Indexing Social Security for Longevity

Republicans met behind close doors to discuss the issue, Democrats such as Joe Lieberman have indicated support for the proposal, Bruce Bartlett of the NRO is pushing it, Columnist Robert Samuelson is certainly behind it, Greenspan is on board, Gramlich is with Greenspan, other Democrats have indicated it’s on the table as soon as private accounts are off. I’ve discussed the costs and benefits of the proposal here.

The issue is raising the retirement age. Dressed up in its new fancy name of longevity indexing, it’s the latest trial balloon - but if the polls are to be believed - this balloon won’t float for long:

In Overhaul of Social Security, Age Is the Elephant in the Room, By Robin Toner and David E. Rosenbaum, NY Times: … the House and Senate, struggling to produce Social Security legislation this summer, are beginning to confront the longevity issue.... But of all the options to shore up Social Security's finances, [raising the retirement age] ranks as one of the most unpopular ... In a New York Times/CBS News Poll ... nearly 8 out of 10 respondents ... oppose raising the age when people are eligible for Social Security benefits. … One way to address the problem - and the direction some lawmakers seem to be heading in - is an automatic adjustment in the retirement age or the benefits received at each age to reflect increases in life expectancy. … For individuals, such a change, called indexing for longevity, would be little different from a direct increase in the retirement age or a specified reduction in benefits … But for the system … it would make a big difference because the changes would be automatic and would not require new laws. It might also be politically attractive because politicians would be relieved of the responsibility of periodically voting to raise the retirement age or to cut benefits. Adjusting the system for longevity would not contribute much to solving Social Security's solvency problem over the next 30 years or so … But over 75 years and longer, he said, it would have an important effect. Still, pollsters question whether even a gradual adjustment based on life expectancy will sell. "You can call it indexing for longevity in Washington, but in America it's raising the retirement age," said Mr. Garin, the Democratic pollster. Mr. Bolger, his Republican counterpart, said, "There's no appetite for anything related to age among the public."

Comments from old site

Comments

[TDM]

The economists and politicians mostly support raising the retirement age because Social Security is relatively better as social insurance than it is as an investment. The people don't want to be told when to retire, which is why the retirement age should be abolished.
The retirement age should be replaced with a more flexible age/benefits curve. For better investment, private accounts should be like Phillip Longman's early retirement accounts where contributions raise the age to collect benefits rather than reducing benefit levels. Pozen style progressive indexing could apply to both retirement age and benefit levels to achieve solvency. We could keep a reasonable retirement age for low earning non account participants while reducing government spending on high earners and account participants who don't need government benefits while preserving the old age insurance aspect of Social Security.

Workers should work until their benefits are enough to live on, or until they have enough savings in private and other accounts to last until their benefit levels increase to sufficient levels for continued retirement.

[anne]

Interesting comment.

"Pozen style progressive indexing could apply to both retirement age and benefit levels to achieve solvency."

There is already solvency, but what are you thinking just here?

[Tommy Shelby]

Valid Comment. I expanded on it a couple decades ago. Our problem is semantic; we are using the wrong words. The gut issue is "Intergenerational Equity". "Equity" is politically correct!! Ditch "Retirement"; it is antiquanted anyway. (Incidently, actuarial equivalence in benefits should also be extended to 80 or more). Here is my latest discussion:

The Problem With Social Security

The word is tossed all around that Social Security is about to crumble. Why, it is said that in another 10 years it will start losing money! And it will be broke within 35 years! Even the word “crisis” is mentioned.

Who says that Social Security is in trouble? Is that a “crisis” now with all of thirty five years to go? Who cares? Heck, 35 years away looks the other side of eternity for us benefit recipients; even the nearly recipients.

Remember that back in ‘84 when it nearly went broke Congress jumped in and fixed it at the last minute. They said they fixed it permanently. Congress will fix it again like always when it finally goes broke. That’s the way our system works. Why should we worry now?

Well a few of us, including some politicians, care about the country and its future generations. Some of us would like to see a permanent fix rather than another temporary patch. Among all the possible solutions, a couple of real simple changes would change the picture entirely.

Why is there a problem anyway if it was fixed permanently in ‘84? Now that we are on the subject let’s take a look at it.

Social Security’s actuaries, who do these projections, blame the problem on “Demographics”. That’s a big word for “people” that covers fundamental design flaws in the program. If the number of taxpayers were constant and deaths at each age were constant and immigration was negligible, then it would be a perfect system (except perhaps for inflation).

But too many people are living too long. More and more beneficiaries with fewer and fewer taxpayers. The system’s actuaries blame declining birth rates. In summary, the system is not designed to adapt to changing circumstances.

The program was created in the 1930's and used the 1930 decennial census to measure life and death rates and set the funding mechanism. The world was a whole lot different then. Taking just a measurement of life span, in 1930 a person entering the work force at, say, age 21 had a life expectancy of about 46 more years. At age 65 a person could expect 12.4 more years then.

