Does this withdrawal "rule" make any sense?

[GravitySucks]

There are more up than down years/months/days so im thinking in the long run its a losing strategy.
But does it help you sleep at night? Instead of my usual quarterly withdrawals in November 2019 I thought things were getting frothy so took 2020 expected expenses off the table. (This only brought me to a 59/41 AA so not far off, a rebalance before my band was met I rationalize.) Kicked myself till March. Felt like a genius March and April.
I'll go back to quarterly withdrawls after 2020 cash is spent. Which with Covid might not be till July.

 

[Rustic23]

Great year, you withdraw $300,000. How does this effect MAGI an Taxes?

 

[pb4uski]

Great year, you withdraw $300,000. How does this effect MAGI an Taxes?

That's out of left field... where did the OP ever suggest that he was withdrawing $300k? He's only withdrawing $100k.

He's talking about rebalancing... so no withdrawal this year but essentially squirreling away cash for future years... no tax implications is the rebalancing occurs in tax-deferred or tax-free accounts.

DawgMan is pretty savvy so I'm sure he has the tax implications figured out.

 

[Animorph]

Tinkering, just for an exercise right now. Let's say you define your expenses for the year as $100K and that represents a 3% WR. While you have already pulled your $100K for this years expenses in early Jan, you expect 2022 expenses to be similar. Come end of Jan, due to a fast market run up, you have hit a 3% return ($100K in unrealized gains). Does in make sense to take the chips off the table the day of the year you see your returns produce the next years (and perhaps the year after's expenses in big bull market), rebalance, rinse and repeat? I've been inclined to do my year's expenses withdrawal/rebalance 1 time a year and deal with the market returns/yield once, as opposed to cashing out. None the less, something to think about. Anyone employ this "rule"/strategy or something like it?

It makes some sense, but not quite. I do this, but I use 6% "real" gains plus 3% inflation gains for a total of 9%. Otherwise your principal and withdrawal aren't keeping up with inflation and you are not preparing any cushion for a down year.

My deterministic projections use 9% a year and an inflation adjusted withdrawal at the start of the year. If my real portfolio is able to match my projected portfolio value at the time of the next projected withdrawal, I'll go ahead and make that next withdrawal. Even if it is a year early. In fact, I'm a couple of years early already. Aggravated by the fact that we don't really spend everything I projected.

The flip side is that if we get a bear market instead, I'll start reinvesting some of the excess cash and shift my AA a little to favor more equities. And if things are still down when I must make a withdrawal then I withdraw monthly and only as much as we need.

Hopefully this all minimizes the impact of the sequence of returns. We've been living off of our portfolio only, delaying all other income. With substantial income expected to come later, I think we usually withdraw about 6% of the portfolio each year. So we have a fairly substantial risk if returns are low for a long time. In fact they have been all over the place.

I think this scheme only works well if you start at a reasonable portfolio value and projected gains. We have managed to have mostly excess cash and have done the monthly withdrawals one year, so the scheme has worked fine for us.

 

[Animorph]

As far as taxes go, I'm happy to simply store "withdrawn" money as bonds (FXNAX Total Bond Market) in a Roth or IRA account. I sell the desired amount in equities and buy bonds, all within retirement accounts with no tax consequences. I take those bonds out of the AA calculation, so it's set aside and earmarked for expenses. It can then sit in whatever retirement account it was in until I need it. When we hit a new tax year I sell taxable equity shares (we have no taxable bonds) with an eye to any of the AGI limits I'm following at the time. At the same time I sell the same amount of the earmarked bond shares in the retirement account and buy equities, or enough equities to balance the AA now that the earmarked bonds are once again included in the AA.

So I sell equities and buy bonds in a retirement account, wait until I can handle a taxable sale, then sell the equities I really wanted to sell in the taxable account. At the same time as the taxable sale, I sell the extra bonds I bought in the retirement account and buy equities.

It may cost some growth in the retirement account due to the extra bonds, though that could be a benefit if the market heads down. This will be made up by growth in the taxable account from the shares not yet sold, or a tax break not available in the retirement account if the taxable equites lose value before the sale. I usually try to do this in a Roth account since that is where we will withdraw from once the taxable accounts run out. I figure it mainly affects the date of the taxable to Roth switch, but has minimal effects on the combined balances.

 

[big-papa]

Why is that? Most of us don't slavishly adhere to an x% withdrawal rate, so why would we slavishly adhere to a rebalancing policy?

Its as simple as "feeling" that the market has had a good run and gotton ahead of its skis and rebalancing then to lock in some profits... it might work out or it might not.

