4% rule. Leaving money on the table?

[smitcat]

It's neither optimistic nor pessimistic; they are raw statistics. Notice the words "average," "2/3rds of the time," and "median." These are indisputable statements on the final data run through every possible year of retirement.

The part below the line is what concerns me. Losing much of your stash in the first 15 years of retirement. This appears to be NOT inflation adjusted. Supporting my FireCalc findings that the 4%er lost 80%, in real terms, before things got better late in old age. If they lived that long. Again, an 80% drawdown isn’t my definition of “4% worked”. It's a Pascal's Wager. Very likely a 4%er will do very well. But the consequences are huge (at least a partial financially failed retirement) if the 1966-81 type risk shows up.

"But the consequences are huge (at least a partial financially failed retirement) if the 1966-81 type risk shows up."
Not of any or a combination of these actions are employed......
- back off spending 4%+ inflation each year if/when markets dip
- when a large % of income is fixed and has cola (SS. annuity, pension)
- when you consider the 'smile' use of funds in retirement posted above
- when you lay these financial odds against the odds of your life cycle (length) and solve for 'happiness'

[Leesbro63]

The part below the line is what concerns me. Losing much of your stash in the first 15 years of retirement. This appears to be NOT inflation adjusted. Supporting my FireCalc findings that the 4%er lost 80%, in real terms, before things got better late in old age. If they lived that long. Again, an 80% drawdown isn’t my definition of “4% worked”. It's a Pascal's Wager. Very likely a 4%er will do very well. But the consequences are huge (at least a partial financially failed retirement) if the 1966-81 type risk shows up.

"But the consequences are huge (at least a partial financially failed retirement) if the 1966-81 type risk shows up."
Not of any or a combination of these actions are employed......
- back off spending 4%+ inflation each year if/when markets dip
- when a large % of income is fixed and has cola (SS. annuity, pension)
- when you consider the 'smile' use of funds in retirement posted above
- when you lay these financial odds against the odds of your life cycle (length) and solve for 'happiness'

I pretty much agree with this. I’m just saying that I think the risk of a 1966 retirement is too often dismissed to the category of asteroid risk. Something you live with and try to ignore because you can’t do anything about it. But I think a 1966 risk is much greater than asteroid risk. And, as you point out, there are SOME things you can do about it. Like doing lower than 4% for large taxable portfolios and perhaps TIPS for many other types of portfolios.

[BrooklynInvest]

To answer your question about budgeting, the money flows out of the investments every qtr into the cash account bucket and from the cash now earning 3.75% to the bank account savings. From savings to checking for bills. So regardless if I draw from the investment or not, unless I cut the budget, that full amount flows from the cash reserve. Therefore, income is stable.

I set a floor that can be adjusted over time based on expenses. For example, if I start out with 1 million (in our case more) the floor is 1 million. Therefore, when the investments go below, say 950k. Nothing is taken from the investments 0%. The withdraw lowers the cash reserve balance.


John

Got it. Got it. So the source of the withdrawal varies based on your rules but the amount doesn't... but could. Thank you.

[smitcat]

"But the consequences are huge (at least a partial financially failed retirement) if the 1966-81 type risk shows up."
Not of any or a combination of these actions are employed......
- back off spending 4%+ inflation each year if/when markets dip
- when a large % of income is fixed and has cola (SS. annuity, pension)
- when you consider the 'smile' use of funds in retirement posted above
- when you lay these financial odds against the odds of your life cycle (length) and solve for 'happiness'

I pretty much agree with this. I’m just saying that I think the risk of a 1966 retirement is too often dismissed to the category of asteroid risk. Something you live with and try to ignore because you can’t do anything about it. But I think a 1966 risk is much greater than asteroid risk. And, as you point out, there are SOME things you can do about it. Like doing lower than 4% for large taxable portfolios and perhaps TIPS for many other types of portfolios.

Just one rhetorical example of perhaps a potential few dozen.....
If we get $100K from SS and draw $100K from investments and need to cut back by a large 20% what is the actual/real change?
What are the real changes we need to make for that year(s) ....is it back to $180K? Or is it really less than that after considering taxes?
What real impact does this make in our plans/life/happiness compared to waiting years to have XX which can still be challenged as 'enough'?

[HMSVictory]

Depending on the age you retire you could be leaving a TON of money on the table or depleting the entire account.

Compare the retirement cash flow needs of someone retiring at 42 vs 68 (claiming full SS). Its all very dependent on your particular situation.

This is why I highly recommend using retirement planning software like Boldin. You can model nearly anything.

[tibbitts]

As I understand the 4% rule you would be living off of your nest egg interest

Ex: $1M = $40K annually

If the principal (Nest Egg) never decreases, is the expectation that you die and leave the Nest Egg to someone?


Are there any other mythologies that include trying to leave the world with a zero balance?

I think you would have been better off if, instead of all these responses, we had just said that your understanding is completely wrong and you need to learn more about the "rule" and what it implies (or not.)

[dcabler]

It's neither optimistic nor pessimistic; they are raw statistics. Notice the words "average," "2/3rds of the time," and "median." These are indisputable statements on the final data run through every possible year of retirement.

