25 times income needs?

[Gillette]

I read on here in another post that a good rule of thumb is 20-25 times gross income. This is the first time I've heard that and I consider myself fairly well read and financially savvy. Anyway, just curious if that has an age constraint? i.e. if you're 50 and have 20-25 times gross income, great, definitely FIRE. or if you're 35 and have the 25x, nope you'll be FIRE too long, back to work for you. Also, if you can recommend reading material that would be great. Thanks.

 

[Rich_by_the_Bay]

I read on here in another post that a good rule of thumb is 20-25 times gross income. This is the first time I've heard that and I consider myself fairly well read and financially savvy. Anyway, just curious if that has an age constraint? i.e. if you're 50 and have 20-25 times gross income, great, definitely FIRE. or if you're 35 and have the 25x, nope you'll be FIRE too long, back to work for you. Also, if you can recommend reading material that would be great. Thanks.

It's 25 x initial expenses, not gross income. It's just a guideline and is the amount at which the 4% SWR should usually work long term.

 

[azanon]

Yes, retirement length matters. If firecalc does not take it into account, then it is logically flawed. Of course a retirment length of, say 40 years, has a greater chance of failing than one that's 10 years long. I realize the 4% withdrawal rate is designed to work indefinitely, but one cannot deny the fact that added length of retirement logically increases risk.

 

[Andre1969]

IIRC, the reason that 4% is thrown around as the accepted number is because when all the calculations get run, at 3% it's almost impossible to ever deplete your savings, even if you lived forever, but at 5% the chances of failure are way too high. So 4% was a good happy medium, almost guaranteed to sustain a 30 year retirement, but with a decent chance of surviving indefinitely.

Interesting, how just a 2% difference can sway you from a strong chance of failure to financial immortality.

 

[Rich_by_the_Bay]

Yes, retirement length matters. If firecalc does not take it into account, then it is logically flawed. Of course a retirment length of, say 40 years, has a greater chance of failing than one that's 10 years long. I realize the 4% withdrawal rate is designed to work indefinitely, but one cannot deny the fact that added length of retirement logically increases risk.

I follow the common sense that retirement length should matter, but I'm not sure it is completely right.

4% is designed to get you through ups and downs with low risk of depletion. But the longer the duration of retirement (say out 25 years or more), the greater the likelihood that returns will approach the historic mean and thus 4% may be increasingly overly conservative. This increases the likelihood it will last indefinitely the farther out you are. In fact, most scenarios yield a large ending balance.

Maybe Dory and others smarter than I am can help with this.

P.S. Just did a few firecalcs using 1mm nestegg and 40k expenses, rest at default. FC does give better safety for shorter durations but it's surprisingly small. In real-world scenarios where you kick in SS, it is a very small factor. Without SS here is what I got:

20y: 100% survival
30: 94%
40y: 90%
50y: 86%

 

[rs0460a]

Here's a link from the oft quoted "bible" 

http://www.efficientfrontier.com/ef/998/hell.htm

The key quote is "Only at $40,000 (4% of the initial amount) withdrawal rates do things look a little less grim. All strategies holding 50% or greater stock survive the 30 year period. However, even this route was one wild ride." as related to the 4% reference.

- Ron

 

[ladelfina]

Sounds like just the longer a period you look at, the smaller an impact (backtracked historical) negative periods will have overall. If your concern is 100% survivability then you might not be able to afford even one bad spell in 10 years. Expand that out to 40 or 50 and you could have a couple bad spells, but also enough positive years in which to make up losses. That is my non-smarter-than-you comprehension. I cannot agree with azanon.


edit: I just realized I should have said survivability maintaining principal.

 

[unclemick]

Hmmmm

A little defense during hard times can work wonders - REHP Homepage shows the effect of adding  some TIPs.

I am old school Norwegian widow - I watch the current yield of what's in my portfolio like a hawk - a tad under 3% nowaday's - as a fall back defensive postion.

