I am a retired independent pension consulting actuary (FSA and AAA) with nearly 40 years experience in all aspects of pensions, including over two decades working for two of the world's largest pension consulting firms and employers of actuaries in the world. I have worked for all types and sizes of firms, profit and non-profit, pubic and private, including some of the largest in the world.
Unlike many of my actuarial associates, I also have extensive investment knowledge and experience, including and especially in stock investing---further back than when I first decided to become an actuary.
I have been on a crusade to save Social Security for a number of years now---writing on this subject, giving lectures, and doing a lot of online Web participation in this arena, worldwide.
Social Security has the following problems:
1. It has extremely high costs---at least double, perhaps three times what other similar defined benefit pension plans in private industry cost. This is due to the PayGo system of financing which produces an extremely low amount of investment returns. In private industry, the assets produced by actuarial advance funding generate returns which provides more than half of the cost of the benefits; Social Security's system provides less than 5%. Retirement systems in private industry do not work on a PayGo basis because no plan can provide decent benefits without substantial investment returns. This has been known for a very long time. PayGo makes Social Security the most expensive possible way to provide for pensions, and taxpayers can no longer afford this. It is patently absurd to think otherwise.
2. There is substantial intergenerational inequity. This is also due to the PayGo system---a system that paid far more to earlier generations than to current ones, and which especially jeopardizes the coming baby-boomers, the children of those that survived the Depression., and fought WWII to save democracy and America.
3. The objectives of the system, as determined through the defined benefit formula, becomes subservient to the demographic and economic changes---the exact opposite of the way all other defined benefit pension plans in private industry work. In the private pension industry, demographics and economics affect the plan all right, but only in the long term, never in the short term. The asset buildup is far more important and these assets together with actuarial smoothing techniques provide the necessary cushion that prevents constantly going back to the well to make changes. In a highly politicized environment like our Congress, this is the wrong way to do things.
4. PayGo lacks financial integrity, discipline, and together with 5 below, accountability. An annual actuarial valuation report (AAVR), using long established and necessary actuarial cost methods to develop the proper amounts of funding, together with laws, are the means by which advance funding and the necessary discipline takes place. The 75 year forecasts used by the SS Administration actuaries are not used by anyone in private industry, because they are very difficult to analyze, involving mass quantities of numbers, and also include new entrants---open group forecasting is the technical name. Present values are universally used in AAVRs because they can easily be understood. No one uses open group, because it can mask important things. Thus, they can be extremely deceptive. AAVRs when done each year provide a moving picture of funding progress and also self-correct. They do the latter by providing the actuarial gains and losses created by departures of any assumption from the prior year's actual experience. These amounts are then amortized. Over time they also serve to prevent assumptions which can otherwise easily be manipulated. When was the last time you saw a detailed gain and loss analysis in a report produced by the SSA actuaries?
5. Social Security lacks legal constraints on Congress and does not contain the simple legal protections of all other defined benefit pensions in private industry---as Congress itself mandated for those plans in 1974. These legal protections include the all-important definition of what an accrued benefit is, and prevent cutbacks of that benefit. There are no such protections anywhere in our system. Social Security benefits can be cutback anytime for any group of participants by Congress.
The good things about the current system, which must be preserved, are:
1. The defined benefit type of plan is perfect for large-scale systems with highly diverse populations, such as the US. It targets the benefits directly where they are most needed--no waste.
2. The level of benefits produced by the current formula is not at all excessive, and is about right, perhaps somewhat on the meager side, if anything. They are not too rich at all. This is nothing short of amazing, given the lack of correct accounting for it. It contradicts the claim of conservatives that the reason that Social Security is too expensive is the benefits. It is the financial structure---the PayGo part---that is at fault for the high costs, not excessive benefits. Most defined benefit pension plans in private industry provide higher benefits at substantially less cost. If you wish to see what excessive pensions are, check out Congress's pensions---defined benefits that typically can provide pensions greater a few years after retiring than the individual got in pay ---high pay, I hasten to add---before retirement. This is also after being a participant for much less than the 35 years it takes to fully earn a Social Security pension. If you wish to save taxes by cutting benefits, I suggest we start here by setting a good example, rather than by such things as cutting Social Security.
3. The benefit formula, by providing greater income replacement ratios for lower paid people, helps reduce the increasing wealth gap in the US, a very serious problem discussed in many learned journals, including Scientific American's June issue.
To fix it we must make it a true defined benefit pension system, and add the following components to the defined benefit formula:
1. Advance fund it using long-established actuarial techniques that have been proven to work and whose absence is the central reason for the financial problems the system currently has.
2. Provide laws with teeth, to protect plan participants by preventing the plan sponsor from doing the wrong things at the wrong time and to prevent benefit cutbacks for benefits earned or accrued to date.
3. Separate the system from politics as much as possible, the way the Federal Reserve was separated before it began to work properly, but more so.
