Question about calculating the 4% rule

[Seattle]

Very quick question. I am doing some planning for ER and I am using the 4% with drawl rate as my guide for planning ER. But I am confused.

Now my question - do you include your total net worth in that 4% calculation, in other words, I have both pre and post tax accounts (401K, IRA and post tax investments).

Do I apply that 4% with drawl rate to my entire portfolio for ER or only to the post tax accounts? I am really confused about that.

Let me put it a simple way. Lets say I have $5M in total pre and post tax investments ($4M in post and $1M in 401Ks and IRAs) and I early retire at 50 years old and plan on living on a 4% withdrawl a year. Does that mean a 4% withdrawl against the entire $5M (or $200K a year) or just a 4% withdrawl against just the $4M ($160K)? I dont plan on touching any of my 401K or IRA savings until I am 70

My quality of life in retirement is going to be based on this answer because living on $200K a year will be easier then the $160K a year

Any advice you guys can give a novice on this so I can plan the right way as I plan my ER would be extremely appreciated.

 

[Texas Proud]

Total portfolio.

 

[2017ish]

As Texas Proud states, it is based on total portfolio.

BUT (kind of a big but!) if you are adamant about not drawing on the tax deferred accounts for twenty years, it could pose a problem. You might be fine with your 80% in regular/post accounts, but for those of us with 80% in tax deferred accounts....

In any event, don't forget that the "4%" includes any taxes that you might pay in your spending.

 

[big-papa]

Generally, it's the total portfolio, but you choose whether to pull from your taxable accounts first, your deferred accounts first, or some blend depending on your age (IRS rules on withdrawals from deferred accounts) and tax situation.

Also, it's adjusted each year for inflation. If it's year 1, you make your first withdrawal on, say, January 1st at 4%. During that year, say there is 3% inflation. On year 2 on January 1st your withdrawal would be 4*1.03=4.12% And so on.

Tons of caveats here.
1. Lots of discussion about whether 4% is too high these days. You can look that up and decide how conservative you want to be/can be.
2. What to use for inflation? For backtesting, you can use historic CPI tables. Once you're in retirement, use your own, personal inflation rate.
3. Withdrawing only a year's worth of funds on January 1st, adjusted to *last year's* inflation puts you at risk depleting that year's withdrawal early if inflation that happens during the year is greater than whatever interest rate you earn in the account you're spending from during the year. Good to have a buffer.

There are many articles on this withdrawal method and many articles on other, variable withdrawal methods. In the end, you get to trade off a "steady" stream of withdrawals that have a non-zero chance of depleting your entire accounts vs. a variable stream of withdrawals that won't deplete your account but have a nonzero chance of having years whose withdrawals are less than what you might need to live on. Either of these non-zero chances can be reduced by living below your means (LBYM).

Oh and beware of "ultimate", "best", adjectives when describing withdrawal methods. If there was one, we'd all be following it.

Big-Papa

 

[Texas Proud]

Big Papa...

I think technically your thinking is not correct..... in your example you are putting the inflation on the 4% and coming up with a higher ratio....

From what I read.... the inflation is put on the calculated number.... so, as an example if your 4% comes out to be $100,000.... and there is 3% inflation then the next year you take out $100,000 X 1.03 = $103,000... no matter what the ratio to the portfolio is at the time....

If the market went down it could be 6% of your portfolio.... if it went up it might be 3%....


I do not want people to think that the ratio is the thing that is being recalculated.....

 

[pb4uski]

+1 The first year withdrawal is adjusted for inflation, not the percentage. So if you have $5m when you retire it would be $200k the first year and $206k the second year (assuming 3% inflation), $212k the third year, etc.

Also, 4% is probably ok for someone retiring today at age 65, but a slightly lower % would be prudent, particularly if one is retiring significantly earlier than 65.

 

[audreyh1]

I only do the calculation on investments that are specifically marked for retirement and from which I withdraw - my retirement investment portfolio and IRAs.

You can do the calculation against your combined taxable and tax-deferred investments. It doesn't matter if you are only withdrawing from the taxable accounts at first.

 

[nash031]

Danger using the term "net worth" depending on definition - your home equity, for example, shouldn't enter into it. I think you already get that, but just covering it anyway.

 

[Lisa99]

I don't worry about the 4%. I use a combination of Firecalc, i-ORP and common sense (i.e., if the market is falling we take less) to tell me what we can withdraw.

