Professor Martin Feldstein's recent Wall Street Journal article, "Private Accounts Can Save Social Security" is a false alarm. He declares:
There are now three employees paying Social Security taxes to finance the benefit of each retiree. That number will fall over the next three decades to only two employees per retiree. This would require either a 50% rise in the Social Security tax rate to maintain the existing benefit rules, or a one-third cut in projected benefits to maintain the existing tax rate.
Yes, those projected ratios of employees to beneficiary numbers are (very) roughly correct. The consequences he attributes to them are not. He reaches them by ignoring other crucial variables -- notably marvelously improved employee productivity, which in turn generates higher pay, which translates into higher Social Security program revenue per employee. Professor Feldstein focuses solely on those remitting payroll taxes under the Federal Insurance Contributions Act (FICA), omitting to mention that Social Security has two other dedicated revenue streams: income tax on half the Social Security benefits of high earners and interest on the trillions the Social Security Trust fund lends to the U.S. Treasury. Those data are not hard to find; the Social Security trustees' annual reports describe them under "Highlights". (See page 3 of the trustees' 2010 report.)
To reach his conclusions, FICA contributions, which are determined as a specified percentage of each employee's pay, would have to stall over the next three decade. But economic data show otherwise. Each successive wave of workers produces more and earns more per capita because employers constantly respond to pressure to provide improved technology, enabling employees to generate more goods and services and earn more. Just as one example, the non-stop computer revolutions (yes, plural) repeatedly multiply what the average employee produce and earns, generating larger and larger FICA contributions. As recently as May 4, Intel announced yet a new advance -- chips that will be as much as 37 percent faster and require half the power. So future workers will generate more goods and services for all of us to share, more FICA per capita and at lower cost.
The 2011 Statistical Abstract of the United States (Table 641) shows that between 1990 and 2000 non-farm employee output per hour increased by over 20 points (using 100 in the early 1990s as the base) and manufacturing per hour output expanded more than 45 points while real (adjusted for inflation) per hour compensation expanded by about 15 points. These crucial elements continued to expand until the 2008 onset of the recession. Even then hourly output in 2008 and 2009 was about double that in 1990. Bottom line: the 2009 employees generated almost 30% higher per capita income than their 1990 counterparts.
But there's more. Social Security FICA income has exceeded benefit payout since 1983 until this past year, building its reserves to about $2.4 trillion (and growing); trust Fund resources enabled the payment of full benefits. Current actuarial projections indicate that trust fund reserves can make such contributions to benefits in the amounts needed until 2037.
But the cap placed on earnings subject to FICA, in effect, exempts high income from FICA, putting a brake on more ample trust fund accumulations. Removing the cap and crediting the additional earnings for benefits would almost offset the entire projected funding shortfall. (See, National Academy of Social Insurance, "Fixing Social Security: Adequate Benefits, Adequate Financing", page 29.) Even the co-chairs of the Fiscal Commission advocate raising the cap (although they would use a slower boat and not go as far).
Alternatively, raising the FICA rate by one percentage point for both employees and employers would offset most of the shortfall by itself. Projected earnings would improve by a greater amount, making such an increase completely affordable. Much the same effect would be achieved by raising the employee and employer FICA rate by 1/20th of one percentage point cumulatively over twenty years. Both removing the cap and boosting the FICA rate as described would make possible benefit improvements -- which are highly desirable given that other sources of retiree income have declined, proven unreliable and tend to shrink with age.
Professor Feldstein makes it sound as if Social Security is doomed to failure by demographic changes. But that's far from the case. Modest changes like those described, which have wide support, would assure Social Security's future -- without trimming benefits. Instead, he plumps for private accounts, omitting that they would incur substantial non-benefit costs. In a 2004 paper, Austan Goolsbee, then a University of Chicago economics professor, using analyses by the Congressional Budget Office and the Government Accountability Office, concluded that the George W. Bush version of private accounts "would lead to a massive increase in payments of financial fees to private financial management companies." His estimate of the net present value of such payments -- $940 billion.
Dr. Feldstein asserts that the private accounts he advocates are like those proposed by President George W. Bush. Perhaps that's all we need to know. But, as I heard one Republican U.S. Senator declare when he and I were on a public panel: "Social Security private accounts are like taking the mortgage money to bet on the races." Not a great idea.
In sum: Dr. Feldstein in Wall Street Journal-mode is no more reliable a guide to Social Security than Dick Cheney is a safe hunting companion.