The part below the line is what concerns me. Losing much of your stash in the first 15 years of retirement. This appears to be NOT inflation adjusted. Supporting my FireCalc findings that the 4%er lost 80%, in real terms, before things got better late in old age. If they lived that long. Again, an 80% drawdown isn’t my definition of “4% worked”. It's a Pascal's Wager. Very likely a 4%er will do very well. But the consequences are huge (at least a partial financially failed retirement) if the 1966-81 type risk shows up.

I'm not sure how we can even say "very likely". This is 100% backwards looking. How can we possibly know that any actual, future sequence of return will be encompassed in the results of past sequences of return that led to the level of success that 4% (or any other number) might have had?

Cheers

[smitcat]

The part below the line is what concerns me. Losing much of your stash in the first 15 years of retirement. This appears to be NOT inflation adjusted. Supporting my FireCalc findings that the 4%er lost 80%, in real terms, before things got better late in old age. If they lived that long. Again, an 80% drawdown isn’t my definition of “4% worked”. It's a Pascal's Wager. Very likely a 4%er will do very well. But the consequences are huge (at least a partial financially failed retirement) if the 1966-81 type risk shows up.

I'm not sure how we can even say "very likely". This is 100% backwards looking. How can we possibly know that any actual, future sequence of return will be encompassed in the results of past sequences of return that led to the level of success that 4% (or any other number) might have had?

Cheers

If you do not use history as a general guide ....what do you substitue for the guidelines?

[dcabler]

I'm not sure how we can even say "very likely". This is 100% backwards looking. How can we possibly know that any actual, future sequence of return will be encompassed in the results of past sequences of return that led to the level of success that 4% (or any other number) might have had?

Cheers

If you do not use history as a general guide ....what do you substitue for the guidelines?

It's one thing to use history as a very general guide, which I think is unavoidable. It's another to use historical statistics to apply to the future. This is not physics - at least not the type of physics where the analysis is tractable.

Cheers.

[smitcat]

If you do not use history as a general guide ....what do you substitue for the guidelines?

It's one thing to use history as a very general guide, which I think is unavoidable. It's another to use historical statistics to apply to the future. This is not physics - at least not the type of physics where the analysis is tractable.

Cheers.

Please add more detail as to how to use historical statistics without applying them to the future.
What do you use to apply for the future?

[Leesbro63]

If you do not use history as a general guide ....what do you substitue for the guidelines?

It's one thing to use history as a very general guide, which I think is unavoidable. It's another to use historical statistics to apply to the future. This is not physics - at least not the type of physics where the analysis is tractable.

Cheers.

We are on the same side of this argument from two different angles. You are saying that history cannot be trusted. I'm saying that even if we trust history, the 1966-1981 historical period seems to be incorrectly minimalized as merely a "one off". Either way, it appears to me that the 4%ers are at greater risk than they believe.

[Scorpion Stare]

Life expectancy was far lower back then. The 65 year old would be lucky to make it 70.

That”s a bit of an exaggeration. Life expectancy for a 65-year-old American in 1970 was 15.2 years (source). In 2021 it was 18.4 years (source). Average remaining lifespan for a 65-year-old increased by about 3 years over those five decades.

[rockstar]

Life expectancy was far lower back then. The 65 year old would be lucky to make it 70.

That”s a bit of an exaggeration. Life expectancy for a 65-year-old American in 1970 was 15.2 years (source). In 2021 it was 18.4 years (source). Average remaining lifespan for a 65-year-old increased by about 3 years over those five decades.

A little. But I know more people that died before 80 than those that made it into their 90s. Folks here have an unrealistic expectation for how long they’ll live.

[bonesly]

What do you use to apply for the future?

To dcabler's point, nobody knows, so it's all guesswork-at-risk.

I don't like using the historical sequences compared to Monte Carlo modeling of the historical distribution since picking a random start date for a sequence, or even picking a random year from the small (from a statistical perspective) data set of about 90 years of returns, always results in the same worst case. A random draw from the distribution does not have that limitation as noted HERE, and choosing your risk as a low percentile (1st, 5th, 10th, 20th, etc.) seems more insightful that "what was the worst over the last 90 years," because that assumes the future worst case will never be worse than the historical worst case, but distribution modeling says there's a low but non-zero chance it could be worse (-60% vs -43.8%).

[itnetpro]

I think everyone is way overthinking this!

The 4% rule, in my opinion, should simply be used for a starting point to benchmark a variety of methods to include something more customized to specific needs, there are two many variables in each investors life to suggest any one method is better/worse than the other.

In general, the loose interpretation of the 4% rule could work just fine for those with a higher tolerance for risk and maybe not so well for the more conservative investor.

I would venture a guess, anyone who frequents these forums and has above average assets to consider the 4% rule, is smart enough to determine what best fits their needs.

Why are we telling anyone what’s right or wrong for them? Perhaps it’s better to make sure that the OP understands the rule with no judgement and encourage an examination into all methodologies to find what fits best.

