Question about SSA retirement benefits estimate

Social Security benefits are figured on the "average indexed monthly earnings." This is figured by adjusting each year's covered earnings to account for inflation, and then averaging the highest 35. In doing this, they project forward to retirement using the last year's income. So if your wife's income dropped last year, they may be using a smaller number for several years than in previous projections, resulting in a smaller estimate. See

more details and examples. Dave

The explanation is written out clearly right there on the statement you received. It's the section labeled "How Your Benefits Are Estimated". Did you read it?

You seem to have that backwards. According to a 2001 General Accounting Office (GAO) report to Congress "IRS Telephone Assistance" (GAO-02-212), the "tax law correct response rate" was 79%, not 25% as you suggest.

You don't have to wonder -- the statement you received clearly explains that the benefit is _not_ based on taxes paid, and also what it _is_ based on. It also explains that unlike the estimate, the actual benefits cannot be determined until you actually apply for them.

-Mark Bole

Thanks. Your summary and the information at that URL was very helpful.

That is not my understanding of their algorithm. Instead, I believe they take the actual earnings for the highest of up to 35 years worked, multiply each by a factor based on average wage indexes for that year and for the year in which the person turns 60 (which I believe is intended to restate the all earnings in the same age-60 future dollars), sum those future-value amounts and divide by the actual months worked. That is the average indexed monthly earnings (AIME).

The monthly benefit (aka primary insurance amount, PIA) is based on the sum of percentages of portions of the AIME; for example, 90% of the first $680, 32% of the next $3420, and 15% of any amount over $4100. The cut-off amounts ("bend points") -- $680, $3420 and $4100

-- depend on the year in which the person turns 62, as well as on how long after that age the person waits before drawing benefits. Finally, the PIA may be reduced depending on how soon before the full retirement age (e.g. 66) the person begins drawing benefits.

Whew!

Yes. But I think it is not so much because her lower earnings were in the last year, but simply because the lower earnings (had they occurred in any or the 35 years counted) drags down the average.

Hmm, does that mean that it is better to stop work cold turkey rather than to taper off (earning less), as many people do?

I believe so, if my understanding of the algorithm is correct.

Thanks again for the pointer. It is exactly what I needed.

[...]

Your understanding is incorrect. As stated, the estimate of your benefits on your annual statement does project forward using the most recent level of income. Please read the statement.

If by "better" you mean getting more government welfare money for less work, the answer is no.

Try this link:

From a financial planning viewpoint, you are almost certainly better off continuing to work at a reduced level than stopping "cold turkey". Not only might you increase your eventual SS benefits, but with earned income you have many more opportunities for tax-advantaged savings and health care, plus of course continuing to meet current living expenses without having to use retirement savings.

-Mark Bole

Common sense should tell you this is wrong. Consider two people, one who worked for 10 years (to qualify for Social Security), and one who worked 35; assume the first 10 year's earnings were the same, and the second person worked at reduced pay the next 25 years.

Who paid in more to Social Security? Who should thus get more benefits? But whose average (the way you computed it) is higher?

Answers: The person who worked longer (but not more than 35 years) paid in more and should get more back. The first person's average is higher, because you chose to divide by 120 (10 years * 12 months), rather than 420 (35 years * 12 months).

The flaw is that you don't divide by the number of months worked, but by the number of months in 35 years: "When a worker has fewer than 35 years of earnings, Social Security law requires that zero years be included, which reduces the average. For example, a women [sic] with only 25 years of lifetime earnings has her retired-worker benefits computed using those 25 years (indexed) plus 10 zero years, but still the total is divided by 420 to determine the AIME."

No. Se above. However, if you've already reached 35 years, and subsequent years are at lower wages (adjusted for inflation), then you don't gain any Social Security benefit by paying in those extra years. You may not be better off (from a Social Security perspective) by stopping work, but you are no worse off.

Mark Freeland snipped-for-privacy@sbcglobal.net

I don't disagree with your examples or analysis. But from a financial planning standpoint, I think it's important to emphasize that Social Security is NOT a pension plan or an investment, and your payroll taxes are just that: taxes, with no direct impact on your ultimate benefits.

Consider that many people can receive benefits without ever paying any SS taxes, such as ex-spouses after ten or more years of marriage, survivors of deceased retirees, or dependent children of living retirees.

Or consider a person who earns $180K from one employer in a year vs. another person who earns $90K from each of two employers in the same year. The employer share of SS taxes will be roughly double for the second wage-earner but will not change his benefit calculation at all compared to the first.

Or consider that two people who paid identical taxes on identical earnings before retirement may not receive the same net benefits due to other income in retirement (resulting in higher taxes on SS income, or resulting in lower benefits due to earnings before full retirement age).

Trying to analyze and maximize eventual SS benefits prior to early retirement age is one of the least productive or useful financial planning activities I can imagine.

-Mark Bole

That there is a function mapping your payroll deductions (not your employers') into your (social security) annuity payments IMHO shows that there is a direct, arithmetic impact on ultimate benefits. There are also additional possible payouts, as you note below. The benefit one receives is insurance - the assurance that those payments will be there if qualifying events occur:

"The social security measure gives at least some _protection_ to thirty millions of our citizens who will reap direct benefits through unemployment compensation, through old-age _pensions_ and through increased services for the protection of children and the prevention of ill health."

FDR Statement Signing SSA, Aug 14, 1935.

