Poll:To those who retired or retiring very soon - what is your comfy withdrawal rate?

What is your approximate withdrawal rate during retirement?

  • Around 2.0% or less

    Votes: 77 31.4%
  • Around 2.5% +/-

    Votes: 37 15.1%
  • Around 3.0% +/-

    Votes: 55 22.4%
  • Around 3.5% +/-

    Votes: 34 13.9%
  • Around 4.0% +/-

    Votes: 29 11.8%
  • Around 5.0% +/-

    Votes: 10 4.1%
  • Around 6.0% or higher

    Votes: 3 1.2%

  • Total voters
    245
  • Poll closed .
Now that I better understand your point, perhaps we actually are in agreement after all. I have never viewed the Trinity Study as setting forth a hard and fast spending rule every year. Rather, it is a guideline for how much one should accumulate in assets up-front to support a 30 year retirement. As you note, the balance in my tIRA is not all mine; some belongs to the tax man. In preparation for retirement, I tried to take that into account. In my view, there are two main ways to do so: 1) discount the starting balance in the tIRA by some marginal tax rate or 2) gross up annual spending for taxes.

But that merely helps me determine when I have enough in my war chest to pull the plug . Once said plug has been pulled and I am engaged in the hurly burly of actually going through retirement, I will of course try to maximize the net value of my assets after taxes. So if I can Roth convert now and pay 12% tax, it would be better to do that than take an RMD of the same amount in a few years and pay 22% tax. And that would be true even if doing so results in a withdrawal rate greater than 4% this year, because the tax arbitrage is increasing my net assets and should, therefore, be improving my position vis-a-vis the base case from the Trinity Study (assuming I used 22% as the base case tax rate in my planning).

Yup, yup, I agree. This is pretty much how I thought/think about it.

In practice, in my case, I accumulated assets until I could have a WR of ~3%, including taxes for any withdrawals from tax-deferred. Then, I separately make a determination on whether or not to make Roth conversions, and pretty much ignore what that does to my WR in the year of conversion.
 
Last edited by a moderator:
I don't really think in terms of withdrawal rates. I think it's a poor indicator of retirement readiness or ongoing survivability.

For one, it changes significantly over time as various income streams come online, like pensions and SS. So there is no one figure for most early retirees. Couples of different ages will have multiple different figures at various times...

I completely agree about the various income streams evolving over time. It can still make sense to figure your net withdrawal rate in your current retirement income phase.

Having said that, I never focused on withdrawal rate. My focus, especially through my late 60s and now early 70s has been on my AGI, which impacts both my income tax and my IRMAA tier.
Fortunately, only a portion of my AGI gets *spent* each year in retirement; the remaining portion gets invested.

I have three parts of my portfolio from which money can either be withdrawn or invested into:
1) my tax-deferred 403(b) - withdraw only, can't contribute anymore.
2) my Roth IRA
3) my taxable account

Later on, I'll go through my records upstairs for last year and determine the dollar amounts in/out of these three accounts as step one in determining my withdrawal rate for 2021...
 
Another interesting approach... There's a member here (I think it might be RunningBum, not sure) who gets around a lot of this confusion by counting ALL retirement resources in the denominator, including the NPV of pensions, SS, and the like. Then for the numerator, you don't count WITHDRAWALS. You count total expenses.

His method ignores the timing complications associated with various income streams coming online at various times. This gives a true reading of retirement readiness on Day One (in the form of a WR). None of this... "Well it's high now, but it'll come down when SS starts, so I think everything is cool." Just compare your expenses to ALL the resources that will support your retirement.

I do it that way, but I don't have a pension. And it does help me get around the confusion that you mention.

The approach has been mentioned by multiple members here, including me.
 
This year I'm looking at about 2.5% of my portfolio, but the last 4 years I have been closer to 5.5% because of my daughter's school tuition and rent. That is finally over, in fact today was her last day of clinic, two days ago, my wife, my son, and I, went to see her, as her last three patients for her dental schooling. She replaced a filing in a wisdom tooth for my wife, it was a big filing and a difficult restoration, she said if my wife was younger, she would just pull it out to prevent future repairs, but since she's older, it won't need to much more attention over her lifetime.

This wisdom tooth repair took a while and made my appointment late, so I kept complaining about these dentists are always over booking not caring about their patients time.:cool: I also mentioned she looked at that molar and saw a Mercedes payment. (note: she didn't get paid for the school clinic work, in fact it cost about $100 a day for her to work at the school clinic, even though the school clinic did charge [-]customers[/-] patients, although at a reduced rate.) Ya, I can be sarcastic dad, $300,000+ in tuition, I think I earned the right! But it's over now, graduation ceremony is in a couple weeks and she has a job lined up a few days after graduation. Congratulations to my baby! :dance:
 
Our withdrawal rate has ranged between 2.6% and 4.1% in the 8 years of retirement... mostly just under 3%. This is based on the original nest egg value, adjusted for cpi - which is the basis for the 4% rule/trinity study.

