant discretionary stimulus you may want to then -- and those things build in additional types of triggers. >> we have a time constraint. we will take literally a five-minute break. i will start in six minutes. if you want to hear the good ,2p֖ stuff you have to be back here in six minutes.

the 114th congress gavels in at noon eastern. watch live on c-span 2 and track the gop-led congress and have your say as events unfold on the c-span networks, c-span radio and c-span.org. new congress, best access c-span. you can join the conversation on what congress should focus on over the next two years on our facebook page. today's question is, your top issues for the 114th congress. here are a couple of the comments so far.

carol brown suggests rebuilding our country, our infrastructure is starting to fall apart around us. dave russell writes, attack the debt, even if it means a government shutdown. there isn't an election for nearly two years. let us know your thoughts on facebook.com/c-span. also taking part in the brook. ings hutchens forum was office director dougless. his remarks are just under an hour. one of the things we have tried to do in this panel in which we intend to do ink t the future is to use the opportunity

to get views from other countries to see what we could learn from what they do and to be sure so we can continue to tell them how they ought to run their place much more like we run ours even though we seem to have lost a little moral authority on that lately. and the other thing is that we have tried to bridge the academic and the practical here. this is not the national bureau of economic research. we are not afraid to say what ought to happen. but we also think that it is important that policymakers hear the best ideas of academics but also that academics learn something about what it is really like to practice policy here in washington. so for the final part of our event today we have invited two people who are really more qualified than almost anybody else to talk about the question of uncertainty and how policymakers react on it. we start with the director of british office of budget

responsibility in october of 2010. he was previously with the institute for fiscal studies in the international monetary fund but he spent a decade as a journalist with the london independent before finding more productive ways to use his time. after he speaks he will be joined by doug elmandorf and among his previous activities was serving as senior fellow here. they will join me on the stage and we will invite both previous panelists and the audience to ask questions or make contributions. >> thank you very much. good morning. it is a great pleasure to be here and see further reminders that there is productive life after journalism.

what i'm going to do today is give you a little background to how the relatively newly created obr is trying to address some of these issues in recognizing uncertainty in the goals we have been given and how it is reflected into policy. let me start with a brief institutional introduction for you. we are an independent institution created in 2010. we are accountable simultaneously to the executive and to parliament in the uk. we have been given four main tasks to fulfill, the first of which is to produce five year ahead forecasts twice a year for the economy and the public finances. and secondly, to use those forecasts to judge progress towards the explicit fiscal rule "átju set itself. that is one rule in terms of a particular measure of structural budget deficit and another rule for the trajectory of debt to

gdp. that discussion of uncertainty should be couched in terms of what that is telling you about the government's progress. the third thing is scrutinize scoring of tax and in particular welfare and social security spending measures and then we also assess the long-term outlook for the public finance as much as cbo does looking over a 50-year horizon and what we can learn from the balance sheet. putting that into u.s. context and if you think about key differences from this structure i guess one point to note is that all of our analysis covers the entire public sector and not just the equivalent of federal government level. we are confined only to producing projections on the basis of the current policy of the current government. we are forbidden from looking at policy options. and the third thing is that the

executive, the administration in the uk does not publish its own projections. we are the provider of the official projections and for the government to comment on what we say and not comment on the figures they produce. in that context thinking about how uncertainty coming into our work i think in a background it is important to bear in mind that in budget setting in the uk the executive is very powerful relative to parliament compared to the relationship between the administration and congress here and the treasury department in the uk is powerful relative to cabinet departments. the rationale for creating this was very much to try to remove politically motivated wishful thinking from official forecasts rather than particular options. when we set out on this task the key objectives increased transparency and to emphasize uncertainty. given our need to produce a

highly disaggregated fiscal forecast transparency necessarily means a lot of detail and a lot of point estimates. why emphasize uncertainty? first policy should reflect uncertainty. we can offer richer assessment of progress towards the target. the first person to run the organization permanently and i didn't want to tie the success of the institution to the accuracy of our forecast. we address uncertainty in the way that we do our work and our outputs both in narrative qualitative sense and quantitatively. in the narrative it means being as explicit as we can in the forecast presentations and publications we do and therefore

