A more realistic way to calculate for your retirement savings

[itsme2]

I have spent quite some time looking for the amount of retirement savings needed. However, I am always perplexed by the way most publications came up with their magic numbers. For example, traditional retirement planning has no problem of identifying the annual expense ($50k, $80k, etc.). Then they quickly throw an arbitrary factor (from 8x to 30x) to come up a rough lump sum ($1M , $2M, etc.). Most of them devoutly use 4% withdraw rate.

It doesn't take the later-on SS into account. Therefore, the required lump sums are always too large and too conservative. Because of these typical published estimates and withdraw rules, no one feels safe to ER if the nest egg worth is less than $1M, even though, according to a recent finding, the average value of retirement accounts for people between 55 and 64 is only $148,900.

So, let's think about it. If you cannot retire with $800k, how can those with less than $150k ever retire?

I am going to show my way of calculation.

 

[mathjak107]

we can survive living in a tent and fishing. it is the wants ,comforts and needs that vary for all of us.

i can tell you here in ny 150k would not generate enough to pay my healthcare costs even if i lived in that tent.

 

[gauss]

My father retired with less than about $50k in savings. He received a $400/month pension and Social Security of about $1,200/month (which I believe he started at age 62). He had employer provided retiree health insurance. He may have had some small real estate agent income until he reached his 70s.

On this income he:

Had a mortgage payment on his condo, lived in the suburbs, had his own car, air conditioning when needed, was active in his church/singles group, traveled internationally a few times, had cable TV, Internet, was popular with ladies in the ballroom dance circles etc.

I don't think he was ever below 150% of the Federal Poverty Level.

For his rather small level of savings he certainly lived life to his fullest IMHO.

-gauss

 

[REWahoo]

Oh boy, new math!

 

[petershk]

WGauss: I think what makes a lot of people nervous is the disappearance of many of those things. I think the total value of that is around 900k at a 4% withdrawal rate.

Imagine if the company kills the health package or the pension or of SS goes down or inflation goes way up.

Of course there's always what its but if you have 50k and any one of those things goes away it could be ugly fast...

...beyond that... If a good chunk of seniors are in the same situation then a large % could get crushed by the same events and that could be fairly catastrophic...

Sent from my HTC One_M8 using Early Retirement Forum mobile app

 

[haha]

ISo, let's think about it. If you cannot retire with $800k, how can those with less than $150k ever retire?

I am going to show my way of calculation.

Don't bother man, it has much more to do with attitudes and emotions than calculations.

Hold back if you want to, let 'er rip if you're feeling bold.

Ha

 

[jerome len]

It's amazing the way different threads attack what's needed for retirement.
Just last week someone asked how they could cut their budget from the over $100,000 per year they needed to retire. Others tell us how they happily live on less than $30,000 a year. My SS is well over $30,000 a year and I could not live on it....not even close. The value of this blog is to learn......reading what others have done and learning from it. Me? I worry about having a stroke, needing full time care and having enough to pay for it. And, I'm healthy....I want to travel, that's expensive.....I like enjoying my DW at a restaurant for dinner.....that costs a little. NY and California cost more than Florida or Texas.....that changes what we need. My goal have been to live off of interest and dividends.....owning dividend mutual funds that own stocks that have increased their dividends each year. I'm assuming that and SS will increase for inflation. Yes, I worked hard, saved my money and take nothing from govmet but SS. Now.....I have a distant relative that retired with less than $50,000........a family member now has dementia....so, Medicaid pays the bills. .....and, all of us that pay taxes share the bill for all of those that didn't save. But.....would I want to live in a little apartment.....go to whatever "home" medicaid will pay for? Absolutely not! I have too much pride and would hate to be in that situation. Again, my thanks to all that share their stories in every thread I read.....It really helps as I plan for my and my family's future.

 

[itsme2]

(Sorry, I could not stay up to finish it last night)

I came up a way to better estimate the needed retirement saving, not only taking SS, but also investment annual return rate, inflation rate, and SS COLA as inputs, into account.

Here are the assumptions to “simplify” my calculation.