Non cpi adjusted, but based on original portfolio, between 2.7% and 4.6... again, mostly under 3%.

When I look at percentage of the portfolio at the beginning of the year, it is between 2.7 and 3.4. That's because the high spend year was last year - high value portfolio, but significant medical expenses for our 21 year old son's ameloblastoma surgery. Looking at another high withdrawal when his jaw is recovered enough to get dental implants to replace his removed teeth from the surgery.

I don't do a set amount - just try to stay within reason and live pretty frugally.
 
We do variable withdrawal of 4.x%, but retired when our portfolio would have supported us nicely on 3%. ( Still ignoring social at ages 70, which is flirting with 100k per year, which is looking pretty paranoid.)

Coming up on 5 years retired (62 and 61 as of today), and majority of spending remains travel. (I treat tax on Roth conversions as not present spending; rather, it is simply reallocation of future assets, or prepayment of future taxes...)
 
... I have three parts of my portfolio from which money can either be withdrawn or invested into:
1) my tax-deferred 403(b) - withdraw only, can't contribute anymore.
2) my Roth IRA
3) my taxable account

Later on, I'll go through my records upstairs for last year and determine the dollar amounts in/out of these three accounts as step one in determining my withdrawal rate for 2021...

Ok, I went through my records from last year (2021) which is my last year before starting RMDs.

1) I withdrew a total of $42,800 from my tax-deferred 403(b), including Roth conversions.

2) I put $33,192 into my Roth IRA. This was all Roth conversion money since I have zero earned income.

3) I put $57,500 into my taxable investment account. This came from excess retirement income (pension/annuities + SS).

So I had a negative withdrawal rate for the year; I'm not going to compute a percentage from this data.

This year I started RMDs, taken as equal monthly amounts for what it matters.
My total RMD for this year is something less than the $42,500 I withdrew last year, mostly for Roth conversion.
And so it goes...
 
I think it can change.

If you retire early, you might need 5% but if you later get SS and drops you down to 2%, that initial 5% might be OK. But I'd be modeling it to see how much would be left to determine that SWR at the time the SS kicks in.
 
I've yet to get into withdrawals in my 13 1/2 years in retirement.

But now it's time for RMD's.

And don't you know I'm thrilled to make withdrawals with the stock marketing dropping like a rock.

I just hope I've got a fund that's not so stock like--for withdrawals before EOY.
 
I do it that way, but I don't have a pension. And it does help me get around the confusion that you mention.

The approach has been mentioned by multiple members here, including me.

This is my primary measure of effectiveness for my retirement model.

Now (56) through age 95:

$3,595,141 Income (includes COLA pension and starting SS @ 70)
$4,816,315 Total Expenses

-$1,221,174 = Savings required

If I have more than the savings required, then I am good to go. Less than that and we have a problem.

I also take a look at other key periods like 56-70 to make sure I don't go negative before then. Once SS kicks in, the pension and SS cover all expenses.

I do all my calculations in today's dollars. Easier to interpret. I also just use 0% real for my savings. Keeps everything clean.

You could use a discount rate on the expenses and/or an inflation rate on the income. I have done that and it doesn't much matter for my case.
 
Last edited:
My withdrawal rate is what the Gov says for my annual RMD (4+%). Otherwise, we live on the SS as it comes in.

This is the best way to do it if it works for you.
We are doing the same.
 
Over 2-1/2 years in retirement, no pension, no SS yet, our spend rate varies between 2% and 3.5%. Hoping to increase next year with a new car and international travel. Oh, and DW plans to apply for her SS benefits next year too.
 
Originally Posted by Gumby View Post
Now that I better understand your point, perhaps we actually are in agreement after all. ...
Yup, yup, I agree. This is pretty much how I thought/think about it.

In practice, in my case, I accumulated assets until I could have a WR of ~3%, including taxes for any withdrawals from tax-deferred. Then, I separately make a determination on whether or not to make Roth conversions, and pretty much ignore what that does to my WR in the year of conversion.

I was in agreement with Gumby's first reply, it is a tax, you paid it, so it is 'spent' and should be in your withdrawal calculation.

But after reading Out-to-Lunch's reply, that tax is really is more like a transfer of money between accounts. Since you anticipate it will save you money in the long run, it's really an 'investment'. A current 'expense' with the anticipation of future 'income'.

