Does The 4% Rule Work In Today’s Markets?

[mathjak107]

i screwed it up though . i have to re figure it .  the cash is a declining balance since by the end of year 1 it is gone and you are no longer holding 2 years . i have to see how to handle it . on fircalc commercial paper i think should be cash instruments but i have to confirm that

[arebelspy]

i screwed it up though . i have to re figure it .  the cash is a declining balance since by the end of year 1 it is gone and you are no longer holding 2 years . i have to see how to handle it . on fircalc commercial paper i think should be cash instruments but i have to confirm that

So you aren't saying always have 2 years in cash (i.e. a 92%/8% equities/cash)?  You're saying start with 2 years, spend it all regardless of market conditions, allow the equities (the other 23x/92% of your portfolio) to grow over those two years, then go 100% stocks?

Please confirm.  If not, please clarify exactly what withdrawal plan/AA you mean, because a few vague sentences aren't helping the communication here, trying to understand.  :)

EDIT: That doesn't quite make sense either, because you say "by the end of year 1 it is gone"--do you mean gone completely, or down to one year after a year, and then gone completely after 2 years?

[pbkmaine]

https://www.kitces.com/blog/are-cash-reserve-retirement-strategies-really-necessary/

[arebelspy]

the cash is a declining balance since by the end of year 1 it is gone and you are no longer holding 2 years .

So you aren't saying always have 2 years in cash (i.e. a 92%/8% equities/cash)?  You're saying start with 2 years, spend it all regardless of market conditions, allow the equities (the other 23x/92% of your portfolio) to grow over those two years, then go 100% stocks?

Okay, I went ahead and tested this scenario.

I ran a cFIREsim scenario of having saved 23x expenses in 100% equities, and 2x expenses in cash, by putting in that 23x expenses as the portfolio (920k on 40k spending in equities, 80k, or two years spending, in cash).  To model this I put in a portfolio of 920k, 100% equities, then I added a single income of 40k in years 1 and 2.  So that offsets the 40k spending those years, as that 80k cash is spent down, and the portfolio is left to grow the first two years.  All other stuff (fees, etc.) left default.

I also ran a default cFIREsim of 75/25 stocks/bonds, entire 1MM portfolio invested, no cash infusion, as well as a 100% stocks scenario.

I then downloaded the spreadsheets and found the median and average of the year two end portfolio for each scenario.

Cash scenario:
Average: 1,109,048.311
Median: 1,098,287.19

75/25 stocks/bonds scenario:
Average: 1,084,244.348
Median: 1,081,889.04

100% equities scenario:
Average: 1,113,412.208
Median: 1,099,959.26

Looks like straight equities > equities + cash > 75/25.

The interesting thing though is that 75/25 has a lower failure rate (i.e. higher success rate, as I posted earlier in the thread), so it's not just about the most money.

Either way though, if you want a lower failure rate, it doesn't seem like cash is the way to go, and if you want more money two years after ERing, ditto.

I'm not seeing a scenario where investing 23x your income and keeping 2x in cash to start out your ER, then spending down that two years in cash is beneficial.  Nor am I seeing one where keeping two years in cash forever is beneficial.

Other than, as we said, psychologically.

If it helps you sleep at night, great, more power to you.  But I don't see the math where it helps your success rate, or leaves you with more money.

[ender]

A lot of this is interesting because of the psychological factors.

Relying on a HELOC to me seems very risky, particularly if you are carrying a meaningful mortgage - if the market drop is correlated with a housing price drop, you may not be able to get a HELOC. Of course whether to carry a mortgage into ER also is a psychological factor. Mathematically and historically for the same total net worth you are better off having a mortgage, due to similar reasons as the "cash drag." That does not mean it is necessarily a worse strategy for you, depending on how strong the psychological factors are.

[Retire-Canada]

Relying on a HELOC to me seems very risky, particularly if you are carrying a meaningful mortgage - if the market drop is correlated with a housing price drop, you may not be able to get a HELOC.

How much equity you have in your home vs. how much HELOC you want to use as a FIRE buffer is an important consideration. If you have a $500K home and a $300K HELOC you can safely plan to take out $50K/yr for 2 years without much risk that property value changes will affect you.

You are also never forced to use your HELOC it's just one tool amongst the many you have like cashing in investments, getting a PT job and/or lowering your spending. The beauty with a HELOC is it costs you nothing until and unless you decide to use it so I would say it's a great tool to have.

Personally I have a LOC not tied to my home equity that I could live off for for 1.5yrs of lowered spending living. A change to my home equity would have zero impact on this LOC.