Once you're managing your asset allocation by "feeling" you're now market timing, in my opinion. More than sufficient info out there about how that works out in the long run. If one "slavishly" follows an algorithmic method of Dynamic Asset Allocation, it might also not work out either, though at least one can prove that it might have worked out in the past over pretty long periods. Not sure one could make the statement that the future might rhyme with the past when you're talking emotions...

 

[OldShooter]

Once you're managing your asset allocation by "feeling" you're now market timing, in my opinion. ...

Well, sure. Every trade is market timing to some degree. So what?

The justifiably bad rap that "Market Timing" (note caps) has is based on the fact that has been shown to be an unsuccessful overall portfolio strategy.

We market time every year by deciding to defer selling until we need the money and by making a decision whether to sell equity or fixed income. (Our withdrawals are generally small enough relative to the portfolio that they don't affect AA.) Can we do this in an optimal way, calling tops and bottoms? No. But on balance we tend to hold equity sales until near the end of the year, which 100 years of market statistics would predict that on average prices will be 5-10% higher than at the beginning of the year. We win some, we lose some, but that kind of market timing is not Market Timing IMO.

 

[CardsFan]

OK, I made a snide comment earlier (It just doesn't matter).

My plan is simple. We have an AA. We take all distributions in cash. Siphon off some of that cash for spending. Invest the rest in what makes sense for the AA. If the market goes spastic (is that PC to say?), we sit on our hands and do nothing.

May not be the best approach to pile up the assets, but we don't need the best approach. Just want it simple and reliable.

Could something happen to change this thinking? Sure. then we can adapt.

 

[big-papa]

Well, sure. Every trade is market timing to some degree. So what?

The justifiably bad rap that "Market Timing" (note caps) has is based on the fact that has been shown to be an unsuccessful overall portfolio strategy.

We market time every year by deciding to defer selling until we need the money and by making a decision whether to sell equity or fixed income. (Our withdrawals are generally small enough relative to the portfolio that they don't affect AA.) Can we do this in an optimal way, calling tops and bottoms? No. But on balance we tend to hold equity sales until near the end of the year, which 100 years of market statistics would predict that on average prices will be 5-10% higher than at the beginning of the year. We win some, we lose some, but that kind of market timing is not Market Timing IMO.

I don't see how deferring withdrawals until they're needed is in any way market timing. You're not changing your AA and you're not using feelings to decide. This is only needs based withdrawals. Similarly, I don't see how waiting until the end of the year, because history tells you that prices will be higher, is market timing, either. IMO this is similar to choosing a fixed AA to begin with - we looked at what the market has done in the past, figure out a combination of assets that would work for us, and hope it does something similar in the future, and we move on.

Anyway that's my admittedly narrow view of market timing - feelings based as opposed to something algorithmic like Dynamic Asset Allocation. I've been on other forums where pretty much everything, at one time or another, including buy and hold has been attached with the scarlet letters "MT" - Shame! Things like that are debated endlessly because I suppose we humans can't help but categorize everything, even when we all don't agree on what the categories actually are.

 

[pb4uski]

I think you meant narrow minded view of market timing.

 

[big-papa]

I meant exactly what I said, no more, no less.

I think you meant narrow minded view of market timing.

 

[DawgMan]

Just catching back up here on the this string. Let me try and be a little clearer on my "thought experiment". These are just hypothetical numbers for illustration. Let's say you have $3.33M of investable assets in your 60/40 AA that you generally would rebalance and pull your 3% WR ($100K expenses) at the end every year. And let's say you have underwritten a 5% average annual return in your AA return assumptions and all the calculators say you are fine and dandy for 30 - 40 yrs, in fact, you will most likely double+ your money upon the dirt nap. A we all know, markets go up and down at anytime in any given year (use 2020 as eg). What I was suggesting, was the thought (and that's all it is right now) of adding a "rule" which says if the market has a run up of 3% (or maybe more specifically it basically adds my annual expected expenses of $100K of profit... balance shows $3.33M + $100K), should that trigger a "pull the money" rule/rebalance at that time? If you pull the money, your $3.33M AA is still in theory working for you, but you locked in your gain. My thought was your $3.33M is still working for you as you had hoped during the balance of the year so your future returns/ability to fund your RE life should not be in jeopardy. I suppose if you did apply this rule you could rebalance 2 ways: 1) Conservatively - don't count the additional cash you pulled for future expenses (whether you keep this in cash/short term bonds/other) in your AA rebalance or 2) More Aggressive: Do include the cashed out money in your fixed allocation which means you are probably upping your equities at this point.