• Disregard the different tax brackets
• Disregard whether it is Roth (after tax) or non-Roth(pre tax)
• You and your spouse retire at the same time and take SS at the same time.
• Your primary home is paid off
• Pension is omitted, but can be combined with SS.

 

[itsme2]

(removed by myself. info duplicated)

 

[itsme2]

Here are the variables:

• r1: Investment annual return (say r1 = 1.06)
• r2: Inflation rate (say r2=1.03)
• r3: SS COLA (say r3 =1.02)
• a: Annual expense for the first year. Each year add by r2
• b: Your savings right before ER (401k, IRA, Pension, etc.)
• c: The combined SS of what you and your spouse can withdraw (say c=$40000). You can check this site: Retirement Estimator to get better estimates.
• n1: The number of years before you start to withdraw SS. If you decide to retire at 60 and begin to withdraw SS at 67, n1 = 67-60=7
• n2: The number of years after you begin to withdraw SS. You can assume n2 = 87-67 = 20

The biggest question most people have is:

If I need $a per year for living and retire tomorrow, how much will I need my retirement savings $b?

Here is the formular:

b=((a*r1^(n1+n2-1)*(r2/r1)^n1*((r2/r1)^(n2)-1)/(r2/r1-1)-c*r1^(n1+n2-1)*(r3/r1)^n1*((r3/r1)^(n2)-1)/(r3/r1-1))/r1^(n2)+a*r1^(n1-1)*((r2/r1)^n1-1)/(r2/r1-1))/r1^n1

Below are ten examples as the results of the above formular:

DIY

Since I believe most people will feel overwhelmed with the lengthy formular, you can follow the steps below to do you own calculation with Excel:

Step #1 - Fill in the values in exact cells as shown below (except for cell B2):

Step #2 - Copy & paste the string below into cell B2:

=((B1*B4^(B7+B8-1)*(B5/B4)^B7*((B5/B4)^(B8)-1)/(B5/B4-1)-B3*B4^(B7+B8-1)*(B6/B4)^B7*((B6/B4)^(B8)-1)/(B6/B4-1))/B4^(B8)+B1*B4^(B7-1)*((B5/B4)^B7-1)/(B5/B4-1))/B4^B7

Step #3 - For every input you enter into B1, B2 is your needed retirement savings right before you retire.

Note 1: I will welcome any feedback to further improve my calculation.
Note 2: Whoever is interested in how I derived this formular, I will be glad to show with.

 

[ERD50]

(Sorry, I could not stay up to finish it last night)

I came up a way to better estimate the needed retirement saving, not only taking SS, but also investment annual return rate, inflation rate, and SS COLA as inputs, into account.

Here are the assumptions to “simplify” my calculation.

• Disregard the different tax brackets
• Disregard whether it is Roth (after tax) or non-Roth(pre tax)
• You and your spouse retire at the same time and take SS at the same time.
• Your primary home is paid off
• Pension is omitted, but can be combined with SS.

Have you looked at the historical calculator/reporters like FIRECalc? They do all that and more.

firecalc.com

-ERD50

 

[itsme2]

Have you looked at the historical calculator/reporters like FIRECalc? They do all that and more.

firecalc.com

-ERD50

Maybe I missed something from firecalc.com. It just gives me a range of curves with a probability based on a lump sum (still a shortcoming without taking SS into account), correct?

BTW, I have been attempting many times, to post the remaining part of my calculation with using POST, EDIT or REPLY. But, it was either blocked or shown briefly (then was removed). Do you know why? As you can tell, I am new to this site. I PMed the Admin, but didn't get a reply. But, someone was obviously watching and manipulating my posted content.

 

[ERD50]

Maybe I missed something from firecalc.com. It just gives me a range of curves with a probability based on a lump sum (still a shortcoming without taking SS into account), correct? ...

No, not correct. Yes, you are missing a lot. Go to the second tab, "Other Income/Spending" and you can enter SS, pensions, and (wait for it....), 'other spending and income'.

Check out all the tabs.

-ERD50

 

[target2019]

Maybe I missed something from firecalc.com. It just gives me a range of curves with a probability based on a lump sum (still a shortcoming without taking SS into account), correct?

