FIREes with cash for down markets: what's your plan? Withdraw Algorithm?

[Villanelle]

This is the part of FIRE that still confounds me.  I'm trying to figure out our withdraw plan.  In theory, the idea of holding cash (6 mo?  12 mo?  18mo? expenses) appeals to me.  If the market is down, we spend from our cash to avoid selling low.  Got it.  But what does that look like in reality?  How do you know when to spend the reserves, when to replenish, and how to replenish? 

How do you define "low" market, and spending your cushion?  When do you replenish that cushion?  I'd like to start working on sort of an IPS but for withdraws, so it is planned out and I don't need to make decisions as a reaction to the market.  But I'm having trouble figuring out what "use cash when the market is down" would actually mean.  I see people here frequently mention that's their plan or their approach, but not with specifics.

How do you do this? 

I'm really looking for something like the following: "Have 18 months spending in cash (or CD ladder, or similar).  Withdraw on set schedule,[ which will probably be quarterly withdraws for us] unless the market is down >15% on planned withdraw day.  In that case, spend cash until market is back to at least 10% of previous high or cash is down to >6 months.  In that case, make next planned withdraw on schedule unless market is still down >20%.  In that case, spend down to 3 months cash, then withdraw regardless of market. Start replenishing cash until back up to 18 months, by adding 50% to each withdraw, as soon as market is at -5% of peak or better.  If market is +10 or more, double each withdraw until back at 18 mo cushion."

I just made that up, but I think that's the kind of detail I need in order to just execute the algorithm I decided on during unemotional times, rather than having to make choices when I'm in the thick of it.  Even if your plan isn't that detailed, how do you handle this stuff?  So many plans seem to involved cash for bear-times, but without specifics that seems vague to the point of being useless.  At least for the way my mind (and emotions) work.  I want an algorithm, preferably one I can program into my spreadsheet, so I don't have to do any significant thinking or deciding when I'm in the thick of it. 

[dividendman]

This is what I do, and it's not as complicated as market movements.

From my IPS:

   - To implement a rising equity glidepath of investments
      - this means starting with VTI/VEU/BND of 48%/27%/25% *this was at retirement start"
      - reducing the bond allocation by 2% (1% to each domestic and international, until international gets 30%, then all domestic) a year for the first 10 years
      - the final breakdown after 10 years will be 65/30/5
      - keep 1 year of expenses in short term bonds/cash (included in the bond allocation)

Review process
   - Investments will be reviewed monthly (as well as liquidated for expenses as need be), liquidation will bring as much balance as it can

So... basically it's a rising equity glidepath. When markets tank, you naturally are going to take money out of your bonds/cash allocation because it'll be out of whack. If markets rise, then you might need to actually sell tocks and move it into cash/bonds.

You don't need any trigger-points for the market. Just set aside some of your bond allocation in short term/cash, and then when you need money, check your allocations, and take money so it makes the allocation go in the right direction.

I do an actual rebalance twice a year on fixed dates as well.

[secondcor521]

The questions you asked and the issues you posed are why I eventually ended up skipping the whole cash cushion thing.

I keep very minimal cash on hand.  Just enough to avoid overdrafting my checking accounts and enough to pay my bills for the next few weeks.  I withdraw from taxable to checking monthly.

Everything else is in investments which I rebalance to my target AA if they are out of whack when I check.  My target AA is 98/2 and out of whack is 25% off of centerline.

I've never had any actual use for a cash cushion except to maintain cash flow if I lost my job while I had a mortgage and child support.  Mortgage is paid off and child support is done, so that's no longer an issue.

My approach does have the risk where I might have a somewhat unexpected, relatively large expense that throws off my tax planning somehow.  I view this as low likelihood and low impact.

[Villanelle]

This is what I do, and it's not as complicated as market movements.

From my IPS:

   - To implement a rising equity glidepath of investments
      - this means starting with VTI/VEU/BND of 48%/27%/25% *this was at retirement start"
      - reducing the bond allocation by 2% (1% to each domestic and international, until international gets 30%, then all domestic) a year for the first 10 years
      - the final breakdown after 10 years will be 65/30/5
      - keep 1 year of expenses in short term bonds/cash (included in the bond allocation)

Review process
   - Investments will be reviewed monthly (as well as liquidated for expenses as need be), liquidation will bring as much balance as it can

So... basically it's a rising equity glidepath. When markets tank, you naturally are going to take money out of your bonds/cash allocation because it'll be out of whack. If markets rise, then you might need to actually sell tocks and move it into cash/bonds.

You don't need any trigger-points for the market. Just set aside some of your bond allocation in short term/cash, and then when you need money, check your allocations, and take money so it makes the allocation go in the right direction.

I do an actual rebalance twice a year on fixed dates as well.

This makes sense to me and I like the relative simplicity.  Unfortunately, I don't think I can make it work for us since we have almost no bonds.  DH's inflation-adjusted pension takes the place of that in our AA.  Otherwise, I'd likely just do something like this.  Since we are almost all equities (>10% bonds, which DH likes to keep even though I'd ditch or at least reduce it), we can't manage market highs and lows by changing where we withdraw from.

[Loren Ver]

So I do mine very differently and it is odd and a work in progress since the last few years have been odd.  I'll still share it in case you can take something useful from it.  DH and I have been adapting as we continue to learn and grow.

We have two cushions.

First:
We take out the cash we need a year ahead of time.  So our cushion is that I am spending 2022 money now in 2023.  Now in Dec of 2023 I will take out 2024 money. 

Why December?  Well we are on ACA and when there is turbulence (lots of growth) in the markets our mutual funds that have never paid out capital gains decide to pay out capital gains the last few weeks of December (two of the last 4 years, but only 3 of the last 23).  So to make sure we don't increase our income more than we need to, I wait until I see what our income will be and then take out what we need for the next year.  Then we have a pile of cash for the year, plus any buffers we want to have etc.  This is not what we started doing when we FIRED, but changed to this over time.

Simple.

Second
We have a money market with our cushion more of what we are talking about.  It holds some cash in it.  it's original goal was to hold enough cash to cover two years of mortgage payments and OOPM insurance in case the market tanked and we needed to cover our two biggest costs we didn't want to miss in DISASTER happened.  This money would only be employed if we first ran out of our year of money and then couldn't make our mortgage and then started to hit our out of pockets costs on insurance.  Note, we would still be on ACA and would have to hit the minimum income that year so even in a down market we are taking out a year of money.  So the longer we were retired the less this was making sense.

So now, this money market money is mostly used to cover that tight time towards the end of the year before we pull out money but might have a big expense like an unexpected vacation or need a new car or home repair.  Then we back fill as needed come December. 

Let me know if you have any questions or if I can help.

I, personally, don't think down markets are all that scary.  But I also invest in some volatile funds. 

Loren

[bacchi]

I follow McClung's (Alternate) Prime Harvesting strategy, which keeps as much invested in equities as possible. It's similar to what dividendman is doing but stock is only sold when the market is on a tear.

1) Sell enough bonds to generate (1 year of) cash.
   a) If there aren't any bonds to sell, sell equities.
2) When the market is up a lot (+20% is a simple method), sell equities to buy more bonds.

Ideally, stocks generate outsized returns and are trimmed to refill the bond bucket. Historically, in most cases, the bond allocation grew to a very large percentage of the asset allocation.

[mistymoney]

This is the part of FIRE that still confounds me.  I'm trying to figure out our withdraw plan.  In theory, the idea of holding cash (6 mo?  12 mo?  18mo? expenses) appeals to me.  If the market is down, we spend from our cash to avoid selling low.  Got it.  But what does that look like in reality?  How do you know when to spend the reserves, when to replenish, and how to replenish? 

How do you define "low" market, and spending your cushion?  When do you replenish that cushion?  I'd like to start working on sort of an IPS but for withdraws, so it is planned out and I don't need to make decisions as a reaction to the market.  But I'm having trouble figuring out what "use cash when the market is down" would actually mean.  I see people here frequently mention that's their plan or their approach, but not with specifics.

How do you do this? 

I'm really looking for something like the following: "Have 18 months spending in cash (or CD ladder, or similar).  Withdraw on set schedule,[ which will probably be quarterly withdraws for us] unless the market is down >15% on planned withdraw day.  In that case, spend cash until market is back to at least 10% of previous high or cash is down to >6 months.  In that case, make next planned withdraw on schedule unless market is still down >20%.  In that case, spend down to 3 months cash, then withdraw regardless of market. Start replenishing cash until back up to 18 months, by adding 50% to each withdraw, as soon as market is at -5% of peak or better.  If market is +10 or more, double each withdraw until back at 18 mo cushion."