the implicit risks. the idea that monetary policy will evauchbl in line with. we also take care to identify what we see as being specific economic risks that may effect the fiscal forecast in any particular presentation we do so what happens if instability reasserts itself. what happens in the new taper tantrums in terms of the global outlook for unloosening of monetary policy and big ) uncertainties of what is the path for productivity and real wage growth. in the uk more than here one of the key puzzles is why productivity growth has been so weak and whether that will reverse itself and how. we then in addition try to identify specific fiscal risks conditioned on the economic forecast. a key one recently is to look at

what is driving effective tax rates. so the actual tax rate or the amount of revenue you are raising per pound in our case of the particular tax base you are talking about. one of the reasons for our recent forecasting errors is not that we overestimated the aggregate size of wages and salaries of labor income but also the average tax rate on that has been less than expected partly because of the distribution of pre-tax wages. other things that we look at will central and local governments stick within the budgets for discretionary spending. this is a set itself. and then finally to cbos work what uncertainty we see around the scoring of particular policies. and we provide, we don't provide what chuck would like a

confidence, explicit confidence interval around the estimates from particular scorings. what we do is provide ranking from each policy that we sign off based on the quality of the data that under pins it, the nature of the modeling you have to do to get to the estimate and the nature and uncertainty surrounding the behavioral response. this is a subjective exodus but listing the sources of uncertainty under lines how difficult it would be to put a confidence interval around a particular estimate. in addition we try to illustrate uncertainty quantitatively. we quantify the uncertainty around our central forecast and with particular reference to the chances of hitting the targets that the government has set itself.

i guess one thing the combination of having an independent forecast and explicit fiscal rules set up by the government is in a sense we smoke out policymakers receive preference of how much margin for error they want to put in in achieving targets against the forecast one way of saying how much you want to overachieve. we use three main techniques to illustrate uncertainty implied by past forecast errors, sensitivity to key economic determinants and scenario analysis. the key thing is not just to say the projections are uncertainty and what ways does it have to be wrong for the targets to be imperil. just to put a few pictures this shows you a fan chart of probabilities for a particular measure of the budget deficit. we do mechanical sensitivity analysis so you take your central forecast for the targeted fiscal variables and say what

difference does it make if there is more capacity in the economy, faster or slower gdp growth and in the central forecast what if government borrowing costs are higher or lower. i think in that context if you were looking for one major source of uncertainty both around the achievement of the targets and around our fiscal projections more generally it's the uncertainty as to what is the level of potential gdp. what is the path of real gdp to which you expect the economy to tend if the central bank is doing its job. if you can see here almost everybody assumes the potential output is lower than the path of the crisis but a lot of uncertainty around how far low. scenario analysis to highlight particular key debates or critiques of our forecast that may be put forward at any given time and illustrate some areas that the public may misinterpret. what difference would it make if productivity growth was higher or lower or monetary policy tightens more rapidly than you anticipate, the fiscal

consequences depend on because the economy is doing better than expected or whether it is driven by risk or something like tñárçr we can add additional information from measures of a balance sheet. the uk has whole of government account prepared on commercial accounting basis including list of contingent liabilities, liabilities that are assumed not in the central forecast to crystallize but whether nonnegligible chance, things like public health care system, challenges to decisions on tax collection. there is also a list of unfortunately unquantifiable reliabilities which we note there is not a lot to do with that. my favorite is apparently the government has a requirement to return the land for the channel tunnel to a suitable condition if it ceases to operate. quite what you are supposed to

do, fill it in, plant flowers around the entrance or quite what i don't know but it is in there anyway. in terms of the long term projections most of it so far on the medium term i think the story here is very similar to the projections that cbo produces. it allows you to take on board demographic change, calculate debt trajectories and have sensitivity analysis based on different population paths, different relationship between interest rates and growth and look at specific interests on the tax revenue side for example whether there are long term trends in the average tax rate. let me note we see as the flip side of emphasizing uncertainty is that creates responsibility on us to learn from our forecasting errors after the event.

we have specific publication we produce each year separately from the forecast in which we look and see how much is because policy changed after the forecast, classification changes, how much because economic terminants were different or because specific fiscal forecasting areas are different? it is not reassuring in the sense that you end up being wrong but this year we were wrong three years ago from completely different reasons we ]éq9ñ were wrong last year. let me leave it there. happy to take questions. thank you. [ applause ] >> there we are. okay. aq

great. thank you. thank you all. thank you. i will start. one quick correction for david. i have been director for six years, not eight. it may feel like eight but only six. my colleagues and i are acutely aware of the uncertainty of the budgetary and economic estimates we provide to the congress. we view our estimates as representing the middle of the distribution of possible outcomes. we frequently explain estimates that way to members of congress and their staffs and we discuss risks to our estimates. i take chuck's concerns seriously. in my experience the members of congress are quite aware of the 2aimuejt in fact, it is not too uncommon to think we got the sign of the estimated effect wrong in addition to the specific magnitude.

