You can search for articles in back issues of Contingencies from July/August 2000 to March/April 2009 using the search box to the right. Simply type in subject words, author's name, or article title and click search. To search for articles from May/June 2009 to the present, go to the current digital issue of the magazine and use the search function on the left of the top navigation bar.

Google Custom Search


PDF version

No Promises

I enjoyed the May/June 2005 issue of Contingencies, especially the pieces on Social Security developments. Here are some thoughts on this matter, relating to the equity of intergenerational transfers and the inappropriateness of the administration’s proposed benefit formula.

The fundamental defect with Social Security is that it’s based on the principle that each generation is forcibly extorting funds from its descendants. The rationale we tell our descendants is that when they get old, they will have the right to follow our example and extort from their descendants. This arrangement is inherently unstable, and it’s evil.

There is no reason to believe that every generation will follow obediently in their parents’ steps. This extortion actually made a certain amount of sense when the program was introduced in the 1930s, but it has been corrupted by later Congresses and is completely out of order in today’s world.

Moreover, the way the administration is portraying it’s plans, using the worst possible words to describe them. Future projected benefits are not “promises.” Everyone recognizes that the present schedule of benefits is beyond the ability and willingness of the public to support. There are no promises.

Further, the administration’s method of describing the proposed benefits is to start with the current “promises” and then deduct something to recognize taxes going to the personal account. Once you use the concept of deduction, the whole battle is lost.

What you must do is first remove the idea of the present benefit schedule. That is nothing but an illusion. Then you must define how much the fixed government guarantee will be in the future. Finally, you must add the amount of estimated income to be provided by the personal account. The result is the total amount the retiree is entitled to.

You must add, never subtract. No one can figure it out if you subtract. The objective should be to make every generation self-supporting, with a substantial portion of their accumulated earnings in their own personal accounts, and not subject to extortion by congressional action.

Finally, you must not make the existence of a personal account an option. The personal account should be standard. The option should be whether the account has variable or fixed-dollar benefits.


Accounting for Women

I wish to commend you on your discussion of Social Security reform and presentation of various aspects of the issue. However, one thing that you did not focus on was how the system works for the majority of older Americans: women.

Social Security is important to all Americans, but there are special issues for women, particularly those who are alone in old age. Proposals for personal accounts are particularly troublesome for these groups. A change to personal accounts will create winners and losers. Widows, divorced women, and surviving minor children are likely to be big losers.

Benefits are earned as people work and are paid over their lifetimes, in retirement, after disability, and to survivors. Social Security benefits are based on the work history of those family members who are employed. Even though many women are in the labor force today, their lifetime work and earnings history is very different from men’s. Men retiring in 2000 worked a median number of 44 years compared with 32 for women. The median earnings of full-time male workers in 2002 were $38,884 compared with $29,680 for women. In 2002, 18 percent of unmarried women over 65 were poor compared with 4 percent of married people (Just the Facts on Retirement Issues 10, April 2004, Center for Retirement Research at Boston College).

Women’s life histories after age 65 are also very different. Women live longer, are less likely to remarry if widowed, and more often require paid long-term care. Many widows and divorced women will be alone for 15 years or more and depend heavily on Social Security. Without Social Security, many more of them would be destitute. Private accounts shift the emphasis from social insurance to property rights and would not meet the needs of widows and divorced women well. Just as happens with other funds, widows might find that little was left of the Social Security account after the death of their husbands. It is my hope that the group evaluating Social Security reform options will focus on the differences in circumstances among single- and dual-earner couples, male workers, female workers, widows, divorced people, and other groups and evaluate how each demographic segment will be affected by a proposed change.

Other benefits for families are also at risk. The existing Social Security system pays important benefits to children of deceased members, surviving spouses after retirement age, and surviving spouses with minor children. Social Security death benefit protection has a net present value equivalent to a life insurance policy with a face value of $403,000 for a young average earner with a spouse and two young children in 2001. The average monthly Social Security benefit for a widowed mother with two or more children was $1,909, or about $22,900 a year in January 2004 (Uncharted Waters, page 158, National Academy of Social Insurance, 2005). Exchanging a significant part of these death benefits for a property right to the accounts that might be accumulated would be a very bad deal for the families that have depended on these benefits.