We're retiring early (54 and 50) so our withdrawal rate before SS is higher than 4%. But we both were high earners, will be taking SS at 62, and are willing to gamble that SS will be there when we need it. Once SS starts we'll be drawing <2% of our portfolio, so it should all balance out.

Having said all that if the market goes into a free fall anytime soon we recognize we'll have to cut way back on spending or work part time until the market recovers.

 

[big-papa]

Sorry I wasn't clear - the first withdrawal is 4% of the initial portfolio when you started retirement. Inflation is adjusted on that particular dollar amount after that. So the calculation I showed is correct, but it's always relative to the initial portfolio when you started retirement, not what remains in the portfolio.

Big-Papa


Sent from my iPhone using Early Retirement Forum

 

[H2ODude]

Well, I retired 4 years ago and if I stick with this rule I'd be pulling (wild guess) something slightly over 3% today (of the original retirement date portfolio) since portfolio has done well since then. I've always wondered....wouldn't it be legitimate to recalculate the 4% on TODAY's value? Or does this "rule" require the discipline to base today's pull based on the 4 year old portfolio to account for the fact that bad years ahead must be accounted for? After all, just because I retired 4 years ago am I sentenced to a lower income of the same portfolio value as someone retiring today? Going forward we both face the same future, eh?

I'd think that's ok to adjust up since if it was the other way around (4 bad years) I'd be adjusting the % down; just couldn't stomach not adjusting after bad performance.

 

[pb4uski]

Agreed - your post makes sense if the % are in relation to retirement date portfolio amount.

 

[pb4uski]

Well, I retired 4 years ago and if I stick with this rule I'd be pulling (wild guess) something slightly over 3% today (of the original retirement date portfolio) since portfolio has done well since then. I've always wondered....wouldn't it be legitimate to recalculate the 4% on TODAY's value? Or does this "rule" require the discipline to base today's pull based on the 4 year old portfolio to account for the fact that bad years ahead must be accounted for? After all, just because I retired 4 years ago am I sentenced to a lower income of the same portfolio value as someone retiring today? Going forward we both face the same future, eh?

I'd think that's ok to adjust up since if it was the other way around (4 bad years) I'd be adjusting the % down; just couldn't stomach not adjusting after bad performance.

Similar story here... 4% of my retirement date portfolio balance would be 3.1% of my current portfolio balance. We don't withdraw a fixed percentage, rather just a fixed amount per month... our retirement "paycheck" and I have not changed it since we retired 4 years ago but we do additional withdrawals as needed (garage renovation, cars, or whatever).

I agree that it is legit to "fresh start" WR as the portfolio increases and stand pat if it decreases. There was a thread on that issue recently.

You wouldn't adjust after bad performance, you would just stay with 4% of the last reset date amount, adjusted only for inflation.

 

[big-papa]

Well, I retired 4 years ago and if I stick with this rule I'd be pulling (wild guess) something slightly over 3% today (of the original retirement date portfolio) since portfolio has done well since then. I've always wondered....wouldn't it be legitimate to recalculate the 4% on TODAY's value? Or does this "rule" require the discipline to base today's pull based on the 4 year old portfolio to account for the fact that bad years ahead must be accounted for? After all, just because I retired 4 years ago am I sentenced to a lower income of the same portfolio value as someone retiring today? Going forward we both face the same future, eh?

I'd think that's ok to adjust up since if it was the other way around (4 bad years) I'd be adjusting the % down; just couldn't stomach not adjusting after bad performance.

Yep, there's a variant called "retire again (and again)" that basically says that if you've started with a certain percentage and adjusted for inflation and saw that your portfolio is increasing beyond inflation after a few years, then pretend that it's "year 1" again and reset. It's the nature of the so-called 4% rule that one year plus or minus can make a noticeable different in the starting withdrawal.

The variations of withdrawal methods are literally infinite.

big-papa

 

[Animorph]

Well, I retired 4 years ago and if I stick with this rule I'd be pulling (wild guess) something slightly over 3% today (of the original retirement date portfolio) since portfolio has done well since then. I've always wondered....wouldn't it be legitimate to recalculate the 4% on TODAY's value? Or does this "rule" require the discipline to base today's pull based on the 4 year old portfolio to account for the fact that bad years ahead must be accounted for? After all, just because I retired 4 years ago am I sentenced to a lower income of the same portfolio value as someone retiring today? Going forward we both face the same future, eh?

I'd think that's ok to adjust up since if it was the other way around (4 bad years) I'd be adjusting the % down; just couldn't stomach not adjusting after bad performance.