1993, age 49 via layoff - had 16.6 times expected income needs. Messy - spent severance pay(16 wks, unemployment 13 wks), sold and consume duplex proceeds, small pension at 55(non cola), 1 yr temp work, took roughly 8% in lumps and bumps out of dividend stocks - and then there was the wonderful decade of the nineties so we didn't tap our IRA portfolio's. Lived high on the hog and cheap in the LA swamp.

So at 63 with early SS - I am forced, I say forced to spend quadruple my prior income - while I'm still young enough to bogie.

Interest rates were higher and valuations a tad lower back then. Just looking at my portfolio and FireCalc runs - 4% looks good in today's environment depending on what's in your portfolio. History proved me waaaay too conservative.

Remember - more than one way to skin a cat - you have way more control over what you do than Mr Market.

heh heh heh - 13th year of ER - party on - but keep your defense handy.

 

[azanon]

Sounds like just the longer a period you look at, the smaller an impact (backtracked historical) negative periods will have overall. If your concern is 100% survivability then you might not be able to afford even one bad spell in 10 years. Expand that out to 40 or 50 and you could have a couple bad spells, but also enough positive years in which to make up losses. That is my non-smarter-than-you comprehension. I cannot agree with azanon.

If your early "hit" was so damaging, that it cannot recover, another 30 years wouldn't help either unless what you're referring to is the growth phase where you are still working and putting new money into investments.  I believe you are confusing the growth stage with the retirement stage.

If the hypothetical is that no more earned income will be reinvested, then this completely an issue of mathematics.   A fixed amount of money invested has a greater chance of running out with more time than with less.   This is matter of fact, not something open to debate.

When a financial advisor asks you how long your retirement timeframe is, as part of an analysis to determine if you have enough money to retire, "longer time" always means you need more money to retire, not less.   There are tons of retirement calculators on the net to confirm this rather obvious truth.

 

[Rich_by_the_Bay]

If the hypothetical is that no more earned income will be reinvested, then this completely an issue of mathematics. A fixed amount of money invested has a greater chance of running out with more time than with less. This is matter of fact, not something open to debate.

So, $1mm (fixed) earning 7% and withdrawn at 3% has a greater chance of running out at 30 years than at 10?

Not sure I follow.

 

[azanon]

So, $1mm (fixed) earning 7% and withdrawn at 3% has a greater chance of running out at 30 years than at 10?

Not sure I follow.

Does there exist an investment that can pay you a net 4% guaranteed, no matter what happens? If one does exist, then i agree, time is not a factor.

I presume you do follow the scenario of realizing nothing is guaranteed, and that a wrench is more likely to be thrown into your well-thought out plan over 30 years than it is over 10 years.

If all i needed was 4% from my portfolio over just 10 years, man i wouldnt bat an eye before i'd retire. But if it had to last 40 years, you bet your ass i would think twice about it. 40 years is a heck of a long time for this world to evolve into something that doesnt even begin to resemble what we have today. Just think; only 10 years ago people were under the believe that the market could give them a consistent 12% for the rest of their lives.

 

[azanon]

Here's a link to an article i read recently at CNN (written recently as well), which suggests retirement length/longevity is something that should be taken into account, and does matter: http://money.cnn.com/2006/08/04/pf/expert/expert.moneymag/index.htm

 

[Dawg52]

Here's a link to an article i read recently at CNN (written recently as well), which suggests retirement length/longevity is something that should be taken into account, and does matter:   http://money.cnn.com/2006/08/04/pf/expert/expert.moneymag/index.htm

Yes, longevity does matter. My mother is 88 and going strong. Her sister is 94 and doing well. My dad died at 84. So it makes sense that I plan for the long haul. But my brother died at 56, cancer. So you have to have a reasonable balance in your planning. I certainly want to enjoy my retirement which includes spending some money on traveling, but you do have to protect yourself in case you are one of the lucky ones to live a ripe old age. 

 

[ladelfina]

If firecalc does not take it into account, then it is logically flawed.   Of course a retirment length of, say 40 years, has a greater chance of failing than one that's 10 years long.   I realize the 4% withdrawal rate is designed to work indefinitely, but one cannot deny the fact that added length of retirement logically increases risk.

azanon, I am not sure I get what you are saying. Firecalc does take into account length of retirement.