I just made that up, but I think that's the kind of detail I need in order to just execute the algorithm I decided on during unemotional times, rather than having to make choices when I'm in the thick of it.  Even if your plan isn't that detailed, how do you handle this stuff?  So many plans seem to involved cash for bear-times, but without specifics that seems vague to the point of being useless.  At least for the way my mind (and emotions) work.  I want an algorithm, preferably one I can program into my spreadsheet, so I don't have to do any significant thinking or deciding when I'm in the thick of it.

nords has this covered! he keeps 2 years in cash, spends throughout the year, and if the market is > last dec 31, he takes another year out of the market and puts to cash.

maybe search some more exact wording from his historical posts.....

[GilesMM]

I don't really time the market as you describe as it requires "cash drag" which long term hurts you.


That said, if you find yourself in a historic law, markets have dropped 30% from highs, there is blood in the streets and it appears things have changed forever, everyone is selling in a panic, it  wouldn't be a terrible time to buy some more stock.

[Villanelle]

This is the part of FIRE that still confounds me.  I'm trying to figure out our withdraw plan.  In theory, the idea of holding cash (6 mo?  12 mo?  18mo? expenses) appeals to me.  If the market is down, we spend from our cash to avoid selling low.  Got it.  But what does that look like in reality?  How do you know when to spend the reserves, when to replenish, and how to replenish? 

How do you define "low" market, and spending your cushion?  When do you replenish that cushion?  I'd like to start working on sort of an IPS but for withdraws, so it is planned out and I don't need to make decisions as a reaction to the market.  But I'm having trouble figuring out what "use cash when the market is down" would actually mean.  I see people here frequently mention that's their plan or their approach, but not with specifics.

How do you do this? 

I'm really looking for something like the following: "Have 18 months spending in cash (or CD ladder, or similar).  Withdraw on set schedule,[ which will probably be quarterly withdraws for us] unless the market is down >15% on planned withdraw day.  In that case, spend cash until market is back to at least 10% of previous high or cash is down to >6 months.  In that case, make next planned withdraw on schedule unless market is still down >20%.  In that case, spend down to 3 months cash, then withdraw regardless of market. Start replenishing cash until back up to 18 months, by adding 50% to each withdraw, as soon as market is at -5% of peak or better.  If market is +10 or more, double each withdraw until back at 18 mo cushion."

I just made that up, but I think that's the kind of detail I need in order to just execute the algorithm I decided on during unemotional times, rather than having to make choices when I'm in the thick of it.  Even if your plan isn't that detailed, how do you handle this stuff?  So many plans seem to involved cash for bear-times, but without specifics that seems vague to the point of being useless.  At least for the way my mind (and emotions) work.  I want an algorithm, preferably one I can program into my spreadsheet, so I don't have to do any significant thinking or deciding when I'm in the thick of it.

nords has this covered! he keeps 2 years in cash, spends throughout the year, and if the market is > last dec 31, he takes another year out of the market and puts to cash.

maybe search some more exact wording from his historical posts.....

I'll look for them.  Thanks.  Like Nords, we will have that sweet military pension.  Depending on where we end up living, just the pension should cover at least most of our expenses, at a not-especially lean FIRE.  (Given the massive differences in COL between areas we might land, it's hard to say, but at a minimum, the pension will do most of it.)  So in really lean times, we could probably live almost indefinitely without withdrawing.  But I still need to have a plan and that continues to confound me.  What you describe seems simple, which I like, but also specific enough that I'll never have to decide, when in the thick of things, what to do and when. 

Maybe I also need to decide how much cash I want to hold. 

[curious_george]

We have about one year of expenses in IBonds. This will hopefully be two years before I retire.

Most of our expenses are covered by dividends from REITs and food manufacturers.

Our plan is simple.

First - we write down the shiller pe ratio the day I retire. This is X.

If I retire and shiller pe ratio of stock market goes higher than X, we sell from stocks and use dividend + stock sales to fund our living expenses.

If I retire and shiller pe ratio of stock market goes down, then we use dividends + IBonds to fund our living expenses until either IBonds are depleted or shiller pe ratio goes back above X again.

I basically want to have enough dividends and IBonds to ride out the first down market after retirement without selling shares.

[mistymoney]

This is the part of FIRE that still confounds me.  I'm trying to figure out our withdraw plan.  In theory, the idea of holding cash (6 mo?  12 mo?  18mo? expenses) appeals to me.  If the market is down, we spend from our cash to avoid selling low.  Got it.  But what does that look like in reality?  How do you know when to spend the reserves, when to replenish, and how to replenish? 

How do you define "low" market, and spending your cushion?  When do you replenish that cushion?  I'd like to start working on sort of an IPS but for withdraws, so it is planned out and I don't need to make decisions as a reaction to the market.  But I'm having trouble figuring out what "use cash when the market is down" would actually mean.  I see people here frequently mention that's their plan or their approach, but not with specifics.

How do you do this? 

I'm really looking for something like the following: "Have 18 months spending in cash (or CD ladder, or similar).  Withdraw on set schedule,[ which will probably be quarterly withdraws for us] unless the market is down >15% on planned withdraw day.  In that case, spend cash until market is back to at least 10% of previous high or cash is down to >6 months.  In that case, make next planned withdraw on schedule unless market is still down >20%.  In that case, spend down to 3 months cash, then withdraw regardless of market. Start replenishing cash until back up to 18 months, by adding 50% to each withdraw, as soon as market is at -5% of peak or better.  If market is +10 or more, double each withdraw until back at 18 mo cushion."

I just made that up, but I think that's the kind of detail I need in order to just execute the algorithm I decided on during unemotional times, rather than having to make choices when I'm in the thick of it.  Even if your plan isn't that detailed, how do you handle this stuff?  So many plans seem to involved cash for bear-times, but without specifics that seems vague to the point of being useless.  At least for the way my mind (and emotions) work.  I want an algorithm, preferably one I can program into my spreadsheet, so I don't have to do any significant thinking or deciding when I'm in the thick of it.

nords has this covered! he keeps 2 years in cash, spends throughout the year, and if the market is > last dec 31, he takes another year out of the market and puts to cash.

maybe search some more exact wording from his historical posts.....

I'll look for them.  Thanks.  Like Nords, we will have that sweet military pension.  Depending on where we end up living, just the pension should cover at least most of our expenses, at a not-especially lean FIRE.  (Given the massive differences in COL between areas we might land, it's hard to say, but at a minimum, the pension will do most of it.)  So in really lean times, we could probably live almost indefinitely without withdrawing.  But I still need to have a plan and that continues to confound me.  What you describe seems simple, which I like, but also specific enough that I'll never have to decide, when in the thick of things, what to do and when. 

Maybe I also need to decide how much cash I want to hold.

I think I'd yolo with 100% stocks in that case!

Maybe 20k or so in ibonds for larger irregular expenses or other "somehtings" that may crop up.

[Telecaster]

The questions you asked and the issues you posed are why I eventually ended up skipping the whole cash cushion thing.

I keep very minimal cash on hand.  Just enough to avoid overdrafting my checking accounts and enough to pay my bills for the next few weeks.  I withdraw from taxable to checking monthly.

This.  Although I keep more cash than that.  But the notion of keeping a cash cushion in order to avoid selling stock in a down market is one of those ideas that just needs to die.  Over any reasonable length of time--like a retirement, this method does not improve portfolio survivability or increase portfolio performance.  Mathematically it can't. 

Another problem it is that it is impossible to actually implement in real life.  For example, the market is currently down from its December 2021 peak.  So time to spend cash, right?   But it is up from the October 2022 bottom.  So time to sell stock and replenish cash.  Or is it?   There is no way to really know if you are in a bull or bear market until after the fact. 

[lutorm]

Our strategy (as of Wednesday this week, we are FIRE!) is to have 5 years of living expenses in T-bills, inspired by https://www.kitces.com/blog/accelerating-the-rising-equity-glidepath-with-treasury-bills-as-portfolio-ballast/. The T-bills will get spent down, implementing a rising equity glidepath, and five years from now we'll hopefully have mitigated most of the immeidate post fire SORR and can just spend from our investments and not worry about it.

[Ron Scott]

We’re 60-40 +-. We try to avoid manufacturing taxes and don’t have enough room in our deferred accounts for the bond allocation. So we set up a muni ladder in our taxable account with maturities spread fairly well throughout the years. We spend forced interest and dividends in taxable plus whatever we need from the maturing munis. When RMDs kick in we’ll suffer the consequences and spend some too vs. goosing the muni ladder.

We keep a couple, 3 months in cash, considering it working capital more than emergency money.

Hate selling stocks. When they’re up you get taxed, when they’re down you lose money. Always a problem.

Hate dividends; prefer buybacks. Don’t want some corporate board deciding to create taxable income for me whenever. Reduce shares outstanding, let the price pop, and let me decide how to deal with the gains.

We are not rebalancers. With the growth in equities and drawdown on munis we stay close to where we started, and don’t worry about risk and returns enough to get technical with it.

Never lived on a budget. Always seemed like an excuse to spend to us.