Again, all part of a thought experiment that seemed to have some physiological and perhaps mathematical benefits I just wanted to bang it around... "quasi-bucket-ish" approach I suppose. Right now I incorporate a loose 5% bandwidth to alert me if I need to rebalance before the end of the year, but otherwise, deal with any withdrawals once a year based on whatever overall AA return occurred.

 

[mrfeh]

I meant exactly what I said, no more, no less.

I agree with you. Some comments strike me as market timers attempting to justify their actions through obfuscation.

 

[OldShooter]

I meant exactly what I said, no more, no less.

I agree with you. Some comments strike me as market timers attempting to justify their actions through obfuscation.

As Humpty told Alice: "When I use a word," Humpty Dumpty said, in rather a scornful tone, "it means just what I choose it to mean—neither more nor less."

So when you guys use the terms "market timing" and "market timer" please enlighten us as to exactly what you mean.

For example, when I decide to pull the cash to pay the IRS in December instead of pulling it in January and holding it until December, is that "market timing?" Am I then a "market timer?"

 

[Winemaker]

I do what Dawgman says in #37. I like to keep about 3 years of WDs in cash, I pull some out in January, then pull out remainder of annual draw in December. Once my nest egg "earns" my annual draw, I sell enough to rebalance and put that in cash. So far, so good and this is 5 months into year 7. Nest egg up twice the amount withdrawn since 2014.

Perhaps it's the same thing as taking your winnings off the table when playing blackjack. But I don't gamble, other than a occasional lottery ticket.

 

[pb4uski]

I think what DawgMan said in #37 makes perfect sense... for those that slavishly insist that there be a set rule, let's say that one defines the "rule" as rebalancing when the portfolio exceeds $3.43m ($3.33m starting balance + $100k).

So if someone decides to rebalance at $3.39m (up $95k) then they are a market timer according to some... ditto if they wait until the total is $3.48m (up $105k).

I think it is silly that so few here slavishly adhere to a 4% rule or 3.5% rule or whatever but when it comes to rebalancing that some insist on a crazy degree of precision... otherwise you are a dirty market timer.

 

[RunningBum]

Let's say you have $3.33M of investable assets in your 60/40 AA that you generally would rebalance and pull your 3% WR ($100K expenses) at the end every year. And let's say you have underwritten a 5% average annual return ...

What I was suggesting, was the thought (and that's all it is right now) of adding a "rule" which says if the market has a run up of 3% (or maybe more specifically it basically adds my annual expected expenses of $100K of profit... balance shows $3.33M + $100K), should that trigger a "pull the money" rule/rebalance at that time? If you pull the money, your $3.33M AA is still in theory working for you, but you locked in your gain. My thought was your $3.33M is still working for you as you had hoped during the balance of the year so your future returns/ability to fund your RE life should not be in jeopardy....

Again, all part of a thought experiment that seemed to have some physiological and perhaps mathematical benefits...

I still say no. You can certainly find some scenarios where it has mathematical benefits, but I can find more where it doesn't. You said yourself your assumption is a 5% annual (average) return. By taking money out at 3%, as you say you could still have your $3.33M working for you, but you could instead have $3.43M working for you getting that other 2% (on average) return.

Another issue...you're not going to have $3.33M for perpetuity. Some years you'll lose money, or never get 3% ahead, so at the end of the year you have less than $3.33M.

There's also the matter of inflation. If your needs and wants are constant, you'll be gradually increasing that $100K withdrawal every year. I guess you could pull your next year withdrawal after 3% return the first year, then 3.06% (for a 2% inflation year) the next year, and so on, but it may be less likely to get that higher return the longer you go.

Now a 3% withdrawal is pretty much bullet proof so it's very likely this strategy will work, but if leaving as much money as possible to heirs and charity is a secondary goal, this method will likely not leave as much as just doing the normal 3% withdrawal at the end of the year for the next year. I base that on the assumption that over the long haul, the market historically has gone up , not down, and likely will continue to do so, so the optimal plan is to keep your money invested rather than on the sidelines.

 

[CRLLS]

In my mind, you are a "market timer" if you act based on a whim. For example, you think that the fast runup is going to reverse and you want to take expected withdrawals out now. Not that that is a bad thing, it is just not part of your normal plan.

If setting aside next year's withdrawals based on some financial occurrence is somehow part of your written (or unwritten) plan, then it is not "market timing". Further, it is not wrong, or market timing if you change or add a trigger definition. No plan is written in stone. They do change as we get wiser and situations change.