Depends on how you do it. A permanent bump that is adjusted for inflation each year probably increases your chances of failure. There are a few historical failures at 4%, and you would be tempting fate by continually trying to have a big bear market right after you start taking your initial 4%. The FIRECalc success statistics are not valid for this case.

With the original 4% rules you have a good chance of the market rising in the first few years, giving you the safety margin that avoids future portfolio failures.

Now if you want to take 4% of the portfolio value each year, taking the cuts with the raises, that's again a different case. You won't fail, but you might be low income at times. But it's a reasonable approach if that works for you.

If you use a "100% safe" withdrawal rate, like 3% initially adjusted for inflation, then you could reset your 3% each year without too much concern about portfolio failure. Mainly because there are no historical failures (under the standard 30 years and 50/50 portfolio) with a 3% initial withdrawal rate.

 

[big-papa]

Yep, there's a variant called "retire again (and again)" that basically says that if you've started with a certain percentage and adjusted for inflation and saw that your portfolio is increasing beyond inflation after a few years, then pretend that it's "year 1" again and reset. It's the nature of the so-called 4% rule that one year plus or minus can make a noticeable different in the starting withdrawal.

The variations of withdrawal methods are literally infinite.

big-papa

Infinite like: Kitces ratcheting rule (discussed elsewhere in this forum), the more complex Guyton-Klinger decision rules, the Scott Burns 6%/90% rules.

 

[audreyh1]

Well, I retired 4 years ago and if I stick with this rule I'd be pulling (wild guess) something slightly over 3% today (of the original retirement date portfolio) since portfolio has done well since then. I've always wondered....wouldn't it be legitimate to recalculate the 4% on TODAY's value? Or does this "rule" require the discipline to base today's pull based on the 4 year old portfolio to account for the fact that bad years ahead must be accounted for? After all, just because I retired 4 years ago am I sentenced to a lower income of the same portfolio value as someone retiring today? Going forward we both face the same future, eh?

I'd think that's ok to adjust up since if it was the other way around (4 bad years) I'd be adjusting the % down; just couldn't stomach not adjusting after bad performance.

Some of us don't use the initial portfolio, and don't adjust for published inflation. Instead we use the value of the portfolio at the end of each year. However, this means that some years, if the portfolio drops, we take a cut in income, which is what the original scheme was set up to avoid. If you are comfortable with the occasional cut in income, this is a viable approach. Personally, I was never comfortable with increasing my income by inflation regardless of what the portfolio did. I would rather follow the portfolio performance, even if that means I have to tighten the belt occasionally.

 

[Davis65]

The 4% was based, as others have noted, on the original portfolio, with adjustments to the withdrawal amount for inflation. There is nothing wrong with calculating your withdrawal on the current value of the portfolio, but 4% has no meaning or basis in this approach. 4% would be a random or arbitrary percentage. Your withdrawal percentage in this case should be related to your income from the portfolio, not a figure taken from a different methodology.

 

[Independent]

Well, I retired 4 years ago and if I stick with this rule I'd be pulling (wild guess) something slightly over 3% today (of the original retirement date portfolio) since portfolio has done well since then. I've always wondered....wouldn't it be legitimate to recalculate the 4% on TODAY's value? Or does this "rule" require the discipline to base today's pull based on the 4 year old portfolio to account for the fact that bad years ahead must be accounted for? After all, just because I retired 4 years ago am I sentenced to a lower income of the same portfolio value as someone retiring today? Going forward we both face the same future, eh?

I'd think that's ok to adjust up since if it was the other way around (4 bad years) I'd be adjusting the % down; just couldn't stomach not adjusting after bad performance.

Suppose your original 4% withdrawal rate was based on a 95% confidence rate. Your first 4 years have been good. Suppose that when we look historically (or at Monte Carlo simulations) and select only the paths that started with results as strong as you've seen in the last 4 years. We'll probably see a 100% success rate for those paths, using the original withdrawal amount throughout.

So, if you stick with your original (inflation adjusted) dollar amount, you seem to have moved from a 95% success plan to a 100%* success plan.

You can take that course. Or, you can increase your withdrawal amount to put yourself back into a 95% success situation.

This may seem too vague, I'll give an example. When I looked at FireCalc, I saw one year (maybe 1960) when the following 5 years were very strong. A retiree staying with the original withdrawal amount did just fine. But, a retiree ratcheting up to the maximum 95%, 25-year, withdrawal amount in 1965 would have run out of money (because the years soon after 1965 were poor enough that 1965 was one of the 5% failures).