[arebelspy]

Relying on a HELOC to me seems very risky, particularly if you are carrying a meaningful mortgage - if the market drop is correlated with a housing price drop, you may not be able to get a HELOC.

How much equity you have in your home vs. how much HELOC you want to use as a FIRE buffer is an important consideration. If you have a $500K home and a $300K HELOC you can safely plan to take out $50K/yr for 2 years without much risk that property value changes will affect you.

Not necessarily; during 2008-9, many banks closed or cut back people's HELOCs.  It's a tool I'd put in place, and would use if it was available, but also wouldn't count on to be there when you need it.

[Mmm_Donuts]

The way cFiresim works, you're essentially holding one years expenses in cash at the start of each year by default. If you are holding two years cash, you need to adjust the cash allocation percentage down slightly because of this. And for those who would never withdraw that much cash in one go, it's a good thing to keep in mind as it may affect your results.

[TomTX]

Relying on a HELOC to me seems very risky, particularly if you are carrying a meaningful mortgage - if the market drop is correlated with a housing price drop, you may not be able to get a HELOC.

How much equity you have in your home vs. how much HELOC you want to use as a FIRE buffer is an important consideration. If you have a $500K home and a $300K HELOC you can safely plan to take out $50K/yr for 2 years without much risk that property value changes will affect you.

Not necessarily; during 2008-9, many banks closed or cut back people's HELOCs.  It's a tool I'd put in place, and would use if it was available, but also wouldn't count on to be there when you need it.

I don't believe my CU did.

Here's how my house situation would be under this scenario:

House value $300,000
Mortages/laons: $0 (paid off house)
HELOC current amount $0
HELOC max amount $200,000 (because why not? CU doesn't care up to 80% LTV)
HELOC rate: Prime -0.5% (so, half a point under prime, currently 3.00%)
HELOC draw period: 10 years
HELOC payback period: 20 years

The cost to set it up is maybe a couple of hours, and there is no fee associated if my assumed value doesn't exceed the tax value.

This is just one of many possible contingencies. Get a part time job. Get hobby income. Ramp up a side gig Craigslisting stuff. Et cetera.

[Retire-Canada]

Not necessarily; during 2008-9, many banks closed or cut back people's HELOCs.  It's a tool I'd put in place, and would use if it was available, but also wouldn't count on to be there when you need it.

If your plan is to borrow a small % of the equity in your house via a HELOC you are at far less risk for this. In my example above that amount was 20% of the equity. That seems pretty safe.

You can also do what I do and have a LOC not tied to home equity at all.

[arebelspy]

This is just one of many possible contingencies. Get a part time job. Get hobby income. Ramp up a side gig Craigslisting stuff. Et cetera.

Sure.  That's always true, and tangential to this discussion.

[Classical_Liberal]

The way cFiresim works, you're essentially holding one years expenses in cash at the start of each year by default. If you are holding two years cash, you need to adjust the cash allocation percentage down slightly because of this. And for those who would never withdraw that much cash in one go, it's a good thing to keep in mind as it may affect your results.

I did not realize that, thanks. 

Also, doesn't cFiresim maintain a constant asset allocation?  So if you allocate a certain amount of cash (80K on 1Mil) it will continue to maintain that allocation even in the historically poor performing equity years.  Correct me if I'm wrong, but what mathjak107 is suggesting is during a bear on equities, one would then change AA through drawdown by only spending the cash, allowing equities to recover (theoretically), before readjusting AA.  This is a very similar approach to drawing on a HELOC.  The difference being with a HELOC the cash is not sitting stagnate without returns, but no need to sell any equities to make minimum payments when stocks are down.

I remember reading a livingaFI blog post where he modeled the two years cash in several bear markets to see how it's work out.

Edit: heres that post https://livingafi.com/2014/05/28/drawdown-part-4-examples/

[steveo]

The interesting thing though is that 75/25 has a lower failure rate (i.e. higher success rate, as I posted earlier in the thread), so it's not just about the most money.

I think that this is the most important to me personally. I don't want the most money. I want to be able to remain retired without going back to work.

[mathjak107]

The way cFiresim works, you're essentially holding one years expenses in cash at the start of each year by default. If you are holding two years cash, you need to adjust the cash allocation percentage down slightly because of this. And for those who would never withdraw that much cash in one go, it's a good thing to keep in mind as it may affect your results.

I did not realize that, thanks. 