My personal withdraw method relied on the 4% rule heavily for benchmarking what works best for our needs…

John

[nisiprius]

It's an insoluble problem for an individual investor who is unwilling to use insurance products.

It's just another risk/return tradeoff. The higher the withdrawal rate, the higher the historical chance of running out of money. The lower the rate, the safer--but with less available to you to support your standard of living, and an a larger amount of money left on the table. The 4% rule historically would have both had a 5% failure rate and also would on the average would still have left a lot of money on the table.

The only way out is to form a pool of a large enough group of investors to have the law of averages work. This is basically insurance companies do, and various studies suggest that they rake off something like 2-10% of the premium for themselves. But in the days when inflation-adjusted annuities were available, it was a clean solution to the problem--you could get that 4% and maybe a bit more, inflation-adjusted. By design you left no money on the table. And the tradeoff was that you indeed left no money on the table--no money for your family or other legatees.

[smitcat]

What do you use to apply for the future?

To dcabler's point, nobody knows, so it's all guesswork-at-risk.

I don't like using the historical sequences compared to Monte Carlo modeling of the historical distribution since picking a random start date for a sequence, or even picking a random year from the small (from a statistical perspective) data set of about 90 years of returns, always results in the same worst case. A random draw from the distribution does not have that limitation as noted HERE, and choosing your risk as a low percentile (1st, 5th, 10th, 20th, etc.) seems more insightful that "what was the worst over the last 90 years," because that assumes the future worst case will never be worse than the historical worst case, but distribution modeling says there's a low but non-zero chance it could be worse (-60% vs -43.8%).

If you are saying it's just another application of past history 'padded' for a potential worse scenario then fine.
The basis is becomes past history.

[rakish_weasel]

I am targeting 2.75%, and I hope to teach my son to follow the same.

The perpetual withdrawal rate isn't that far from the 4% rule.

Mr Bogle himself says in his books that your dividend/yield checks are really your baseline budget, and this is much closer to the perpetual withdrawal philosophy.

Why not leave something to your family or a charity while ensuring perpetual access to the cash flow for yourself?

To me the perpetual withdrawal rate is the single most important future in all of BH style investing!

I don't disagree with this in theory -- the main challenge with this extra-conservative approach is that saving a sufficient amount of money, particularly if one is not an HCE or equivalent, is that it would likely require an extremely frugal lifestyle over many decades to achieve. Not that there's anything wrong with this in the least. That said, there are going to be many folks who would probably view that tradeoff as tilted a bit too far towards the "delayed gratification" end of the scale, in terms of balancing future/retired financial security against enjoying life in the here & now.

If I sound like I'm disagreeing with you, I'm not intending that -- but 2.75% is quite a lean WR. If one has a multi-decade career that can accommodate that (few employment gaps, high income/salary, manageable tax situation, healthy parents, etc.) then I'm all for it. But I do think that, say, 3.25% (which essentially tracks to 99% success in most all scenarios) or even 3.5% would be plenty safe.

As always, though, personal finance being exactly that -- personal -- each of us will have a different definition of what makes us comfortable re: retirement drawdowns. If I can ever reach the point where 2.75% WR gets me my anticipated retirement-years expenses, I'll be over the moon -- on the other hand, I doubt I would work an extra 4-5 years in my 50s/60s, or however long that might take, to achieve that WR instead of my rough plan of 3.25%.

-rw

[dcabler]

What do you use to apply for the future?

To dcabler's point, nobody knows, so it's all guesswork-at-risk.

I don't like using the historical sequences compared to Monte Carlo modeling of the historical distribution since picking a random start date for a sequence, or even picking a random year from the small (from a statistical perspective) data set of about 90 years of returns, always results in the same worst case. A random draw from the distribution does not have that limitation as noted HERE, and choosing your risk as a low percentile (1st, 5th, 10th, 20th, etc.) seems more insightful that "what was the worst over the last 90 years," because that assumes the future worst case will never be worse than the historical worst case, but distribution modeling says there's a low but non-zero chance it could be worse (-60% vs -43.8%).

Unfortunately, the future is under no obligation to have the same returns distribution that the past did, either.

Cheers.

[bonesly]

If you are saying it's just another application of past history 'padded' for a potential worse scenario then fine.
The basis is becomes past history.

It's not "padded," it's just that a random sample from the historical distribution of returns is a more accurate model than a random sample from the array of historical returns. Yet a lot of the Monte Carlo models out there use a random index into the array (Portfolio Visualizer, FiCalc, FireCalc all use an index by default, but PV can use a distribution which is why I like it best), or they don't pick each year's return at random from the array and instead just choose a random starting year and run a 30y period (which is likely worse from a statistical perspective than a random index for each year as there are fewer 30y rolling periods so an even smaller sample size to make inferences from).

Unfortunately, the future is under no obligation to have the same returns distribution that the past did, either.

"All models are wrong, some are still useful" -- George Box (famous statistician)

I think a model that produces the exact same worst case on 1,00 trials is less useful than one that produces a different worst case on each set of 1,000 "random" draws, especially if you're concerned about 1st or 5th percentile results (rather than actual worst case). However, I also think "max drawdown" is a misleading metric that a lot of folks here have latched onto as very important.

