Using FIRECALC to determine Pres Value of SS

[moguls]

I've been wondering how to best factor in the value of future Social Security income when estimating my portfolio value needed at ER. I've used various present value mechanisms and decided to play with FIRECALC to see how I could use it to estimate impact of SS on my starting portfolio.

First off, I know many don't believe they can count on SS at all, but I believe there will still be some income for me, albeit reduced from current estimates.

So here's what I did.

I set up FIRECalc with basic assumptions of my annual withdrawal over 40 years and continued to adjust the starting portfolio value until I found the point where the portfolio value became 100% survivable. I assumed no other adjustments throughout the 40 yr span.

I then set up FIRECALC using the same annual withdrawal, but this time entered the appropriate adjustments for my social security income (using conservative numbers). I then continued to rerun FIRECALC reducing the initial portfolio value until it dropped below 100%.

By subtracting the portfolio value from the second exercise from the portfolio value in the first I got an estimated value as expressed in current portfolio terms for the Social Security income stream.

Good news for me was that the value was much higher than my estimates from other means. So, in my case, SS income could be substantially less and still keep me well within comfortable SWR.

Any thoughts on this process? Flaws in my logic?

Mogus
(240 days and counting)

 

[sgeeeee]

moguls,

I've played around with estimating the value of my ss and pensions. You can go at this a lot of different ways and get a fairly large range of answers -- depending on your assumptions. I haven't gone through the exercise you describe, but I expect that you should get a fairly reasonable answer.

Here's another fairly simple way you can get an estimate if you want to do a sanity check on the number you already found.

1) Use the 4% rule to come up with an approximate value for your social security effective the date you start to collect. Simply take your estimated pay and divide by 4%. So if you expect to collect $10,000/year starting 10 years from now, you can figure that those payments would be worth ~ $10,000/(0.04) = $250,000 in the year 2013. The 4% rule is handy for this calculation since it assumes an inflation adjustment in the payments and also because it is a value that has a built in historical "guarantee" that it has always been enough to avoid depletion of the nest egg regardless of the economic times. Both of those conditions approximate the "guarantees" of the social security system. (I know . . . I know . . . those social security guarantees may not be worth much in a few years).

2) Compute a current value of the lump sum amount calculated in step 1. To do this you have to assume an interest rate that you believe you could make on the money if you included it in your portfolio today. This is where you really get a lot of variation. If you actually treat social security as part of your investment portfolio, then do you consider it part of your fixed income balance? part of your equity balance? as a little of both? or do you ignore social security when computing your investment balances? Depending on how you compute your overall portfolio balances, you can probably justify anywhere between ~3.5% to ~10%. I have played with values all over the map, but believe that a value of about i=4% to 5% makes sense. The current value of the amount in step 1 is then computed by dividing it by the quantity (1+i)^n where n is the number of years before you start collecting your social security. So in the example above, if you use i=5% and n=10, the $250,000 amount is worth $153,478 in today's dollars.

 

[unclemick]

Hey Salaryguru
I too played with pension and ss numbers in the 90's. The gist of the article that started me off was to count the calculated lump as fixed income so you could take more risk with equities. I think I was using .06(6%) then. Also wasn't discounting present value and I had apples and oranges - pension was fixed and ss has a COLA. I had already ER'd in 93 at age 49 so I didn't change our basic 50/50 index fund portfolio's - BUT any extra money went into dividend paying DRIP stocks(93-present) so that my fixed pension has help from dividends. So it was intuitively influenced if not not numerically precise. Using FIREcalc - ? should I also try some different equity/fixed mixes or will that just confuse the issue?

 

[sgeeeee]

If you consider social security as a low risk benefit, then it makes sense to use bond-like interest assumptions when calculating a present value of the benefit. But that implies that you should also consider it as part of your fixed income investment balance when you run FIREcalc.

 

[Ted]

Attempting to calculate the "present value" of future social security benefits is an exercise similar to the apocryphal story of the stupid olympic athlete who won a gold medal and had it bronzed.

The great thing about social security, from the standpoint of retirement planning, is that it promises an income stream that is adjusted for inflation. As I have said before, I think that the benefits will be pared back somewhat in the future to keep the burden on future workers of financing the program more acceptable, but I still regard social security as the most predictable of all retirement benefits.

Converting social security benefits to a present value is a meaningless exercise because (1) a person does not have the option of taking these benefits as a lump sum and (2) the present value depends on the number of years that benefits will be paid, which depends on how long a person will live, which is unknowable.

As an academic exercise, a person could get a rough idea of the present value for a "typical" retiree of their age by using your life expectancy, and for a discount rate using the real (inflation-adjusted) rate of return on government bonds, which is about 2.5%. But it would be silly to include this value with other financial assets in estimating what rate of withdrawal could be sustained in the future. FIRECalc has a provision for including annual social security benefits (inflation-adjusted) in the simulation of possible future cash flows. Just use that!