BTW, I have been attempting many times, to post the remaining part of my calculation with using POST, EDIT or REPLY. But, it was either blocked or shown briefly (then was removed). Do you know why? As you can tell, I am new to this site. I PMed the Admin, but didn't get a reply. But, someone was obviously watching and manipulating my posted content.

I'd like to see your calculation.

BTW, your post may be hitting a filter of some kind. It is not necessarily that someone is watching.

 

[Alan]

I'd like to see your calculation.

BTW, your post may be hitting a filter of some kind. It is not necessarily that someone is watching.

It is an automated filter for new members that traps posts that contain links. Part of the site's spam filter. Once you have enough posts then links posted will not get set aside for review.

 

[itsme2]

I'd like to see your calculation.

BTW, your post may be hitting a filter of some kind. It is not necessarily that someone is watching.

My calculation finally showed up again at #10.

Your id reveals your intended ER year as 2019. We are in the same boat, except you joined this site years earlier than me. Hopefully, there will be no catastrophic market correction in coming years.

 

[Midpack]

My calculation finally showed up again at #10.

Your id reveals your intended ER year as 2019. We are in the same boat, except you joined this site years earlier than me. Hopefully, there will be no catastrophic market correction in coming years.

No disrespect intended, but you really are missing all that FIRECALC offers, look at all the tabs. It does everything you are trying to do, and considerably more, while providing probability of success for as many scenarios as you want to view. It does take a little time to fully grasp what it does & doesn't do and how it works, but you'd be very hard pressed to write something yourself that does more.

And if you want more than "the curves," you can output to an Excel spreadsheet to see much more detail.

What we do know for sure is a constant rate of return and a constant rate of inflation, as your spreadsheet is based on, WON'T happen...

 

[rodi]

Echoing Midpack here.

You can enter all the info you suggest, into firecalc, and get out a LOT more... Including how you'd have done historically based on real world inflation rates (which vary from year to year rather than being a fixed, deterministic number) and real world investment returns (which also vary from year to year - with some NEGATIVE years. Your proposal doesn't take into account the possibility of cycles of negative returns. That's a VERY real risk - and if it happens early in your retirement can kill your nest egg.

Explore the tabs in firecalc - there's a whole lot more than just the first page of nest egg and annual spending and # of years till you die.

 

[karluk]

Yeah, itsme2, your model of using constants for portfolio growth, inflation, and SS COLAs is way too unsophisticated to generate reliable forecasts. You should take the trouble to investigate sequence of return risk. Reading up on that subject will give you some idea as to how your calculations might be accurate and still lead to retirees running out of money in real life.

If you want feedback on outright errors in your model, I would say that one thing that jumps out at me is that you seem to positively encourage people to use different values for inflation and SS COLAs. As was discussed in http://www.early-retirement.org/forums/f28/ss-at-fra-vs-at-70-unexpected-results-76531.html, inflation is typically measured by CPI-U and SS COLAs are based on CPI-W. Since the two statistics are based on slightly different weightings of the same input data, it's almost impossible for inflation and SS COLAs to differ by more than a trivial amount, especially over long periods of time.

 

[itsme2]

Echoing Midpack here.

You can enter all the info you suggest, into firecalc, and get out a LOT more... Including how you'd have done historically based on real world inflation rates (which vary from year to year rather than being a fixed, deterministic number) and real world investment returns (which also vary from year to year - with some NEGATIVE years. Your proposal doesn't take into account the possibility of cycles of negative returns. That's a VERY real risk - and if it happens early in your retirement can kill your nest egg.

Explore the tabs in firecalc - there's a whole lot more than just the first page of nest egg and annual spending and # of years till you die.

Following your suggetion and Midpack's, I studied again with FireCalc and it does appear it includes lots of the factor and tested with historical data.

However, I came up my formular mainly for my own retirement planning and prior to learning the availability of FireCalc. I still believe most of people here are too conservative and seeking unnecessarily large lump sum before feeling safe. My formular can offer another approach.

I agree the rate of return and inflation will never remain the same. The annual expense like assumed rates should not be fixed as well. If we have a fantastic year with the market, we can definitely spend a little more.

