What's your withdrawal strategy?

[conwy]

For the more pedantic among us (I consider myself that way) I think it can be calming and helpful to have some kind of plan around how to take money out in early retirement.

After a lot of varied research, investigations and thinking, I think I'm settling on a withdrawal strategy that works for my lifestyle and expenses.

I'm calling it the "waterfall" for lack of a better term.

The basic structure is like this:

1. At the top of the waterfall: diversified stocks portfolio. The most churn and volatility in the short-term, but the best bet on long-term real returns (or at least, capital preservation). This one has most of my money – around 70%, though it will fluctuate in the short-term.
2. In the middle of the waterfall: inflation-linked bonds. These are in a ladder, mostly concentrated in the next 5-10 years, but with a little at the long end, to take advantage of convexity. These bonds, including coupon, are enough to cover my living expenses adjusted for CPI. These are around 25% of the portfolio.
3. At the bottom of the waterfall: high-interest savings account. This is enough to cover me until the next inflation-linked bond in the ladder matures, minus the inflation-linked bond coupon income, which I already get paid quarterly from my existing bond holdings, so no need to hold it in savings. Currently around 2 years of living expenses, with a little padding for emergencies. These are around 5% of the portfolio.

How I plan to manage it is something like this:

Every quarter, or after a stretch of above-average stock performance, I'll sell a small percentage of the stocks, enough to cover one quarter of living expenses.

• If stocks are up since the last sale, I'll draw down ~3-4% of the cost basis, plus personal realised inflation (as measured by my budget, which I will do monthly, based on examining my bank statements).
• If stocks are down since the last sale, I'll draw down ~2-3% of the cost basis, plus personal realised inflation (measured similar to the above, but perhaps factoring in some budget cuts in response to the lower returns)

By "cost basis", I mean the original nominal amount of the stocks, averaged over a period of time, to smooth out volatility. For example, if at the time of early retirement, my stocks were worth $1M on average over the prior 6 month period, then I will take $1M to be the nominal cost basis.

By "personal realised inflation", I mean the lower of CPI or personal spending, compounded over the time since early retirement. For example, if I retired in 2023, and inflation was 3% in 2023 and 4% in 2025, then the inflation adjustment will be 1.03 x 1.04 = 1.0712 = 7.12%. So in that case, I will withdraw 7.12% of 1 million or $71,200. Well actually, a quarter of that, so really 17,800. The goal is to keep my living expenses at or below CPI. Of course this might be tricky in the short term, but the idea is to try and keep it this way as much as possible over the long-term.

With the one quarter of living expenses withdrawn, I will invest it immediately into the next step on the inflation-linked bond ladder, such that it covers that future quarter's living expenses after CPI.

For example, suppose the year is 2026 and we're in January. If my next bond is maturing in 2036, and has enough to fund 2 quarters of living expenses (6 months) then I will top it up to 3 quarters (9 months) by buying additional units of that bond.

In this way, there will always be a buffer of 5-10 years between my stock valuation and my living expenses. This buys me time to shift my income, location, lifestyle, etc. in any way necessary to deal with any long-term fluctuations.

-----

Note: I'm motivated to write this mainly to get my thoughts out clearly so I can analyse them, but also to get feedback from anyone else's who's interested/curious. This is not intended to sway anyone else's withdrawal strategy. There are pros and cons to all strategies. This is just what I feel will work best for me and my own needs and personality.

-----

What's your withdrawal strategy?
Are there elements in this plan you agree or disagree with?
Do you track your withdrawals or do you prefer to take a more relaxed approach and leave it to the market?