We don’t rely on historical returns to make future predictions. We trust investment disclaimers that tell you not to and rely on our assumptions of likely political and Fed reactions to prolonged low real returns.

Don’t plan our lives so our last dollar is spent on the morphine drip. Seems morbid and way too much work.

No more Roth conversions. Have enough in there to handle any future big spending without manufacturing taxes. Regardless of the scenarios we envision, paying taxes upfront always seemed like a gift to our daughter vs. a tax savings to ourselves. That medicine just tastes too bad.

Sometimes we feel like we could end up with more at the end of the year if we got more technical about things. But out accountant tells us we are highly efficient and we don’t have the patience to run it like a business. A dollar here, a dollar there, pretty soon you’re talking about $2…

[Nords]

Thanks, @mistymoney! 

@Villanelle, the two years of expenses in cash is intended to protect people against sequence of returns risk.  It’s generally only needed during the first decade of FI. 

This post goes into greatly excessive detail on how to invest that two-year cash stash (in money markets & CDs) and how to draw it down. 
https://militaryfinancialindependence.com/2014/02/20/how-should-i-invest-during-retirement/
The post is intended for people who want an algorithm, and after 10 years you can stop doing it.  By then you’ll be much more accustomed to your FI lifestyle and ready to drop the two-year cash stash anyway.  Or keep doing it if it makes you feel better-- the emotions of behavioral financial psychology will derail math & logic every time.

Since you have an inflation-fighting military pension anchoring most of your expenses (the asset allocation equivalent of I bonds*), your investment asset allocation can afford to be high in equities.  (I’d recommend 70%-80%, or even higher if you don’t mind the volatility.)  It’s more important to be able to sleep well at night, but we’re comfortable with having all of our investments in VTI and staying >95% equities.  After over 20 years of FI, we favor the simplicity of VTI instead of managing a stash of CDs.

With most of your expenses covered by the pension, your withdrawal rate might not reach the 4% Safe Withdrawal Rate.  Even if it did, your military pension will rise each year with inflation while your investments (high in equities) will (over a decade) grow faster than inflation.  At the end of your first decade of FI, your spending will have risen at about the rate of inflation while your investments will have grown faster than inflation.  If you measure your withdrawal rate all over again for year #11, it’d be lower than when you started FI-- and even Karsten Jeske would agree that a 3.5% withdrawal rate is good for 60 years. 

*(The I bond analogy is flawed because they mature at 30 years, but the bond-like logic works for making asset allocation decisions.)

[Villanelle]

Thanks, @mistymoney! 

@Villanelle, the two years of expenses in cash is intended to protect people against sequence of returns risk.  It’s generally only needed during the first decade of FI. 

This post goes into greatly excessive detail on how to invest that two-year cash stash (in money markets & CDs) and how to draw it down. 
https://militaryfinancialindependence.com/2014/02/20/how-should-i-invest-during-retirement/
The post is intended for people who want an algorithm, and after 10 years you can stop doing it.  By then you’ll be much more accustomed to your FI lifestyle and ready to drop the two-year cash stash anyway.  Or keep doing it if it makes you feel better-- the emotions of behavioral financial psychology will derail math & logic every time.

Since you have an inflation-fighting military pension anchoring most of your expenses (the asset allocation equivalent of I bonds*), your investment asset allocation can afford to be high in equities.  (I’d recommend 70%-80%, or even higher if you don’t mind the volatility.)  It’s more important to be able to sleep well at night, but we’re comfortable with having all of our investments in VTI and staying >95% equities.  After over 20 years of FI, we favor the simplicity of VTI instead of managing a stash of CDs.

With most of your expenses covered by the pension, your withdrawal rate might not reach the 4% Safe Withdrawal Rate.  Even if it did, your military pension will rise each year with inflation while your investments (high in equities) will (over a decade) grow faster than inflation.  At the end of your first decade of FI, your spending will have risen at about the rate of inflation while your investments will have grown faster than inflation.  If you measure your withdrawal rate all over again for year #11, it’d be lower than when you started FI-- and even Karsten Jeske would agree that a 3.5% withdrawal rate is good for 60 years. 

*(The I bond analogy is flawed because they mature at 30 years, but the bond-like logic works for making asset allocation decisions.)

As always, thanks!

We are currently at about 10% bonds (bond funds).  I'd be at 100%, but DH gets a say as well. ;)

Depending on where we land, it's entirely possible the very generous pension will cover 100% of our expenses, but even if it does, I don't want to leave easy money on the table or stress about the market, but I also don't want to have to spend the first Tuesday of every month doing complex math.  This solution seems like it might be a good balance of both.   At the very least, it gives us a solid plan from which we can tinker for the our personal comfort levels.  Thank you! 

[mistymoney]

Thanks, @mistymoney! 

@Villanelle, the two years of expenses in cash is intended to protect people against sequence of returns risk.  It’s generally only needed during the first decade of FI. 

This post goes into greatly excessive detail on how to invest that two-year cash stash (in money markets & CDs) and how to draw it down. 
https://militaryfinancialindependence.com/2014/02/20/how-should-i-invest-during-retirement/
The post is intended for people who want an algorithm, and after 10 years you can stop doing it.  By then you’ll be much more accustomed to your FI lifestyle and ready to drop the two-year cash stash anyway.  Or keep doing it if it makes you feel better-- the emotions of behavioral financial psychology will derail math & logic every time.

Since you have an inflation-fighting military pension anchoring most of your expenses (the asset allocation equivalent of I bonds*), your investment asset allocation can afford to be high in equities.  (I’d recommend 70%-80%, or even higher if you don’t mind the volatility.)  It’s more important to be able to sleep well at night, but we’re comfortable with having all of our investments in VTI and staying >95% equities.  After over 20 years of FI, we favor the simplicity of VTI instead of managing a stash of CDs.

With most of your expenses covered by the pension, your withdrawal rate might not reach the 4% Safe Withdrawal Rate.  Even if it did, your military pension will rise each year with inflation while your investments (high in equities) will (over a decade) grow faster than inflation.  At the end of your first decade of FI, your spending will have risen at about the rate of inflation while your investments will have grown faster than inflation.  If you measure your withdrawal rate all over again for year #11, it’d be lower than when you started FI-- and even Karsten Jeske would agree that a 3.5% withdrawal rate is good for 60 years. 

*(The I bond analogy is flawed because they mature at 30 years, but the bond-like logic works for making asset allocation decisions.)

Had not had the nuance on it being only the first 10 years! Link is a good read even if you don't have the military pension going on.

[Catbert]

 Two thoughts:

For starters on you taxable accounts, set mutual fund dividends and internal capital gains to payout rather than be reinvested.  Then you can choose whether to reinvest/rebalance or spend the cash without generating additional taxable events.

Look at how much your portfolio has increased semi-annually (or quarterly or yearly but not monthly) and compare that to what you've estimated you'll get in an average year.  If it's above that percentage scrape the excess gain and use as cash or bonds.  For example, if you estimate your portfolio *should* increase by 10% a year and it's increased 15% mid-year then take the 5% extra gain out of stocks and put in your bond or cash bucket.  This does not take into account things like ACA eligibility or the seemingly random huge uncontrollable cap gains payouts.

As someone with pensions/SS/rental real estate I do the first suggestion but the second one is just theoretical.

[MrGreen]

This isn't particularly relevant to the OP but I'm going to tack it on here anyway in case it would help someone younger.

For us, being FIREd in our 30s, this is largely based upon other determining factors. Our tIRAs make up the bulk of our stash. Because of the age constraints on those accounts, we want our brokerage accounts to grow as aggressively as possible. We DON'T want our tIRAs growing any more aggressively than they have to because that can create tax headaches with RMDs later in life. So, in general, we hold our bond allocation in tIRAs. By effect, Roth accounts are all equities as well, though this is somewhat irrelevant because we can't spend anything but Roth principal until 60.

So we are 100% equities in brokerage accounts. We sell them every month to replenish our checking account and, if need be, we'll convert some bonds to stocks in our tIRAs to keep the AA in line. That's a less frequent occurrence because otherwise we're essentially rebalancing monthly. We also intentionally generate taxable income because we're now in a position that if we don't, we'll actually pay more taxes over our lifetime by letting balances build up until RMDs. This also improves access to our money by moving it from tax-deferred accounts where we can't spend it without penalty to after-tax accounts that we'll be drawing from exclusively for the next 20+ years.

Statistically, keeping our brokerage accounts 100% equities and only selling as we need the money outperforms other alternatives. If the brokerage balance is small enough though this can feel scary until one's Roth conversion pipeline is fully established. Then it's potentially irrelevant. Psychologically, making small withdrawals every month feels a lot easier than trying to be comfortable with big lump-sum withdrawals, especially during times of high volatility.