i think the issue at hand is not whether we should communicate the existence of uncertainty but whether or attempts to discuss and to quantify that uncertainty provides particular additional help to the congress in the choices that it makes. we have worked hard in the past few years to quantify the uncertainty of more of our analysis. there are important limitations currently on our ability to do that quantification and to help legislators make effective use of such quantification. let me begin by discussing the things that we do and talking about the limitations we see undoing more. we quantify the uncertainty of our estimates in a number of contexts. first, when we estimate the macroeconomical effects we regularly provide a central estimate and a range. the ranges allow for uncertainty about the responsive labor supply, effects of changes in budget deficits on national savings and other factors.

and these ranges are intended to cover roughly two thirds of distribution of possible outcomes. one example is estimates of short term and long term economic effects of alternative paths for federal debt. second example is in our long term budget outlook. we regularly include projections for alternative scenarios. in july we showed what would happen to the budget if four key underlying factors, interest rates and growth of health care costs differed from the values known -- similarly, we published additional information for long term projections for social security that includes range of possible outcomes based on historical year to year variation in key demographic and economic factors. this suffers from weaknesses that peter identified for similar projections from social security actuaries. a third form of quantification occurs in some analysis of specific federal policies. this example is estimates on

effects of unemployment, of raising minimum wage and provided similar ranges for estimates of the effects on federal budget of adopting a more competitive system for medicare. a fourth and slightly different way in which we quantify uncertainty is to analyze how actual outcomes differed from our efforts. we do a recurring evaluation of our economic forecasting errors, which i've shown here. we have an evaluation of our revenue forecastings coming out shortly. we published as obr does rule of thumb for how much budget outcomes vary forgiven various. vast majority of cbo estimates. i think there are three principle reasons for that. first, the congressional budget process requires point estimates of the budgetary effects of proposed legislation. bill hogan put this in a more pithy wayen i will. they provide allocations to funds.

the house and senate budget committees track budgetaried. the statutory pace you enacted in 2010 and various parliamentary rules of the house and senate are enforced using point values. a range that encompassed some values and others that did not would not be useful in following those procedures or enforcing those rules so we need to provide point estimates. one might argue that the congress should change its approach to explicitly reflect uncertainty but developing comprehensible figures seems quantity daunting. a second reason cbo usually does not provide ranges. one obstacle is that most of the models and estimating techniques we use don't readily yield

uncertainty. not probability models. another object that cal is underlying research on which we draw generally provides only limited information about uncertainty. for example, our analysis of the effects of raising the minimum wage drew on dozens of studies so we could form an educated sense of uncertainty related to the time, the place, the modeling approach used as well as to sampling uncertainty. our analysis in other areas, for example, the effects of prescription drug use and other medical spending was based on just i handful of studies done in that area. forming a sense of uncertainty would have been quite difficult. further related challenge is the lack of time. we're often rushing to finish

analyses and especially the cost estimate before congressional action on an issue is expected to occur. doing the additional modeling or gathering the additional information needed to quantify g, uncertainty can take considerable time that would delay the release of particular estimates and would also delay our turning to other analyses or estimates of other proposals. and the third main reason we usually don't provide estimates is we're still developing ways to help legislators make effective use of our quantification of uncertainty. is part of the challenge is providing ranges for estimates sometimes muddies rather than ny enhances general understanding of our analysis. for example, when we report ranges people who like numbers to be big tend to pick the top end of the range. people who like that number to be small tend to pick the bottom

end of the range. that can make the public discussion of our analysis quite confusing and we have a limited ability to clear up that confusion. another part of the challenge is it is often unclear how legislators might respond to the quantification of uncertainty. and we in the analytic community more generally have developed only limited guidance in this area. surely it is useful for legislators to be aware of the uncertainty of budgetary and economic estimates. but as i said, we think they f ) are. the harder question is what else might they do we particular quantification of uncertainty we could provide. some legislators might want to adopt policies only very likely to increase or decrease a variable of interest or hit a particular target. this is the example robert gave in the uk. if you're trying to hit a particular target for the budget deficit, you might want to pick policies that have a certain probability of actually hitting that target. we did an analysis, for example