There is a variety of paths to solving Social Security’s long-term problems. The American Academy of Actuaries ( looks at 10 options in Actuaries Look at Options for Reforming Social Security. I personally would start by indexing retirement ages. Life spans have increased a great deal since Social Security started, and normal retirement ages have increased just a little. Option 1 is to raise the retirement age for full benefits to 70 by 2030 and keep adjusting the age as people live longer. That option would solve 68 percent of the problem based on current actuarial estimates. Heather Jerbi provided her solution in the May/June Contingencies. I wish Contingencies had included more on the options to encourage everyone to build a solution. I encourage everyone to look at the Social Security game on the Academy website.

There is one more thing that I would have liked to see in your issue: simulations showing account balances at retirement and how much income they might produce. One of the arguments for account balances is that they will help people without other savings to get started. In understanding the importance of this argument, it will be helpful to see how large these account balances might be and how much retirement income they will produce. I believe that many people will be disappointed by the size of the account balances. In addition, when people think of account balances, they think of savings available for a variety of life’s needs. It is important to be clear that these are retirement assets, not available before retirement.

It is important to remember that a strong employer retirement system is important for Americans. Policy needs to support and encourage employer-sponsored defined benefit and defined contribution plans. Without such plans, the pressures on Social Security as a source of income grow greater.


Land of Absurd and Foolish

Once again the realm of the absurd takes over our thinking (“Assessing Social Security’s Future,” May/June 2005). We selectively apply projections according to the needs of politicians, rather than the needs of the people. And we fail to realize that projections get vague very quickly.

I recall that before my retirement we dropped 10-year business projections, and then even five-year projections, because things were too indeterminate that far away. But in this land of the absurd and the foolish, we gravely consider beyond 75 years, as if we know what we’re talking about.

What about our U.S. federal budget, which, if projected for 75 years with today’s facts, will reduce us to a totally dependent nation, owing everyone our lives and sustenance. And I’m not talking about beyond 75 years.

And environment factors, which really should be projected more than 75 years, can turn our world into a living hell or kill us. But in this instance, no one can imagine anything longer than next year.

Political thinking is dangerous when it ignores reality and facts.


Refining Projections

James Lockhart provides an informative commentary relating to the financial projections for our Social Security system. However, the large range of results in the three traditional projections may help us understand the reasons why there is disagreement on the extent of Social Security’s financial ills and corresponding remedies.

If we focus on the intermediate projection, significant changes on the revenue or benefit side (or both) would be called for. However, there is a vocal minority who believe that the intermediate projections are based on overly pessimistic assumptions, particularly as they relate to future GNP annual growth. These supporters of more robust economic assumptions reject both the intermediate and high-cost projections in favor of the more optimistic assumptions found in the low-cost projection. Under the low-cost scenario, the trust fund never becomes exhausted. So, the optimists say, why worry? If it ain’t broke, don’t fix it.

In an effort to better understand and interpret the divergent results under the three projections, Mr. Lockhart says that the actuaries have supplemented their traditional forecasts with stochastic projections. This analysis appears to show that the low-cost projections produce results that are outside the 97.5 percentile. Indeed, if such results have a very low probability, perhaps that number is getting too much respect and prominence. It’s then possible that consideration should be given to have the 90.0 percentile and 10.0 percentile labeled respectively as the low- and high-cost projections. By recharacterizing these benchmarks, the new narrower divergences would perhaps facilitate a better understanding of the issues and a greater consensus on actions to be taken.

In order to consider a possible future change in how these projections would be reported out, actuaries would have to be reasonably confident in the stochastic methodology used. Mr. Lockhart writes that “the Office of the Actuary ... is contributing to refine their methodology.” It’s not clear, however, what types of refinements it’s considering and whether they’re material. I believe the readership of Contingencies would be interested in these answers and would like to see continued discussion on this important issue facing our nation and our actuaries.


Defined Risk

Please allow me a joint response to the “Bait and Switch” editorial and to Mr. Al Johnson’s letter, “The Last DB Plan?” both of which appeared in the May/June 2005 edition of Contingencies.

With respect to “Bait and Switch,” I find it quite amusing that you attempt to express concern over a bait-and-switch strategy in an editorial in which you relate the war in Iraq with President Bush’s Social Security reform proposal. Who is the “bait and switcher” and who is trying to mask his “real agenda,” President Bush or the Contingencies editor?

I have long felt that reporters and opinion writers should begin each article with a declaration of their political leanings. Yours is obviously extreme left wing, and mine, from the perspective of a devout liberal, is admittedly extreme right wing. While I am not in agreement with all of President Bush’s policies, I wonder if there is anything that the president has done right from your perspective.