Disclaimer: If anyone looks at my journal you'll see that our current brokerage AA is a different than what I've typed above. This is an artifact of the house purchase we made earlier this year and a decision to borrow money from a family member and take out I Bonds rather than use that cash for the down payment. We were 100% equities in our brokerage accounts prior to that and will return to it as soon as the interest rate on I Bonds falls below the interest rate of the loan we have with our family member.

[mistymoney]

This isn't particularly relevant to the OP but I'm going to tack it on here anyway in case it would help someone younger.

For us, being FIREd in our 30s, this is largely based upon other determining factors. Our tIRAs make up the bulk of our stash. Because of the age constraints on those accounts, we want our brokerage accounts to grow as aggressively as possible. We DON'T want our tIRAs growing any more aggressively than they have to because that can create tax headaches with RMDs later in life. So, in general, we hold our bond allocation in tIRAs. By effect, Roth accounts are all equities as well, though this is somewhat irrelevant because we can't spend anything but Roth principal until 60.

So we are 100% equities in brokerage accounts. We sell them every month to replenish our checking account and, if need be, we'll convert some bonds to stocks in our tIRAs to keep the AA in line. That's a less frequent occurrence because otherwise we're essentially rebalancing monthly. We also intentionally generate taxable income because we're now in a position that if we don't, we'll actually pay more taxes over our lifetime by letting balances build up until RMDs. This also improves access to our money by moving it from tax-deferred accounts where we can't spend it without penalty to after-tax accounts that we'll be drawing from exclusively for the next 20+ years.

Statistically, keeping our brokerage accounts 100% equities and only selling as we need the money outperforms other alternatives. If the brokerage balance is small enough though this can feel scary until one's Roth conversion pipeline is fully established. Then it's potentially irrelevant. Psychologically, making small withdrawals every month feels a lot easier than trying to be comfortable with big lump-sum withdrawals, especially during times of high volatility.

Disclaimer: If anyone looks at my journal you'll see that our current brokerage AA is a different than what I've typed above. This is an artifact of the house purchase we made earlier this year and a decision to borrow money from a family member and take out I Bonds rather than use that cash for the down payment. We were 100% equities in our brokerage accounts prior to that and will return to it as soon as the interest rate on I Bonds falls below the interest rate of the loan we have with our family member.

hey Mr. Green...this has me confused. What is the purpose of bonds in the tIRA if they aren't ever there to spend if the stocks are down? I don't understand the role of a large bond allocation if you are selling stocks to live on for 20 more years.

[ChpBstrd]

Cash is yielding about 5.25% right now, with a 2.25% real premium over CPI. That's very attractive today, but these are unusual times. Usually the real interest rate on risk-free assets is around 0%.

So in today's world, it probably makes sense to have a fat cash or bonds cushion, but if conditions changed back to where they were a few years ago, it would become expensive to maintain a large allocation that is earning next to nothing in real terms. What we'd like to do is have an algorithm (which we can backtest) which stops us from selling stocks if they are down.

First we have to define down. Down has to represent a decline of a certain size from a certain baseline. My suggestion is to define the size as a -20% drop, and the baseline as from the S&P500 all-time high. There are many other possible definitions such as -15% compared to twelve months ago but just make it big enough not to trigger the rule every couple of years.

Second, we must decide how to implement the goal of not selling stocks if they are down. It is worth noting here that plans to withdraw only from bonds are functionally the same thing as plans to change one's asset allocation. I.e. "Withdraw only from bonds if the stocks-are-down trigger is activated" has roughly the same outcome as "increase the AA to stocks by 4%/year and decrease the AA to bonds by 4% per year while the stocks-are-down trigger is activated" for a portfolio with a 4% WR. The former phrasing is simpler though.

Third, we need to think realistically about the scenarios in which we would apply our plan. E.g. after the 2000 bear market, the S&P500 wasn't higher than 20% down from its all time highs until 2004. So we would need 3-4 years of cash/bonds to withdraw if we were to avoid selling stocks in that bear market. That would have left us with a 100% stock portfolio in 2004, which was not a bad place to be! After the rule was triggered in 2008, we would need to have sufficient bonds to live off of until the S&P was down less than 20% from ATH's in 2010. So 3-4 years of risk-free assets seems like the minimum requirement.

Next we need to think about whether we hold a stock-heavy portfolio forever after spending down our bonds, or if we eventually return to our original AA. If in 2003 or 2004 we started selling stock to rebuild our bond/cash AA we would still be selling stocks very near the bottom, and the net result would not be that much different than if we just sold stocks in 2001 or 2002 for living expenses. I.e. the S&P500 was at the same level in 2001-2002 as it was in 2004. A glidepath back up to the original bonds/cash AA would be scarcely different than having no algorithm at all, because in 2004 you'd start selling stocks for living expenses plus selling stocks to rebuild your bond/cash allocation. So the only way the strategy works is if our algorithm says to hold a stock-heavy AA for some time after our stocks-are-down trigger has been released.

The danger to someone making this decision in 2004 is obvious in hindsight. If by 2008 they weren't fully loaded with another couple years of living expenses in bonds/cash, the GFC was about to hit their stock-heavy portfolio hard! A rule such as "return to the original XX/XX asset allocation when the S&P500 achieves a new all-time-high" would have only barely gotten them out of experiencing the GFC with an aggressive AA and no bonds/cash to fall back on. Maybe a safer rule would have been "return to the original XX/XX asset allocation when the S&P500 is 5% from an ATH", but still it looks like luck has a lot to do with the algo being triggered at the right time.

Is it luck though? Stocks tend to rebound in the years after large losses like 2001 or 2008. A simple rule like "return to the original AA 3 years after the stocks-are-down algo is no longer triggered" might allow the stock-heavy allocation soar for a few years, but not long enough to put us at risk of the next crisis.

So to summarize, here's an algo that would have worked well over the crisis-prone past quarter-century:

1) Make all withdraws from the bond+cash allocation if the S&P500 is >20% down from its ATH.
     a) If bonds+cash is less than 6 mos living expenses, return to making withdraws evenly across AA categories.
2) Return to original AA three years after the stocks-are-down algo is no longer in effect by selling stocks and buying bonds as necessary. Also resume withdrawing evenly across AA categories. 

Even if you only succeed with part 1, and keep a stock-heavy AA forever thereafter, you have effectively dodged a sequence of returns risk event and are better off than had you rode it out. Because you already dodged the sort of event that might happen once a decade, valuations are better, growth is more likely, and your overall risk profile is lower, justifying the aggressive AA. 2001-2008 was a tricky double-crisis, so part 2 is only in place to protect against rapid back-to-back SORR events like occurred in the 1930s, 1970s, and 2000s. 

[MrGreen]

@mistymoney the bonds are just a volatility damper. Selling equities in a brokerage and then selling bonds to buy those same equities in a tIRA is effectively the same thing as selling bonds, just in a location where we're not actively spending money. This does mean our brokerage balance is volatile, which can be physcholgically challenging to see if you need that money. Though I'd hazard that most very early retirees are utilizing the Roth conversion pipeline as part of their spending strategy before age 60 so once the pipeline is fully funded the withdrawals from brokerage accounts will decrease significantly or cease altogether. Another benefit of this approach is that once the Roth pipeline becomes the main source of spending, this leaves a 100% equities brokerage to grow as aggressively as possible. This inevitably means a large bucket of cash available to you at any time, with only part of it (gains) being taxable. Whereas tIRA and Roth balances are not entirely available without significant penalty.

[RedmondStash]

I've discovered I'm more comfortable with a larger cash cushion, even with the drag. So I've got roughly a 75-20-5 AA. That may change someday.

I don't formally rebalance, but when I need more cash, I check my AA and sell whatever's overweighted. So far, that's always been stocks. I don't use a formal schedule, just keep a close eye on expenses and available cash. I always keep at least a year's cash available, often closer to 2. Most years, I sell chunks of stock 2-3 times a year in taxable, maintaining the AA by rebalancing in retirement accounts as necessary.

In terms of figuring out whether the market is "up" or "down," back in 2016, I created a spreadsheet projecting out a flat 7% increase over the next 50 years. That's been immensely helpful to me in keeping my nerve and having a yardstick to compare my current portfolio against.

It hasn't happened yet, but if my portfolio ends up dropping below the values projected for that 50-year span, and I need cash, I plan to spend down the cash. And then the bonds. Not sure if I'd spend them down to 0, especially the bonds.

I'm not particularly bothered about selling stocks in a "down" market as long as I'm still ahead of that 50-year projection. So far, even if the market is down, the stocks are up quite a bit relative to purchase price.

I really do mostly just try to maintain my AA: in accumulation phase, by what I purchased, and in FIRE, by what I sell for expenses.

[Telecaster]

hey Mr. Green...this has me confused. What is the purpose of bonds in the tIRA if they aren't ever there to spend if the stocks are down? I don't understand the role of a large bond allocation if you are selling stocks to live on for 20 more years.

