FREE NEWSLETTER
SOCIAL SECURITY retirement benefits are one of the most complicated topics in financial planning. As you try to figure out how much you might receive, there are thousands of rules, different types of benefit and numerous scenarios to evaluate.
And then there’s the impact of COVID-19.
It turns out that this year’s economic slump, which caused the economy to shrink by a tenth in the second quarter, may interact with Social Security’s methodology to hurt those who turn age 60 in 2020.
It’s a tad confusing, so bear with me while I explain. Social Security benefits are based on your lifetime earnings. To make sure earnings can be compared across time, your actual earnings from each year are adjusted, or “indexed,” to account for differences in average wages from one year to the next. This puts each year’s earnings on a comparable footing.
Social Security then calculates your “average indexed monthly earnings,” or AIME, based on your 35 years with the highest index-adjusted earnings. Think of this as your average career earnings. Your AIME is then used to calculate your base benefit amount as of your full Social Security retirement age, which will be either age 66 or 67. This base benefit is also called your primary insurance amount, or PIA. It’s the amount all other benefits are based on, such as spousal and family benefits.
With me so far?
This is where things get funky—and it has to do with the methodology that Social Security uses to calculate your all-important AIME, that measure of your average career earnings. Each year, Social Security calculates an Average Wage Index, or AWI, based on economywide earnings for that year. Your lifetime earnings are then indexed to the economy’s earnings as of the year you turn age 60. Any earnings received after age 60 are not indexed, but instead they enter the benefit formula in nominal terms.
This creates a huge problem for those who were born in 1960 and hence reach age 60 this year—an issue highlighted in a recent paper by Andrew Biggs and published by the University of Pennsylvania’s Wharton School. What’s the problem? To calculate each year’s AWI, the Social Security Administration divides aggregate payroll earnings for that year by the number of workers who receive W-2 tax forms.
If you turn age 60 this year, all of your earnings from earlier years will be indexed based on the Average Wage Index for 2020. But so far, 2020 has seen record unemployment and a serious economic downturn. The number of W-2s issued shouldn’t change much—remember, lots of people were working at the beginning of the year—but aggregate payroll earnings will likely be significantly lower, because so many of these workers got laid off. This will cause 2020’s AWI to be far lower than expected. That, in turn, means that when your lifetime earnings are indexed to this year’s AWI, you’ll end up with a lower AIME—and hence a lower Social Security benefit.
According to Biggs’s paper, “For 2018, the most recent year for which data are available, the AWI was $52,146. The 2019 Social Security Trustees Report estimated that the nominal AWI for 2020 would be $56,396, or 8 percent higher. This is highly likely to be overstated due to the Coronavirus-related economic downturn.”
Biggs estimates the reduction in AWI could cause annual retirement benefits for workers born in 1960 to be lower by between 13.6% and 14.3%. He calculates that a medium-wage worker might see his or her annual retirement benefit fall by $3,900, while the highest earners could see reductions of some $6,500 per year.
The paper goes into a lot more detail about the benefit formula and some secondary impacts of the reduced AWI. I’d recommend reading it if you’re interested in gaining more insight than I’ve provided in this (relatively) simple summary.
We won’t know 2020’s actual AWI until the end of the year. It could go up or down, depending on economic conditions. What’s the fix? That’s in the hands of Congress. My advice: Contact your local senator and member of the House of Representatives, and let them know about the problem.
Richard Connor is a semi-retired aerospace engineer with a keen interest in finance. Rick enjoys a wide variety of other interests, including chasing grandkids, space, sports, travel, winemaking and reading. His previous articles include Refi or Not, Working the Numbers and Summer Job. Follow Rick on Twitter @RConnor609.
Want to receive our weekly newsletter? Sign up now. How about our daily alert about the site's latest posts? Join the list.
Absolutely fascinating. While reading I flashed back to Social Security benefits for those born in 1919 being less than 1918 or 1920 due to a legislative quirk, but which had a significant impact to those of that birth year. (My father was born that year so I had occasion to know of it). Anyway, I mention the 1919 birth year quirk because for a few years there was an effort to request Congress correct that seemingly random, unfair quirk. (Which came to naught.) Anyway, I will also red the cited article. Thank you.
Way back when I was a teenager in the ’70s, I thought “2020”—the year I would turn 60—had such a far-off, futuristic ring to it. Now that it’s here, it’s all too real. I think we have an “interesting” decade ahead of us.
Interesting discussion of a very complex topic Richard, thanks. After looking over that paper you referenced, I was rather surprised at the complex fixes that the author proposes. I would think that they would be a pretty difficult sell, as some of them would require basic changes to the whole system, which of course make lots of people nervous. A much simpler fix that I’ve seen proposed would be one that simply uses the previous year’s AWI data to smooth out the bump. And going forward, comparing the current year’s AWI to the previous year’s AWI, and using the higher of the two, should minimize big differences for future retirees.
I agree completely.
Thanks, Richard! Always wondered how SSA calculated benefits. Dropped a note to our representative. It might be hard to get action on this, but worth trying.
Interesting insights into the wage indexation…. one would think that the number of pay periods might enter into the annual calculation too, I think that would also likely offset the $ wage differential. I also like the simplicity of using the higher of current or prior year indexation, proposed above.