[IlliniDave]

I'm not withdrawing yet but I set myself up to be able to do a liability matching portfolio. Basically something like 1-2 years expenses in cash, ~10 years in bonds, the rest in stocks.* Where I can do it, all dividends/coupon payments go into cash.** Spend the cash first, sell bonds to replenish cash above the the dividends supply to to keep the cash level sufficient, probably yearly, let the bonds dwindle, let the stocks grow. Over time it may or may not be desirable to sell stocks to replenish bonds. Other people will look at the total they anticipate spending over their entire retirement and put that much in bonds/cash, and keep the rest in stocks for rainy day/unanticipated expenses/the occasional big splurge/legacy, with the plan of spending the bonds and cash down to zero on the last day of their life if they guessed their future expenses correctly. That probably works better for a more conventional retirement. My situation is atypical in that I have a pretty good base income level from a retirement annuity that started when I took ER, and will later be augmented by Social Security (maybe). So the 15 years of expenses in bonds is probably overkill.

* For other reasons I'm somewhat more conservatively invested right now.
** Much of my stash in in tax-deferred savings plans for which the only option is reinvesting of dividends into the fund that pays them. I only hold mutual funds.

Overall, that sounds reasonably close to what you are considering in terms of strategy: cash to spend now, bonds to stabilize resources for the intermediate term, stocks to maximize growth potential to keep the longer-term funded. It's not a bad idea to try to bias stock sales to be greater during periods of stronger growth, but I probably won't bother. I'll be forced to withdraw from the retirement accounts every year eventually, which forces me to sell the stocks (though part of the proceeds will likely be reinvested in stock funds in a taxable account). I'm already 60 so that juncture is not far off.

[conwy]

Additional thoughts that just came to me ...

Certain items can be purchased and somewhat "locked in" in advance, from several weeks up to months.

High calorie, non-perishable, sealed food items such as canned beans, oatmeal, peanut butter, etc can be stored from weeks up to months, providing a massive hedge with low risk of theft or other loss. I already do this, keeping 2-3 weeks' consumption of canned beans, oats, peanut butter and tinned fish in storage and ordering a new batch every 1-2 weeks.

Frozen foods - vegetables, berries, etc - can be stored but that needs to be balanced with the energy cost of a freezer.

Rent can also be locked in, via a fixed-term lease or AirBnB for shorter term leases.

Clothing is even easier - just buy and hold high quality items like quality boots, jeans, etc. and they will tend to last at least 5 years if not longer, with proper maintenance, hedging clothing cost inflation.

[conwy]

I'm not withdrawing yet but I set myself up to be able to do a liability matching portfolio. Basically something like 1-2 years expenses in cash, ~10 years in bonds, the rest in stocks.* Where I can do it, all dividends/coupon payments go into cash.** Spend the cash first, sell bonds to replenish cash above the the dividends supply to to keep the cash level sufficient, probably yearly, let the bonds dwindle, let the stocks grow. Over time it may or may not be desirable to sell stocks to replenish bonds. Other people will look at the total they anticipate spending over their entire retirement and put that much in bonds/cash, and keep the rest in stocks for rainy day/unanticipated expenses/the occasional big splurge/legacy, with the plan of spending the bonds and cash down to zero on the last day of their life if they guessed their future expenses correctly. That probably works better for a more conventional retirement. My situation is atypical in that I have a pretty good base income level from a retirement annuity that started when I took ER, and will later be augmented by Social Security (maybe). So the 15 years of expenses in bonds is probably overkill.

* For other reasons I'm somewhat more conservatively invested right now.
** Much of my stash in in tax-deferred savings plans for which the only option is reinvesting of dividends into the fund that pays them. I only hold mutual funds.

Overall, that sounds reasonably close to what you are considering in terms of strategy: cash to spend now, bonds to stabilize resources for the intermediate term, stocks to maximize growth potential to keep the longer-term funded. It's not a bad idea to try to bias stock sales to be greater during periods of stronger growth, but I probably won't bother. I'll be forced to withdraw from the retirement accounts every year eventually, which forces me to sell the stocks (though part of the proceeds will likely be reinvested in stock funds in a taxable account). I'm already 60 so that juncture is not far off.

Interesting stuff, IlliniDave, thanks for sharing. Good to see that you have a plan around this.

You mention the mandatory withdrawals since you're 60. My parents have a similar situation. Actually that's probably not a bad way to go for me too, since the research seems to say that market timing doesn't work.

