We don’t have a Social Security problem; we have an unemployment problem.

As our political class wrangles about the federal budget deficits and debt, there are persistent wrong-headed proposals to raise the Social Security retirement age and to make annual cost-of-living adjustments ("COLA") smaller. We can restore actuarial balance to Social Security by simply restoring full employment, and if we can't or won't restore full employment we should lower the retirement age and raise the COLA.

The Social Security "Problem"

The current problem with the Social Security retirement system is that in recent years taxes paid into the system have been too small, when supplemented by scheduled drawdowns from the accumulated trust fund, to pay all of the scheduled benefits over all of the next 75 years. At present and projected receipt levels, the trust fund will be exhausted in about 25 years, after which current receipts will be only large enough to fund about 77% of the scheduled benefits out to the planning horizon. This is not occurring because nobody noticed that there were a lot of baby boomers who were going to retire but because the percentage of working-age Americans in employment has been unexpectedly shrinking since 2000. Here, courtesy of the St. Louis Fed, is the employment-to-population ratio for Americans 16+ years old.

If the employment-population ratio had continued at the 2000 level, the trust fund would still be increasing, and current receipts plus trust fund drawdowns would never become insufficient to pay presently scheduled benefits. With full employment there is no Social Security funding problem at all. None. All scheduled benefits can be paid for at least 75 years without any tax or benefit changes. In other words, we don't have a Social Security problem, we have an unemployment problem.

In the political arena, we should insist on solving the immediate and devastating unemployment problem and not be drawn into discussing Social Security changes now to prevent a crisis a quarter century from now. If we are going to discuss Social Security at all, we should be demanding changes that make it better for America instead of giving credence to proposals that will both exacerbate unemployment and shrink the safety net.

The retirement age should go down, not up.

The proposal to raise the retirement age is intended to improve Social Security finances by both reducing total benefits paid and collecting more Social Security taxes from seniors who defer retirement. The unintended* but unavoidable consequence of doing that is that senior citizens who keep working will not be making room in the work force for young people, who must then be supported by their parents or some other government program or left in the streets.** This is not a fanciful or trivial problem or one caused by the Great Recession; it has been happening for at least 10 years.

Those 55 and older have been staying in employment longer each year since 1993 and have hardly given up any jobs in the Great Recession. In stark contrast, the employment ratio for 16-24 year-olds has dropped like a rock since 2000. The employment situation for young adults would be improved by lowering the retirement age to get seniors out of the workforce sooner. If there is going to be chronic unemployment, the least bad option is to have unemployed seniors instead of unemployed youth.

The COLA should be changed to CPI-E which recognizes the spending patterns of seniors.

Another proposal to lower the Social Security expenditure rate is to base annual cost of living adjustments on a new consumer price index, "chained CPI." This differs from the currently applicable CPI-W index in that the chained series adjusts the relative weights in the "basket" of goods and services on which prices are tracked to take into account consumers' changes in buying patterns. For example, when consumers can no longer afford steak, they buy hamburger, and when they can't afford that they move successively to bologna and low-protein staples. Thus, a smaller COLA forces lower spending and then (mis)calculates that a smaller COLA enables the beneficiary to stay even. Although the real reason for wanting to move to chained-CPI is simply to reduce Social Security outlays, proponents argue that it is more accurate than CPI-W and therefore fair to everybody. Not so.

If accuracy and fairness is the goal, we should adjust Social Security benefits by an index, CPI-E, which is based on the unique consumption patterns of seniors and has already been developed by BLS. Seniors spend relatively more on health care and housing and less on transportation and education, for example. Switching from CPI-W (which specifically excludes the spending patterns of retirees) to CPI-E (E for elderly) would generate somewhat larger annual increases in Social Security benefits as health care continues to be one of seniors' fastest increasing costs.

Be competent negotiators for a change.

Tactically, those who want to protect Social Security and solve broader economic and social problems, should be demanding a lower retirement age and a switch to CPI-E because the obvious compromise between that and a higher retirement age and chained CPI is the status quo. On the other hand, if progressives only defend the status quo, a compromise with proponents of a higher retirement age and chained CPI is more unnecessary economic and social distress.