I don't wish to speak for @Mr. Green but (and clarifications and corrections welcome) what he is doing is rebalancing.  Your total portfolio doesn't know about how it is a rebalanced across various accounts.  It just acts as one unit (disregarding tax implications of course).   @Mr. Green is looking at the future tax implications of his withdrawals and rebalancing, which is pretty high-wire stuff that doesn't get discussed enough. 

Rebalancing almost never improves performance, and indeed typically reduces performance but it does reduce volatility.  Lowering volatility can be a good thing in the withdrawal phase.

Spending cash or bonds when the market is down is market timing.  Market timing almost never provides any positive benefits.   It is something to avoid if you want to maximize your withdrawal rate. 

[MrGreen]

To provide an example, our Vanguard brokerage accounts were recently 100% VTSAX. We had bonds in our tIRAs such that our overall asset allocation was 80/20 stocks/bonds. Let's say the market is down 20% and we don't like selling equities. We sell 5k of VTSAX out of our brokerage to fund our regular expenses. At the same time we convert 5k of VBTLX to VTSAX in our tIRAs. From the perspective of our overall stash, we just sold 5k of bonds.

As to the tax ramifications, if we're using specific ID in our brokerage accounts, which I do, we can choose which batch of VTSAX shares to sell from. We bought them all throughout our accumulation phase so we have everything from highly appreciated shares to ones that are underwater because they were the last ones bought before FIRE. I can choose which shares to sell depending on what I want my income for the year to look like. Maybe I'm trying to maximize my Roth conversions this year so I sell shares closest to break even. This minimizes capital gains and leaves more room for conversions if I'm looking to hit a specific income number. Maybe the market is so depressed that I don't want to maximize Roth conversions because I want a larger tIRA balance to allow for more conversions for longer over the 20+ years (until age 60) I'll be living off those conversions. In that case maybe I sell more highly appreciated shares to help bring down my overall capital gains because it leaves me with more basis if I need a big lump sum in the future.

Odds are extremely likely that I do not need to sell VTSAX shares deeply in the red (in brokerage) unless I've done significant capital gains harvesting previously so I'm not worried about selling VTSAX out of brokerage and making the VBTLX/VTSAX swap in tIRA. In this specific example, there is no rebalance. It is however shifting how much of our stash is in brokerage vs. tIRAs. Plus it's more tax efficient because bonds will throw off more ordinary income. This can limit Roth conversions if you are trying to keep income below a certain number, or just raise your taxes overall by making more of the income ordinary rather than capital gains.

I get really nerdy on this particular topic. It makes me all giddy inside. I know, something is wrong with me.

[secondcor521]

To provide an example, our Vanguard brokerage accounts were recently 100% VTSAX. We had bonds in our tIRAs such that our overall asset allocation was 80/20 stocks/bonds. Let's say the market is down 20% and we don't like selling equities. We sell 5k of VTSAX out of our brokerage to fund our regular expenses. At the same time we convert 5k of VBTLX to VTSAX in our tIRAs. From the perspective of our overall stash, we just sold 5k of bonds.

As to the tax ramifications, if we're using specific ID in our brokerage accounts, which I do, we can choose which batch of VTSAX shares to sell from. We bought them all throughout our accumulation phase so we have everything from highly appreciated shares to ones that are underwater because they were the last ones bought before FIRE. I can choose which shares to sell depending on what I want my income for the year to look like. Maybe I'm trying to maximize my Roth conversions this year so I sell shares closest to break even. This minimizes capital gains and leaves more room for conversions if I'm looking to hit a specific income number. Maybe the market is so depressed that I don't want to maximize Roth conversions because I want a larger tIRA balance to allow for more conversions for longer over the 20+ years (until age 60) I'll be living off those conversions. In that case maybe I sell more highly appreciated shares to help bring down my overall capital gains because it leaves me with more basis if I need a big lump sum in the future.

Odds are extremely likely that I do not need to sell VTSAX shares deeply in the red (in brokerage) unless I've done significant capital gains harvesting previously so I'm not worried about selling VTSAX out of brokerage and making the VBTLX/VTSAX swap in tIRA. In this specific example, there is no rebalance. It is however shifting how much of our stash is in brokerage vs. tIRAs. Plus it's more tax efficient because bonds will throw off more ordinary income. This can limit Roth conversions if you are trying to keep income below a certain number, or just raise your taxes overall by making more of the income ordinary rather than capital gains.

I get really nerdy on this particular topic. It makes me all giddy inside. I know, something is wrong with me.

You're probably already aware, but be careful about selling VTSAX at a loss in the brokerage and then rebalancing into VTSAX in the tIRA within 30 days.  That would create a wash sale and the loss in the taxable would be disallowed.  Since the basis gets added to the replacement shares in the tIRA, and tIRAs don't have tax basis, the loss is essentially gone forever.

[Much Fishing to Do]

Our strategy (as of Wednesday this week, we are FIRE!) is to have 5 years of living expenses in T-bills, inspired by https://www.kitces.com/blog/accelerating-the-rising-equity-glidepath-with-treasury-bills-as-portfolio-ballast/. The T-bills will get spent down, implementing a rising equity glidepath, and five years from now we'll hopefully have mitigated most of the immeidate post fire SORR and can just spend from our investments and not worry about it.

I'm similar to this.  Plus given about 40% of our spending will be funded by dividend and ltcg distributions from our taxable that I no longer reinvest, this 5 years of cash should really last more like 8.  I like to not have to worry about any selling of investments over this period (beyond the automatic distributions) and also considered this a way to do a sort of rising equity glidepath before I had ever heard of that term.  I like I'll not have to be always figuring gains affects from selling at this early point in retirement (esp its affect on ACA subsidies and taxes) After that 8 years is up I'll have all kids off the dole, be a lot closer to medicare, SS, and an at age where I'm only really planning for a 20 year retirement anyway (I've never seen family member make it to 80, so though I realize its entirely possible I dont think it worth it to hold back on anything for my 82 yo self....).

[mistymoney]

To provide an example, our Vanguard brokerage accounts were recently 100% VTSAX. We had bonds in our tIRAs such that our overall asset allocation was 80/20 stocks/bonds. Let's say the market is down 20% and we don't like selling equities. We sell 5k of VTSAX out of our brokerage to fund our regular expenses. At the same time we convert 5k of VBTLX to VTSAX in our tIRAs. From the perspective of our overall stash, we just sold 5k of bonds.

As to the tax ramifications, if we're using specific ID in our brokerage accounts, which I do, we can choose which batch of VTSAX shares to sell from. We bought them all throughout our accumulation phase so we have everything from highly appreciated shares to ones that are underwater because they were the last ones bought before FIRE. I can choose which shares to sell depending on what I want my income for the year to look like. Maybe I'm trying to maximize my Roth conversions this year so I sell shares closest to break even. This minimizes capital gains and leaves more room for conversions if I'm looking to hit a specific income number. Maybe the market is so depressed that I don't want to maximize Roth conversions because I want a larger tIRA balance to allow for more conversions for longer over the 20+ years (until age 60) I'll be living off those conversions. In that case maybe I sell more highly appreciated shares to help bring down my overall capital gains because it leaves me with more basis if I need a big lump sum in the future.

Odds are extremely likely that I do not need to sell VTSAX shares deeply in the red (in brokerage) unless I've done significant capital gains harvesting previously so I'm not worried about selling VTSAX out of brokerage and making the VBTLX/VTSAX swap in tIRA. In this specific example, there is no rebalance. It is however shifting how much of our stash is in brokerage vs. tIRAs. Plus it's more tax efficient because bonds will throw off more ordinary income. This can limit Roth conversions if you are trying to keep income below a certain number, or just raise your taxes overall by making more of the income ordinary rather than capital gains.

I get really nerdy on this particular topic. It makes me all giddy inside. I know, something is wrong with me.

thanks so much, this makes sense to me :)

strategies I would never have thought of.

[MrGreen]

@mistymoney The idea of keeping bonds in tax deferred accounts came from JL Collins. He mentions it in his Stock Series post about asset allocation.

@secondcor521 I am familiar with the wash sale rule across account types. It really makes you pay attention if you're planning on taking losses and reinvesting dividends anywhere. Luckily we've only only time we've ever sold for a loss we've used it to balance out gains so we could make a large withdrawal without generating any taxable income. At this point I don't anticipate ever needing to sell shares at a loss for regular living expenses, even if the market were to fall 40%.

[Blissful Biker]

Posting to follow.  I intend to have a cash buffer worth two years of expenses when I FIRE but haven't get defined what would trigger using it.  Appreciate the input.

[mistymoney]

Posting to follow.  I intend to have a cash buffer worth two years of expenses when I FIRE but haven't get defined what would trigger using it.  Appreciate the input.

yes - this seems tricky to me. On one hand, spending as soon as the market begins to really weaken, 2 years would cover most downturns. On the other hand - when 2 years doesn't cover the downturn, one would have been best off not spending the cash piece the first year.