Likewise, some people worry about what if the market crashes again like what we experienced in 2008. If that happens, I believe most people are ok to reduce their withdrawal without depleting the nest egg too soon.

 

[pb4uski]

Itsme2, your calculation makes a lot of sense with a HUGE exception. It ignores what is called sequence of returns risk. You can google it, but in short, if your 6% return is large negative returns in the early years and very healthy returns in the later years, even if they average out to 6% for a 30 or 40 year period, the fact is that between withdrawals and negative returns in the early years, you end up running out of money and eating catfood.

That is why your calculations are coming up with a much higher withdrawal rate that you'll see as being safe with other calculators. In short, you need more that what you calculate to protect yourself from possible adverse sequence of returns scenarios.

 

[Midpack]

However, I came up my formular mainly for my own retirement planning and prior to learning the availability of FireCalc. I still believe most of people here are too conservative and seeking unnecessarily large lump sum before feeling safe. My formular can offer another approach.

What is your basis for believing most people here are "too conservative?" You're right that those who use FIRECALC and similar tools to estimate their retirement portfolio needs, are hedging to the high side by design, but it's not an uninformed or arbitrary choice. Planning on a 95% success rate for example, means a 5% chance of failure (historically) and a 95% chance of having more than enough. But most of us looking at a 20-40 year time horizon, begin with placing a much higher priority on not running out of funds than on leaving a surplus. One can easily ratchet up spending as the retirement proceeds (or save for end of life costs, LTC etc.). It would be pretty difficult late in retirement to radically reduce spending once it becomes evident you're going to fall short of funds.

FIRECALC results provide a probability of success based on actual returns and inflation from 1871 through present, and what the range of possible outcomes would have been throughout those periods.

And for anyone who might be interested in your approach, what probability of success does your formular offer? It's relatively simple to project the average portfolio needed to support a given spending for a given number of years, assuming a real rate of return - but that may give you a 50% probability of success. Good enough?

 

[Independent]

Following your suggetion and Midpack's, I studied again with FireCalc and it does appear it includes lots of the factor and tested with historical data.

However, I came up my formular mainly for my own retirement planning and prior to learning the availability of FireCalc. I still believe most of people here are too conservative and seeking unnecessarily large lump sum before feeling safe. My formular can offer another approach.

I agree the rate of return and inflation will never remain the same. The annual expense like assumed rates should not be fixed as well. If we have a fantastic year with the market, we can definitely spend a little more.

Likewise, some people worry about what if the market crashes again like what we experienced in 2008. If that happens, I believe most people are ok to reduce their withdrawal without depleting the nest egg too soon.

I'm not sure how you arrived at the bold statement.

Maybe a numeric example would help. Suppose we try a very simple retirement. A single person, with

a: Annual expense for the first year: $40,000

c: SS of what you can withdraw:$18,000.
n1: The number of years before you start to withdraw SS: 0
n2: The number of years after you begin to withdraw SS: 30

You can specify r1, r2, and r3, plug them into your formula, and get a value for b.

Then, you can specify what number "most people here" would use for b, and show why you think it is "too conservative".

 

[daylatedollarshort]

I just use a return of 1% over inflation for asset income (long term TIPS have historically been around 2% + inflation) in a spreadsheet. We also have a low rate, fixed rate mortgage we will not pay off unless we move offset by a couple of non-COLA pensions (liability matching). We have some stocks, but aren't betting the farm on them.

100 years of U.S. stock history, most of it in an era with less U.S. income inequality and without flash boys, derivatives, computerized trading programs and global conglomerates, don't give me warm fuzzies on the ability of U.S. centric Monte Carlo programs to predict 50 years into a global future economy. I think the Triumph of the Optimists conclusions, taking many countries into account, are more likely to be correct regarding probabilities of returns going forward (stocks for the long term are generally a good bet but the long term might mean a long term past our remaining life spans).

We take a Zvi Bodie / post 2088 Bill Bernstein retirement investing approach.

 

[marko]

I hate to think that this guy spent hours and hours putting together and testing a model that may/may not be more/less accurate than the dozen other calculators immediately available online for free.

Besides, there's always a certain margin of error in all these calculators, so I'd wonder if it's better/worse by only 0.0002% or something.