Krugman and Some "Other" Social Security Reform
Paul Krugman correctly points out the flaws with an argument of the form, "there is no trust fund, so the system will be in crisis in 2017." But his response is not a good argument against reforming the system now.
I acknowledge that we should be careful about demonizing the word "crisis." The use of the word "crisis" should be construed as an attempt to focus people's attention on how economic and fiscal relationships will change as we transition to a society with so many more retirees relative to workers. For example, in 1994, the World Bank published a very influential research report
Ten to fifteen years ago, this was not viewed as particularly controversial. In the interim, we missed our chance to make progress during the Clinton administration, as the momentum that was building (e.g., "Save Social Security First") for phasing in reform gave way to impeachment proceedings. I supported this approach and was looking for ways to make it happen. As the Clinton Administration gave way to the Bush Administration, we got a Social Security Commission that had a very tough task before it. It was supposed to "strengthen" Social Security but was not supposed to collect more revenue to do it. I would take the Commission's recommendation (Model 2) over the status quo, but that's mainly because I have a small-government view of the world. Others who don't share that view are going to push back, and they did, with sufficient force to block the whole thing.
But none of this means that reform is not a pressing issue today. For example, Krugman states, "As Kevin Drum, Brad DeLong, and others have pointed out, the SSA estimates are very conservative, and quite moderate projections of economic growth push the exhaustion date into the indefinite future." As I pointed out at the time, this gives a very odd definition to "quite moderate projections of economic growth."
You can look at the sensitivity analysis for the growth in real wages in Table VI.D4 and see that increasing the projected rate of growth in real wages by 0.5 percentage points (around the baseline growth rate of 1.1% per year) shrinks the 75-year actuarial deficit from 1.70% to 1.12% of taxable payroll. That gets us about a third of the way toward a zero balance over 75 years and is a necessary but not sufficient condition to support Krugman's claim. If continued linear extrapolation is valid, then we would need to add about 1.5 percentage points to the real wage growth rate--over 75 years--to get the balance to zero. That's sustained real wage growth of 2.6% per year for 75 years. Krugman should come out and say that such a number is "quite moderate" if that's what he means. Seems pretty optimistic to me.
So I return for now to the same place we started: The population is aging. The aging population will place larger fiscal demands on workers in future generations. We can see this demographic challenge now. We should work to face up to it now.
OK, but what about Medicare/Medicaid?
Andrew, Another of Krugmans points is that the other entitlement elephant in the room is bigger and stomping his feet more loudly. Shouldn't that be our number one concern? I checked your Vox Baby archives and see no category for this issue. I am curious as to why? I like McCain's position of eliminating the tax loophole for business to employee complensation. That seems like it has some real potential to drive significant numbers to HSAs. My take is that much of the inflation problem is due to price inelasticity driven by insurance middle men. The 'market' who pays gets no benefit from the service. The people who benefit don't see the cost. There is real incentive for insurers to create red tape and poor service as a tool to drive usage or at lest repayment down. HSAs allow the consumer and provider to interact directly wrt cost. Once we can get a sufficient number on HSAs, the providers will start to compete on cost as well as service. A better database would stimulate competition on quality as well.Liebman-MacGuineas-Samwick Non Partisan Social Security Plan
Liebman-MacGuineas-Samwick Non Partisan Social Security Plan One perfectly good answer is that Prof. Samwick has a fully formed Social Security plan in hand. On the other hand it is a little difficult to get a handle on how you attack Medicare's funding gap given that that was increased by the introduction of Medicare Part D under President Bush's leadership. Which is by the way where Bush lost Bruce Bartlett who I say takes the general position of 'When you are already in the hole, stop digging'. Not that I agree with Bruce B, Part D was a much needed improvement to Medicare that itself needs to be improved. Done right Part D can go a long way to improving the overall health of Part A, it was somewhat crazy to have a program structured so that you couldn't get heart medication paid for until or unless you were hospitalized, this turns medical cost control on its head.Addressing Medicare
I didn't object to having a Medicare Part D. I objected to it being done in a way that passes so much of the costs to future generations. Inside policy circles, introducing a $16.6 trillion unfunded obligation with Part D completely undermined the argument for reforming a (then-) $10 trillion unfunded obligation in OASDI. There is good research based on geographic differences in Medicare expenditures relative to diagnoses and outcomes that there is unnecessary expenditure in Medicare. (See http://www.dartmouthatlas.org/) It is also widely known that chronic conditions like diabetes are not managed efficiently in the Medicare population. There is plenty of room to cut expenditures without sacrificing quality of care. But I don't think the political will exists to do that--if we cannot even balance Social Security on paper, how could we trim expenditures on medical care in practice? So we should face that and be ready with another approach. Keeping with my views on old-age transfer programs, I would raise the age of full benefit eligibility and allow those 65 and over to buy their way in.Actuarial Deficit