[secondcor521]

Posting to follow.  I intend to have a cash buffer worth two years of expenses when I FIRE but haven't get defined what would trigger using it.  Appreciate the input.

yes - this seems tricky to me. On one hand, spending as soon as the market begins to really weaken, 2 years would cover most downturns. On the other hand - when 2 years doesn't cover the downturn, one would have been best off not spending the cash piece the first year.

Yup.

The majority of the time, a cash cushion is a drag on the portfolio.  When it's not a drag, it's confusing to implement in practice:  when to use?  when to refill?

Some people choose to have a cash cushion right at FIRE-time due to SORR, which can make some sense.  As an alternative, I just worked until I hit 100% historically safe and therefore did not have to (in theory) worry about SORR - you can safely play Russian roulette 1000 times if there are no bullets in the gun.

See my and @Telecaster's posts on this thread.

[Ladychips]

As an alternative, I just worked until I hit 100% historically safe and therefore did not have to (in theory) worry about SORR ...

@secondcor521, what rate is that? Inquiring minds want to know...

[Nords]

Posting to follow.  I intend to have a cash buffer worth two years of expenses when I FIRE but haven't get defined what would trigger using it.  Appreciate the input.

yes - this seems tricky to me. On one hand, spending as soon as the market begins to really weaken, 2 years would cover most downturns. On the other hand - when 2 years doesn't cover the downturn, one would have been best off not spending the cash piece the first year.

I get these questions all the time, and here's some discussion points.

First, it's only two years' expenses in cash.  With the 4% Safe Withdrawal Rate, that's only 8% of the overall asset allocation, and it's not likely to make a significant difference in the portfolio's total return.  The absolute "worst" case (100% equities) would be only getting 92% of the stock market's upside.  If you have a high-equity asset allocation then it still might be 80% equities, 12% bonds, and 8% cash.

Second, it's only for the first decade of FI.  By the end of the decade, even with keeping 8% in cash, investments should still grow faster than inflation and you'll be immune to sequence of returns risk. 
As an example, someone starting their FI with $1M would spend $40K during the first year.  At the end of the decade, spending would have grown with inflation but would still be $40K in today's dollars. 
During that decade, a high-equity portfolio (at least 60% stocks) would grow faster than inflation.  If the withdrawal rate at year #11 dropped from 4% over that decade to begin year #11 at 3.5% (which EarlyRetirementNow would agree is bullet-proof from sequence of returns risk), then the new investment value at year #11 to support that (in today's dollars) would have to be $40K / 3.5% = $1.143M. 
To reach that value in a decade, it'd have to grow at an annual compound rate of only 1.4%/year faster than inflation.

Third, nobody knows how long the market downturn will last.  There's no way to predict it, but no prediction is necessary.
The only thing you know is that you'll be able to let your cash stash get you through two years without selling equity shares.  Even if you start selling equity shares at the beginning of year #3 of the downturn, you've already given your equities two years of breathing room by skipping sales of those shares.

Finally, after two years of a bear market or a recession, you still have the options of:
- cutting spending or
- seeking part-time work or
- if you're eligible, starting Social Security at age 62. 

Our first decade of retirement was 2002-2012, which meant that we got to test this two-year cash stash through two significant recessions. 

Ironically, we didn't cut our spending during those recessions because corporations couldn't raise prices.  Better yet, we were able to score bargains in our lifestyle areas like home improvement (contractors) and travel (discounted airfares & lodging).  We spent the money we would normally spend and actually received more value from it.

A financially-independent person seeking part-time employment during a recession doesn't need much money.  Spending $40K/year means that even $5000 is more than 10% of your annual expenses.  During a recession, employers don't want to hire people full-time because they can't afford them.  However they'd love to hire part-time employees because they don't need to give them benefits like 401(k)s or health insurance.  This means there are plenty of part-time jobs during recessions, and it doesn't take much part-time work to earn the money to give your investments time to recover.

If you're financially independent, maybe you should actually be hoping for a nasty recession during that first decade.  You'll probably get more value for your spending, you'll have more opportunities to work (if you want), and you won't be at a full-time job where you're worrying about layoffs or picking up the bigger workload after others are laid off.

Based on the last century or so, a nasty recession will probably happen during that first decade anyway.

If a 4% SWR portfolio fails from sequence of returns risk, it still lasts for at least 20 years (most of the failures happen after year 24).  This means that someone reaching FI before their early 30s (30 years from Social Security) would have to be concerned about a recession so prolonged that their investments couldn't survive even after using the cash stash, then (2-3 years later) cutting expenses and seeking part-time employment, and then not making it to Social Security at age 62.

If we encounter a recession that bad, I'd hate to be stuck in the workplace environment of a full-time job.  Why, for that sort of meteor-strike extinction-level catastrophe, it'd take something like... oh, I don't know... maybe a global pandemic.

On another popular FI forum, we'd refer to this level of concern as the analogy of keeping your pants around your waist by relying on a belt, a set of suspenders, and a nailgun.

To summarize:
Once you reach the tripwire of FI (assets of 25x annual expenses), then how much longer would you feel obligated to trade your life energy (which you might not have) for more money (which you almost certainly will not need)?

[secondcor521]

As an alternative, I just worked until I hit 100% historically safe and therefore did not have to (in theory) worry about SORR ...

@secondcor521, what rate is that? Inquiring minds want to know...

There is a large body of work and discussion on this topic.

At the time (2015), I personally used 4.08% as 100% historically safe based on indexing to the CPI-U per the chart at https://retireearlyhomepage.com/restud1.html.  I actually worked a bit longer and at the time semi-accidentally ignored the fact that I had some other income sources besides my portfolio, so I think I started FIRE in spring 2016 at somewhere around a 2% effective WR%.

It does depend on what assumptions you make, especially what AA you use, what approach you take (historical or Monte Carlo), what time frame you want to plan for, whether you think the future will be better or worse than the past, and probably lots of other things which get discussed here and at other FIRE boards extensively.

I personally assumed an AA of between 90/10 and 100/0, a historical approach based on US data, a 40 year retirement horizon, and that the future would be as good as or better than the past.  But everyone gets to decide for themselves.

Oh, and there is a bit of the luck of the draw.  When I retired on 2/19/2016, the S&P500 closed at an adjusted 1917.78.  Today it closed ~2.5X that.  So today my net WR% is about 1.38%.

[MrGreen]

As an alternative, I just worked until I hit 100% historically safe and therefore did not have to (in theory) worry about SORR ...

@secondcor521, what rate is that? Inquiring minds want to know...

There is a large body of work and discussion on this topic.

At the time (2015), I personally used 4.08% as 100% historically safe based on indexing to the CPI-U per the chart at https://retireearlyhomepage.com/restud1.html.  I actually worked a bit longer and at the time semi-accidentally ignored the fact that I had some other income sources besides my portfolio, so I think I started FIRE in spring 2016 at somewhere around a 2% effective WR%.

It does depend on what assumptions you make, especially what AA you use, what approach you take (historical or Monte Carlo), what time frame you want to plan for, whether you think the future will be better or worse than the past, and probably lots of other things which get discussed here and at other FIRE boards extensively.

I personally assumed an AA of between 90/10 and 100/0, a historical approach based on US data, a 40 year retirement horizon, and that the future would be as good as or better than the past.  But everyone gets to decide for themselves.

Oh, and there is a bit of the luck of the draw.  When I retired on 2/19/2016, the S&P500 closed at an adjusted 1917.78.  Today it closed ~2.5X that.  So today my net WR% is about 1.38%.

I didn't realize our FIRE dates were so similar. We've had the same luck with respect to how big out stash has gotten because of the market returns in the years immediately following FIRE and some real estate sales that were more lucrative and occurred sooner than I would have thought possible. Our WR is also below 2% now. However, I also keep in the back of my mind that the huge returns we had early on could easily be followed by a period of worse than average returns that will revert our overall performance toward the mean. It might be easy to think we're 6-7 years in and under 2% so let's increase spending. That is a trap! Maybe after a few more years we can safely know we've outrun a period of poor performance but I'm not quite there yet. And that's assuming there's anything we'd even care to increase spending on.

[secondcor521]

As an alternative, I just worked until I hit 100% historically safe and therefore did not have to (in theory) worry about SORR ...

@secondcor521, what rate is that? Inquiring minds want to know...

There is a large body of work and discussion on this topic.

At the time (2015), I personally used 4.08% as 100% historically safe based on indexing to the CPI-U per the chart at https://retireearlyhomepage.com/restud1.html.  I actually worked a bit longer and at the time semi-accidentally ignored the fact that I had some other income sources besides my portfolio, so I think I started FIRE in spring 2016 at somewhere around a 2% effective WR%.

It does depend on what assumptions you make, especially what AA you use, what approach you take (historical or Monte Carlo), what time frame you want to plan for, whether you think the future will be better or worse than the past, and probably lots of other things which get discussed here and at other FIRE boards extensively.

