Firmly as I support the principles of the President’s Social Security reform plan, I expect to be lukewarm about their initial instantiation. Most of the work of filling in the details will fall to young “policy wonks”, who are at least as strongly attracted to complexity and compromise as to the opposite sex. Reform, Mark I, is likely to be a confusing blend of the old system and the new, phased in over a generation, never fully attaining the goal of “privatization” and replete with anomalies created to skirt this or that short-term problem (often converting it into a long-term problem in the process). We have a foretaste in the incomprehensible Medicare prescription drug benefit, which succeeds in displeasing everybody.
A major difficulty in formulating programs of this magnitude is that the wonks tend to be more interested less in where they are going than in how they will get there. It may be useful to begin with the objective. Once the transition has been completed, what would a sustainable retirement income system look like?
The key to reform is to minimize the dependence of each generation of retirees on the generosity of its successor. Workers should set aside a portion of their income for consumption after they leave active employment. A libertarian regime would let them figure that out for themselves, but we are taking government paternalism for granted here. Saving and investment will be mandatory, just as income transfer between generations is today.
An individual who consistently saved and invested the 12.4 percent of his wages that he currently pays in FICA taxes would be very likely to accumulate over his working lifetime enough capital to provide a decent basic provision for retirement. For the median worker, Social Security benefits under the present system equal 40 percent of pre-retirement income. How difficult would it be to match that through individual investment?
As an illustration, I did calculations for someone who begins working at age 25 and retires at 65. I assumed that his pay would increase over time but at a decreasing rate, reflecting the typical pattern of rising through the ranks followed by attainment of a plateau (7.5 percent annually, in constant dollars, for 10 years, then 5 percent, 3 percent and one percent for each of the next ten). I applied the rule of thumb that ten dollars at age 65 is sufficient to buy one dollar of lifetime income. Based on those assumptions, a real investment return of a mere 1.4 percent per annum produces a 40 percent replacement of income at age 65. The historical real rate of return for stocks over the past 75 years was 6.5 percent. If our sample worker enjoyed that result, he would have more income in retirement than while he was working.
Figures like these suggest that substituting private capital accumulation for unfunded government promises is not an absurd idea. We may, however, wish to add a few “safety nets” to take account of the vagaries of the financial markets and the variety of career patterns.
One precaution, already put forward by Administration spokesmen though obscurely described, is the establishment of what is called in the private sector a “floor-offset plan”: At the time of retirement, a notional benefit would be calculated based on the factors currently used by Social Security (essentially, wage history and length of service). The value of that benefit would then be compared to the balance in the worker’s private account. If the account was smaller, the government would add money to bring it up to the “floor” benefit.
This protection against downside risk for workers carries risks for the government. A market downturn could generate unexpected liabilities at an inopportune time, and individuals would have incentives to make excessively adventurous investments, on the theory that the government was insuring them against catastrophic loss.
Steps can be taken to minimize these hazards. First, the floor benefit can be set at a level where it is a true safety net, that is, somewhat lower than the present level of Social Security benefits. While Social Security, as originally conceived, was an equal leg of a “three-legged stool” (along with employer-provided pensions and voluntary savings), it has developed into by far the longest leg for most retirees. With private accounts, it still would be, but the guaranteed floor ought to serve the more modest original purpose of warding off destitution. (Note that, because the Social Security formula is graduated downward, the safety net is strongest for the lowest wage earners, who of course are most likely to need it.)
Second, investment options can be constrained to weed out high-risk choices. That policy would in any case be congruent with the paternalism of the system as a whole. The limitations don’t need to be draconian, just enough to enforce diversification, with perhaps a little attention to duration (
Third, the ups and downs of the market can be smoothed by converting accounts into annuities over, say, a ten-year period before retirement age, thus reducing the risk of retiring at a moment when the value of one’s account is at a temporary low point.
The outline that I have just given is what Social Security might have looked like had it been designed in a time of prosperity rather than in the midst of the most severe depression in our nation’s history, when the prime consideration was getting checks out to impoverished oldsters. As others have pointed out, FDR himself did not see the structure fashioned under emergency conditions as permanent. When he introduced the concept, he said that it “ought ultimately to be supplanted by self-supporting annuity plans”, which is simply the 1935 lingo for personal accounts funded by workers’ own contributions.
It’s a sign of how hidebound liberalism has become that President Roosevelt’s epigones cling tenaciously to an accident of history as if it represented a profound principle. No longer, it seems, are conservatives the ones standing athwart history yelling, “Stop!” or elevating ideology about the best interests of the elderly.
Comments