At any rate, I fully agree with you that "the future is under no obligation to have the same returns distribution that the past".

[dcabler]

It's an insoluble problem for an individual investor who is unwilling to use insurance products.

It's just another risk/return tradeoff. The higher the withdrawal rate, the higher the historical chance of running out of money. The lower the rate, the safer--but with less available to you to support your standard of living, and an a larger amount of money left on the table. The 4% rule historically would have both had a 5% failure rate and also would on the average would still have left a lot of money on the table.

The only way out is to form a pool of a large enough group of investors to have the law of averages work. This is basically insurance companies do, and various studies suggest that they rake off something like 2-10% of the premium for themselves. But in the days when inflation-adjusted annuities were available, it was a clean solution to the problem--you could get that 4% and maybe a bit more, inflation-adjusted. By design you left no money on the table. And the tradeoff was that you indeed left no money on the table--no money for your family or other legatees.

Yep, insurance products can be part of the solution, even though there are no truly inflation adjusted annuity protects available today.

But I think a broader way of thinking about it is to at least consider something that has some resiliency should certain risks materialize, instead of trying to guess the probability that those risks might materialize. By "resilient", however, this doesn't necessarily mean that you are unaffected by it if it happens. And it doesn't mean that the final outcome is better should those risks never materialize. Sometimes resiliency has a cost.

And it's not about being able to address every materialized black swan event that has happened in history or any future black swan event that might only be limited by your imagination. It's not even about being able to address all bad events that might not rise up to being classified as black swans. It's just about at least considering those things that, relatively speaking, might improve outcomes if you are willing to pay the price to use them. If you're not willing to pay that price or, perhaps, are unable to do so, then I think there is at least some value in having thoughtfully considered them.

Regarding withdrawing from a portfolio, one thing that many have brought up on the forum are withdrawal methods based on amortization. One of the methods discussed on this forum is VPW. VPW, like all of the amortization methods discussed on this forum can be set to have a $0 terminal portfolio value at some age/year. VPW internally uses a fixed rate for amortization that is based on long term trend returns of stocks and of bonds. Again, the future is under no obligation to follow this trend, so if your portfolio's returns stay below this trend for a long period of time, your withdrawal values will drop. But your portfolio will still last exactly as long as you planned for it to last. In that sense, compared to something like the 4% rule, you have mitigated against the possibility of prematurely emptying your portfolio before the age/date you planned. Within VPW itself, mitigation against poor outcomes can be made, such as setting a nonzero terminal portfolio value and/or extending the year/age farther out. The cost of doing these is lower overall withdrawals.

Regarding a source of income, another simple example is if considering an insurance product, is to realize that even insurance companies can fail. And state protections are limited. The advice given on this forum, which might apply, would be to spread the risk across multiple insurance companies. I know of no way to know how many insurance companies would be "enough". But I have to believe that using 2 insurance companies has less risk than using only 1. Or that 3 has less risk than 2. But I also can't help but believe that a total collapse of this industry would have ramifications far beyond the loss of the annuities themselves. So, yes, using an insurance product is another way to add resilience to your income stream via risk pooling. And besides having other sources of income, adding resilience to this source of income might be to spread the risk of insurance company default by purchasing annuities from multiples companies.

TIPS ladders and I-bonds are other products that are brought up as well as ways to mitigate against inflation risk.
Not to mention SS or Pensions.

That point is that I'm not sure how one can look at the past results and state that if one follows the same way of generating spendable income, that one can state what the probability of a bad outcome is going to be going forward. But if one chooses to do so, there are ways to construct things that will spread the risk around and at least reduce the probability of whatever you consider a bad outcome to be, even if you can't know by how much in advance and which might not come for free.

Cheers.

[dcabler]

It's one thing to use history as a very general guide, which I think is unavoidable. It's another to use historical statistics to apply to the future. This is not physics - at least not the type of physics where the analysis is tractable.

Cheers.

Please add more detail as to how to use historical statistics without applying them to the future.
What do you use to apply for the future?

I spent probably a decade looking at withdrawal methods and how to structure things before landing on what I'm actually doing now that I'm retired. And I can backtest with the best of them! I have no idea how many spreadsheets I have sitting on my hard drive for the seemingly endless simulations and experiments I've run over the years.

For me, it was a combination of thinking about relative risk, how risk manifests itself and whether I can change that, and whether I had "enough" to retire all the while trying my best to minimize relying on past results.

Like many, I started by looking at SWR. I also looked at what I consider to be "hacks" to SWR to try and improve outcomes. And there are many, from Guyton-Klinger decision rules, Kitces ratcheting method, Gummy's sensible withdrawals, Hebeler's autopilot, etc. The problem to me wasn't that using a lower SWR would reduce the odds of failure or that any of these hacks might improve outcomes. Rather the problem to me was the nature of failure itself - that is, running out of money before I planned. And I didn't want to rely on anything that wasn't systematic to avoid running out of money.

I briefly looked at a fixed % of remaining of portfolio since such a method would theoretically last forever. But that required me to figure out a percentage. And I had no way to do that without directly using past statistics and perhaps adding some padding to it.  