of changes had resulted in savings for the department of defense. we concluded it probably had. similar analysis could be helpful but particularly in those specific cases where the congress was trying to hit a particular particular. as an example, lawmakers might want to adopt policies with a smaller variance of budgetary effects to reduce risks to the federal budget. for example, understanding the extent of uncertainty about future federal spending that arises from uncertainty about life spans might affect whether policy makers would want to index eligibility ages for certain programs to life spans. another situation legislators might want to adopt policies with a larger variance of future budgetary outcomes as a means of experimenting to identify the best policies. i think analysts can help sort out those sort of situations but

it's a complicated problem. as a final example, legislators might want to respond to greater uncertainty about long-term. as allen argued. on the other hand, ledge tors might want to respond to greater uncertainty by focusing less on projected long-term outcomes at some horizon. more research, like allen's paper and other papers being presented here and being written elsewhere can help guide policy makers in that issue as well sm1vy policymakers driblgt use is limited at this point. that n turn, leads us to devote only a limited portion of our time to reducing such estimates. there are also important limitations on our current ability to do that and help legislators make effective use of that limitation. thank you.

>> thank you very much, clear and interesting presentations. i'm only going to have a few questions and we can make encourage people who were on earlier panels to join in the conversation. robert, tell me if i got this right. do you have to express an opinion about whether the government has a greater than 50% probability of meeting its fiscal targets? >> yes, we do. that, therefore, requires us to prod;@"ip oint estimate median forecast in order to say whether they fall on the right side of that. yes, we do. it's interesting the formal fiscal targets are not the only thing politicians are interested in aiming at. we have two formal fiscal targets at the moment and one isn't achievable and the other isn't binding. so, actually, the government is focused on other things like the particular date on which it expects to balance the overall budget, what the path of the

deficit is on a year by year basis. you're absolutely right. that's the formal role. it's not necessarily the thing policy makers are most worried about most of the time. >> i see. there's been -- deb talked some about what this is like from the policy maker side. do you find -- chuck suggested maybe the british are just different, that they're actually interested in this uncertainty or is this just make you more popular with the manski crowd? >> i think they are. it is this combination of the fact that you have a clearly defined target and independent forecasting because essentially you are requiring the policy makers, even if only by reveal preference to say by how much do they want to overachieve the on/ central targets which is essentially asking them to say how much built do you to want take in the future. i think in terms of the impact - on policy changes, we have adopted an index of the state pension age to life expectancy,

the government sent out a general principle of people expecting to spend a third of adult life in retirement but there will be a committee every five years to translate this into a precise set of ages which allows you to take into account the fact that may not be what you want to be aiming at that point in time. i think there is a general willingness to embrace this sort of thing. obviously, the bank of england has been doing this for quite a long time that it presented for monetary policy. you know, the issue about -- it's important the issue on uncertainty doesn't give you an excuse to dodge the responsibility to come up with your best judgment. the bank of england used to be -- measure vin king used to be so resistant to providing estimates. the consequence was members of our former professor rushed home, photocopied the inflation report, blew it up as big as they could and tried to back out the point estimates from the

middle of the probability distribution. i think there is a general acceptance, but, you know, real politics has not been suspended in the face of it. >> it seems to me that a couple things, one is -- one reason to talk about uncertainty you can say to the members of congress, i told you so, even if you weren't looking and you can cite the day in which you testified before the budget committee. but by giving long-run forecasts are you conveying something to policymakers that we don't really know that might n a different state of the world than we have now, lead hem to do something that's harmful? >> we've wrestled how much weight to give to different forecast horizons. although our long-term projections go out 75 years, the report we talk about talks about the next 25 years.

the 50 years beyond that we have greatly reduced the emphasis on recognizing that we have less and less knowledge about what will happen to that horizon. takes the projection over the next 25 years because the challenges that the country faces are not just short-term challenges and the policy responses that are being weighed are not just short-term responses. they are gradual responses, in general. some cases responses that wouldn't have much or even most of -- much if at all until when moved beyond the ten-year window we focus on. >> it's very important for us to look out over that -- over a longer period than the standard ten-year budget window, but we are very transparent in the greater uncertainty as we go out. and in this last long-term budget outlook this year, this range, has a debt to gdp ratios