I would also assume that my political leanings are in the opposite direction of Mr. Johnson’s. Mr. Johnson obviously has more faith than I do with respect to the government providing what he calls Social Security’s “guaranteed income.”

The repeated mantra that the current Social Security system provides “guaranteed” income, quite frankly, drives me up a wall. Throughout the history of the Social Security program, this “guarantee” has been violated with regularity. Imagine an insurance company providing pension or annuity “guarantees” where the insurance company can change premium rates at will (Social Security taxes), delay benefit periods at will (delay Social Security’s retirement dates), and reduce or eliminate “guaranteed” benefits at will (lower and/or tax and/or eliminate Social Security benefits for all or selected beneficiaries). If anyone could be fooled into purchasing such a product, the company and its executives would soon find themselves drowning in lawsuits and criminal charges. Talk about bait and switch!

With respect to Mr. Johnson’s lament over the termination of defined benefit pension programs in the private sector, I have similar concerns. There is nothing terribly “defined” for most employees about defined benefit pensions. Unfortunately, I am quite a bit older than Mr. Johnson and have participated in defined benefit pension programs for more than 30 years. When I become eligible for my defined benefit in a couple of years, my retirement income from my defined benefits will be less than 5 percent of my final year’s salary. Why? Because I had the audacity to change employers twice (i.e., three employers) over my 30+ working years. If only I could go back and get that risky defined contribution plan in lieu of my “guaranteed” pension benefits.


Lost Keys

I would like to respond to “Can Education Substitute for Actuarial Exams?” by Brown and Cicotello, in your March/April issue. The article purports to offer evidence, but all it shows is that pass rates are higher for candidates enrolled as students than for non-students. Why would the existence of a correlation in the range of 10 percent or 20 percent make academic coursework an effective substitute for actuarial exams? And what does correlation have to do with causation? The authors never address these questions.

Perhaps they suppose it’s obvious that academic coursework causes candidates to do better on the exams and so become better prepared for the profession. Actually, the causation works in the opposite direction; the existence of the exam causes teachers to “teach to the test,” which makes students more likely to pass the test. But what would teachers teach to if there were no test? If there were no test, there would be no accountability, and they would teach whatever they thought best and in whatever way they thought best within the syllabus guidelines. The results would be unpredictable and non-uniform. It’s this very kind of situation that is causing states to introduce exit exams before graduating students from high school and is causing nationwide testing, as in the “No Child Left Behind” program. Testing introduces accountability that simply would not be there if it were left to accrediting agencies and well-meaning deans and principals.

Perhaps the authors think that studying causes people to pass the exams and so prepares students for the profession. If studying is occurring, then preparation for the profession is taking place, and exams are really unnecessary. But it’s really the exams that are causing people to study, and without the exams many students would find that studying isn’t really necessary.

The folly of the authors’ proposal is revealed in how the incentives are set up. University administrators, like it or not, have budgets to meet and need to find paying students to fill their classes. Right now, the students fill the actuarial classes because they believe they will give them a better chance to pass the test and so enter the profession. It doesn’t take long for a teacher’s reputation to get out regarding whether his students have a higher or lower passing rate. If the teacher doesn’t produce passing students, the students will go elsewhere.

Now suppose there were no test. Then students would find the easiest teacher whose passing grade would gain them entry into the profession. It’s just human nature to take the easiest route to the intended destination. Perhaps the authors think that students wouldn’t figure this out and would study just as hard without the test as they now study with the test. The simple answer to this is “antiselection.” It’s everywhere, and it’s malpractice for an actuary not to take it into account.

The students are the paying customers, and they aren’t going to sign up with instructors who flunk two-thirds of the class. If you don’t believe me, visit your favorite institution of higher education and see what the growth areas are, where programs and instructors are being added. I’m sure things have changed a lot since my day when the hot areas were the sociology of pop culture and golf. Then ask yourself why the growth is occurring where it is.

All of the other arguments having to do with travel time, what the Europeans are doing, the risk of losing the best and brightest to other professions, etc., are really irrelevant. If you hate the idea of difficult exams discouraging the best and the brightest, would you like it better if entry barriers were lower and the best and the brightest completed the requirements only to be passed over in favor of people who are more fashionable and better at the art of confabulation using business buzzwords?