I then discovered what was originally called "the actuarial method" which is amortization. First with a paper that discussed a method called ARVA by Waring and Siegel, then VPW from longinvest, and finally a 2 part series on the Bogleheads blog from Siamond. My own withdrawal spreadsheet is heavily influenced by Siamond's blog post. Later on, ABW and finally TPAW from Ben Mathew also appeared on bogleheads. What attracted me to amortization methods was that instead of all of the risk piling up in the very last withdrawal like 4%/SWR does, it spreads it out over all withdrawals, resulting in withdrawals that vary. So instead of a risk of being fully depleted, which I would consider to be a disaster, the risk is now that any single withdrawal might be too low to keep the lights on.

Note: amortization withdrawals are based on the same math used for loan calculations. And these calculations require a "rate". For withdrawals, the rate is a real rate of return. VPW, for example, uses fixed numbers based on long term worldwide real returns for stock and bonds. Other methods might use some sort of predicted future returns based on CAPE, or what's published by various research firms and investment houses. When calculating withdrawals, you always use the latest predictions. Note: this is definitely a number that is one way or another extracted from historical returns. Historically, at least, it doesn't appear that there has been a huge sensitivity to the rate chosen due to the fact that there is at least a little bit of self-correction in the way the math operates. Future? Who knows.

After thinking on this more, I ultimately decided that I wanted to start everything with a base of fixed, real income. What that means is that I'm relying on what is effectively a TIPS ladder. 3 ladders, actually: 2 SS bridges and 1 that lasts until we're in our mid 90's. This covers the vast majority of our nondiscretionary spending. This plus distributions from our stock and a small withdrawal from our stock will cover all of our nondiscretionary spending. It would really require an unprecedented event for the amortization based withdrawal from stocks to ever be so low that, when added to the other sources of fixed income would be such that we'd have to scramble to make up the difference.

The decision to retire always started with running my amortization spreadsheet and looking at the results to see if what popped out was "enough" to pay our nondiscretionary expenses with sufficient upside for lumpy and discretionary expenses.

Then there are contingencies
- We have I-bonds that start to mature in our mid 80's. Here the current plan is that if good health is smiling on at least one of us, we'll consider purchasing SPIAs with this money. Or perhaps it can be used to supplement long term care expenses or even extend the TIPS ladder. The point is we have options.
- We have sufficient discretionary spending capacity such that if the current projected SS shortfall is realized as an actual benefits reduction, we can pull from that to make up the difference. We aren't even coming close to spending the full amount we could today from the discretionary bucket, so we just keep what we don't use invested.

Bottom line is that there's no escaping looking at the past in some way. And for us that means
- The source of the "rate" for amortization, though the withdrawals seem fairly insensitive to this having a pretty wide range - at least for our needs
- The fact that what we do need to withdraw from stocks for nondiscretionary spending is fairly small - small enough that if this withdrawal was too small to complete what we need for spending, we'd probably just take the extra out from discretionary and keep going if that ever happened.

Personal finance is, as they say, personal. This is just an example of how we chose to put things together based on the risks we care most about and how our portfolio progressed during our accumulation years. The future, as they say, is uncertain and my imagination can easily come up with scenarios that can break this - and none of them involve an asteroid. The price to pay to mitigate those scenarios is more than we're willing to pay.

Cheers.

[sf_tech_saver]

I am targeting 2.75%, and I hope to teach my son to follow the same.

The perpetual withdrawal rate isn't that far from the 4% rule.

Mr Bogle himself says in his books that your dividend/yield checks are really your baseline budget, and this is much closer to the perpetual withdrawal philosophy.

Why not leave something to your family or a charity while ensuring perpetual access to the cash flow for yourself?

To me the perpetual withdrawal rate is the single most important future in all of BH style investing!

I don't disagree with this in theory -- the main challenge with this extra-conservative approach is that saving a sufficient amount of money, particularly if one is not an HCE or equivalent, is that it would likely require an extremely frugal lifestyle over many decades to achieve. Not that there's anything wrong with this in the least. That said, there are going to be many folks who would probably view that tradeoff as tilted a bit too far towards the "delayed gratification" end of the scale, in terms of balancing future/retired financial security against enjoying life in the here & now.

If I sound like I'm disagreeing with you, I'm not intending that -- but 2.75% is quite a lean WR. If one has a multi-decade career that can accommodate that (few employment gaps, high income/salary, manageable tax situation, healthy parents, etc.) then I'm all for it. But I do think that, say, 3.25% (which essentially tracks to 99% success in most all scenarios) or even 3.5% would be plenty safe.

As always, though, personal finance being exactly that -- personal -- each of us will have a different definition of what makes us comfortable re: retirement drawdowns. If I can ever reach the point where 2.75% WR gets me my anticipated retirement-years expenses, I'll be over the moon -- on the other hand, I doubt I would work an extra 4-5 years in my 50s/60s, or however long that might take, to achieve that WR instead of my rough plan of 3.25%.

-rw

This is all fair, and it raises one mindset shift as you scale the size of your portfolio.

