"Retiring the Earnings Test—At Last"
Federal Reserve Bank of St. Louis Central Banker, Summer 2000.
After more than 60 years of older Americans having their Social Security benefits reduced because they earned too much income at work, the burden has finally
been lifted. In April, a new law eliminated the Social Security retirement earnings test for those between 65 and 69 years of age. Prior to its repeal, the earnings
test for persons in this age group subtracted $1 in benefits for every $3 in earnings above $17,000 (in 2000) until all benefits were exhausted. Now,
those in this age group will continue to receive full benefits, regardless of their earnings from work.
Will this move prompt Social Security recipients to work more hours because their benefits won't be affected? This is difficult to answer. Those who already earn
enough to reduce their benefits to zero are much less likely to change their work habits. In fact, these people might actually work less because they will now have
higher monthly incomes without working more—an "income effect."
How the work habits of those who are working and receiving a reduced benefit—but still receiving some benefit—will change is more difficult to pin down. The
less someone in this situation earns, the more likely he is to increase his work hours. Why? Because he views eliminating the penalty as if it were a wage increase—a "wage
effect." On the other hand, the more he earns—that is, the closer he is to reducing his current benefit to zero—the less "wage effect" and more "income
effect" he feels. Which one prevails depends on the balance of the two, since one effect encourages more work, while the other favors less.
Those who consciously restrict their work hours to avoid the penalty are the most likely to work more because the wage effect is very strong for them. Those already
earning less than $17,000 should experience no change in work hours because their benefits were not being reduced anyway.
How many people, then, will this change actually affect? The Social Security Administration reports that approximately 800,000 beneficiaries aged 65-69 (of more than 7 million
in total) lost some or all of their benefits under the earnings test in 1999. This figure includes an estimate of the number of people who chose to delay retirement—that is,
not even file for Social Security—because they knew that they earned too much to collect any benefit. It doesn't include an estimate of how many consciously earned less
than the exempt income amount to avoid the penalty.
Yet in terms of the number of additional hours people might choose to work, this last group, if it exists, is undoubtedly the most important. How important? In a recent
article, economist Leora Friedberg noted that some older workers have consciously adjusted their work practices to stay just below the exempt income amount. In fact, the
gap between those earning just below and just above this threshold is noticeable and important. Eliminating the penalty, Friedberg argued, would increase average hours
worked for those near the threshold by about 5 percent. This translates into an average of about two additional work hours per week from each of these workers.
In the grand scheme, however, the total increase amounts to much less than 1 percent of all hours worked in the economy. Nevertheless, the importance to most economists
of eliminating the earnings test lies not in the additional hours some may choose to work—even in these times of tight labor markets—but rather, in the removal of a government-imposed disincentive to work. Sixty years ago, the earnings test probably seemed like a good idea. More recently, it had begun to appear anachronistic. Change
can be for the better.
For additional information on this subject, see Zaretsky's article "Working
in the Golden Years and Paying for It—The Retirement Earnings Test" in the January 2000 issue of the St. Louis Fed's The Regional Economist.