Also, doesn't cFiresim maintain a constant asset allocation?  So if you allocate a certain amount of cash (80K on 1Mil) it will continue to maintain that allocation even in the historically poor performing equity years.  Correct me if I'm wrong, but what mathjak107 is suggesting is during a bear on equities, one would then change AA through drawdown by only spending the cash, allowing equities to recover (theoretically), before readjusting AA.  This is a very similar approach to drawing on a HELOC.  The difference being with a HELOC the cash is not sitting stagnate without returns, but no need to sell any equities to make minimum payments when stocks are down.

I remember reading a livingaFI blog post where he modeled the two years cash in several bear markets to see how it's work out.

Edit: heres that post https://livingafi.com/2014/05/28/drawdown-part-4-examples/

the problem with helocs a side from interest and payments during a cash crunch  is banks can  cut helocs just when you need them most . if real estate falls helocs  can be  cut or suspended .  i would use a heloc only as emergency money at best . i would never use it as a proxy for my own cash being spent monthly in a downturn with an unlimited potential for an extended  time frame .

i rather pull 4% out of my investments once the cash is depleted  than the additional money to include principal and  interest in my payments  on a potentially increasing balance  when things are taking an extended beating

[Retire-Canada]

Other than, as we said, psychologically.

If it helps you sleep at night, great, more power to you.  But I don't see the math where it helps your success rate, or leaves you with more money.

The whole cash buffer vs. HELOC/LOC vs. 100%-invested-just-ride-it-out debate sounds very much like the paying down your mortgage faster vs. paying the mortgage slower and investing debate. There is a heavily emotional component to one option that for some people out weighs the mathematical/statistical benefits of the other option.

[arebelspy]

Other than, as we said, psychologically.

If it helps you sleep at night, great, more power to you.  But I don't see the math where it helps your success rate, or leaves you with more money.

The whole cash buffer vs. HELOC/LOC vs. 100%-invested-just-ride-it-out debate sounds very much like the paying down your mortgage faster vs. paying the mortgage slower and investing debate. There is a heavily emotional component to one option that for some people out weighs the mathematical/statistical benefits of the other option.

Yup, definitely.

It's absolutely worth it for some people to pay off the mortgage, keep a cash buffer, etc.

It's just disingenuous when they try to argue that's the better approach mathematically, because data has shown it isn't, historically, and there's no reason for that to be "different this time."

IMO, the best thing to do is explore the options, understand them thoroughly, and choose the best one for you, accepting its downsides (whether the downside is you'll have a portfolio crash harder and need more nerves of steel to ride it out, or the downside is you have to work longer, or whatever).  Trying to convince yourself of a certain approach having no downsides is dangerous.  Learn, accept where the flaws in your plan are, and then you can be prepared for them, and help compensate for them.

[mathjak107]

the cash is a declining balance since by the end of year 1 it is gone and you are no longer holding 2 years .

So you aren't saying always have 2 years in cash (i.e. a 92%/8% equities/cash)?  You're saying start with 2 years, spend it all regardless of market conditions, allow the equities (the other 23x/92% of your portfolio) to grow over those two years, then go 100% stocks?

Okay, I went ahead and tested this scenario.

I ran a cFIREsim scenario of having saved 23x expenses in 100% equities, and 2x expenses in cash, by putting in that 23x expenses as the portfolio (920k on 40k spending in equities, 80k, or two years spending, in cash).  To model this I put in a portfolio of 920k, 100% equities, then I added a single income of 40k in years 1 and 2.  So that offsets the 40k spending those years, as that 80k cash is spent down, and the portfolio is left to grow the first two years.  All other stuff (fees, etc.) left default.

I also ran a default cFIREsim of 75/25 stocks/bonds, entire 1MM portfolio invested, no cash infusion, as well as a 100% stocks scenario.

I then downloaded the spreadsheets and found the median and average of the year two end portfolio for each scenario.

Cash scenario:
Average: 1,109,048.311
Median: 1,098,287.19

75/25 stocks/bonds scenario:
Average: 1,084,244.348
Median: 1,081,889.04

100% equities scenario:
Average: 1,113,412.208
Median: 1,099,959.26

Looks like straight equities > equities + cash > 75/25.

The interesting thing though is that 75/25 has a lower failure rate (i.e. higher success rate, as I posted earlier in the thread), so it's not just about the most money.

Either way though, if you want a lower failure rate, it doesn't seem like cash is the way to go, and if you want more money two years after ERing, ditto.

I'm not seeing a scenario where investing 23x your income and keeping 2x in cash to start out your ER, then spending down that two years in cash is beneficial.  Nor am I seeing one where keeping two years in cash forever is beneficial.