Whether this is a good plan or not is a different discussion. Personally, I don't think so, at least for my situation. YMMV

 

[GravitySucks]

Your explaination in post 37 makes it seem more that your rebalance point on the upside is 3%, clears it up too me. Sure, why not? I rebalance at 5%, some yearly. There's no law saying what it has to be.

 

[deserat]

Thanks for linking the amortization plan for withdrawals at Bogleheads.

To the OP, I've been playing with the VPW calculator. It is different from the amortization model in that it assumes the rate of return and withdrawal percentages in the model based on your AA and age, as well as other streams if income. However, what I like about it is you can choose the frequency of calculation for withdrawal: monthly, quarterly, or annually. It also gives you a moderate worst case scenario in which the stock portion of your portfolio takes a 50% hit, so you can see how your withdrawal amounts would adjust.

As above, the developer of the spreadsheet is also running a forward test every month with the method also using a 6 month cash account buffer to smooth the monthly withdrawal amounts.

In any case, what I find interesting/comforting is that some of the fear of not having enough or 'do I have to take my chips off the table' can be mitigated with the built in visualization of that scenario in the model/spreadsheet.

 

[audreyh1]

The VPW spreadsheet also shows you max drawdown based on history which is really important for a withdrawal method based on portfolio value each year, as that affects your income/allowed withdrawal. During the worst case scenarios your portfolio is shrinking for quite a while in real terms, therefore so is your annual income. Just a potential something you have to figure out how to live with if you chose one of these % remaining portfolio type methods. Firecalc doesn’t provide this, although it does provide worst case, average and max ending portfolio values for a given time period - also valuable. I had to look at those generated scatter plots of runs and tease out which were the starting years for the worst scenarios, and run those to get lowest portfolio value during the period.

 

[RockLife]

Not worry, just chasing the best mouse trap!

That's what I've learned in almost 4 years of retirement. it's not about "best" any more. It's about managing and maintaining. So, no greed, I take what the markets give me and I'm happy. I'm not chasing perfection, because that can lead to poor decisions. In this case, it sounds like you're doing fine, just don't get caught up in over-achieving.

Just my advice. You asked for it, and got exactly what you paid for it. Good luck!

 

[audreyh1]

DawgMan is a bit famous around here for perhaps excessive analysis. But he writes great thought provoking posts as he works he way through them!

 

[RockLife]

DawgMan is a bit famous around here for perhaps excessive analysis. But he writes great thought provoking posts as he works he way through them!

I'm good with that. I did it myself and still have a tendency to over-analyze. BUT, I'm learning to do less of it which actually makes for worrying less. Out of sight, out of mind.

 

[deserat]

The VPW spreadsheet also shows you max drawdown based on history which is really important for a withdrawal method based on portfolio value each year, as that affects your income/allowed withdrawal. During the worst case scenarios your portfolio is shrinking for quite a while in real terms, therefore so is your annual income. Just a potential something you have to figure out how to live with if you chose one of these % remaining portfolio type methods. Firecalc doesn’t provide this, although it does provide worst case, average and max ending portfolio values for a given time period - also valuable. I had to look at those generated scatter plots of runs and tease out which were the starting years for the worst scenarios, and run those to get lowest portfolio value during the period.

Ah, a respite from the Gamestop posts here and at BH

You are correct...one has to know the assumptions for each of the models/calculators being used. Being an engineer, I have looked/used all of them. I've come to several general conclusions:

1) starting the saving habit early and being consistent over time has had more of an effect, outside of any large windfall-bolus, on my projected safe withdrawal amount

2) pension income matters a lot more than I thought...it is a stable basis for the streams of income one might have in retirement

All of the models I've used have predicted SWR within $5-10k of each other....could be many things, as most use similar assumptions or algorithms to I may not be varying my inputs much.

For me, I try to focus on what I control...what has been gratifying with running the models is that there is a fairly good correlation for me that my long term behavior will help me as I age and wish to transition to a different model, ie, not being an employee, and yet not require me to significantly alter my behavior, ie spend less. If the models are true, I will need to very carefully read the 'Blow that dough' thread to find some ideas for spending

To the OP-understand your angst-as above you can see sometimes I act on it. From the responses here, there's an app/model/spreadsheet for that if you wish to analyze some more.

I suspect lockdowns and limited opportunities to blow off steam in other ways is being manifested in analysis paralysis....I am antsy for my old lifestyle of gym going, skiing, playing badminton, possibly flying and traveling..I've crocheted too many afghans...