And, above all, we must assiduously avoid the quick easy 'solution' of privatizing it by converting it to a defined contribution plan using individual accounts. Such a solution would create enormous new problems, in addition to utterly failing to provide for the social needs of our citizens.
It has been being pushed heavily by the retail sector of the financial service industry for a long time. It is their dream, but if ever enacted, would soon become our nightmare---and theirs as well.
That sector of the industry would stand to gain hundreds of billions of dollars unnecessary to invest the kind of money needed to advance fund it under a typical defined benefit approach.
In short, privatize it, but keep it a defined benefit (DB) plan. Then add established and well-known actuarial and legal components that have been found to work well, and are needed in all such plans. No defined benefit pension plan can ignore these forever, not even Social Security.
The following are the eight reasons why we need to avoid privatizing it; i.e., by making it an individual account plan, known as a defined contribution (DC) plan:
1. The spread of the results would leave significant portions of the population behind---the very ones who need the system the most. This is unavoidable when you have large numbers of people investing and bearing their own results. The sharing of investment results under DB plans (when advance funded, of course) avoids this. In fact, a DB plan, when they have all three components in place, are perfect examples of the whole being greater than the sum of its parts. You simply cannot reproduce its advantages, using a DC approach. And you avoid the DC plan approach disadvantages completely. It also should be noted that most large employers have DB plans as their main plan, and most of these are integrated, or dovetailed, directly or indirectly, with Social Security. This means that any failure of Social Security to deliver adequate benefits will increase employer costs, often by a lot. Alternatively, they could of course, weaken or simply drop their pensions entirely---something, I thought, was not the oft-stated goal of our government.
2. Because of worse investment returns to be expected by the general population, individual accounts would cost at least 40% more than making it a true defined benefit pension system per dollar of benefit delivered. Investing for retirement is the single most difficult thing any individual will be asked to do in his or her lifetime, and professional money managers in defined benefit pension systems do it far better. And this does not even include the much higher expense of administration, nor the much higher investment management fees of DC plans, either of which by themselves is sufficient reason to avoid the individual account approach.
3. DC plans that cover large numbers of people have extremely high administrative costs. This fact is usually ignored or significantly downplayed by proponents of privatization. What's more, this burden will fall heavily on employers who are being asked to pick up the costs in all of these proposals. Small employers, the generators of most new jobs in the US in recent years, would be especially hard hit.
4. The costs of managing the money, in terms of fees, commissions, and transaction costs would be enormously higher under a DC plan than under a comparable size fund managed under a DB plan. The former is at least 10 times as high, and could easily be more than 100 times as high or more. The retail sector of the financial service industry heavily promotes stock trading and even speculation in long-term investing, the absolute wrong thing to do. This industry sees privatization as the long awaited pot-of-gold at the end of the rainbow---but for them, not for us. They have long been actively involved in contributing large amounts of money to right-wing conservative organizations like the Cato Institute. They often compound this insult by also providing egregiously incomplete, incorrect and highly misleading comparisons with Social Security.
5. It would exacerbate the aforementioned dangerously increasing wealth gap in the US. Recent attempts to assuage this element by putting 'minimum' benefits in for lower paid people, virtually always ignore the costs of such proposals, which is usually considerable. In addition, such minimums have the perverse effect of creating both freeloaders (when made voluntary, as in Chile, for example), or by encouraging excessive risk. No need to do this in a DB plan, where the investment results do not determine any individual's benefits.
6. It would increase stock market volatility, dangerously so. Speculation and frequent trading, promoted by many in the retail sector of the financial service industry, because of the fees they collect, has become the bane of much of the investing public, and this accentuates market volatility. Most large professionally managed DB plan funds---the wholesale sector of the financial service industry---trade far less, sometimes by specific direction from the plan sponsor, calming markets. This excessive trading is very difficult if not impossible to control in DC plans.
7. There would be a high probability of a stock market bubble, followed by the inevitable market crash. This results also from a strong propensity for the retail sector to promote momentum-based investment strategies to individual investors as a means to higher investment fees. These strategies, by their very nature, can create bubbles. Most large DB pension funds pursue value-oriented investment strategies and are contrarian, thus mitigating market bubbles. Contrarian strategies, by their very nature, are difficult to do for individuals, since they call for a high degree of careful research, discipline and buying stocks when they are temporarily out of favor. In a hyper speed world in which many people have too little time, lack a basic knowledge of historical returns, and are arithmetically challenged, to say nothing of the arcane and difficult world of the mathematics of finance, it is a game best left to the pros. Mutual funds are good for diversification and even professional management, but not for the all-important asset allocation decision, which, by far, is the single most important ingredient of market returns. There are also a bewildering number of such funds---more than there are stocks on the NY Stock Exchange---leaving the individual a daunting task. And please spare me the spectacle of a bifurcated Congress doing this---a Congress that can scarcely pass an effective campaign finance reform bill, in large part because of the effect of too much influence of money in the first place. Who needs that in a multi-trillion dollar major league important program?