You factored in your annuity and social security also. I've not really factored the government pension into my plans, as it is means-tested in Australia and I wouldn't qualify with my current assets. There aren't any Australian bonds I can purchase beyond 2050 either.

I guess my plan is that, by keeping my stock withdrawals low and adjusting along the way with lower spending, occasional work, etc., I can preserve the stocks and bond ladder as a kind of "perpetual portfolio". But it gives me tremendous peace of mind knowing the government pension is there just in case my stocks get utterly wrecked, which is not out of the question over 50 years.

It's good that you have the annuity and social security, for similar reasons. They can hopefully allow you to minimise your stock withdrawals in order to maintain the value of your stock portfolio for as long as possible.

[Scott 2]

The hedging via hard goods strikes me as a robustness strategy. Not bad, but it can carry high overhead and be potentially brittle.

I think what's favored here tends towards resilience - the mindset Taleb promotes in antifragile. IE have enough flexibility and sources for your nutrition, that supply chain disruptions aren't a big concern. Instead you can celebrate the chance to learn new recipes.


My withdrawal strategy is pretty simple. I manually constructed a version of the Vanguard Life Strategy fund, as my entire portfolio. I allow a variable withdrawal rate, using a ceiling based on relative market valuations (cape10) and my qualitative sense of the economy.

Each year I take what I need in cash. I'm following a rising equities glide path, so that typically comes from selling my bonds. Then, I rebalance the international vs. domestic allocations, back to my original ratios.

For the portion of my investments in a taxable account, I do allow dividends to flow into cash. That's only a few thousand dollars per year. I'll also fill out my desired taxable income via Roth conversions. The goal is just below 2x poverty line, to manage health insurance costs.

I own my home and vehicle, reducing cash flow needs. Cash is kept in a high yield savings account.

[2Birds1Stone]

Definitely not trying to reinvent the wheel here......

Retired with a 60/40 portfolio, planning a 10 year glidepath to 90/10.

Rebalance once a year if necessary.

Spend whatever we spend, it's naturally <3% so the exact amount doesn't matter.

Reassess annually if spending aligns with values/goals, adjust as necessary.

[fingeek]

Rebalance every 6 months, which maintains 5% cash - enough to cover yearly living costs+.

[C40]

I've found the important universal parts to be:
- Tax prevention/avoidance
- Maximizing ACA subsidy
- IRA ladder for getting money out of retirement accounts tax-free

And then it's:
- Have investments safe enough. (conservative.. balance allocation.. be ready for market drops... however you choose)
- Still spend less than your income so your NW grows and you feel secure
- Perhaps - doing some things that result in money flowing in

[IlliniDave]

Interesting stuff, IlliniDave, thanks for sharing. Good to see that you have a plan around this.

You mention the mandatory withdrawals since you're 60. My parents have a similar situation. Actually that's probably not a bad way to go for me too, since the research seems to say that market timing doesn't work.

You factored in your annuity and social security also. I've not really factored the government pension into my plans, as it is means-tested in Australia and I wouldn't qualify with my current assets. There aren't any Australian bonds I can purchase beyond 2050 either.

I guess my plan is that, by keeping my stock withdrawals low and adjusting along the way with lower spending, occasional work, etc., I can preserve the stocks and bond ladder as a kind of "perpetual portfolio". But it gives me tremendous peace of mind knowing the government pension is there just in case my stocks get utterly wrecked, which is not out of the question over 50 years.

It's good that you have the annuity and social security, for similar reasons. They can hopefully allow you to minimise your stock withdrawals in order to maintain the value of your stock portfolio for as long as possible.

Just wanted to clarify, here in the US mandatory withdrawals now begin at age 72 now, iirc (it had been 70 up until 5-6 years ago or so). I've got 11-12 years until then. It's set up on a schedule based around life expectancy--small percentages initially but they grow each year and if I'm remembering correctly by around age 110 one would have to completely empty out the account. Deferred income taxes are collected annually as the account is drawn down. I think it's pretty common for withdrawals to get increasingly large over time, but that depends on how much the remaining balance gains or loses. So the water is a little muddied.