in 25 years that vary by factor of 2 to 1. it turns out essentially all that range that we show has debt which is as large or larger which is unusually high level by standards of entire history of the country and i think there's important information in that. >> finally, do you ever find kñ yourself thinking, i wish members of congress wouldn't pay so much attention to the scores in deciding whether something is a good policy or not? >> yes, i think this is a standard view among my .3m@ colleagues that too much weight is given in some cases to hitting a particular number that we recognize as uncertain. the legislators recognize is uncertain. i think that's a by-product of ongoing legitimate concern about the trajectory for the federal debt. we often think in particular circumstances or estimates are

given too much weight relative to particular considerations. >> let me ask if any of the authors or respond epts or earlier panelists want to weigh in. >> i'm so quiet. you can't hear me. >> we want to make sure we can get every word down and we can hold it against you in your next confirmation hearing. >> it won't be you holding it against me. no, i just want to follow up on doug's question there, which is -- and maybe it's for both, which is there probably are a lot of policies that you do think are good policies but part of the problem may be that when the congressional budget office says no savings, it's not reflected as we're not certain enough about it to do savings but that it would produce no

savings. and so it undercuts. and i think my question is more on kind of health care reform, where my guess is there may be areas where you could say, here are a number of things that probably move in the right direction, might do good. we don't have the certainty. one question is, is there a way of expressing without a score, an opinion or certainty these things could have a positive effect because you may have things that are -- become almost delegitimized by lack of their score, although i don't think that would necessarily be the intention. >> you raise an important pi1÷ gene, and i'm not sure what we can do about it. when we analyze, for example, the independent payment we wrote a long discussion about how this is a group that would have certain avenues open to it and it might look for things it

could do and it wasn't clear what the group would look to, wasn't as successful as they would be. our estimate of its budgetary effects is explicit problemistic estimate. other proposals for specific changes. we analyze, for example, a few years ago the effects of raising the cigarette tax on the federal budget has direct revenue consequences. also would make the population healthier. and many people expect that would improve budget outcomes. we did an elaborate analysis that show it might not improve budget outcomes, factors going both directions. i think you came through very clearly in our work that some things that seem good, at first blush, might turn out to be good or not good. i think that comes through what we're doing. i don't know how -- what else we can do. i think particularly in cases where it's hard to understand what the right point estimate of effects of legislation would be, to quantify the uncertainty would be much harder. what robert talked about, which i found interesting, were qualitative characterization of

the uncertainty of proposals. that's not something that we do and i personally at least haven't thought about it, although maybe other people have. i see bob, others in the audience, who may have thought about this. i'm not sure if that would help or not. i'm not sure what our capability would be. we publish 500 written cost estimates this year. we probably did ten times as many informal estimates. i don't want to sound like i'm just focused on running the trains but i am focused on running the trains.r&0f we need to find mechanisms that are feasible and very short -- very short timetables for a wide range of legislation. >> that was the one you showed and you had marked in yellow. is that something you were asked to do or is that your invention in order to avoid the problem that doug finds himself accused

of? >> i stole this from the australian parliamentary -- >> we'll have that next year. >> we developed it beyond that. i think what we previously had done, we have a slightly odd system in the uk that the government -- we publish the forecast. the government publishes its own scoring which we have scrutinized before they announce them. they know whether we'll say we agree with this or not and now having introduced this thing about what they'll say about the uncertainty around it. previously all we had done was to say, here is the list of 80 measures announced in this ú3:%ñ particular package. we think the following three are particularly uncertain. you put a little language around that. i think we felt it would be much more useful to have a more systematic and transparent way of doing this. people are, of course, quite interested if a particular

package consists of a lot of well-identified giveaways to the general public, highly certain, paid for by highly uncertain -- >> we never do that here. peter diamond. >> any scoring rule is open to some kind of gaming. and and with presentation of all the details, the gaming is evidence so people who dig in know about it and yet the scoring rule plays a big role. a lot of our concerns has been, as it were over too much. i wish you had included somebody today who did the reverse. what was life like before the cbo? what was it like without scoring when everybody was making up their own numbers? i certain remember back -- the issue, social security, before we got indexing for inflation, so the issue was how much should we raise benefits this year and what are the political pressures and what kind of limits can we

set on it so scoring set the limit but it was all made up scoring because the system had no automatics and the scoring b@ú$ was made up -- here's a formula. here's a number. so congress could hide behind the number when beneficiaries wanted bigger things i think scoring really matters, even as imperfect as it is. >> pass the mike back to representative cooper. >> i love the british way of whole of government accounting and also on a commercial basis, which i take to mean a cruel accounting. here austerity was caused by state and local employees as federal government was trying to do the right thing. so few of our colleagues have any idea of what a fiscal gap is, much less how large it is. >> the other panelists? steve goss?