8. It would lessen the efficient deployment of capital, compared to DB plans. These efficiencies are vital to keeping our capitalistic system healthy and creating new jobs. Individuals simply do not have access to the enormous range of investments that major pension funds have. It is interesting to note that some of the very same people who claim to promote efficient markets, also push privatization of Social Security, while totally ignoring this major point.
OOOOOO
Social Security has often been portrayed as a partial pay-as-you-go defined benefit system. True enough, I suppose, if you believe that the small pond near my home is like an ocean.
Size counts in this business---a lot.
The current Social Security system has assets about two times one year's benefit pay out, invested in Treasury securities, by law. This is no more than 5% of the accrued benefit obligations of the system.
No one knows the exact amount because it is not calculated anywhere in the SSAs reports. The numbers apparently range from 4 to 14 trillion dollars, the high end the number if a low interest discount is used, the small end if a higher rate is used. The rate of course is supposed to reflect the long-term investment strategy of the assets---and we have none---a vicious circle if ever there was one.
A typical DB pension system in private industry also has around that much in assets relative to one year's pay out, and also in short-term treasuries. The difference is that the latter plans also have from 30 to 60 times as much invested in other assets as well---a high proportion invested in stocks and in a variety of other investments as well. These assets are more than 100% of the accrued benefit obligations of these plans.
These assets are sometimes portrayed, as Bob Reischhauer and others of the 'Hold-the-Liners' policy, say are not necessary for Social Security because it is
1. permanent, and
2. the government has taxing authority, thus it can always increase taxes to make up any shortfall.
To it's permanence, I say, if it so permanent why are we here discussing how to maybe cut back benefits to keep it solvent and why are there no laws making it permanent? Why, for that matter,are there no laws (at a minimum) defining accrued benefits that cannot be taken away?
And of course, the government's taxing authority is quite limited, as much today than at any time I can remember. In fact, this great country was founded on a revolution when it threw overboard some British tea in Boston harbor, after a series of taxes like the stamp act tax and the tea tax was imposed against the will of the colonies.
Defined benefit pensions (with actuarial advance funding) had their origins as early as 1904. However, they grew rapidly back beginning in the period after WWII, when actuaries left the safety of the insurance industry to become independent pension consulting actuaries for large employers.
These companies asked them two questions---and both are vital:
1. How do we provide pensions to our employees based on their worth to us and their need, and
2. How do we do this in the least expensive way, but in a systematic, responsible way.
The answer to the first was by a defined benefit formula based on pay and service, and designed to satisfy needs at retirement by using the concept of replacement ratios; i.e., pensions that replaced a certain percent of pay at retirement, with the percent higher for the lower paid people.
The answer to the second was to invest for those defined benefits such that the amount contributed by the employer, together with the assumed investment returns exactly pays for the benefits provided under the plan formula.
The cost methods used by actuaries then--now known as Actuarial Cost Methods, typically spread out then payments over long periods of time, and typically were level or near level % of pay. Virtually all such plans had to deal with past service liabilities, similar to the one that is not even computed by social Security----and they dealt with them successfully too.
You cannot deal with a problem by sweeping it under the rug and not even computing it. Not calculating it does not make it disappear---only from view perhaps, until it snaps back to bite you. Better the devil you know and calculate, than the devil you don't know and won't calculate.
The use of a cash accounting system has contributed mightily to the problem. But we don't need to have the entire federal government change to accrual accounting to fix Social Security. The determination of the level or near level contributions using standard techniques in every annual actuarial report produced in such reports in the private sector is the right way to go. It is the only way to go.
There are 6 methods in common use and the one I recommend strongly is one called the Entry Age Normal or a variant thereof.
Essentially this method computes the normal cost as a level % of pay as if the plan had always been in effect from the first day participants begin accruing benefits.
It then determines the accrued benefit liability as the sum of these normal costs accumulated to the current date as if the system had been accumulating and investing these normal costs since that fictitious date. By subtracting the current asset amount from this number, you get an unfunded past service liability that must be paid off, typically as a level mortgage payment over a long time such as 40 years.
Adding the current year's normal cost to this payment, with appropriate adjustments for interest depending on when the payments are made, completes the annual required contribution.
I think you and those who have defended the system might be pleasantly surprised to find out that the cost of paying for both the past service liability and the current costs might not be all that much greater than the current taxes---perhaps 3-5% more---or about 300 billion or so each year.
When this is done---the Ben Franklin bonus comes in: The ongoing normal costs should drop down to no more than half and probably one third the current 12.6% combined employer/employee tax. I have never seen a large defined benefit pension system (with typical demographics) be out of the range of 4-6% of payroll.
