This is not strictly a tax topic, although it involves income tax reporting. Since the forum is not exactly overwhelmed with traffic right now, hopefully this discussion can be indulged.
The following article raises some interesting points about how the average wage index (AWI) is calculated. It was actually written back in May, but for some reason I came across it more recently.
https://www.forbes.com/sites/janetno...take-a-big-hit
TL;DR version: because an unusually high number of workers will not receive a full year's worth of wages this year, the number used to adjust past wages for inflation will be abnormally low this year, dropping below the prior year's number for only the second time this century (and probably much longer than that).
Why is that? Well, I did some research and actually had to go find the regs related to Social Security. I found this very surprising info at ยง 404.211
"(3) For years after 1990, all remuneration reported as wages on Form W-2 to the Internal Revenue Service for all employees for income tax purposes, including remuneration described in paragraph (c)(2) of this section, plus contributions to certain deferred compensation plans described in section 209(k) of the Social Security Act (also reported on Form W-2), divided by the number of wage earners."
So, what I thought was some kind of semi-sophisticated measure of wage inflation is actually no such thing. It is a pretty crude and inaccurate measure that literally takes the sum of all wages reported on all W-2s for the year, divided by the number of people who received W-2s (multiple W-2s received by one person still counts as one person).
Hopefully the myriad problems with that are clear. First, it ignores self-employed workers altogether. Second, it doesn't take into account which wage earners worked a full year (let's call it 2,000 hours) versus those who worked more or less than that. So in 2020, if many people who were on the payroll in Jan and February are no longer on the payroll for most or all of the rest of the year, their wages will bring the average way down. Third, as students of statistics know, an average is often less meaningful than a median, as it can be strongly affect by outliers, such as those multi-million dollar CEOs we often read about.
The Forbes article is surprising accurate and detailed, I found, but the expert they quote omitted one key fact, even though he does talk about the "bend points" that apply. He says the 2020 number would result in a 9% reduction in SS benefits from the pre-coronavirus projection. But because of those very same bend points, he is wrong. For those already in the third tier, where the average indexed monthly earnings only count 15% toward the benefit calc, the effect will be signficantly smaller than a 9% reduction. For example, I did a calc on my own earnings, and found that if the AWI in my "base year" (two years before turning age 62) had been 9% lower, my benefit would only go down by less than 4%.
The author's suggestion is that Congress set a special limit that would prevent the base year AWI from ever going down, much the same as they do for benefit recipients in a year when the COLA does not keep up with Medicare deduction increases.
I have a better idea: instead of AWI, use the federal minimum wage. So, for example, if the minimum wage was $2 in 1975, and it is $8 now (approximate numbers), then the inflation factor would be 4. This has the benefit of being actually tied to hourly wage (a standard unit of measure) regardless of how many part time or full time workers there are in any given year. And it also provides Congress an incentive to keep the federal minimum wage at a reasonable level, because now it would affect not only current workers, but future benefit recipients.
The following article raises some interesting points about how the average wage index (AWI) is calculated. It was actually written back in May, but for some reason I came across it more recently.
https://www.forbes.com/sites/janetno...take-a-big-hit
TL;DR version: because an unusually high number of workers will not receive a full year's worth of wages this year, the number used to adjust past wages for inflation will be abnormally low this year, dropping below the prior year's number for only the second time this century (and probably much longer than that).
Why is that? Well, I did some research and actually had to go find the regs related to Social Security. I found this very surprising info at ยง 404.211
"(3) For years after 1990, all remuneration reported as wages on Form W-2 to the Internal Revenue Service for all employees for income tax purposes, including remuneration described in paragraph (c)(2) of this section, plus contributions to certain deferred compensation plans described in section 209(k) of the Social Security Act (also reported on Form W-2), divided by the number of wage earners."
So, what I thought was some kind of semi-sophisticated measure of wage inflation is actually no such thing. It is a pretty crude and inaccurate measure that literally takes the sum of all wages reported on all W-2s for the year, divided by the number of people who received W-2s (multiple W-2s received by one person still counts as one person).
Hopefully the myriad problems with that are clear. First, it ignores self-employed workers altogether. Second, it doesn't take into account which wage earners worked a full year (let's call it 2,000 hours) versus those who worked more or less than that. So in 2020, if many people who were on the payroll in Jan and February are no longer on the payroll for most or all of the rest of the year, their wages will bring the average way down. Third, as students of statistics know, an average is often less meaningful than a median, as it can be strongly affect by outliers, such as those multi-million dollar CEOs we often read about.
The Forbes article is surprising accurate and detailed, I found, but the expert they quote omitted one key fact, even though he does talk about the "bend points" that apply. He says the 2020 number would result in a 9% reduction in SS benefits from the pre-coronavirus projection. But because of those very same bend points, he is wrong. For those already in the third tier, where the average indexed monthly earnings only count 15% toward the benefit calc, the effect will be signficantly smaller than a 9% reduction. For example, I did a calc on my own earnings, and found that if the AWI in my "base year" (two years before turning age 62) had been 9% lower, my benefit would only go down by less than 4%.
The author's suggestion is that Congress set a special limit that would prevent the base year AWI from ever going down, much the same as they do for benefit recipients in a year when the COLA does not keep up with Medicare deduction increases.
I have a better idea: instead of AWI, use the federal minimum wage. So, for example, if the minimum wage was $2 in 1975, and it is $8 now (approximate numbers), then the inflation factor would be 4. This has the benefit of being actually tied to hourly wage (a standard unit of measure) regardless of how many part time or full time workers there are in any given year. And it also provides Congress an incentive to keep the federal minimum wage at a reasonable level, because now it would affect not only current workers, but future benefit recipients.
Comment