please. >> thank you. first of all the point of scoring before cbo, at least on social security, there was. peter, we'll have to have a longer discussion about what happened prior to the automaticing because there was serious consideration of that. my question is for robert. robert, you took my breath away with one of your drafts where it looked as from your 2008 projection, i hesitate to say forecast, to your 2014 projections, you had a lowering of your estimates of potential gdp for 2020 of approximately 15% with no apparent trajectory in the more recent projection to be closing that gap back. so, it looks like you've taken a 15% bite off potential gdp, at least through 2020. is that your presumption going forward and doesn't that sound like an awful lot of uncertainty? the question is, do you think your 2008 was that bad that now you really had a need for that much correction? >>. >> the first one to make 2008 projection was before we existed

so that was the published projection when the forecast was still in political hands.ré although that said -- that said, i would not say that if it wasn't in political hands, you would have been very much closer to where we are now. but i think this has been true since the crisis hit in the uk. and i think it's more so here than it was to start with. the whole notion of how big is the hole your fiscal consolidation program has to fill in is entirely wrapped up in this potential gdp has moved down from the projected path. basically for the uk, you multiply that loss by 0.7 and that tells you what your increase in the structural debit is. i hope the chart showed, there's a lot of uncertainty amongst other people looking at this as to how big this decline is. and for large bodies, very

difficult to explain quite why it should have been that large. but basically in response to the fact that, you know, we've had a weak economic recovery, the business indicators are saying there's not very much spare capacity, you've got a very weak productivity performance, all of that suggests there's been a big4éfq÷ on the other hand, that view is challenged by the fact we really still don't have any wage growth to speak of. so, can that really be that narrow? but i think it's something that's important for us to be up front about, you know, we're in the middle of a ten-year fiscal consolidation program based on filling a gap between one number you can't estimate or, indeed, observe directly, and another one. >> how about potential here? >> so, we revised down our estimate of potential for 2017

by 8% or so relative to our projections in 2007. we think that some of that is due to recessions and weak recovery specifically. we think some is due to developments up through 2007 that just weren't apparent to people at cbo in 2007. of course, as the period since the last cyclical peak goes on, as the recovery occupy more and more of the time we're looking at, it's harder and harder to disentangle what part of the revision is due to developments in this business cycle or just a mistake in assessment by us before the business cycle. the down revision is substantial. and i think that does provide appropriate caution about taking a particular -- a particular point estimate at face value. >> i'm a journalist, so i don't know how productive this will be, but -- >> don't start before you ever ask. >> this is for mr. allendorf. i'm more familiar with jct's estimating methodology but the

big debate has been over whether or not to account for macro dmik factors or tend to project them. right now they're mostly just basing it off your agency's baseline. and i was wondering what your thoughts were on whether doing the -- accounting for these factors would reduce uncertainty or increase it? >> let me quick up some confusion about dynamic scoring our scope for incorporating macro economic effects is limited by the fact that estimates of changes in tax provisions are the responsibility of the staff, the joint committee on taxation, not a cbo. a second thing to emphasis is that our estimates and jtc's estimates incorporate lots of kinds of behavioral responses. they're not static. but they don't generally incorporate macro economic ó effects. another thing to emphasize, though, is when legislation is being discussed or proposals are

being weighed, that would plausibly have large macro economic effects. then and if time allows, then we, jct, do estimates of those macro economic effects. we do this every year in analysis of the president's budget. we did it this year for our analysis have chairman ryan's budget plans. we do it now every year in our long-term budget outlook. jct did it this year for their analysis of chairman camp's tax reform proposal. so this sort of work is not knew to us. we spent a lot of times in our last few years on improving the models of the effects of fiscal policy on the economy and in writing reports that spell out how we do that sort of modeling. so, doing those kind of analyses for legislation that would plausibly have noticeable macro

economic effects and when there is time in the congressional process for us to do it, that's fine. we're doing that now. i think the question for members of congress is how they want to use that information. that's really up to them to decide what role they want to give these macro economic effects and their feedback to the luj when they consider legislation. but that's appropriately up to the members of congress to decide. >> another question? yes, henry aaron. >> this is a question for you, doug, with respect to the alternative fiscal scenario projection which in many quartz is regarded as, perhaps, more realistic than the extended baseline.