There is a subtle shift between retirement and wealth management once you cross, say, $7M in your portfolio (to throw a crude number out there).

I've frankly gotten lucky to work at a few tech companies where the paydays gave me this situation -- but what I love about cranking my target withdrawal rate down is that even though I recently broke the 'magical' $10M, when I see that is merely a $275k a year cashflow before taxes it keeps me hungry and frugal.

[kd2008]

....what I love about cranking my target withdrawal rate down is that even though I recently broke the 'magical' $10M, when I see that is merely a $275k a year cashflow before taxes it keeps me hungry and frugal.

You know better than anyone that this is just score keeping. Money you are earning has no actual utility. Congratulations on your success.

[Mr. Rumples]

The 4% "rule" was useful getting to retirement; however, now there I use a different method. It may work, it may not but I am looking at a 7% withdrawal rate this year, but the amount flexes.

I take the current year end balance (not including emergency and other funds) the five year average inflation rate and my (not a generic) average five year return then see how much I can spend with my remaining years - I use 100 as my final age. Leftover $ is fine.

I do it with this yearly: https://www.mycalculators.com/ca/retcalc2m.html

Thus if I am 92, I would use eight as my number of years in retirement, the average five inflation rate is about 4.3% and my return five year average is 10.41%. Insert the money and see what comes up.

[sf_tech_saver]

....what I love about cranking my target withdrawal rate down is that even though I recently broke the 'magical' $10M, when I see that is merely a $275k a year cashflow before taxes it keeps me hungry and frugal.

You know better than anyone that this is just score keeping. Money you are earning has no actual utility. Congratulations on your success.

Thank you kindly.

Score keeping? Maybe so, but as somebody living in a city where the homes in nice neighborhoods routinely go for $5-7M its helped me stay content in my nice 2-bedroom 1,400 sq foot condo ($1.9M) vs. imagining I can easily afford a $6M house and all of the costs that go with it!

The BH philosophy helped keep me grounded and I count on it to continue to do that. If I keep working now, it's primarily to pay cash, etc, for a slightly larger home for my family. That's the remaining utility.

[Saintor]

Who has the best crystal ball?

If one invested 1 000 000 in S&P500 until 1999 and started pulling 50 000/yr inflation-adjusted since then would have run out of money...

Then 4% looks better... of course, many scenarios would have prevented going out of money.

[abuss368]

You know better than anyone that this is just score keeping. Money you are earning has no actual utility. Congratulations on your success.

Thank you kindly.

Score keeping? Maybe so, but as somebody living in a city where the homes in nice neighborhoods routinely go for $5-7M its helped me stay content in my nice 2-bedroom 1,400 sq foot condo ($1.9M) vs. imagining I can easily afford a $6M house and all of the costs that go with it!

The BH philosophy helped keep me grounded and I count on it to continue to do that. If I keep working now, it's primarily to pay cash, etc, for a slightly larger home for my family. That's the remaining utility.

Hi sf_tech_investor -

Your post speaks to me and it has impact.

I appreciate your financial strategy and it makes a lot of sense.

I too am trying to do this with dividends only. Jack Bogle always called this “income risk” and “dividend checks in a mailbox”. Love those interviews (on YouTube).

That you can do this with Total Stock (VTI) and US High Dividend (VYM) is admirable. The yields are much lower than International stocks.

Best.
Tony

[MikeWillRetire]

Are there any other mythologies that include trying to leave the world with a zero balance?

Nobody actually uses the 4% rule. It’s a rough planning guide……

I am one of those that uses the 4% rule. Fortunately, I have a pension and SS, and that covers all of my base expenses. My 401k is for my fun money, and I use the 4% inflation adjusted withdrawal for it.

[Cletus Davenport]

Nobody actually uses the 4% rule. It’s a rough planning guide……

I am one of those that uses the 4% rule. Fortunately, I have a pension and SS, and that covers all of my base expenses. My 401k is for my fun money, and I use the 4% inflation adjusted withdrawal for it.

So that’s cool! How long have you been doing it, and how do you like it? What do you think of the criticisms or the downsides of this?

[sf_tech_saver]

Thank you kindly.

Score keeping? Maybe so, but as somebody living in a city where the homes in nice neighborhoods routinely go for $5-7M its helped me stay content in my nice 2-bedroom 1,400 sq foot condo ($1.9M) vs. imagining I can easily afford a $6M house and all of the costs that go with it!

The BH philosophy helped keep me grounded and I count on it to continue to do that. If I keep working now, it's primarily to pay cash, etc, for a slightly larger home for my family. That's the remaining utility.

Hi sf_tech_investor -

Your post speaks to me and it has impact.

I appreciate your financial strategy and it makes a lot of sense.

I too am trying to do this with dividends only. Jack Bogle always called this “income risk” and “dividend checks in a mailbox”. Love those interviews (on YouTube).

That you can do this with Total Stock (VTI) and US High Dividend (VYM) is admirable. The yields are much lower than International stocks.

Best.
Tony

Hi Tony,

You are a diligent researcher and investor as always. Hope you have been well.