So the system design set the ratio of post retirement life to total working and post retirement lifetime to be 27.26%. Call it the “Generation Ratio”. By 1980, improvements in life span had raised this ratio to 30.11%. In 1980 the life expectancy of a 71 year old exceeded that of a 65 year old in 1930! By 1984, the full benefit age should have been raised at least to 70. Clearly, there had been a change in the balance between generations.

In 1984 our courageous Congress punted! It set the full benefit age to rise incrementally to only 67 against better advice. And that will only be reached in 2022.

In 2000 the Generation Ratio (at age 65) was 30.91%, a 13.4% increase in 70 years. In 2020 the Generation Ratio at the then full benefit age of 66 8/12 will be 29.5%. Is there any wonder why we are running out of money?

One simple change would be to set the Generation Ratio at a constant percentage and tell the Social Security Actuaries to recalculate the Full Benefit Age quinquinnielly to be effective beginning the fifth year following the recalculation. That would produce a regular systematic adjustment similar to CPI adjustments free of political decision making while making it more adaptable to the demographics. Setting the Generation Ratio about at the 27% of th the original ratio would fix about 70% of the problem.

Financially, the system is actually designed to operate similar to a lot of individual households. Lots of us hope to have enough cash coming in at the first of the month to pay the month’s bills with maybe a little to spare. Social Security is intended to do the same thing but on an annual basis. It likes to keeps enough in the bank to pay about a year’s benefits. Most all of the money collected in the past has already been used to pay benefits. The funds that the Administration has on hand at any time is a tiny fraction of the amount of benefits workers have already earned. The fundamental problem is that the taxes collected just do not match the outgo each year because of changing demographics.

After the ‘84 revisions in the law, more was coming in than going out. But the excess is losing ground and will continue to do so for the foreseeable future. Nevertheless, the bank balance is staggering in actual dollar amounts. For 2006 SSA paid $546 billion in benefits. Funds on hand are $2.0 trillion dollars. That is less than $7,000 per person in the US.

Some politicians claim that the Social Security’s funds have been absconded with by other politicians. That is so naive as to be deceptive. In truth, the Social Security Administration puts out at interest available funds not immediately needed for benefits, just like individuals do. And here lies one of the problems.

By law, the Administration must invest only in U S Treasury Bonds. Social Security is required to buy and the Treasury is required to sell special bonds. Since this is obviously not an arms length (public) transaction, the interest rate is based upon a formula using the minimal rates on Treasuries sold to the public in the capital markets. The argument for it says that this minimizes market risk. Of course, the interest rate in this case has nothing to do with market risk; it is the U S Treasury’s mandated backing that makes the difference.

.For Congress this is a dream windfall situation. Congress’ greatest mission appears to be to spend money. The general fund of the government gets all this excess cash pouring in from these bonds that it has to spend. Piles and piles of real cheap money to spend.

One real simple correction is to modify the interest rate formula. After all, it is an inside deal anyway. Why should Congresses have this sweet, nearly secret deal to borrow from us? While earmarks are unstoppable, at least they should pay us a decent rate when they force us to loan them money! According to the 2007 Trustee’s report, the program will go broke in 2041. The interest credited is about 4.6% per annum. Adding a constant incremental step of 1.0% (100 basis points to 5.6% currently) to the rate would defer insolvency by an additional 6 years. Or, adding an incremental step of 2.1% would make it solvent for 75 years (the projection period)! But, alas, this would force greater accountability on Congress and make its windfall more expensive to justify. Good arguments favoring this can be made from both sides of the aisle.

A politically popular solution is to just raise taxes. Some have pointed out, “There is no limit to how high the government can raise taxes”. After all, that has always been done in the past. Originally the tax was only 2% of payroll (with half of it hidden from the workers). Incrementally, especially in the 1980's, it has been stepped up to now 12.6% of payroll (half of which is kept secret from the knowledge of the producing worker). Nevertheless, there are untoward consequences to that stratagem.

Lots of ideas have been bandied about such as other investment media, private accounts, increasing the tax rate, raising the wage base, etc. The two simple solutions suggested here neither require nor prohibit any of them. They will eliminate uncertainty in the System without an increase in the Social Security tax and will have beneficial side effects. Its short coming is to fail to increase government’s grip on workers’ lives. Some folks think that is a good thing.

Thomas H. Shelby III, FCA (Age 72)
November 8, 2007

Sources:

= The 2007 Annual Report of the Board fo Trustees of the Federal Old-Age and Survivors and Federal Disability Trust Funds

= Bell, Felicitie C & Miller, Michasel L. Actuarial Studies 116 and 120, Office of the Chief Actuary, Social Security Administration

= AARP Bulletin, October 2007, Vol 48 No. 9, p.30.

= Private correspondence and discussions with former Chef Actuaries of Social Security.