Our Social Security is not means tested, but it has been mismanaged. If there's no effort to fix it benefits will have to be lowered starting around 2035, which is a year after I intend to start collecting. So it's not a bad idea to at least examine scenarios where any outside income a person anticipates dries up.

The idea of an LMP is really just an offshoot of the oft-touted 60/40, withdraw/spend 4% strategy. For various reasons people might have an interest in refining both the "40" and withdrawal/spending percentage. Retiring before traditional retirement age is one of the considerations, as is a person just not having a stomach for seeing too much asset volatility with out the security of a paycheck. Also varying portfolio sizes relative to anticipated need can be a driver. So a straightforward way to address that is to try to make an estimate of total lifetime needs and invest that amount "safely" (typically bonds and cash) and put the remainder in riskier assets (typically that would be stocks). That could lead to nearly 100% in bonds/cash, or nearly 100% in stocks, depending on circumstances. The weakness of it in my view is the difficulty in coming up with a good number for how much a person might need or want to spend over a period of 30, 40, 50 years. I actually ran the numbers under many circumstances and unless I baked in a ton of unfortunate happenings or a ton of luck, the midpoint of what I felt were the reasonably conservative calculations was equivalent to something like a 70/30 allocation and 3% average withdrawal rate. When I saw a convergence of a couple strategies I felt good about it. I tend to think of it in the LMP framework because it's comforting to see a pile of low volatility assets that barring the unforeseen should more than meet my day-to-day needs going forward. And it doesn't hurt to have a largish backstop that could reasonably be expected to grow over time.

I don't hold any individual bonds, those are all in bond mutual funds where when individual bonds mature the fund manager rolls them into new bonds. That does take me out of the market for the better inflation-indexed US bonds, which a person can only purchase directly from the treasury in limited amounts. It was laziness that kept me from doing that, really, along with the fact that as far as I know they can't be purchased within a tax advantaged retirement account which is where the bulk of my assets are at this time.

I think you should be able to achieve your goal of a relatively permanent goal with the strategy you are considering.

[J_]

I have no withdrawal strategy. I act simply as C40

- Still spend less than your income so your NW grows and you feel secure
- Perhaps - doing some things that result in money flowing in

[frommi]

I use dividends and if more is needed the rebalancing of expensive assets into cheap assets to free up some cash. It probably is highly dependend on your investment strategy.

[2Birds1Stone]

Just wanted to clarify, here in the US mandatory withdrawals now begin at age 72 now, iirc (it had been 70 up until 5-6 years ago or so). I've got 11-12 years until then.

75 if you're born after 1960

[jacob]

I'm not pedantic or at least not about this

I'm not withdrawing. For me it's all dividends where the allocation is determined by tax-optimization and the portfolio is selected for "unlikely to cut dividends, not because track record, but because fundamentals" (I've been to a few rodeos by now).

I have 1.5x of my expenses covered by qualified dividends from taxable accounts and 2.8x my expenses covered by tax-deferred accounts which become available and get added in 10 years from now. Starting 13 years from now, SS will cover 2.0-2.9x my expenses depending on when it's taken and gets added on top again.

#tapwater

[IlliniDave]

75 if you're born after 1960

Thanks, I didn't realize that. I must not have read carefully enough. That's good news in that I have an extra 3 years before I have to figure out how to set up RMDs in my account.

[Western Red Cedar]

I think it can be calming and helpful to have some kind of plan around how to take money out in early retirement.

I think it goes beyond calming and helpful. Having a well-defined withdrawal strategy is an absolute necessity.

[Western Red Cedar]

This is what I developed prior to leaving work:

Withdrawal Strategy:

Rising equity glide-path in first 3-5 years (up to 93-95% equities excluding pension)
Roth IRA conversions up to standard deduction in low or no-income years
Withdraw from 457b until age 59.5
Consider accessing Roth IRA between 59.5 and 65
Access pension at 65
Access SS at 67-70 depending on the size of the portfolio

Other Considerations:

Utilize geographic arbitrage to maximize potential portfolio growth in early years and limit sequence of returns risk
Consider periodic employment or volunteer opportunities to alleviate pressure on the portfolio
Explore opportunities for passive income to supplement portfolio withdrawals
Consider returning to full-time work for a period to save for a large capital purchase (land, house, etc…)

*I realized that 93-95% equities is likely too aggressive for me, but having at least a couple years of expenses in cash/highly liquid assets allows me to SWAN. The plan will likely change as employment opportunities present themselves.

