While
there is generally agreement across a broad range of the political
spectrum that Social Security is unsustainable in its current form,
there is less consensus about what to do to assure long-term stability.
Most democratic policy makers, including Presidential contender Joe Biden, would increase tax revenue available for benefits by raising the wage base — the amount of earnings subject to Social Security taxes. (For 2020, that amount is $137,700.) Others would like to contain the costs of the program. Various bipartisan proposals would do both.
For
those focused on the expense of the system, one of the favored
approaches is to change the measure of the cost-of-living adjustment, or
COLA. The COLA is an essential component of the program. Under current
law, Social Security automatically increases benefits annually based on
the Consumer Price Index for Urban Wage Earners and Clerical Workers
(CPI-W) to offset the impact of inflation. Reformers would change that
process to use the Chained CPI for Urban Consumers (C-CPI-U) to offset
the rise in the cost of living.
The appeal of the shift of the COLA to a C-CPI-U to the spend-less-crowd is easy to understand. It saves money. The Congressional Budget Office estimates that this policy option would save the system more than $100 billion over the coming decade. That savings comes disproportionately from those with the highest benefit levels. Moreover, supporters of the change sell it as a more accurateAccording to the SSA this policy option would address about 20% of the shortfall.
measure of inflation than the current index used. If you listen long
enough, the solution not only makes the system more stable, but actually
improves the inflation adjustment process.
The
problem is that C-CPI-U is not a better measure of inflation. The index
does not even measure true inflation. It actually measures the cost of
living, which in part reflects the behavioral response to inflation.
To
illustrate the impact of behavioral response, two years ago the rising
cost of health insurance forced me to switch to a less expensive
alternative. In order to keep my new insurance cost roughly the same as
my old cost, I had to double my deductible. In my case, it is possible
to say that my cost of living hasn’t changed ---assuming, that is, I
don’t have medical costs that force me to pay the full higher
deductible. But it is insane to say that there was no inflation in
healthcare that year.
While
it is true that in any year the savings from switching to C-CPI-U will
come disproportionately from those receiving the biggest Social Security
checks, there is nothing to suggest that these people are particularly
well off. Moreover, data from the Social Security Administration
suggests that the large reductions from this policy option may well come
from those most in need because the relative importance of Social
Security benefits to recipients’ overall living standard rises with age.
As
people age, most beneficiaries have fewer work options and have used up
some, if not all, of their retirement savings to cover costs that go
over the level of benefits. As a result, we tend to depend on Social
Security benefits for a greater percentage of our total income as we get
older.
The
Social Security Administration's statistics support this thesis. Social
Security benefits account for 90% or more of income for more than 1/3rd
of seniors. That percentage of seniors who fall into that highly Social
Security dependent group nearly doubles between the ages of 70 and 80.
This trend is a great concern because a typical retiree can expect to
live to 85.
Unfortunately, the impact of lower COLAs accumulates over time. So the chained-COLA would apply progressively larger reductions to people as they age. For example, FactCheck.org reports that a reduction of .25% in the annual inflation adjustment will lower
the buying power of a retiree’s monthly check by 3 percent after 10
years. By the time that retiree is 95, the lower COLA adjustment will
have cut the value of the check by 8 percent, relative to what he would
have with a COLA that is 0.25% a year larger.
I am not going to tell you that CPI-W is the correct measure. The fact is
that there isn’t a single best measure for inflation. Inflation is a
theoretical concept that varies by person and by location. Some argue
that an experimental index which takes into account the spending patterns of those 62 and older
(CPI-E) is a more accurate benchmark for Social Security. Among other
things, it reflects the higher share of their spending which goes to
medical care, where costs have been rising faster. Using CPI-E would
tend to make Social Security benefits increase faster.
Bottom
line: the shift to a chained-CPI would fix our old-age insurance
program with a solution that lowers the buying power as one ages. This
option is like buying auto insurance that declines in value as the car
wreck gets worse. Fixing Social Security's finances in the most
responsible manner requires that we look at more than just how much a
change will save.