It seems like you are leaving out one part of the equation, which is the sum you start out with. At some point of accumulation, someone will arrive at essentially zero risk of their money running out in an investment universe modeled on that of the last 100 years or so. (I guess there could be a nuclear holocaust that would disrupt things sufficiently to belie this.) If you think the number that makes you comfortable is 3% and not 4%, you may not be alone. But to say a longer timeframe always has greater risk of failing in the absolute seems unwarranted.

A fixed amount of money invested has a greater chance of running out with more time than with less.

But the amount is never "fixed" because in some years.. most years.. you will be adding with earnings in excess of your withdrawals. See unclemick2.

Plus, you're right, there is no ONE investment that will guarantee 4%+inflation, but I think time has shown you can usually do better overall.. which is why most people here advocate a diversified portfolio.  What would you have us do differently?

DOG52 rightly points out that OUR survivability is far more unpredictable than that of our portfolios!

 

[Mwsinron]

My family they tend to die in their 60's 70's. While my DW's family live a long damn time. Her parents are going strong in their 80s and she has had aunts that made it past 100. Of course watching her family its easy to see why they all stayed active in their later years while mine plopped themselves on a couch

 

[haha]

I follow the common sense that retirement length should matter, but I'm not sure it is completely right.

4% is designed to get you through ups and downs with low risk of depletion. But the longer the duration of retirement (say out 25 years or more), the greater the likelihood that returns will approach the historic mean and thus 4% may be increasingly overly conservative. This increases the likelihood it will last indefinitely the farther out you are. In fact, most scenarios yield a large ending balance.

Imagine that you are offered an option on the S&P. It's an American style option that can be exercised at any time. Would you pay more for a 15 year option, or a 40 year option?

It's the same with retirement duration. A longer time gives a greater risk of large deviations from mean returns. Remember, you can be knocked out at any time, so even if very long term returns are easier to predict (IMO this is questionable when we are talking about the durations that people on this board talk about) any large interim deviation or string of very bad luck could knock you out before this long term would come into being.

Ha

 

[Rich_by_the_Bay]

It's the same with retirement duration. A longer time gives a greater risk of large deviations from mean returns. Remember, you can be knocked out at any time, so even if very long term returns are easier to predict (IMO this is questionable when we are talking about the durations that people on this board talk about) any large interim deviation or string of very bad luck could knock you out before this long term would come into being.

Yes, it does expose you to more knock-out punches, but these are spread out over many years and overall it will tend to regress to the mean, no? If stocks yield 10% on average, you are more likely to see 10% over 30 years than over 3 years (of course you may also see 20% over 3 years, but not likely over 30). Like a coin flip, do it 3 times and you'll always get 0, 33, 67, or 100% results for heads; do it 30,000 times and you'll get pretty close to 50% almost every time.

No, there is no sure-thing in any case.

 

[flipstress]

I'm math-challenged in comparison to the many fine posters here, but I look at this way:

The longer period has more individual or single years in which my returns can vary. The 40-yr period has more single years than the 30-yr period in which my returns can run the gamut from negative to positive, or the economy or other factors can turn foul.

Whatever happens in each of the single years is what will determine the success of my portfolio. So the longer my retirement withdrawal period is, the riskier it is.

Each year's returns or happenings are independent of the next year. Just because I had one bad year already does not mean that there will be an evening out and that next year will be a good one. The bad string could go on for who knows how long, how ever many years they go on. (I could have just used day instead of year as the period of time here; year is just a convenient measure of time.)

 

[haha]

Yes, it does expose you to more knock-out punches, but these are spread out over many years and overall it will tend to regress to the mean, no? 

One knockout is all you need, unless you are working and can replenish your stores

The concept of regression the mean is often misunderstood. Say you flip a fair coin 100 times, and get say 60 heads and 40 tails, or a deviation of 10 successes from the theoretical mean, and a 60% heads, 40 % tails in terms of proportions.

OTOH, flip 1,000,000 times, and you might get proportions much closer to the theoretical mean, 50.85 % heads, 49.15% tails. But your deviation from the theoretical  50% mean is much greater -8500!