actually, one of my questions refers to both projections. during the first session when commenting on the projections of social security and medicare hospital insurance, peter diamond observed that the probability that scheduled benefits would be paid in both programs and current revenues would be collected in both programs was the probability of those joint events was about zero. yet both extend the baseline and alternative fiscal scenario assume that that zero probability event will occur and is fed into the projections. my question is, what do you think about that? and are you bound by instructions or statute to estimate in that way or was this a decision of cbo. the other question relates to the alternative fiscal scenario

only, which is the assumption that revenues will continue beyond the ten-year budget window at 18% of gdp. which is based on historical analysis but historical analysis of a time radically different in character from the one you're projecting. one in which in the future were we to remain at 18% of gdp, congress would have to continually cut tax rates in the face of projected widening deficits, which even given congressman cooper's evaluation of the colleague strikes me as verging on calamny, something they would not do. so, i appreciate if you would comment on both of those limits of the projections. >> so, it is common, as you say, hank, for people to refer to alternative fiscal scenario as the most realistic scenario. i don't refer to it that way. we at cbo don't refer to it that way because many of the cases we've done this, alternative scenario has shown debt just skyrocketing ultimately. that's not a realistic projection. for our purposes, the baseline and then the extend the baseline scenario are meant to correspond

roughly to current law. we think about an -- the alternative fiscal scenario as corresponding more closely to current policies, one might say, and the difference was particularly stark when the 2001-2003 tax cuts were scheduled to expire because members of our baseline and extended baseline followed current law and incorporating those expirations but many, many members of congress were in their own minds starting from a benchmark in which the existing tax rates were continued. and they wanted to understand the effects of policies relative to that sort of benchmark. we put that continuation of the current -- then-current tax 1x rates into the alternative scenario. so, i don't think it's meant to be realistic. so, peter's concerns about this

don't worry me because i think the alternative scenario is to give somebody what would happen if we continue with social tb security benefits and other benefits as they're currently being paid out and we continued with current tax policy.out, and we continued with current tax policy. now, we are required by law to follow certain rules in constructing our ten-year baseline projections. much of this was written into law in 1985. with some modifications since then. so there are rules that we follow. when there are ambiguities, we consult with the budget committees with how to handle those. that refers to the ten-year projection. the baseline projections we try to follow the current law as we go out. current law is for example, a debt ceiling. we don't assume in our

projections the debt ceiling is not raised. the baseline rules are supposed to be about current law, makes certain sorts of am daysccommodations. social security, there's only if there is money in the trust fund. that's the baseline. we follow that same principle for the extended baseline. the alternative scenario is really up to our judgment about what alternative benchmark, what representation of current policies we think will be most useful to the congress in its deliberations. but the goal is not realism. the goal is to capture what many members of congress think of as the current sorts of things going on in the budget. but there's obviously ambiguity in that and we use our best judgment to be specific. and as you know, we lay out the

assumptions we make very explicitly. >> in the baseline you're required to assume that social security benefits are paid even if there's not money in the trust fund? on the tax question? why 18% of gdp? >> on the tax question what we're trying to capture there is what people might think of as current tax policies. in a broad sense. so what's happening -- what's happened over time is the federal government, as you know, has collected revenue equal to about 18% of gdp on average with a variation up and down of a few percentage points. we would like to show the congress what would happen if they continued with the benefit policies that are in place and the tax policies we have now, what will happen in the future. the fact that those lines, spending revenues, diverge, that is the fiscal challenge. that is exact lip the point. that with the sorts of tax policies we've had before, the federal government will not collect enough money to support the sorts of benefits that we have now, given the rising costs

for health care. those lines diverge, is our effort to demonstrate to the congress why important changes in fiscal policy will probably be needed. and so it's not our job to line those up. it's the job of the congress to decide what policy changes over time might line those lines up better. >> with that, first please join me in thanking these two guys for such a good job. [ applause ] i also want to remind everybody that all this and more, is on our website. as you know, these events don't happen automatically. i want to thank carrie, emily parker, and my colleague lee shaner for helping make this such a success. and i'm no more certain about what the right answers are. but i'm a lot more sophisticated about my understanding of uncertainty. and i hope that's true for you

