Once you reach age 62, the only benefit to working longer is increasing your highest 35 years. Your taxable wages are no longer indexed up annually by the NWI. The only variable inputs appear to be your highest 35 years and the age at which you take the annuity. Unless you are really going to boost the 35 year calculation in a last burst of salary increases, there is not much of an impact on the payment.
If we see a lot of wage inflation over the next few years, the folks approaching or beyond 62 are going to be hurt if they depend on the SS benefit in retirement. The NWI portion of the PIA calculation is or will be locked in, and they will be stuck with a lower benefit.
It's hard to put these formulas in words. But I would say this a little differently:
Once you have 35 years of wage history, the only benefit of working additional years is the replacing an earlier lower indexed income year with a later higher indexed income year.
Whether you work past 62, or you don't work past 62, your earlier wages are only indexed up to the year you turn 60. So the gap between (this year's post-60 wage) and the (lowest prior indexed wage) will be somewhat greater than it would have been if past wages were indexed beyond 60. It seems to me this provides slightly more benefit from working (but the benefit is usually small regardless).
Whether you work beyond 62 or don't work beyond 62, the bend point used for your PIA is fixed at the year you turn 62, and prior wages are only indexed up to age 60. However, after applying the bend point formula to get a PIA that would be used if you start benefits at age 62, that PIA is then increased with the CPI growth from 62 to your start year. (And, of course, after starting benefits your benefit continues to grow according to the CPI.)
It seems to me there's a gap in the SS indexing system for the two years from age 60 to age 62. This impacts everyone, regardless of when they quit working or start benefits. People who happen to be born in years that correspond to high wage growth in those two years get lower benefits relative to their wage histories than "equally" paid workers who happen to be born in years that correspond to low wage growth in those two years.
However, the final purchasing power of their benefit depends on the relation of wage growth and price growth. The first group could get higher benefits in terms of purchasing power if wages outran prices in those years.
All this comes from my non-expert reading of the example I linked above
Social Security Retirement Benefit Calculation