They receive payments. The worker received the benefit - the insurance-like coverage of familiy.

Similar to a private pension - if you have other income, then the after tax value of the payments is reduced because the other income puts the incremental pension income into a higher bracket. And so on.

It depends on how one choses to view/classify the benefits; ultimately what you call it doesn't matter.

Amen.

Mark Freeland snipped-for-privacy@sbcglobal.net

[...]

Ah, if only the insurance-like coverage were voluntary -- then I wouldn't be calling it "taxes"! ;-)

-Mark Bole

If it were voluntary, many of the people who need it most would not have it. Then, I guess it would be necessary to bring back "poor houses," "alms houses," or whatever you want to call them. Or, it might be necessary for some people in their old age to lie down in the streets and starve, as has happened in some places throughout history. Civilized society means that many things must be deliberately planned and arranged for everyone's good and not at all voluntary. You might voluntarily choose to shoot your neighbor, but that will never do. Voluntarily failing to save money for old age will not do either.

This might take a bit of work, but Nomail1983 could type the social security history in the online SSA benefits calculator, zero out 2006 and see if there is something quirky in the formula. It would give the same answer as the 2005 mailing because there should be the 2006 income inflator added in.

I've personally found the online calculator to be pretty close to what the mailing says.

To get the most accurate possible simulation of the method actually used by the SSA, download the detailed calculator from

and install iton your PC. Type in her entire earnings history and save it. You cannow do "what if" scenarios as desired.

Please quote what the statement says about this. I still do see anything about "project[ing] forward using the most recent level of income". In fact, I believe your understanding is incorrect. See below.

Thanks; good pointer.

In fact, I believe Thomas's explanation on that web page along with

is similar mine, withjust one important correction. The sum of the highest of at most 35years of earnings is always divided by 420 (35 * 12), not the numberof non-zero months, as I said. But the point is: the monthly retirement benefit (PIA) is based on the AIME, the __average__ index monthly earnings. It is an __average__, not a number "project[ed] forward using the most recent level of income". Please read Thomas's explanation.

Having just received my statement last Friday, I can quote what is on my statement, just below my estimate in the paragraph on the left, bold type: What we assumed

"If you have enough work credits, we estimated your benefit amounts using your average earnings over your working lifetime. For 2007 and later (up to retirement age), we assumed you'll continue to work and make about the same as you did in 2005 or 2006. We also included credits we assumed you earned last year and this year."

I think that clearly states your wife's estimate is based on her continuing to earn at the reduced amount, thus her average earnings will be less than was projected for the previous years.

But will she be continuing to work until age 70 and thus not taking benefits until then? You seem to be comparing this and the last estimate of her SSA Benefit at that age. There was a really good discussion here a few months ago on the financial advantages of starting SSA benefits and varying ages. Age 70 may or may not be the most advantageous time for her to begin receiving her benefit. You may want to google this group. There were links to some good articles, too. I know it opened my eyes.

Elizabeth Richardson

[...]

There are two separate things here: the annual estimate, and the actual benefits calculation once you actually file a claim for benefits. I could be wrong but I think you keep on trying to combine the two.

The estimate calculation DOES use only most recent earnings, not an average, to FORECAST the numbers that WOULD BE used in actual calculation. So for example if you were unemployed last year, that annual statement will show a much lower estimate, similar to the situation for your wife, because it will assume the same lower earnings for all future years until retirement. If you then resume full employment, the estimate on the following year's statement should go right back up to approximately where it was (except possibly for the slight downward drag due to the low income year).

Assuming that you don't yet have your top 35 years of full-time earnings to report, then if you know your most recent reported year was abnormally high or low, you should really just disregard the estimate numbers printed on the statement, since they are based on an assumption (continued future earnings at the same abnormal level) that you yourself probably wouldn't make. You could then use one of the interactive calculators in this situation.

-Mark Bole

I see what you and Mark mean now. And I do now see that explained in the statement. When I read it before, I thought it did not apply because we already have more than our 35 years of non-zero earnings. But I see now that it does apply; and it makes sense to me that it would, since projecting the last year's earnings (subject to Soc Sec tax) forward might replace years with lower earnings.

And yes, that does explain why my wife's estimated future retirement benefits (PIA) took a hit, whereas mine did not, as you explained earlier.

But I especially appreciate your pointer to (and quoting of) the URL above [truncated by Google Groups, not I]. Not only does it explain things more fully, but also it makes it clear that I want to use the online calculator because both my wife and I retired early. By that, I mean we stopped working, not that we started receiving Soc Sec benefits.

I have some more questions germinating; mostly nitpicks and what-ifs. But I need to digest Kaye Thomas's explanations, which Mark pointed to, as well as all the information at ssa.gov, which you pointed to.

Again, thanks very much. Your inputs have been especially helpful and easy to receive.

Thanks. I found the thread "Optimizing Social Security Benefits Withdrawal" starting in Feb 2007. Is that the discussion you have in mind? If not, I would appreciate if you post the thread title, if you can remember.

I'll read through the "Optimizing" thread, and any other one that you mention. But I'm curious: do these threads say essentially the same thing that Kaye Thomas says about receiving Soc Sec benefits earlier and later than the full retirement age?

(Good information at fairmark.com, as usual.)

Scott Burns, a syndicated financial columnsit, has written about choosing the age at which to commence taking Social Security benefits. See

for more articles about Social Security, click on the SocialSecurity link in the Categories pane. Dave