_______

* Of course, there are those who fully intend to weaken Social Security because they want to privatize or eliminate it altogether.

** An additional problem with raising the retirement age is how to support senior citizens who don't keep working those extra years. One cost will be that Social Security disability benefit payments will increase. ADDED 8/21/2011: The disability trust fund will be exhausted in 2017 because of burgeoning applications by people who have lost their jobs and are arguably disabled, according to recent estimates.

Update on Wednesday, August 17, 2011 at 07:35AM by Skeptic

Richard Eskow shows how much a shift to chained CPI would reduce typical benefits after a decade or two and discusses the politics of doing that and/or raising the retirement age.

Update on Tuesday, August 23, 2011 at 10:15AM by Skeptic

Kevin Drum points out that retirees in the top half of the income distribution live about 4 years longer on average than bottom-half earners. Thus, raising the retirement age would cut the benefits of bottom-half earners more than of top-half earners, which goes against the idea of needs testing. In another post Drum links to a 2010 CBO study reporting the projected effect on OASDI of 30 different adjustments that have been proposed. The summary table from the CBO paper evaluates options in terms of their effect on the "actuarial balance," which was then -0.6 percent of GDP. Thus, an option that has a +0.3 change in the actuarial balance fixes half the problem.

Update on Tuesday, July 28, 2015 at 08:50PM by Skeptic

This post says the problem is the failure of wages to keep up with productivity increases, in other words growing inequality, that upsets the Social Security cash flows.

Why teenagers don't want to raise the retirement age

The French parliament is about to raise the national retirement age by two years, and high school students are at the barricades.  Link.  Why?  Truly, I don't know why teenagers do anything, but it's easy to see how this change will adversely affect them in the short term:  During the transition period, there will be fewer jobs open to young people because there will be more sexagenarians hanging on to job slots that would otherwise trickle down to new labor market entrants. 

Older workers have the power of incumbency:  As long as their jobs aren't actually eliminated, they can generally keep working as long as health and motivation permit.  So, when there is a labor surplus (job shortage), the burden falls most heavily on the young who have not yet established themselves in the work force.  This pops out of US age-and-education-dissected unemployment data for the Great Recession: The greatest increase in the unemployment rate has been among college graduates under 25. Link.

Some in the US want to eliminate Social Security entirely on the ground that it is wrong to tax the young and middle-aged to pay pension benefits to the old.  (They don't say whether it is also wrong to tax the middle-aged and old to pay for the even more valuable benefit of free public educations.)  Those who simply want to put US Social Security on an actuarily sounder long-term financial footing propose raising the retirement age but do not discuss the adverse impacts of that on youth unemployment (or other human effects). 

Perhaps some readers will object that I'm assuming the Lump of Labor Fallacy (aka the Lump of Jobs Fallacy), but I'm not. The "fallacy" originated in proposals to reduce unemployment rates by reducing the work week from 40 hours to 35 hours and spreading the work among 12.5% more employees.  Economists pointed out that, for a variety of reasons believed to be present in real labor markets, that wouldn't happen--or at least would not happen to the extent claimed.  The criticism is, I think, correct, but some misuse that fallacy to try to prove the opposite--that nothing will happen to the unemployment rate, which is also a fallacy: The Lump of Labor Fallacy Fallacy?  I don't assume that the number of jobs in the US will not expand if general economic conditions improve, but it would be equally false to say that because the number of people in the work force increases, an equivalent number of jobs will magically appear for them.  In fact, the number of US jobs as a percentage of working age US population has actually been shrinking--not just during the Great Recession but ever since 2000 and is lower now than in 1991. Link.  Whatever the number of jobs existing at any time, the competition for them will be more intense if sexagenarians stay in the work force longer, and the cohort most disadvantaged by that increased competition will be the youngest.

If we're going to increase the retirement age, what's the plan to increase the number of jobs?