 

[sgeeeee]

Converting social security benefits to a present value is a meaningless exercise because (1) a person does not have the option of taking these benefits as a lump sum and (2) the present value depends on the number of years that benefits will be paid, which depends on how long a person will live, which is unknowable.  

It is often instructive and comforting to evaluate your retirement/investment plans from alternative perspectives. This provides, at worst, an additional level of comfort that you are on track . . . and at best, may provide greater understanding of certain aspects of your plan.

FIREcalc is one spreadsheet evaluation tool that is very useful to the RE planner, but it is not the only one and I wouldn't recommend that anyone depend entirely on it's calculations. That's why I suggested the alternative method of evaluating the value of your social security.

Using the 4% rule does exactly what you want to do to evaluate social security as a present value because it accounts for inflation adjustment and virtually no risk. Clearly it is an approximation assuming an ~30 years in retirement, but it works fairly well. Converting the lump sum ss value at initial payout to a current value using long term bond rates will probably overvalue the social security benefit and lead you to believe that you are significantly overweighted in fixed income investments. Using your long term average yield performance number will probably underestimate the value of your social security since it does not appropriately account for the low risk.

That's my reasoning for the evaluation scheme. I've also done this evaluation using annuity calculators and by building fairly elaborate spreadsheets that attempt to consider inflation etc. You can get a fairly significant spread in values, but the exercises can also give you a better understanding of the true value of the benefit and how best to plan the other parts of your investment portfolio.

 

[haha]

I've also done this evaluation using annuity calculators and by building fairly elaborate spreadsheets that attempt to consider inflation etc.  You can get a fairly significant spread in values, but the exercises can also give you a better understanding of the true value of the benefit and how best to plan the other parts of your investment portfolio.

It also whiles away the hours almost as effectively as going to the office, a benefit not to be ignored!

 

[Ted]

(1) It is often instructive and comforting to evaluate your retirement/investment plans from alternative perspectives.  

(2) Converting the lump sum ss value at initial payout to a current value using long term bond rates will probably overvalue the social security benefit and lead you to believe that you are significantly overweighted in fixed income investments.  

(1) I agree with the idea of evaluating possible future financial outcomes from different perspectives -- as long as all of those perspectives are reasonable. But attempting to convert one's social security benefits to a present worth, and then converting back to a future annual withdrawal rate as if it were any other fixed income investment is not logical IMO.

(2) Using the "real rate of return" as a discount rate in calculating present value does have the effect of yielding a relatively high "present value." But this is justified when the future payments can be expected to increase each year. This is the case with social security benefits (and TIP interest and principal gains), but is not the case with interest payments on conventional bonds.

While I have pointed out limitations on FIRECalc in previous posts, and recommended modifications to it, I still regard it as the best retirement calculator that I have seen, and it is easy to run it using different assumptions. One modification that I particularly recommend is to input an "inflated" value for management expenses to, in effect, account for future returns on equities being less, on average, than they were during the period included in FIRECalc's data base.

 

[sgeeeee]

(1)  I agree with the idea of evaluating possible future financial outcomes from different perspectives  -- as long as all of those perspectives are reasonable.  But attempting to convert one's social security benefits to a present worth, and then converting back to a future annual withdrawal rate as if it were any other fixed income investment is not logical IMO.

I don't think that's what anybody is trying to do, Ted. The analysis I've done to put a present value on my own future social security benefits is focused on examining what impact that benefit has on my overall equity/fixed balanced if I were to consider it to be a fixed income type of investment. I think that's what the original post on this thread was looking at too. You don't have to consider your social security in this manner. But if you did, what does it do to your portfolio balances? If your future social security benefit is large, you might end up deciding you are not taking enough risk in your current portfolio.

If you don't care to examine this kind of information, then by all means don't use any of the techniques described. But to declare the effort to be illogical seems misguided.

 

[Ted]

In determining one's "target" allocation of financial assets (generally meaning cash equivalents, stocks, and bonds), it is not unreasonable to consider other assets too. For example, the cash value of a life insurance policy, or a mortgage that you own as the lender, or even your equity in your own home, may be counted as "bonds." And if you want to estimate a present value of your social security benefits and count that as a "bond," there's no harm in that, as sort of an academic exercise.

But when you are retired and trying to determine the most suitable asset allocation to sustain future withdrawals, what is the best way to do it? IMO, it's not to rely upon some "standard" asset allocation -- whether or not that includes assets other than cash, stocks, and bonds. The best approach is to use a calculator like FIRECalc that explicitly incorporates such things as annual social security benefits, and future sales of major assets such as a home, and combines that with data that reflects the variability that can be expected in future returns of cash, stocks, and bonds.