But, OTOH, when it comes to sequence risk, that is a cash-flow hit on the current portfolio, which can add to that risk. Although probably not enough to be a material concern.

-ERD50
 
Similar - we are withdrawing 5% for 5 years and then 1% or less with pensions and SS.
 
But, OTOH, when it comes to sequence risk, that is a cash-flow hit on the current portfolio, which can add to that risk. Although probably not enough to be a material concern.

Maybe.

I still contend that, if you do not alter your overall tax-adjusted AA while making the conversion, SORR does not change. This is because of the commutative law of mulltiplication, as we have discussed before.

I do agree that the conversion could make future outcomes worse due to unanticipated changes in tax situations, laws, and/or market conditions. Or if you alter your tax-adjusted AA as a result of the conversion.
 
Maybe.

I still contend that, if you do not alter your overall tax-adjusted AA while making the conversion, SORR does not change. This is because of the commutative law of mulltiplication, as we have discussed before.

I do agree that the conversion could make future outcomes worse due to unanticipated changes in tax situations, laws, and/or market conditions. Or if you alter your tax-adjusted AA as a result of the conversion.

I can tell you that my model indicates that Roth conversions and the increased taxes they incur earlier in retirement, does increase SORR. Retirement is pretty simple. Money in, money out. If you take more out early to save money later, you increase SORR.

Having said that, it is not a significant increase if your withdrawal rate is less than 4%. You can pretty much call it six one way, half dozen the other.
 
Planned 4% withdrawal rate but have been closer to 3%. Retired 5 years and not drawing SS or pension yet (ages 60 & 61).
 
I can tell you that my model indicates that Roth conversions and the increased taxes they incur earlier in retirement, does increase SORR. Retirement is pretty simple. Money in, money out. If you take more out early to save money later, you increase SORR.

Having said that, it is not a significant increase if your withdrawal rate is less than 4%. You can pretty much call it six one way, half dozen the other.

Does your model hold your tax-adjusted AA constant?
 
Does your model hold your tax-adjusted AA constant?

I don't model tax adjusted AA. Not sure what purpose that would serve. I just maintain my 60/40 AA.

If I convert before age 70, I pay less overall taxes over my model period (age 56-95). If I do no conversions, I pay little taxes from 56-70 but pay quite a bit more taxes over the model period because I will be in the 22% tax bracket when RMDs start vs. the 12% tax bracket now. I am taking a balanced approach and doing modest conversions from now until age 60 (4 years) then increase that until SS kicks in @ 70. If market returns are good over the next 14 years, I may convert more, faster. If this current bear continues, I may convert little to nothing to preserve capital.

I have a full 1040 in my model so it accounts for the conversions and taxes paid. Also has a full SS worksheet to see the impact on how much of my SS is taxed. And it accounts for LTCG because I am doing tax gain harvesting @ 0% LTCG on my taxable account. And if I enter a year of my death (or my wife's), it uses single vs. MFJ data. It was a lot of work incorporating all of this, but it ended up telling me not to worry too much about it because it wasn't a big deal.
 
Last edited:
None of the above. There are 2 reasons. First, taking Social Security and secondly, "are you leaving anything for your kids?".

We had a gap of several years between when we retired and when we started taking SS. It was about 8% before SS and 3% after taking SS.

If you aren't planning on leaving anything to your kids, you can comfortably take a higher percentage each year.

Our 3 sons are doing very well and won't need anything from us.
 
I don't model tax adjusted AA. Not sure what purpose that would serve. I just maintain my 60/40 AA.

Yes, then I would agree that SORR would vary in your model between the convert and no-convert scenarios.

I am not saying that one SHOULD maintain a constant tax-adjusted AA. But, if you don't, the after-tax chararacter of your portfolio changes as you perform conversions or other changes.

It was a lot of work incorporating all of this, but it ended up telling me not to worry too much about it because it wasn't a big deal.

Despite my evident interest in the topic, I heartily agree that it doesn't make a lot of financial difference. One of my main goals is simplifying the financial situation for my DW when I am pushing up daisies.
 
When we first retired our withdrawal rate from tIRA ranged from 3.81% to 7.65% in the first 5 years. Much of that withdrawal was reinvested in home improvements. Since then it's been 5.4% - 5.7%. This is based on delaying SS until age 70, and the expectation of ~$750,000 inheritance within the next 10 years. Recently, some of the tIRA withdrawal has not been spent, but reinvested in iBonds and a Growth LLC supporting local businesses . Not counted in the withdrawal rate (except for taxes) is the amount we have converted to ROTH.
 
Back
Top Bottom