Taxable Wage Base

[Tom Poje]

It was announced to be 87,000, but do you know how that value was determined?

1994 Base $60,600

1992 avg wage index $22,935.42

2001 avg wage index $32,921.92

60,600 * (32,921.92 / 22,935.42) = 86,986.34

round to nearest multiple of $300

next year simply substitute the 2002 avg wage index into the formulas. the other numbers are locked in place.

...just another useless mathematical calculation brought to you by me.

[MGB]

Well, as long as we are talking about useless mathematical calculations -- do you know how the national average wage index is computed? Actually, that was the subject of my doctoral dissertation along with the effects that calculation has on SS and other indices over time.

The NAW is the average of all wages reported on 1040s. Individuals are not limited to the wage base for this, as is often incorrectly assumed. So, the IRS gathers the information, forwards it to the SSA and they do their thing with it. Given that you can file a return as late as October (with the extra 2 month extension beyond the automatic 4 month extension), it takes them until now to gather all of the input for the 2001 number that produces the 2003 wage base that was illustrated above.

Now, the real question is the following (which most people will answer wrong): If everyone's wages went up 5%, how much will the NAW go up? The answer is about 4%, not 5%. The reason is the demographic affects. The cohort group (mostly new retirees) leaving the average have high wages, and the cohort group (mostly young first-time job holders) being added to the average have low wages. Also, there are usually more entering the calculation than leaving it (an expanding population). This drags down the average computation. The thing to remember is that this is the average of wages, not the average of wage increases. Also, there are always "shocks" to the calculation, such as time periods when large amounts of layoffs or early retirement programs occur. One very noticable shock was during Clinton's first year in office. His administration kept touting the 10 million new jobs they created while the Republican retort was that they were all at McDonalds. Although average wage increases were over 4% that year, the NAW only increased less than 1% due to the high influx of low-wage workers. Similar affects have come from recent welfare reform as well as the prior increased participation of women in the workforce at lower wages. Another recent and future effect is all of the retirees that have been forced to return to the workforce at low-paying jobs that are often part-time. They are also included in the average.

What does this mean to SS benefit calculations? Researchers (especially anyone involved with SS reform proposals) constantly use an "average person" calculation in discussing what people get from SS currrently. They describe this as "...assume a person has had wage increases that are the same as the NAW increases." A person actually earning the NAW at retirement would have a SS benefit of about 42% of their wages under this setup and this number is often tossed around. However, if you take into account that the average individual has wage increases 1% per year higher than the NAW increase, the SS benefit is much less (their prior earnings are much lower than previously assumed) for the person earning the NAW at retirement. AND, if you also factor in that the average person retiring is earning more than the NAW, then the real average SS benefit is only about 32% replacement ratio, not 42%.

[Tom Poje]

And I thought I was a bit unstable! Nice response, obviously, you are anything but 'average'.

I see you have a long way to go to get to the ASPA conference.

oh well, I will be there!

[AndyH]

MGB,

I find your comments fascinating and very informative, thank you.

A followup question if I may: Based upon you knowledge, do you think there is any correlation between the wage base increase and the CPI, over either the short term or long term?

Perhaps it's tied to the price of hamburgers, or vice versa!

[MGB]

Yes, there is a correlation over the long run (too many shocks for the short run), but it is a second-order effect. The inflation is correlated to the wage increases, but then there is the demographic effect that makes correlations between the wage increases to the average wage very complex. Isolating the demographic effect over the long run requires inputs of new entrants and exits to the workforce which has major shifts over time due to immigration, economic conditions of the elderly, changes in fertility, etc.