Andrew, wondering if I could get your analysis of the measurement changes due to immigration classification. * Why did they add this feature? I.e., was the prior model fundamentally missing this injection, or is this a statement about the structure of illegal immigration? * How robust is the result to policy changes? And how did they calibrate the history? Separately, I'd like to see a persuasive argument for why the current structure -- viewing SS as a stand-alone program -- is a good one. My instinctual prudence supports the approach. But I'd like more than that. The alternative approach implicitly suggested by Krugman & co. is that we should move to Soc. Sec. funding out of general funds, perhaps more like Part B is structured. What are the pro's and con's of that? I wouldn't have been amenable to that at all 5 years ago, but seeing how popular solutions built exclusively from tax increases are to restoring "solvency", I'm weighing the evidence. Specifically, my concern is that narrow focus on actuarial balance is neither necessary nor sufficient for sound policy, yet the current structure seems to engender that sort of narrowness. "Hey, things improved from 1.9% to 1.7% yippee!" sort of logic fundamentally misses the major issues, but has broad popular appeal to those who don't want to get into the details.Victor those are the important details
And you are missing the Krugman/Baker argument entirely. Neither of them would support moving Social Security over to the General Fund on the model of Medicare Part B/D. Instead they are making two much different arguments. One, in crudest terms if those in charge of General Fund budgeting think repaying the money they borrowed from the Trust Fund will be some impossible burden in 2017 then maybe they should stop borrowing it today. Otherwise you are in effect increasing marginal rates on everyone earning wages under the cap, in effect converting a portion of the current payroll tax into a hidden income tax. (So much for the poor not paying federal taxes argument.) There is a serious question of equity there. Two, it is precisely the improvement in payroll gap that is the issue. The Trustees have always expressed ultimate outcomes in these terms, if the data shows a long term trend of reducing payroll gap projected over the 75 year window from Report Year to Report Year, and it does, then large movements in this measure (and 1.95% to 1.7% is large) raises some serious issues. For example the Social Security Actuaries officially scored Professor Samwick's LMS Plan with results seen in table 2. It is perfectly fair to put up this official projection by the OACT against LMS as scored by that same OACT, those are the important details for forming policy, and not the top down logical analysis your comment seems to suggest. As to why Social Security works best as a standalone program. That was a political calculation related to point one above. Social Security is currently explicitly modeled as wage worker funded insurance for wage workers. It owes nothing to capital because it draws nothing from capital. The Trust Fund and the Payroll Cap both were designed as political insulation against just the kind of backdoor raids the Thompson Plan (among others) suggested. This current Administration likes to separate spending into two categories: spending that we can't NOT afford, spending where billions are no problem, and spending we CAN'T afford where suddenly millions during into unbearable obligations. The military and corporate tax subsidies falling into the first category, social spending into the second category. Keeping Social Security separate is all that is (barely) keeping it out of the second category. Currently SS has dedicated Income some $200 billion a year more than current Cost, a fact that would be obscured if FICA was just another component of the Income Tax.Immigration Assumptions
Victor you will find the Trustees explanation in IV.B4.7 7. Reasons for Change in Actuarial Balance From Last Report This is less than perfectly clear but seems to state that the difference is not in the numbers of total 'other' immigrants but instead in their age adjusted behavior, with relatively higher numbers of younger productive workers contributing while not accumulating enough legal quarters to collect benefits.Fixing Social Security and medical care
Fixing Social Security is very easy, even 43 years after I first said how to do it. Medical care is far harder but needs the same two components that Social Security needs. To fix Social Security we need to make it a real defined benefit pension plan as opposed to the fake Potemkin Village one it is now. To do this we bag the PayGo way we finance it in favor of actuarial advance funding, using a legitimate actuarial cost method called The Entry Age Normal Cost Method instead and add strong laws with teeth that protect the assets and the past service accrued benefits-which by the way were both put into law in 1974 when ERISA was passed for all corporate-sponsored private pension plans--but alas they were done poorly and the sad state of affairs of that once mighty industry is well known. The Entry Age Normal Cost would be around 3-4% of pay and paid by a joint employee-employer tax, replacing the present 10% of taxable payroll, while the annual amortization payment of the initial unfunded past service liability under that same method would come from the federal government's General Account. The total additional annual payments over and above what we are currently paying would be around $50 billion a year for 40 years, and would be entirely invested in a low-cost broad based stock index fund, such as Vanguard's 500 Stock Index Fund. To fix health care we need to have a universal, single payer national health care system that has exactly the same two missing components. There is an important role for private health insurers, but not as insurers, as administrators using Administrative Services Only (ASO) contracts. They would also be judged by how well they prevent fraud. The country would be divided up into 9 regions and these ASO contracts would be competitively bid, selecting two companies for each region, competing against Medicare. There would be the same benefit program for every region and for every person and everyone would pay the same premium--a tiny fraction of the total, which would be paid by the government through the collection of dedicated taxes, the Entry Age Normal Cost, around 3% of pay, would be paid by a joint employee-employer tax, while the amortization of the unfunded past service liability would be paid from the Government's General Account. By separating these costs in this manner for both Social Security and medical care, you will avoid having the current generation of workers pay for the failure of our government to put the system on the right basis 50 years ago and prevent intergenerational conflicts, permit corporations to compete better internationally, make individuals very happy given that the current Social Security payroll taxes are the highest taxes we pay, and get the monkey off the back of US corporations on health care. It will take a decade to implement the needed reforms, but at the end of that period the $2 trillion we now pay--double that of any other nation on a per capita basis and 3-4 times some nations--would be reduced by about 75% in today's dollars, partly from those reforms and partly from the investment returns coming from the actuarial advance funding. As one important example of needed reforms we need to develop and put into place three major national software systems: 1. A personal file on each individual's medical records. 