Update on Sunday, October 24, 2010 at 12:41PM by Skeptic

Another question to ponder:  If we were to create as many new jobs as would be necessary to allow later retirement without diminishing the prospects of the young, wouldn't Social Security actuaries probably calculate that we don't need to raise the retirement age after all? 

Update on Monday, October 25, 2010 at 09:28AM by Skeptic

More reporting on how unemployment hits the young hardest is in this LAT piece and, echoing other reports, it says cohorts that get off to a slow start tend to lag behind throughout their careers. 

As the nation struggles with the aftermath of the Great Recession, few groups have suffered greater setbacks or face greater long-term damage than young Americans — damage that could shadow their entire working lives.

Unemployment for 20- to 24-year-olds hit a record high of more than 17% earlier this year. Even for young adults with college degrees, the jobless rate has averaged 9.3% this year, double the figure for older graduates, according to the Labor Department.

Adding to the impact, surveys by the Pew Research Center indicate, a greater share of workers in their 20s lost hours or were cut down to part-time status than any other age group. And their incomes have fallen more sharply, even as they are far more likely than others to say they are working harder than ever.

Update on Thursday, November 4, 2010 at 11:24AM by Skeptic

More LAT reporting from Paris finds French youth indeed worried about the impact of the retirement age increase on their already-bleak job opportunities. 

They may be showing solidarity with their older compatriots, but for many of the teens, the motivation resonates closer to home. With unemployment estimated at 23% for 15-to-24-year-olds, fresh graduates can feel as though they're on the outside of the job market looking in, and they don't think the government is doing enough to help them.

"They aren't only in the streets because they are worried about their future retirement," said Jean-Baptiste Prevost, head of the largest student union, UNEF, as he led a protest march through Paris. "It's not the future in 40 years; it's the future in two years that they're worried about."

Update on Tuesday, December 7, 2010 at 02:27PM by Skeptic

Another way raising the retirement age fails to work as intended is that there will be more applications for disability retirement, which comes out of the same fund.  That is the prediction of the Government Accountability Office as reported by Mark Thoma. 

Update on Friday, December 24, 2010 at 08:05AM by Skeptic

Commenter Sandwichman has written extensively about the lump of labor fallacy, most recently here. And here (added 12/27/10) and here (added 1/4/11) and here (added 1/4/11).

Update on Saturday, January 15, 2011 at 01:49PM by Skeptic

Floyd Norris reports in NYT that the percentage of Americans 55+ holding jobs has increased 7.6% since the start of the Great Recession.  Meanwhile, the employment rate among those 45-54 declined 4.6%, among those 25-44 the decline was 7.5%, and 16-24 year-olds dropped 13.2%. Three years is much too short a period for these changes to have been substantially influenced by demographic changes. People who had jobs are keeping them, and those trying to break into the work force are being shut out.

The federal government is a big insurance company with an army.

In the 2010 federal spending budget, 20% is for defense and security, 6% is for interest on the federal debt, and 65% is for retirement, disability, medical, education, and other welfare programs. That leaves only 10% for everything else.  Hat tip to Mark Thoma who posted here the following pie chart from CBO.

Intergenerational tax equity

We hear much about how infeasible and unfair it is for young people to pay Social Security taxes that get transferred directly to retirees. A recent study at UC Berkeley finds that the value of the educations the young receive at the expense of tax-paying older folks is worth more than the intergenerational transfer of Social Security benefits.

On average, Americans pay the taxes that subsidize education 30 years after receiving the benefits, the study noted. By contrast, people start drawing their Social Security and Medicare benefits 30 years or so after paying taxes into these government funds. Thus, each education dollar is worth $10 in retirement benefits, according to the study, which is published in the March issue of the journal Population and Development Review.

So, if it's wrong for 40-year-old John Doe to pay taxes that support 70-year-old retirees, was it also wrong for them to be paying taxes 30 years ago to give 10-year-old John a free education?

Replace the bean counters with visionaries.