Other than, as we said, psychologically.

If it helps you sleep at night, great, more power to you.  But I don't see the math where it helps your success rate, or leaves you with more money.

in firecalc

75/25 had a slightly better success rate going out to 30 years

100% equity s had a higher success rate and bigger balance  going out to 40 years .

i would think you guys are more interested in the 40 year results retiring so young

30 years 4%  at 75/25


FIRECalc looked at the 116 possible 30 year periods in the available data, starting with a portfolio of $1,000,000   and spending your specified amounts each year thereafter.
Here is how your portfolio would have fared in each of the 116 cycles. The lowest and highest portfolio balance at the end of your retirement was $-400,986 to $5,679,475, with an average at the end of $1,845,488. (Note: this is looking at all the possible periods; values are in terms of the dollars as of the beginning of the retirement period for each cycle.)
For our purposes, failure means the portfolio was depleted before the end of the 30 years. FIRECalc found that 6 cycles failed, for a success rate of 94.8%.

30 year 100% equity
FIRECalc looked at the 116 possible 30 year periods in the available data, starting with a portfolio of $1,000,000   and spending your specified amounts each year thereafter.
Here is how your portfolio would have fared in each of the 116 cycles. The lowest and highest portfolio balance at the end of your retirement was $-931,017 to $8,509,297, with an average at the end of $2,686,348. (Note: this is looking at all the possible periods; values are in terms of the dollars as of the beginning of the retirement period for each cycle.)
For our purposes, failure means the portfolio was depleted before the end of the 30 years. FIRECalc found that 8 cycles failed, for a success rate of 93.1%



40 years 75/25
FIRECalc looked at the 106 possible 40 year periods in the available data, starting with a portfolio of $1,000,000   and spending your specified amounts each year thereafter.
Here is how your portfolio would have fared in each of the 106 cycles. The lowest and highest portfolio balance at the end of your retirement was $-1,120,937 to $12,410,499, with an average at the end of $2,365,051. (Note: this is looking at all the possible periods; values are in terms of the dollars as of the beginning of the retirement period for each cycle.)
For our purposes, failure means the portfolio was depleted before the end of the 40 years. FIRECalc found that 16 cycles failed, for a success rate of 84.9%.


40 years 100% equity
FIRECalc looked at the 106 possible 40 year periods in the available data, starting with a portfolio of $1,000,000   and spending your specified amounts each year thereafter.
Here is how your portfolio would have fared in each of the 106 cycles. The lowest and highest portfolio balance at the end of your retirement was $-1,624,576 to $23,195,431, with an average at the end of $4,088,042. (Note: this is looking at all the possible periods; values are in terms of the dollars as of the beginning of the retirement period for each cycle.)
For our purposes, failure means the portfolio was depleted before the end of the 40 years. FIRECalc found that 13 cycles failed, for a success rate of 87.7%.

[Retire-Canada]

in firecalc

75/25 had a slightly better success rate going out to 30 years

100% equity s had a higher success rate and bigger balance  going out to 40 years .

i would think you guys are more interested in the 40 year results retiring so young

The numbers you posted are essentially equal in success for both options at either 30 or 40yrs within the limitations of the simulation. I wouldn't object to either asset allocation if those were the success rates. I haven't looked into it any deeper than that nor what the differences between FIRE Calc and cFIREsim are.

The issue I had was holding cash as a buffer due to the cost for portfolio returns - particularly if the plans is to hold cash for the long term. The costs far out weigh the benefits in my opinion. What I am still not clear on with your plan is if you will start with 23x COL invested and 2x COL cash and use the cash for the first two years or if you will start with those assets and keep the 2yrs in cash until there is a crash. With either option are you replenishing the cash allocation after the crash or then going 100% equities?

[mathjak107]

our plan is to always have the current years cash ready to go , and to start filling up the 2nd year once we deplete the current year .

so far it worked well . our first year  in retirement our returns in total  were 1% as markets were pretty flat , while we spent 3.50% . had we used a heloc we would have had 4% interest too paid out .

this year we are up 5% with about 3.50% going out , we are just about down to just 1 years cash left which is next years . had we had a heloc another 4% in interest  would have went out .

no cash only works well after you are in retirement and had your first run up . once you are up comfortably holding cash is not a big issue . you can sell directly from assets up or down .

but the danger is the early years where any down year is like a trader having a string of losing trades  right up front and that starting principal is gone .

most of the ploys done with buckets ,rising glide paths and cash are to protect the early years going in .

in our case i don't need more than 40-50% equity. i don't want the volatility of it anymore as well as don't need it . in fact we are delaying ss just to cut market dependency  .  my wife being a widow once already and getting burned in 2000  does not want to get stuck with  a pile of investments again and no real base income .