[conwy]

I think it goes beyond calming and helpful. Having a well-defined withdrawal strategy is an absolute necessity.

Agreed. You're going to apply some kind of method in any case. It might as well be determined by a well-thought-out strategy.

[conwy]

This is what I developed prior to leaving work:

Withdrawal Strategy:

Rising equity glide-path in first 3-5 years (up to 93-95% equities excluding pension)
Roth IRA conversions up to standard deduction in low or no-income years
Withdraw from 457b until age 59.5
Consider accessing Roth IRA between 59.5 and 65
Access pension at 65
Access SS at 67-70 depending on the size of the portfolio

Other Considerations:

Utilize geographic arbitrage to maximize potential portfolio growth in early years and limit sequence of returns risk
Consider periodic employment or volunteer opportunities to alleviate pressure on the portfolio
Explore opportunities for passive income to supplement portfolio withdrawals
Consider returning to full-time work for a period to save for a large capital purchase (land, house, etc…)

*I realized that 93-95% equities is likely too aggressive for me, but having at least a couple years of expenses in cash/highly liquid assets allows me to SWAN. The plan will likely change as employment opportunities present themselves.

Not advice of course, but I like this approach.

I try to look at my asset allocation as a whole, so including things like government bonds and potential employment income (human capital). I also try to take a kind of utility-based approach, where food shelter and clothing are core.

Thanks to frugality and ERE systems-based approach, I don't really need much income to satisfy my core needs, so I can easily fund all my needs on less than minimum wage if needed. I think national minimum wage provides a nice clear guideline for forming a baseline for future earned income expectations. It's legally enforced in developed countries and typically indexed to inflation.

So many of us have a combination of relatively safe assets:
- Government pension for retirement
- National minimum wage for earned income

Given the above, I think stock-heavy, diversified asset allocations make sense.

I think we have to also keep in mind the risks of not holding diversified equities:

• Your home country might experience unexpected high inflation, say due to an unexpected climate event
• Some unexpected conditions might cause you to move to another country
• Your individual inflation might unexpectedly exceed CPI, say due to an unexpected health condition

So "risk" isn't just volatility, and I think almost any kind of diversification is helpful.

For this reason I add additional diversification beyond equities including:

• Inflation-linked bonds (recently in multiple currencies, not just my home country)
• Small-cap-value tilt in my stock funds

[Scott 2]

I chose the rising equities glide because cape10 was at 35 when I pulled the trigger in 2021. The thought was it'd fall and benefit me to shift allocation over time

It's still at 35.

My allocation shift has been slower than expected. I spent less, international didn't perform all that well, there was some unexpected income.

But at these valuations, I'm not mad about it. I don't understand the appetite to hit 90%+ equities, until something shifts in valuations. Unless maybe an active investor is very confident in their selections.

[IlliniDave]

So "risk" isn't just volatility, and I think almost any kind of diversification is helpful.