I personally assumed an AA of between 90/10 and 100/0, a historical approach based on US data, a 40 year retirement horizon, and that the future would be as good as or better than the past.  But everyone gets to decide for themselves.

Oh, and there is a bit of the luck of the draw.  When I retired on 2/19/2016, the S&P500 closed at an adjusted 1917.78.  Today it closed ~2.5X that.  So today my net WR% is about 1.38%.

I didn't realize our FIRE dates were so similar. We've had the same luck with respect to how big out stash has gotten because of the market returns in the years immediately following FIRE and some real estate sales that were more lucrative and occurred sooner than I would have thought possible. Our WR is also below 2% now. However, I also keep in the back of my mind that the huge returns we had early on could easily be followed by a period of worse than average returns that will revert our overall performance toward the mean. It might be easy to think we're 6-7 years in and under 2% so let's increase spending. That is a trap! Maybe after a few more years we can safely know we've outrun a period of poor performance but I'm not quite there yet. And that's assuming there's anything we'd even care to increase spending on.

If your age in your profile is accurate, you're about 15 years younger than I am.

We may have a period of underperformance in the future, but I see no reason to expect one worse than what we've seen in the past.  If that premise is correct, then 4% is conservative and 2% is extremely so.  If we even just muddle along, then 5% might even be OK.

It does seem weird and uncomfortable to think that we should/could be spending 2.5X what we're currently comfortably living on, though.

Being 15 years older, having lost one parent and in the process of losing the other, plus being in sight of 59.5 and the option of SS at 62, I certainly feel like I can loosen the purse strings.  But the only thing I really feel worthwhile - wasting money doesn't work for me - is experiences with my kids.  I took them all on vacation last week and have plans for more.

[MrGreen]

@secondcor521 my profile age is accurate. I turn 40 in about two months. I suppose some SS adjustment could happen that will affect us but even now it feels reassuring to know that we only have 30 years until the majority of our spending will be covered by what is essentially a pension. One area I've really tried to relax on is Airbnb spending. For the kind of travel we want to do, having a second bedroom and a place that isn't a basement might mean another 1-2k but that also means we have room for my mom to come visit her only grandchild, etc. I lost my dad in 2019 and my last grandparent in 2021 so I feel more aware of how short life can be. There isn't really much else we feel inclined to spend more money on and I, too, can't bring myself to spend it in ways that feel wasteful.

[joemandadman189]

the Mad Fientist has a post along these lines, not exactly but a good read

https://www.madfientist.com/discretionary-withdrawal-strategy/

[reeshau]

I follow McClung's (Alternate) Prime Harvesting strategy, which keeps as much invested in equities as possible. It's similar to what dividendman is doing but stock is only sold when the market is on a tear.

1) Sell enough bonds to generate (1 year of) cash.
   a) If there aren't any bonds to sell, sell equities.
2) When the market is up a lot (+20% is a simple method), sell equities to buy more bonds.

Ideally, stocks generate outsized returns and are trimmed to refill the bond bucket. Historically, in most cases, the bond allocation grew to a very large percentage of the asset allocation.

Great thread, and very interesting to see all the approaches.

I wanted to add this: I invest primarily in individual stocks.  It turns out that a natural outcome of that looks a lot like bacchi's approach above.  I sell stocks as they are overvalued, in my estimation--often, though not always, as the overall market is overvalued.  And I look to buy undervalued companies, which occur more often when the market is undervalued.  So, I regularly generate cash, as I sell individual equities.  For the most part, this cash is dry powder, to buy something else with.  But I do transfer some out to top up my cash cushion, which I try to keep at roughly a year.  (I start together uncomfortable at 6 months', so it is as much a qualitative thing as a calculated risk)

Since I am withdrawing from my taxable account, this general approach is tempered by tax / ACA planning.  The 3/4 of my assets in retirement accounts aren't impacted, of course.

[Ladychips]

As an alternative, I just worked until I hit 100% historically safe and therefore did not have to (in theory) worry about SORR ...

@secondcor521, what rate is that? Inquiring minds want to know...

There is a large body of work and discussion on this topic.

At the time (2015), I personally used 4.08% as 100% historically safe based on indexing to the CPI-U per the chart at https://retireearlyhomepage.com/restud1.html.  I actually worked a bit longer and at the time semi-accidentally ignored the fact that I had some other income sources besides my portfolio, so I think I started FIRE in spring 2016 at somewhere around a 2% effective WR%.

It does depend on what assumptions you make, especially what AA you use, what approach you take (historical or Monte Carlo), what time frame you want to plan for, whether you think the future will be better or worse than the past, and probably lots of other things which get discussed here and at other FIRE boards extensively.

I personally assumed an AA of between 90/10 and 100/0, a historical approach based on US data, a 40 year retirement horizon, and that the future would be as good as or better than the past.  But everyone gets to decide for themselves.

Oh, and there is a bit of the luck of the draw.  When I retired on 2/19/2016, the S&P500 closed at an adjusted 1917.78.  Today it closed ~2.5X that.  So today my net WR% is about 1.38%.

I've read alot of that body of work but always like to hear your thoughts. These particular thoughts tickled me because they sound like the typical forum chorus. It goes like this...

"4% is bullet proof...but I worked a little longer...then I forgot to add some other money/income that I had...and then i made a little money when i didn't mean to...then I spent less than I planned...and I didn't count social security...and then I had a little luck.  I'm telling you, 4% is bullet proof!"

Hahaha, I hope you take this as tongue in cheek because I really do enjoy your posts!!

[secondcor521]

As an alternative, I just worked until I hit 100% historically safe and therefore did not have to (in theory) worry about SORR ...

@secondcor521, what rate is that? Inquiring minds want to know...

There is a large body of work and discussion on this topic.

At the time (2015), I personally used 4.08% as 100% historically safe based on indexing to the CPI-U per the chart at https://retireearlyhomepage.com/restud1.html.  I actually worked a bit longer and at the time semi-accidentally ignored the fact that I had some other income sources besides my portfolio, so I think I started FIRE in spring 2016 at somewhere around a 2% effective WR%.

It does depend on what assumptions you make, especially what AA you use, what approach you take (historical or Monte Carlo), what time frame you want to plan for, whether you think the future will be better or worse than the past, and probably lots of other things which get discussed here and at other FIRE boards extensively.

I personally assumed an AA of between 90/10 and 100/0, a historical approach based on US data, a 40 year retirement horizon, and that the future would be as good as or better than the past.  But everyone gets to decide for themselves.

Oh, and there is a bit of the luck of the draw.  When I retired on 2/19/2016, the S&P500 closed at an adjusted 1917.78.  Today it closed ~2.5X that.  So today my net WR% is about 1.38%.

I've read alot of that body of work but always like to hear your thoughts. These particular thoughts tickled me because they sound like the typical forum chorus. It goes like this...

"4% is bullet proof...but I worked a little longer...then I forgot to add some other money/income that I had...and then i made a little money when i didn't mean to...then I spent less than I planned...and I didn't count social security...and then I had a little luck.  I'm telling you, 4% is bullet proof!"

Hahaha, I hope you take this as tongue in cheek because I really do enjoy your posts!!

Aw, thanks for the kind words!

I still think 4% is bulletproof.  Obviously I've had a more comfortable ride than someone who retired in January 2022.  Although I do remember saying to myself when I did retire in 2016 that I should be able to handle a market drop if one should happen, because I should trust the math.  Fortunately it wasn't needed.

I worked longer because I still had work goals, not really to pad the stash.  Work turned bad, I accomplished my goals, and my Mom got sick all about the same time.  There's a lot of angst about when to pull the plug; I think most people will have a pretty good sense of the timing.

Forgetting about other income and then making money on a side gig (piggybacking) are worth talking about.  It was very easy for me to do what @Nords said above about suspenders, belt, and nail gun, because I wanted to be very sure that I would be OK.  But with all the catastrophic planning we do, we shouldn't be surprised when things turn out better than we planned.  Because by definition, they typically do.

I did actually count SS, partly because earlier on in my FIRE journey I was impatient to get to FI and so I counted everything including the spare change under the sofa cushions (not quite, but almost literally true).  Once my spreadsheet was set up that way, it was easy to continue to include it.  But it's not a very big percentage of the picture for me.

@Mr. Green, the paragraph above about catastrophic planning might apply to you.  I know you like to optimize.  I think splurging a bit with where you are - especially on stuff you value - is something you'll be glad you did.  It's a version of optimizing the value of spent money when it will matter for your kid and not giving them $3M when they're 64.