too. marty and bob and glen would like to take a picture. so if you would come up, i'd appreciate it. and to the rest of you, thank you very much. [ applause ] the 114th congress gavels in tomorrow at noon eastern. we'll see the swearing in of members and the election for house speaker. you can watch the house live on c-span and the senate live on c-span2. with the new congress you'll have the best access the most extensive coverage anywhere. track the gop as it leads on capitol hill and have your say on tv, radio and the web. in the new senate 36 states will have both of their senators from the same political party. 21:bçn states with two republicans and 15 states with two democrats. there will also be a record number of 53 former house members in the senate. c-span recently sat down with a senate historian done ritchie to

discuss the majority leaders in the u.s. senate. >> donald ritchie, thank you for being with us on "american history tv." the senate majority leader is not a position in the constitution. how did the job evolve? >> that puzzles a lot of people. we once had a former speaker of the house contact us and say i was a president pro tem in the order of succession but not the majority leader. the speakers of the house see the speaker as the functional equivalent on the senate side. the speaker of the house is written into the constitution, and the president pro temporary is. no majority leader or minority leader, and that's because the constitution did not anticipate political parties. i think they thought that it was probably going to happen but they didn't want to encourage political parties. so they make no provision for it. and for most of the history of the senate we had no majority leader, from 1789 until 1913,

there was no majority leader in the senate. there were chairman of committees who would sort of take care of things on the floor. there were the chairman of the party conferences who would come out and open things up and close things the way they do today. but nobody sitting down in the front row center seat trying to organize what happened on the floor on a daily basis. and that changed in 1913, when woodrow wilson was elected president. he had a ph.d. in history and political science. he had written his doctoral dissertation about congress. he had very strong ideas about how congress ought to operate. and he also had a very ambitious legislative agenda. so he prevailed on the democrats who had just come into the majority to pick one of their senators to be the leader, to be the person who could take charge of things on the floor. and instead of going to a senior senator, as you would say, for the pro tem.

they had one well known running .lnltz democratic ticket in 1908. this was now 1913. kearn became the first majority leader of the senate. and he functioned just about the way that majority leaders do today. he started out by opening things up during the day, and scheduling things, and closing things down at night. when the republicans came into the majority in 1919, they decided this was a good idea. and so they picked one of their senior centers, henry kevin lodge, to carry on those functions. they didn't actually officially have a title majority leader. they had all sorts of other types, conference chairman and things like that. it's not until 1925 that charles curtis of kansas is officially designated the majority leader. the fact of the matter is that both kearn and lodge functioned the way the majority leader would today. and ever since then, of course the majority leadership has

grown considerably in power. although, if you read the rules of the senate you won't see a lot of mention of the majority leader. a lot of it has just evolved over time. some of it is from -- the powers come from tp@ñs precedence of the senate rather than from the rules. lyndon johnson, who was probably the most powerful of all the majority leaders used to say that his single greatest power was the power of persuasion. and howard baker, a republican who succeeded him a few years later, used to chuckle and say, well, the power to call bills off the calendar is also pretty significant. there's a number of things. and then in 1937, there was a vice president of the united states, john nance garner who had previously served as speaker of the house. and he gave the majority leader the greatest power of them all. not by rule but by a precedent and that was he realized control of the floor was really essential for doing things in an efficient way in the senate, and

so john nance garner said, i will grant the majority leader the right of first recognition. when all the senators are seeking recognition i will always call on the majority leader first, and after that i will call on the minority leader. that means the majority leader can get the floor when he wants it. and that's a huge influence. it has a lot to do with the way things happen in the senate today. that's also one reason you will not see minority senators presiding over the senate. there was a time when both majority and minority freshmen would preside. one day a minority member was presiding and a majority leader sought recognition and that senator called on someone else. it was a little like crack a toe a going up again. there was this huge eruption in the senate. and they said there will never be a minority senator in the chair. except for snow days when nobody else is available, you won't see a minority senator in the chair because that's so essential to

the powers of the majority leader. >> where on the senate floor do we find the majority and minority leaders and how do they work together? >> the majority and minority leaders are front and center. they both occupy the two front row center seats since 1937. the democrats were there a little earlier. the republicans had to wait for a senior senator to retire to move that seat up there. again, around them are all the senators by seniority. the freshmen who just got elected will be way in the back in the corner. and the senior senators are usually down in the center aisle, or towards the middle. and that's so that again if the presiding officer is looking out over the body and a number of senators are seeking recognition, the people front and center, of course are the ones who catch the presiding officer's eye. the people in the back have to shout a little bit and wave their arms to get that kind of attention. so that's one reason people tend to move to the front.