I'm enjoying this thread. Some years back I devoted considerable mental energy to the problem that I tended to think of as getting the most out of finite resources. My quest that led me to where I am today started from a defensive mindset. In short, on the eve of year end holiday break one year I learned there was a real chance that my employment would end two weeks into the new year, and the times were such that the mid-late midlife white-collar cohort was not doing well recovering from being laid off due to lack of employment opportunities. So I spent my time away from work figuring out how long I could last, and my proxy for that was how long could I stretch out my relatively paltry financial resources assuming I could at least get a part time gig bagging groceries at my local Publix or whatever before I defaulted on my mortgage (I was upside down at the time, this was shortly after a divorce in the wake of the so-called mortgage crisis). I hadn't been exposed to ere yet, nor MMM, so I was sort of sitting around trying to reinvent the wheel in isolation. The storm passed and I didn't get RIF'd but I kept that mindset. The challenge of finding an optimal path through the myriad constraints (largely tax laws) and potential outcomes energized me. But in the end over the final 2-3 years of iDave Phase II my FIRE got fat (in big part simply by getting to age 55 while still employed). Long story short I'm sort of boxed in when it comes to trying to game the problem of tax liability for having the good fortune of being grandfathered into an old school pension plan from my employer and having leaned heavily on tax-advantaged retirement accounts. I do get to play around with tax efficiency in my taxable brokerage account, but that's only about 5-10% of the stash. What I say to people on the rare occasions they ask is: definitely take advantage of tax-advantaged accounts, but if you can, definitely try to build up assets in a taxable account in parallel, essentially because they can be very efficient if you opt for ER using the strategies many here are sharing. I'm slightly envious. Although I can play around on the margins in my taxable account such efforts don't really ove the needle, and bereft of meaningfulness the efforts feel mostly feel like accounting drudgery.

Anyway, I've digressed before I even got started.

The sentence I quoted contains an insight that I think is important. Somewhere along the way academic types compressed investing "theory" into the 2-D space of return versus volatility (as captured in a variance calculation) and tried to identify an optimum in that space. If I'm remembering correctly it was initially called Modern Portfolio Theory. And it was a clever method to streamline a complex and complicated subject into something that could be programmed into the computers that were around in the 1970s. To me it was interesting to read about, but the distillation of "risk" into statistical variance never resonated with me. I came to recognize that my short-term "risk" tolerance, i.e., the ability to stomach ongoing variance in my account balances, was fairly high while my long-term risk tolerance was significantly lower, and the prospect of being an octogenarian bagging groceries at Publix made me queasy (yes, I know that's not necessarily the end of the world, but long-term risk tolerance is not an outgrowth of the rational part of my brain). That contributed a lot to my happy willingness to pay the price for fatFIRE. Even the fattest fatFIRE does not mitigate all risk, it just puts certain options on the table.

As an aside, and idea I encountered that really resonated was that as investors of any type, the game is getting paid for putting your money at risk. I think I first heard or read it phrased that way from Jack Bogle, though I don't think it originated with him. It doesn't seem to be a big issue with most of the folks here, but among my colleagues and contemporaries I knew a lot of people that would sell all their stocks based on a Yahoo Finance article they read that morning, as well as a cohort who would repeatedly fall for the honeypot trap of risk-free rewards. I think it's an good thing to contemplate periodically through life.

Some of the risks you listed are rooted outside the bounds of investing. Although money's fungibility does make it a flexible risk mitigator, it has limits, and some risks when manifested can even vaporize it. In thinking about this thread I've started kicking around ideas like: how would one most effectively "withdraw" their human capital assets along side their financial assets in a complementary manner to reduce a broader palette of risks, and for me that generally means longer-term risk (I'm getting old enough that the true long-term isn't something I'm apt to see). That's not necessarily a new idea around here, in my way of thinking it's inherent to resilience and robustness, but I never picked up on the nuance of it being a finite resource we might want to use as prudently as possible while drawing it down as we would a retirement account. Of course expending it to produce financial resources outside of a standard career is something that gets talked about a lot. Attending to good health is a way to increase and extend it to a point, and to lower the burden on financial resources most likely. However, I've always thought about it as something that's largely stove piped away from the strategy of how to deploy financial resources while heading towards the finish line. I wonder if there's a way to synthesize "the withdrawal of everything" strategy and thereby minimize correlations or somehow make the final descent smoother and more palatable as a function of individual temperaments. Social capitol could be worked into the framework, and a certain amount of gaming of the system. Maybe other things? None of that would be outside the ere framework writ large, it's really just looking at it as a withdrawal strategy rather than an extended and varied resource accumulation strategy. It may well be there's zero benefit to that, but it's what's rattling in my brain this morning.