[MrGreen]

@secondcor521 I'm in total agreement. It's one of the reasons I've spent so much effort considering the utility of money, as I like to call it. We could have a very small income and still generate what we need for our base expenses and that would lead to a huge pile of cash as we approach traditional retirement age. But we can go above and beyond that while still paying a very reasonable amount of taxes and even if we don't spend that income every year, it makes it available for spending. Maybe a life event happens in the future and we have a blowout spending year when my mom's mobility is dwindling. You just never know what life is going to throw at you. If all our money is left locked up in tIRAs, we can't make the best use of it. So I'm aggressive with the Roth conversions now, while still being mindful of taxes, knowing that 60 year old me is going to thank 39 year old me for having done so, just like present me is eternally thankful I took the steps that I did years ago.

[Zamboni]

Thanks for starting this, Villanelle. Interesting to see the different strategies, and to be reminded of NORDS blog post.

[Car Jack]

I'll mention that I'm likely the most conservative (financially, I voted for Ted Kennedy back in the day) guy on here and even on Bogleheads.  I'll also note that I retired 2 1/2 weeks ago and I'm well over 59 1/2. 

Approaching retirement, I started hoarding cash.  Less investing in taxable but still maxed my 401k for this year and DW and I both maxed our Roths.  The cash is in high yield savings and CDs.  Doing the calculation, we have 4 1/4 years of retirement spending in cash.  On top of that, we have a bit over 6 years of retirement spending in US Savings bonds (all paper).  I know from experience that I can cash these at my credit union and it becomes available to withdraw immediately, unlike electronic where you need to wait some number of weeks.

So the upside of all this is that the market could truly tank and I'm just fine.  Yes, I'm giving up some investment return but I retired with over 50 times retirement spending in total (including these cash things).  This now allows me to Roth convert without worrying that I need to pull more out to live off of.  I'm on Medicare so my RMD reducing conversions need to be reigned in or I'd be penalized by IRMMA and pay big extras for my health insurance. 

I know I'm not typical.  I'm not retired early, I sort of retired on time as it was almost exactly when I hit FRA in social security.  I'm of course not taking that yet in order to keep my income as low as possible so I can Roth convert as much as possible.

If you are willing to go back to work then of course you can hold a much smaller amount of cash.  I have zero thoughts of ever working again and don't need to.  But if you're barely holding enough to make it 30 years (25 times retirement spending), then sure, get a part time job or a gig job.  Just be careful with anything you drive your own car for and ask your insurance agent.  In my state, it's wicked expensive.

[Villanelle]

I'll mention that I'm likely the most conservative (financially, I voted for Ted Kennedy back in the day) guy on here and even on Bogleheads.  I'll also note that I retired 2 1/2 weeks ago and I'm well over 59 1/2. 

Approaching retirement, I started hoarding cash.  Less investing in taxable but still maxed my 401k for this year and DW and I both maxed our Roths.  The cash is in high yield savings and CDs.  Doing the calculation, we have 4 1/4 years of retirement spending in cash.  On top of that, we have a bit over 6 years of retirement spending in US Savings bonds (all paper).  I know from experience that I can cash these at my credit union and it becomes available to withdraw immediately, unlike electronic where you need to wait some number of weeks.

So the upside of all this is that the market could truly tank and I'm just fine.  Yes, I'm giving up some investment return but I retired with over 50 times retirement spending in total (including these cash things).  This now allows me to Roth convert without worrying that I need to pull more out to live off of.  I'm on Medicare so my RMD reducing conversions need to be reigned in or I'd be penalized by IRMMA and pay big extras for my health insurance. 

I know I'm not typical.  I'm not retired early, I sort of retired on time as it was almost exactly when I hit FRA in social security.  I'm of course not taking that yet in order to keep my income as low as possible so I can Roth convert as much as possible.

If you are willing to go back to work then of course you can hold a much smaller amount of cash.  I have zero thoughts of ever working again and don't need to.  But if you're barely holding enough to make it 30 years (25 times retirement spending), then sure, get a part time job or a gig job.  Just be careful with anything you drive your own car for and ask your insurance agent.  In my state, it's wicked expensive.

Do you have a specific plan for when you'll spend the cash (savings bonds, etc.), and when you will replenish them and at what pace?

What triggered this thread is I so often see people saying they will have 2 years of expenses in cash (or similar) but with no further details, and I wonder what that actually looks like for them, because that info is almost never mentioned.   In your case, you have 10 years in "cash".  But does that means you don't plan ti withdraw any money for 9-10 years?  Or will you withdraw if the market is [some metric] and otherwise use cash?  And will you just go on withdrawing for current expenses once the cash is gone, or will you replenish it somehow?  Or do you plan on winging all this?

[Nords]

Interesting to see the different strategies, and to be reminded of NORDS blog post.

Thanks, @Zamboni, these days I spend most of my time at an ESIMoney .com forum, but I check in here every week or so.

After 21 years of FI, we're still spending at the 4% SWR but we've substantially ramped up our gifting & philanthropy.  We're planning to spend down all of our taxable account and most of our retirement accounts during the rest of the decade, and in 2030 my spouse and I start Social Security at age 70.

[Catbert]

I'll mention that I'm likely the most conservative (financially, I voted for Ted Kennedy back in the day) guy on here and even on Bogleheads.  I'll also note that I retired 2 1/2 weeks ago and I'm well over 59 1/2. 

Approaching retirement, I started hoarding cash.  Less investing in taxable but still maxed my 401k for this year and DW and I both maxed our Roths.  The cash is in high yield savings and CDs.  Doing the calculation, we have 4 1/4 years of retirement spending in cash.  On top of that, we have a bit over 6 years of retirement spending in US Savings bonds (all paper).  I know from experience that I can cash these at my credit union and it becomes available to withdraw immediately, unlike electronic where you need to wait some number of weeks.

So the upside of all this is that the market could truly tank and I'm just fine.  Yes, I'm giving up some investment return but I retired with over 50 times retirement spending in total (including these cash things).  This now allows me to Roth convert without worrying that I need to pull more out to live off of.  I'm on Medicare so my RMD reducing conversions need to be reigned in or I'd be penalized by IRMMA and pay big extras for my health insurance. 

I know I'm not typical.  I'm not retired early, I sort of retired on time as it was almost exactly when I hit FRA in social security.  I'm of course not taking that yet in order to keep my income as low as possible so I can Roth convert as much as possible.

If you are willing to go back to work then of course you can hold a much smaller amount of cash.  I have zero thoughts of ever working again and don't need to.  But if you're barely holding enough to make it 30 years (25 times retirement spending), then sure, get a part time job or a gig job.  Just be careful with anything you drive your own car for and ask your insurance agent.  In my state, it's wicked expensive.

Do you have a specific plan for when you'll spend the cash (savings bonds, etc.), and when you will replenish them and at what pace?

What triggered this thread is I so often see people saying they will have 2 years of expenses in cash (or similar) but with no further details, and I wonder what that actually looks like for them, because that info is almost never mentioned.   In your case, you have 10 years in "cash".  But does that means you don't plan ti withdraw any money for 9-10 years?  Or will you withdraw if the market is [some metric] and otherwise use cash?  And will you just go on withdrawing for current expenses once the cash is gone, or will you replenish it somehow?  Or do you plan on winging all this?

I'm not Jack (obviously), however, if it were me I'd set mutual funds in my brokerage account to pay out capital gains and dividends rather than automatically reinvest.  That should help re-fill the cash coffers.  Within 10 years there'll be SS and RMDs creating a firehose of money.

[2sk22]

Do you have a specific plan for when you'll spend the cash (savings bonds, etc.), and when you will replenish them and at what pace?

What triggered this thread is I so often see people saying they will have 2 years of expenses in cash (or similar) but with no further details, and I wonder what that actually looks like for them, because that info is almost never mentioned.   In your case, you have 10 years in "cash".  But does that means you don't plan ti withdraw any money for 9-10 years?  Or will you withdraw if the market is [some metric] and otherwise use cash?  And will you just go on withdrawing for current expenses once the cash is gone, or will you replenish it somehow?  Or do you plan on winging all this?

I am currently keeping extra cash for several reasons:
- have to pay another two years of college for daughter
- to fund Roth conversions once my wife retires (hopefully in a year or two). This money is held in the form of a 3 year MYGA

Once my wife retires, I will draw down cash until we have two years of expenses.

[Omy]

We fatFIREd 4 years ago and have always been suboptimal when it comes to holding cash. Mostly because of my fear of selling stock in a down market.

We have 15% of our net worth in EE and I bonds, CDs, MYGAs, and high yield savings accounts. Our challenge is that "cash" is increasing instead of being spent down (rents and dividends exceed our spend rate) and we already have a lot (60%) in the stock market so we don't necessarily want to plow more into that.  The remaining 25% of our net worth is in real estate, and we don't really want more in that, either. In fact, we just sold one of our rentals resulting in a lot more cash to put back to work.

My plan to hold a lot of cash for down markets has sort of backfired as we are now swimming in the stuff. It has me considering putting a chunk into an annuity that will provide a "pension' in a few years to replace the rental income.

Our approach has been very "seat of the pants" and has resulted in too much cash on hand, so I really appreciate seeing how others approach this issue.