The president has proposed cutting the cost-of-living adjustment for Social Security by tying it to a “chained” consumer price index that rises more slowly than the current one, but with a partially offsetting benefit enhancement intended to protect vulnerable beneficiaries. In a fact sheet about the chained CPI belatedly released by the Office of Management and Budget, the White House claims: “Because of the benefit enhancement for the very elderly, the Budget proposal would not increase the poverty rate for Social Security beneficiaries, and would slightly reduce poverty among the very elderly according to SSA [Social Security Administration] estimates.”

This claim deserves scrutiny, since the SSA estimates aren’t shown and the description of the “benefit enhancement” was slow to be released and remains sketchy. While we await further information, a quick look at the benefit enhancement casts some doubt on this claim, since retirees with average benefits would see a benefit cut until their late 90s, and few retirees live that long.1

The chart shows how the nominal benefit might increase for a prototypical retiree with the current COLA (tied to the CPI-W), the proposed COLA (tied to the Chained CPI-U), and the chained COLA with the proposed benefit enhancement. The projections assume that the benefit at age 76, when the birthday bump first takes effect, is equal to the average benefit for all retirees, which appears reasonable based on Census data. Though the chart shows the annual benefit up to age 100, life expectancy at age 65 is around 20 years, so what matters for the typical retiree is the area to the left of the shaded rectangle.

The benefit enhancement—a.k.a. the “birthday bump”— eventually equals 5 percent of the average retiree benefit but is phased in over 10 years, starting at age 76 and again at 95. If the bump were in place today, the maximum amount would be around $760 dollars per year, phased in at $76 dollars (1/10 of $760) at age 76, $152 at age 77 (2/10 of $760), and so on.

As shown on the chart, by age 85, the bump is sufficient to offset the COLA cut. Then, strangely, it goes away and reappears at age 95, explaining the weird squiggle. The reason the birthday bump has a bigger effect at older ages is that the average retiree benefit—and therefore the birthday bump—grows faster than the COLA (the projections are based on the assumption that the bump grows by 4 percent per year, based on Social Security wage growth projections).

If most retirees see a reduction in benefits, how can there be no increase in senior poverty? One explanation is that the poverty rate is higher for the beneficiaries age 97 and older who would see an increase in benefits. But since only 4 percent of 62-year-old men and 9 percent of 62-year-old women are expected to live to 97, this seems like a partial explanation at best. An additional explanation is that the benefit enhancement is based on the average retiree benefit, which is higher than the average spousal, survivor, or disability benefit, so the poverty rate may increase among seniors receiving retiree benefits while declining among those receiving other kinds of benefits. A final explanation is that the birthday bump is a flat dollar amount and therefore has a bigger impact on those receiving lower benefits.

The upshot is that the birthday bump may prevent an increase in the share of seniors living below the official poverty line, though this claim deserves further research. Regardless, there can be no doubt that the president’s proposal will reduce the incomes of millions living just above the poverty line and increase hardships among seniors and other beneficiaries.

Endnotes

431