2. A national database that provides the best current medical thinking on the best treatment for every medical ailment--literally thousands of them. 3. A national database on quality and costs for every hospital and clinic and for every doctor and for every treatment, and accessible locally. There will cost billions to develop and maintain but without them we cannot move forward. It is interesting to note that both Microsoft and Google are developing at least the first one and of course in time the nation will have to buy one or the other or perhaps both. It is also interesting to note that actuarial advance funding was developed 150 years ago and moved into the defined benefit pension arena in 1921. The Entry Age Normal Cost Actuarial Cost Method was one of six permitted by ERISA in 1974 and by 1980 it was being used by well over 90% of the major Fortune 500 private pension plan sponsors. But five years later almost no one used it, due to the serious flaws in the laws and accounting rules that have caused that entire industry to collapse. Andy Lang, Fellow of the Society of Actuaries (FSA) and Member, American Academy of Actuaries (MAAA).FDR and Social Security
I have a saying:
If you don't know where you have been, how will you know where you are?
And if you don't know where you are, how will you know where you are going?
And if you don't know where you are going, how will you know when you get there?
In the year before FDR got the Social Security Act passed, 1934, he had three sets of experts look into various aspects of it's feasibility.
One group looked into how the benefits should be determined.
A second group looked into how to assemble the massive records--this was before computers anything remotely like the powerful PCs we have today.
The third group was three pension-consulting actuaries to see how it should be financed.
This type of actuary was relatively new in 1934.
George Buck had left the life insurance industry, the Mothership of most pension actuaries, during WWI, and had set up shop in NY City, and, in 1921 would be awarded the job of establishing the proper way to fund one of the largest defined benefit pension plans in the world, The NY City Employees Retirement Plan.
Buck was not just the first independent pension-consulting actuary; he was the very first person to develop an actuarial cost method.
This is a rigorous mathematical and self-correcting methodology that determines how much to contribute each year, long in advance of it being needed, and then invested, such that this contribution, plus the investment returns earned thereon, compounded over time, would exactly pay the promised benefits (defined by formula)--with said returns paying the bulk of the benefits.
Unlike the way we currently finance Social Security, which would be illegal if it was done in the private sector, an actuarial cost method is always 100% in actuarial balance—by definition. You do not try and reach some sort of ‘actuarial balance’ over 75 years. In fact there is no such thing as that term in the private sector pension valuations.
Anyway, FDR, being Governor of NY State, was very aware of Buck and NY City's pension plan.
So when he hired these three pension-consulting actuaries more than a decade later and they came up with the conclusion, not surprisingly, that a national Social Security system, which would be a defined benefit pension plan in actuality, should also be actuarially advance funded, FDR asked them how much the reserve fund would eventually build up to.
When they gave him a figure it was absolutely huge relative to the size of our economy, which was a disaster at that time, and FDR then turned down the system in favor of a PayGo system.
Why?
Easy--we were in the middle of The Great Depression, the worst the world had ever seen and there was no place to invest the money. (And FDR was also concerned about a little guy with a mustache in Germany).
The 1929 stock market crash had kicked off the depression and also many corporations had failed and so had their bonds--and to top it all off the banks had also failed.
So he made his decision--but had in the back of his mind that he could always correct it later on, when the economy was back on it's feet.
Unfortunately, he died before WWII was over and never did.
The US economy had soared during the war with all of the military hardware we were producing, and then after we moved to a peacetime economy it also did thanks to FDR and Truman's progressive policies, such as subsidizing many of the 20 million returning troops to go to college, not only setting the stage for the great economic expansion that would follow, but also solving the enormous problem of how to find work for them all.
Alas, no Social Security actuary who followed was an experienced pension consultant and therefore never set the system up right or brought it to anyone's attention again.
And so, it has come to pass that we have been on this crazy PayGo system ever since, which hardly gets any investment returns, with many life insurance actuaries supporting privatization and many pension actuaries working hard and making huge money by working for large corporate pension plan sponsors especially following the passage of ERISA in 1974--often by screwing plan participants out of benefits, more than a trillion dollars by my estimation, by taking advantage of very poor laws.
No actuary since has ever said the right way to finance Social Security, or the right way to fix broken pension laws (and broken pension accounting rules too) in the private sector, save for myself.
The private pension system is a mere shadow of itself today, covering less than half the people it one did in 1980 and they have benefits less than half of what they once were.
And Social Security systems the world over, not just here in the U.S. are in deep trouble.