One other way I'm supplementing my investment income this year is buying California Municipal Bonds. Given my tax rate on dividends in this state it was the only way I could start to achieve my goals of living entirely off of my investment income over time (as a baseline). It's helpful that the rates of VCLAX are now 3.5% after taxes and climbing!

I've been using new investments and my income to move from a 90/10 where living off income is harder as you note to hopefully a 70/30 by the end of this year.

Love that Jack image of going to your dividend check at the mailbox its one I've never forgotten as well! Being able to keep the lights on and feed myself off of that has been a focus vs. the probabilistic world of the 4% rule etc.

[trueblueky]

As I understand the 4% rule you would be living off of your nest egg interest

Ex: $1M = $40K annually

If the principal (Nest Egg) never decreases, is the expectation that you die and leave the Nest Egg to someone?


Are there any other mythologies that include trying to leave the world with a zero balance?

Nobody actually uses the 4% rule. It’s a rough planning guide……

But if you did follow it rigidly in the past, and increase your annual withdrawals to account for inflation, then you’d have a very good chance of dying with a lot more than $1M inflation adjusted dollars. This is one of the criticisms of the 4% rule……

Note. It’s for 30 years. Annual withdrawals are adjusted upwards for inflation every year, even if your portfolio is down for the year.

And in only 1 case would,you actually spend all the money.

So consider 4% a number to aim for when you’re a long way from retiring. And as you get closer, ignore it. Do more careful, well reasoned planning and analysis.

If somebody wanted to die with zero dollars, assuming they are single, they could sell all their possessions, and simply buy an annuity with all,of their money. When they die, the annuity payments stop, social security payments stop, pensions stop, etc. but nobody advocates that either…..

Buying an annuity doesn't get you to zero either unless you scrupulously spend the money each month.

[MikeWillRetire]

I am one of those that uses the 4% rule. Fortunately, I have a pension and SS, and that covers all of my base expenses. My 401k is for my fun money, and I use the 4% inflation adjusted withdrawal for it.

So that’s cool! How long have you been doing it, and how do you like it? What do you think of the criticisms or the downsides of this?

I have been retired for 2 years now, and it is working out very well. Since my pension + SS covers my base expenses, I don't think there is anything wrong with using a 4% withdrawal at my age (62).

[Cletus Davenport]

So that’s cool! How long have you been doing it, and how do you like it? What do you think of the criticisms or the downsides of this?

I have been retired for 2 years now, and it is working out very well. Since my pension + SS covers my base expenses, I don't think there is anything wrong with using a 4% withdrawal at my age (62).

Thanks for the information. I certainly don’t think there’s anything wrong with a 4% withdrawal rate, so I hope you don’t think I was criticizing.

I’m personally not worried about the future being worse than the past has ever been. So I’d be okay (in theory) with following the rule, but I just don’t think my spending (or desired spending) will be that consistent.

Enjoy your retirement!

[MikeWillRetire]

I have been retired for 2 years now, and it is working out very well. Since my pension + SS covers my base expenses, I don't think there is anything wrong with using a 4% withdrawal at my age (62).

Thanks for the information. I certainly don’t think there’s anything wrong with a 4% withdrawal rate, so I hope you don’t think I was criticizing.

I’m personally not worried about the future being worse than the past has ever been. So I’d be okay (in theory) with following the rule, but I just don’t think my spending (or desired spending) will be that consistent.

Enjoy your retirement!

Thank you! Good luck with your retirement planning!

[itnetpro]

If using Bengen's 4% rule of thumb guideline, bear in mind that taxes and all investment fees and expenses come directly out of that inflated 4%. The 4% is not what one can spend, it's the amount that can be drawn down. For example, if one were paying an AUM fee of 1% and active management expense ratios of 0.5%, this leaves only 2.5% of the portfolio value for spending and tax payments. Just another reason to always use low cost, low tax drag index funds.

I keep hearing about these excessive fees for actively managed funds and why index’s are better. I own two old very highly regarded actively managed funds I bought when I retired that I’m very happy with. They cost me .16 and .18. Over the years while comparing different actively managed funds, about the most expensive I came across for consideration was .48.

For me, I back tested plenty of equivalent mixes in index funds. They performed great! Often times about the same as mine for SLIGHTLY cheaper. It’s true, an actively managed fund won’t always beat an index over time and management does change. However, I did notice something interesting about a few actively managed funds. During big downturns like 2000 .com 2008 and 2022, I noticed, they seem to mitigate the downside a little better than indexes. In many cases shorter underwater periods.

Low tax drag? Irrelevant in a tax deferred account.

For a buy and hold person like me, that means something in retirement. So while index’s are certainly an excellent choice, actively managed funds are not necessarily bad!

Indexes are one right way to invest, decent low cost, old actively managed another! They retired me at 52…

John

[Rocinante Rider]

I did notice something interesting about a few actively managed funds. During big downturns like 2000 .com 2008 and 2022, I noticed, they seem to mitigate the downside a little better than indexes.