I think of myself as one of the best and the brightest, and I was attracted to the actuarial profession because it actually has standards; there are right and wrong answers, and you can be judged by which kind you produce. Without standards and without a clearly defined right and wrong, you’re left to politics, fashion, personal prejudice, and all the frothiness of society, which cannot produce anything of substance for the companies doing the hiring and also leaves candidates without a star to navigate by.

The authors talk about the importance of broad knowledge, communication skills, adaptability, and business acumen, none of which are tested for in the early exams. All of these are important, but none of them distinguish actuaries from people in marketing, human resources, or many other branches of business activity. Mathematical and statistical competency is what sets actuaries apart from other business professionals and so forms the foundation of the actuarial profession. Given how highly ranked the actuarial profession is, maybe we should be cautious about emulating other professions; maybe it’s time for them to emulate us.

If you give university administrators the keys to the kingdom, don’t be surprised when they sell them on eBay in order to make up for another funding gap. I can assure you it will be very painful to get those keys back.


Discord in the Choir

In his essay favoring partial privatization of the Social Security program via voluntary personal accounts (VPAs) in the May/June 2005 issue of Contingencies, Haeworth Robertson says the present system operates like a Ponzi scheme or chain letter. This type of pejorative may go over well when preaching to the choir but does not do him credit as a former chief actuary of the Social Security. Surely he realizes that the revenue and benefit projections are just projections and the taxes and/or benefit formulas will have to be adjusted from time to time as they have been in the past.

His points about the consequences of a smaller workforce relative to the population are valid but have nothing to do with the privatization issue, nor, by his own admission (and that of the Bush administration), would privatization solve them.

He thinks that the system “...should be revised so that it provides a Tier I subsistence level (on a pay-as-you-go basis,)” with Tier II via VPAs. Many of us think of the current benefit levels as Tier I. Is the carve-out of the VPA portion simply a way of reducing current benefits?

Robertson says that VPAs will improve government accounting. The government doesn’t need VPAs to do that. In fact, the old age, survivors, and disability insurance (OASDI) accounts were not combined with the federal budget until the late 1960s. We can easily exclude them, get true operating deficits, and reflect the OASDI trust fund debt in the national debt numbers.

He lists the potential failings of VPAs and asserts that “all of these possibilities would be guarded against by a properly designed system.” This is truly disingenuous since the essence of the VPA “problem” is that nobody has come up with a system to avert the potential costs and failings. When they do, I, for one, will get on the bandwagon.

He asserts that if the benefits under a revised system are the same, it cannot possibly cost more. (“This seems obvious.”) The method of financing the benefits can have a huge effect on their ultimate cost. Imagine this lose/lose scenario: Borrow trillions of dollars to finance transition; annual cost of money to U.S. Treasury moves to the 10 percent-plus range; inflation à la the 1970s and, ’80s indexed Social Security benefits become more costly; VPA returns from equity markets in the 4 percent to 5 percent nominal range, negative after inflation.

Only in the world of accounting fictions does Mr. Robertson’s assertion hold. I don’t understand his complaint against those who would spread the cost over time. Hasn’t Mr. Robertson ever held a mortgage?

Mr. Robertson concludes that he favors “a system which the majority of citizens can understand and perceive as fair and reasonable.” That would be the current system.


Contingencies (ISSN 1048-9851) is published by the American Academy of Actuaries, 1100 17th St. NW, 7th floor, Washington, DC 20036. The basic annual subscription rate is included in Academy dues. The nonmember rate is $24. Periodicals postage paid at Washington, DC, and at additional mailing offices. BPA circulation audited.

This article may not be reproduced in whole or in part without written permission of the publisher. Opinions expressed in signed articles are those of the author and do not necessarily reflect official policy of the American Academy of Actuaries.

September/October 2005

Pandemic: The Cost of Avian Influenza

Just Whose Wonderful Life Is it, Anyway? The PBGCs Plan to Save Pensionville

The State of Managed Care Insurance: What Goes Up Must Come Down

Inside Track:
Bad Data


Social Security Super Bowl

Policy Briefing:
A Seat at the Table

Pension Simplification: The Factor of 11 and U.S. Pension Reform

Improving the Profitability of In-Force Long_Term Care Insurance Policies

Statistical Miscellany

Blueland Soccer Tournament

Channel Surfing

Past Issues

Contact us

American Academy of Actuaries