Ben Bernanke says federal entitlement programs are unsustainable:

To avoid large and unsustainable budget deficits, the nation will ultimately have to choose among higher taxes, modifications to entitlement programs such as Social Security and Medicare, less spending on everything else from education to defense, or some combination of the above. These choices are difficult, and it always seems easier to put them off–until the day they cannot be put off any more. But unless we as a nation demonstrate a strong commitment to fiscal responsibility, in the longer run we will have neither financial stability nor healthy economic growth.

But a Rand Corporation study sees a way out:

An unprecedented upturn in the number of older Americans who delay retirement is likely to continue and even accelerate over the next two decades, a trend that should help ease the financial challenges facing both Social Security and Medicare, according to a new RAND Corporation study.

Hat tip to Mark Thoma for these. My thoughts:

When actuaries project that raising the retirement age from 65 to 67 (or 71 to 72, as I heard proposed on the radio recently), what do they assume about the number of jobs that will be available? If 65-year-olds keep at it another 5-7 years, will there be that many more domestic jobs for them, or would younger folks be forced into unemployment? I strongly suspect it's the latter. Labor force participation rates for men have been declining since the 1970s and for women since 2000, according to EPI. 

 

If we could put in place policies that cause sustained broad increases in real wages and increasing labor force participation, fiscal problems all over the nation would start to shrink and disappear. On the other hand, if real middle class incomes continue to stagnate and decline, there will be no good solutions to fiscal problems.

Furthermore, Social Security cannot morally be discussed as a mere problem of fiscal management for the national government. There is no realistic prospect that reduced Social Security benefits will be replaced by private savings or any other program because employer benefits and private capacity are shrinking even faster. What's really being discussed is managing the regression to dystopian conditions that our grandparents struggled to put behind us. Our leaders should be thinking the big ideas that will be necessary to prevent that decline, not about how to cope with it.

Why Social Security needs restructuring and Medicare doesn’t

I think I've got the answer to why the Right wants to privatize, downsize, revamp, or even eliminate the Social Security system but seldom inveighs against Medicare, which is in much, much worse shape actuarially. Both programs benefit an almost identical group of seniors, but in addition to that constituency Medicare benefits healthcare providers and insurers who have real political power—on display now in the pending healthcare "reform" legislation.

The continuing education of Marie Antoinette—Social Security

As the federal deficit mounts, there are renewed calls for breaking the promises we've made to the millions of middle-class and poor folks who do or will depend heavily on Social Security benefits. Many who subscribe to the idea that the Social Security system is broke or unsustainable without major reductions in benefits do not realize, I suspect, that dramatic cuts in benefits would be disastrous to a vast majority of Americans. Paul Krugman reports that the one-quarter of the older population whose incomes are above the median but not in the top quarter rely on Social Security benefits for over half their income. On average, they get only 9% of their income from IRAs and other assets, and the bottom half gets even less asset income. Thus, asset prices are not particularly important for at least 75% of retirees. Here's the graph for the 50-75 percentile group.

1,000 years of historical context for Social Security

The Social Security Administration has a fascinating brief history of the societal and political developments that led eventually to the enactment of America's Social Security program in 1935. Although it was enacted during The Great Depression, it wasn't primarily a countercyclical device but was the culmination of the evolution of collective economic security institutions starting with guilds in the Middle Ages. Thanks to Mark Thoma for bringing this to my attention.

Social Security ain’t broke and don’t need fixin’.

This week Obama said entitlement programs will be "a central part" of his Administration's effort to control federal spending. (Interestingly, these remarks at a press conference are not in prepared statements on the Obama transition website.) Clearly, Medicare and Medicaid are underfunded, but Social Security needs, at most, a minor tweak and is probably just fine for the long term with no changes in funding or benefit formulas, according to Bruce Webb and Barkley Rosser. So why did he say it?

Update on Monday, January 12, 2009 at 10:38PM by Skeptic

Not to worry.  Obama's a fixer, not a transformer.  Bruce Webb explains it all at Angry Bear.