Some (many?) actively managed funds maintain cash positions. Thus, unlike an index fund that's in the same market, those actively managed funds are less than 100% invested. This gives those active funds an advantage during market downturns, but it's comparing apples and oranges. An index investor can always choose to allocate whatever amount s/he chooses to cash, etc. One doesn't need to use an active fund to minimize risk during down markets.

Active funds, like yours, with expense ratios of 0.16 or 0.18 have much less of a negative impact, especially if they're managed in a tax-efficient manner and have low portfolio turnover.

It's mathematically certain, however, that a dollar invested in an index fund will always outperform the average actively-managed dollar that's fully invested in the same market.
https://web.stanford.edu/~wfsharpe/art/ ... active.htm

[Moniker]

I spent probably a decade looking at withdrawal methods and how to structure things before landing on what I'm actually doing now that I'm retired. And I can backtest with the best of them! I have no idea how many spreadsheets I have sitting on my hard drive for the seemingly endless simulations and experiments I've run over the years.

For me, it was a combination of thinking about relative risk, how risk manifests itself and whether I can change that, and whether I had "enough" to retire all the while trying my best to minimize relying on past results.

Like many, I started by looking at SWR. I also looked at what I consider to be "hacks" to SWR to try and improve outcomes. And there are many, from Guyton-Klinger decision rules, Kitces ratcheting method, Gummy's sensible withdrawals, Hebeler's autopilot, etc. The problem to me wasn't that using a lower SWR would reduce the odds of failure or that any of these hacks might improve outcomes. Rather the problem to me was the nature of failure itself - that is, running out of money before I planned. And I didn't want to rely on anything that wasn't systematic to avoid running out of money.

I briefly looked at a fixed % of remaining of portfolio since such a method would theoretically last forever. But that required me to figure out a percentage. And I had no way to do that without directly using past statistics and perhaps adding some padding to it.  

I then discovered what was originally called "the actuarial method" which is amortization. First with a paper that discussed a method called ARVA by Waring and Siegel, then VPW from longinvest, and finally a 2 part series on the Bogleheads blog from Siamond. My own withdrawal spreadsheet is heavily influenced by Siamond's blog post. Later on, ABW and finally TPAW from Ben Mathew also appeared on bogleheads. What attracted me to amortization methods was that instead of all of the risk piling up in the very last withdrawal like 4%/SWR does, it spreads it out over all withdrawals, resulting in withdrawals that vary. So instead of a risk of being fully depleted, which I would consider to be a disaster, the risk is now that any single withdrawal might be too low to keep the lights on.

Note: amortization withdrawals are based on the same math used for loan calculations. And these calculations require a "rate". For withdrawals, the rate is a real rate of return. VPW, for example, uses fixed numbers based on long term worldwide real returns for stock and bonds. Other methods might use some sort of predicted future returns based on CAPE, or what's published by various research firms and investment houses. When calculating withdrawals, you always use the latest predictions. Note: this is definitely a number that is one way or another extracted from historical returns. Historically, at least, it doesn't appear that there has been a huge sensitivity to the rate chosen due to the fact that there is at least a little bit of self-correction in the way the math operates. Future? Who knows.

After thinking on this more, I ultimately decided that I wanted to start everything with a base of fixed, real income. What that means is that I'm relying on what is effectively a TIPS ladder. 3 ladders, actually: 2 SS bridges and 1 that lasts until we're in our mid 90's. This covers the vast majority of our nondiscretionary spending. This plus distributions from our stock and a small withdrawal from our stock will cover all of our nondiscretionary spending. It would really require an unprecedented event for the amortization based withdrawal from stocks to ever be so low that, when added to the other sources of fixed income would be such that we'd have to scramble to make up the difference.

The decision to retire always started with running my amortization spreadsheet and looking at the results to see if what popped out was "enough" to pay our nondiscretionary expenses with sufficient upside for lumpy and discretionary expenses.

Then there are contingencies
- We have I-bonds that start to mature in our mid 80's. Here the current plan is that if good health is smiling on at least one of us, we'll consider purchasing SPIAs with this money. Or perhaps it can be used to supplement long term care expenses or even extend the TIPS ladder. The point is we have options.
- We have sufficient discretionary spending capacity such that if the current projected SS shortfall is realized as an actual benefits reduction, we can pull from that to make up the difference. We aren't even coming close to spending the full amount we could today from the discretionary bucket, so we just keep what we don't use invested.

Bottom line is that there's no escaping looking at the past in some way. And for us that means
- The source of the "rate" for amortization, though the withdrawals seem fairly insensitive to this having a pretty wide range - at least for our needs
- The fact that what we do need to withdraw from stocks for nondiscretionary spending is fairly small - small enough that if this withdrawal was too small to complete what we need for spending, we'd probably just take the extra out from discretionary and keep going if that ever happened.

Personal finance is, as they say, personal. This is just an example of how we chose to put things together based on the risks we care most about and how our portfolio progressed during our accumulation years. The future, as they say, is uncertain and my imagination can easily come up with scenarios that can break this - and none of them involve an asteroid. The price to pay to mitigate those scenarios is more than we're willing to pay.

Cheers.

Thanks for sharing your thoughts. I have gravitated to a similar view on withdrawals. Amortization makes the most sense to me.