Fortunately, it's not a matter of staying exactly where you started or going up to the 95% withdrawal rate. There are intermediate points that most people would likely find more comfortable. In fact, if you originally planned for 30 years, and you are now convinced you only need to plan for 26, you might find that your original 4% is 100% successful over 26 years.


* sure, there is no such thing as 100% looking forward. We can only talk about 100% looking backward.

Edit: Oops, I see that Animorph has already covered this.

 

[H2ODude]

Agree with all who have posted since mine. It's apparent that adjustment is OK unless you take the position that if you're retiring after good market performance, perhaps that's not a good time to peg your withdrawals at 4, since the bear lurks. But that's MARKET TIMING! My personal strategy is to adjust annually and be flexible. Needing less that 3% helps a lot, I wouldn't want to be counting on 4+% for bare essentials unless I was already in my 70's. In the end, I believe it's best to use many tools, be aware and know your risks. I do think a rigid aversion to increasing as the portfolio grows ignores the fact that most of the models are based on worst case or close to that (in the early years). So if the worst didn't happen, why ignore your good fortune? Party on!

 

[Chuckanut]

Needing less that 3% helps a lot, I wouldn't want to be counting on 4+% for bare essentials unless I was already in my 70's.

<snip>

I do think a rigid aversion to increasing as the portfolio grows ignores the fact that most of the models are based on worst case or close to that (in the early years). So if the worst didn't happen, why ignore your good fortune? Party on!

+1

 

[Animorph]

And the SWR also increases as the number of years your portfolio has to last declines.

 

[tuixiu]

Danger using the term "net worth" depending on definition - your home equity, for example, shouldn't enter into it.

Depends.

We knew we were going to sell our home when we retired to travel for awhile then become renters, so it made sense to include the value of the home in our net worth.

When our SWR applied to net worth could support anticipated retirement expenses (which includes housing) we knew we'd hit our number.

 

[big-papa]

Some of us don't use the initial portfolio, and don't adjust for published inflation. Instead we use the value of the portfolio at the end of each year. However, this means that some years, if the portfolio drops, we take a cut in income, which is what the original scheme was set up to avoid. If you are comfortable with the occasional cut in income, this is a viable approach. Personally, I was never comfortable with increasing my income by inflation regardless of what the portfolio did. I would rather follow the portfolio performance, even if that means I have to tighten the belt occasionally.

Yep - that's another method with many variants. As I noted in my response, every method has positives and negatives. The positive for
this one is that since you're always withdrawing a percentage of what is in the portfolio now, mathematically, you can not run out of money. But there is a nonzero probability that you may have withdrawals that are below what you need to live on, unless you have another steady source of income (SS, pension, etc.)

Variants include Scott Burns 6%/95% suggestion. Withdraw the larger of 6% of what is in your portfolio or 95% of last year's withdrawal, whichever is larger. Neither of those percentages are cast in stone. But what the method does is limit year-to-year drops in withdrawals if there is an extended bear market.

Another is Guyton-Klinger which has a large set of decision rules that set any given year's % withdrawal from your portfolio.

Then there are other variants which increase the percentage withdrawn year-on-year until the last one is 100%. The RMD method the IRS uses does this as does VPW.

Me? I like a combination. I like a 4%-like rule (with the percentage set much lower than 4%) to provide a steady, inflation adjusted income with a variable added that gives me a bonus when the portfolio does well.

Big-Papa

 

[big-papa]

+1 The first year withdrawal is adjusted for inflation, not the percentage. So if you have $5m when you retire it would be $200k the first year and $206k the second year (assuming 3% inflation), $212k the third year, etc.

Also, 4% is probably ok for someone retiring today at age 65, but a slightly lower % would be prudent, particularly if one is retiring significantly earlier than 65.

Not sure if my clarification made it through, but the percentage does increase, but it's the percentage based on your original portfolio, not what's left in it. Mathematically, it's identical to multipllying the previous year's withdrawal by the previous year's inflation.
Example:
$5M portfolio
4% rule
3% inflation
Year 1: $200,000
Year 2: $206,000 (or $200,000 * 1.03)
Year 3: $212,180 (or R206,000 * 1.03)
Year 4: $218,454 ($212,180 * 1.03)
etc.

$5M portfolio
4% rule
3% inflation
Year 1: 4%*5M = $200,000
Year 2: (4*1.03)%*5M = 4.12%*5M= $206,000
Year 3: (4.12*1.03)%*5M = 4.2436%*5M = $212180
Year 4: (4.2436*1.03)%*5M = 4.3709%*5M =$218454

Big-Papa