If a person were to count social security and these various other items as "bonds," I suspect that they would be inclined to invest an excessively risky amount of their "liquid" assets in stocks. Before retirement, people can afford to do that, but after retirement, it's a different ballgame.

 

[Mr._johngalt]

Very good Ted. And that is why I continue to avocate a
"no common stocks" position for ERs.

 

[unclemick]

I admit to being 60/40 stocks/bonds early in ER intuitively counting ss as 'bond like'-now ten years have passed and we're 50/50 balanced index assuming a 30 yr span (age 90). Not in above 'hobby' money is in dividend stocks. If we start spending the dividends this will act more like a ss stream (some dividend growth) than the ER portfolio. Any thoughts on how to model this on FIREcalc.

 

[sgeeeee]

In determining one's "target" allocation of financial assets (generally meaning cash equivalents, stocks, and bonds), it is not unreasonable to consider other assets too.

Thanks for your approval, Ted.

But when you are retired and trying to determine the most suitable asset allocation to sustain future withdrawals, what is the best way to do it?  IMO, it's not to rely upon some "standard" asset allocation -- whether or not that includes assets other than cash, stocks, and bonds. . .

That certainly makes sense to me if you are retired and either collecting social security or expect to collect it within a few years. If you are years away from retirement, or your are retired at age 49 and don't expect to collect pensions or social security for some time, it makes less sense to me. . . But if it makes sense to you, then it's the right way for you to do it regardless of your circumstances.

If a person were to count social security and these various other items as "bonds," I suspect that they would be inclined to invest an excessively risky amount of their "liquid" assets in stocks.  Before retirement, people can afford to do that, but after retirement, it's a different ballgame.

The existence of a significant, stable social security benefit makes your investment earnings less risky than they otherwise would be -- regardless of whether you are retired or years away from retirement. Some people may choose to exploit the benefit by taking on more risk in the remaining part of their portfolio while others may not.

 

[Ted]

The existence of a significant, stable social security benefit makes your investment earnings less risky than they otherwise would be -- regardless of whether you are retired or years away from retirement.  Some people may choose to exploit the benefit by taking on more risk in the remaining part of their portfolio while others may not.  

I agree.  But people should be aware that the best way to determine the target asset allocation that is "best" for them is to use a reasonably comprehensive calculator like FIRECalc, rather than to use some "standard" allocation that is recommended for retirees by some "expert."  

For example, one "rule of thumb" that I have heard is that the stock percentage of a portfolio should be -- I think it was -- 100 minus the person's age.  Well, in calculating that amount of money, does a person include the present value of their social security benefits, or their home equity, as part of their "portfolio"?  To me, it's an irrelevant question because that's not how I would determine asset allocation.

 

[earlyout]

Social secuity value,

To get an estimate of your SS $, you can go to http. ssa.gov/ and use the on-line calculator. They ask you to input this years income and then they figure increase in salary in future years. Assuming you don't work after 52, or work part-time, then click on the assumption button and then u can enter in "0" for your current income, and get a projection of your future SS $. It at least gives you an idea of the difference in SS $ working to age 62 ( or age 65) vs stop working at age 52 (or some other age).

Also, to help evaluate a lump sum vs a monthly income, or vice versa, you could go to immediateannuity.com, enter in the age, etc and get a value for a pension, social security, etc. There are other tools available, but this is a quick way for an initial assessment.

earlyout

 

[sgeeeee]

Here's an applicable link for this question about how to value SS in planning your retirment portfolio. The article is a Scott Burns column describing a discussion with John Bogle.

http://www.dallasnews.com/business/scottburns/columns/2003/stories/102603dnbusburns.50a9e.html

From the article:
The advice

"First, look at all of your assets. The most important thing is asset allocation – how much is in equity and how much is in bonds. . ."

""Think about everything, not just your investment program. That means think about Social Security. If you're fortunate, it can be $25,000 a year. It's great money. When you do your asset allocation, you want to include that. At 5 percent, that $25,000 is like having $500,000 [in assets]. As a life income, it's more like $350,000. That should be part of your calculation."

Suppose, for example, that your Social Security income is $17,000 a year – a typical figure for a couple – and your 401(k) rollover account and other financial assets are $400,000. Then you start your asset allocation with a conservative value for your Social Security at 14 times the income, or $238,000. Added to your $400,000 in savings, this means you have the equivalent of $638,000. In effect, 38 percent of it is already allocated to a bond equivalent. . ."

By the way . . . thanks to intrcst for posting this link on the MSN REHP boards.

 

[Mr._johngalt]

Hey Salaryguru! This explanation of how to factor in
your SS benefits is excellent. This is exactly the way that I look at it. Couldn't have explained it better myself.