40-50% in equity's gives us all the volatility we want . we have seen single days already at that allocation where we have swung almost 40k in one day .

percentage wise it is not a lot but when you think a 7% move represents 9 years of maximizing our 401k's at catch up it is an insane amount in dollars .

so our plan really calls for matching our allocations to our needs , both dollar wise and mentally wise . we may cut dependency even more and ladder some spia's eventually as part of the overall plan .  we are far more comfortable betting on our longevity as a couple since odds are one of us will go on longer than betting on the whims of markets and rates at these levels .

for us it is no longer about growing richer , it is now about enjoying our lives and not growing poorer . so whatever we do is a balance .

eventually as you reach your 60's the decisions you have to make will all revolve around two choices , do i want more  longevity risk or market and interest rate risk .

as far as how many times savings we went ?

i never looked at it that way ,since i always was an investor since my teens and had enough to retire in my 50's if we wanted  to . we set our budget in a different way since we tailored our lives to fit the budget we had  instead of having the life  in place and then working at  growing enough  to fit that life  .

so what we did is add up all our non discretionary bills . things like rent , insurance , and all the costs that were pretty much written in stone that we wanted to keep in place .  it did not include food ,travel,gifts ,entertainment , etc.

we then doubled that amount we had in non discretionary spending  to allow for discretionary spending . that big 50%  budget in discretionary spending gave us plenty of wiggle room for the awe craps like the more than 20k in dental we just got hit with .

so lets use some hypothetical numbers .

lets say non discretionary was 50k  so our budget would be 100k plus taxes . lets call it 120k .

we would then subtract out pension and eventual social security .  lets say the two are 75k . 

so we need 45k from our portfolio.

a rough check by taking 45k and multiplying it by 25 times says we need a ballpark of  1.25 million to support that at 4% .

using both firecalc and the fidelity planner shows that an allocation of 40-50% equity's should be able to sustain that before we introduce life expectancy and human spending patterns which tend to increase odds even greater .

in practice our draw worked out to about 3.50% and once ss kicks in it will fall to 2% .

the extra ss money from delaying  will refill what we spent down  delaying up front .

that method worked out well for us and we pretty much did  and bought whatever we wanted and it still worked out we did not spend the max we could have if need be .

[arebelspy]

our plan is to always have the current years cash ready to go , and to start filling up the 2nd year once we deplete the current year .

I think that's what most early retirees do, pull their budget beginning of the year (though some may do it quarterly or monthly).

[mathjak107]

we are just more comfortable doing it in one shot  . we know we do not have to make selling decisions through the year . i just added to the above post so you may have read it before i gave you all our details .

we put all dividends and interest towards next years income . we find our distributions from our funds have run from 29k to 69k in a year . it is to hard to keep the income flow going with that kind of spread  so we just put it towards filling next years money and that gives us time to plan how to fund the shortfall rather than do it in real time by trying to use that money today .

all cash we do hold is laddered in cd's and matures monthly going out the two years . we buy them through fidelity so if need be we can click and sell the cd's instantly in their market .

january will be our first refill date since we retired 2 years ago  . so in january we will have the current year in place but then have to see what to sell to start to form the 2nd year we hold in reserve .

[TomTX]

our plan is to always have the current years cash ready to go , and to start filling up the 2nd year once we deplete the current year .

so far it worked well . our first year  in retirement our returns in total  were 1% as markets were pretty flat , while we spent 3.50% . had we used a heloc we would have had 4% interest too paid out .

No.  Your sentence implies you are paying WAY more interest than you actually would by mixing your percentages improperly. Below I convert the HELOC numbers to percent of the portfolio.

Even following your method, your HELOC interest would have been 0.04 x 2.5%  to cover the portfolio shortfall (the first 1% is covered by portfolio growth) if you took it all at the beginning of the year instead of as needed. So, added cost is 0.075% of the portfolio

This brings your spend from 3.50% to 3.575%

If you took it as-needed, it would have been 2% of 2.5%. So, 0.038% of the portfolio

This brings your spend from 3.5% to 3.538%

4% isn't a great rate for a HELOC. My credit union is 3%. So, if you shop around, you can cut the amount even further.

The second year with your 5% gain would have 3.5% used for spend, and 1.5% to pay back the HELOC. You would carry over 1% in cheap debt.