These numbers are pulled out of the air, but some time spent with a random number generator will likely convince you of the principle.

It is absolute deviations that kill you.

Ha

 

[haha]

I'm math-challenged in comparison to the many fine posters here, but I look at this way:

The longer period has more individual or single years in which my returns can vary.  The 40-yr period has more single years than the 30-yr period in which my returns can run the gamut from negative to positive, or the economy or other factors can turn foul.

Whatever happens in each of the single years is what will determine the success of my portfolio.  So the longer my retirement withdrawal period is, the riskier it is.

Each year's returns or happenings are independent of the next year.  Just because I had one bad year already does not mean that there will be an evening out and that next year will be a good one.  The bad string could go on for who knows how long, how ever many years they go on.  (I could have just used day instead of year as the period of time here; year is just a convenient measure of time.)

Very elegantly expressed, Fliptress. You are far too moidest about your math skills. 

Ha

 

[CCdaCE]

Don't forget the "timing of the volatility".

A 50% drop in year 2 is harder to take than a 50% drop in year 28.

-CC

 

[wabmester]

A 50% drop in year 2 is harder to take than a 50% drop in year 28.

Well, this is sort of where the idea of the 4% SWR originates.   Historically, stocks have returned around 10% annually (or 7% real).   But if you retired during a worst-case sequence (e.g., Great Depression or 70's stagflation), your portfolio would have recovered and survived only if you had limited yourself to an initial 4% withdrawal.

On the other hand, if you had retired at the start of a boom, you could have safely withdrawn as much as 10% per year!

So, if you make it safely past the first few years of retirement, you should be able to crank up your withdrawal rate to 4% of whatever your present nest egg size has grown to.   Assuming that the future unfolds as kindly as the past, of course. 

 

[haha]

Well, this is sort of where the idea of the 4% SWR originates.   Historically, stocks have returned around 10% annually (or 7% real).   But if you retired during a worst-case sequence (e.g., Great Depression or 70's stagflation), your portfolio would have recovered and survived only if you had limited yourself to an initial 4% withdrawal.

On the other hand, if you had retired at the start of a boom, you could have safely withdrawn as much as 10% per year!

So, if you make it safely past the first few years of retirement, you should be able to crank up your withdrawal rate to 4% of whatever your present nest egg size has grown to.   Assuming that the future unfolds as kindly as the past, of course. 

There is a paradox here. Say you retire in 2000 with $1,000,000, and an initial withdrawal of $40,000.  If that drops to $800,000 by Fire-Calc SWR theory you can go right ahead withdrawing $40,000, or 4% of your starting amount, plus whatever inflation has done in the interim.

Yet you are saying that if your indexed and diversified investments had increased to $1,200,000, then you could say, "New start, I now get to take 4% of $1,200,000, or $48,000."

What a wonderful world!

Ha

 

[wabmester]

There is a paradox here. Say you retire in 2000 with $1,000,000, and an initial withdrawal of $40,000.  If that drops to $800,000 by Fire-Calc SWR theory you can go right ahead withdrawing $40,000, or 4% of your starting amount, plus whatever inflation has done in the interim.

Yet you are saying that if your indexed and diversified investments had increased to $1,200,000, then you could say, "New start, I now get to take 4% of $1,200,000, or $48,000."

What a wonderful world!

I don't think there's a paradox.   FC basically tells you that 4% works independently of the starting period.   If your net worth grew by leaps and bounds in the first few years of retirement, does FC somehow know that you really retired a few years ago, so the 4% bet is now off?   Nope, 4% is still good from any starting period.

Of course, this assumes that whatever terrible things the market does to you will be offset nicely by wonderful booms going forward.   To me, that represents the biggest danger in believing the 4% dream.   If we have a longer drought or a weaker upside boom after a drought than we've seen historically, then all bets are off.

 

[justin]

This has been discussed here before,

http://early-retirement.org/forums/index.php?topic=8377.msg150284#msg150284

There is also a graph on that link showing the asymptotic nature of the SWR as you increase duration of portfolio. Adding 10 years to portfolio duration reduces the SWR by a small amount.