Ratchet up 4% SWR amount with portfolio growth?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

Marseille07 wrote: Fri Jan 28, 2022 10:54 am
willthrill81 wrote: Fri Jan 28, 2022 10:19 am
marcopolo wrote: Thu Jan 27, 2022 10:51 am My main point is that soon as you get to a situation where the success rate is less that 100%, ratcheting up increases risk of failure.
Yes, ratcheting up as the OP has described it, which is just 'restarting' the '4% rule', is no less safe than the '4% rule' itself. It's actually safer because the retiree presumably has fewer than 30 years of retirement remaining. But if the '4% rule' is in danger, so is ratcheting up.
This is not true. Suppose 1966 was a year that the 4% rule failed. The problem with ratcheting up is that it would cause earlier years (1963~1965) to fail, when the 4% rule itself would have worked for the 1963~1965 retirees.
Please keep in mind what I said: restarting the '4% rule' is no more dangerous than starting with the '4% rule'. If the '4% rule' worked for 30 years for 1966 retirees, and the Simba backtesting spreadsheet indicates that it actually did, though barely, then it means that prior year retirees restarting their withdrawals at no more than 4% of their current portfolio balance were in no worse shape.

Let's look at some real numbers from the Simba backtesting spreadsheet to illustrate this.

Let's say that a 1963 retiree had $1m in a portfolio with 60% TSM and 40% TBM. At the end of that year, the portfolio was up to $1,133,680, and with that year's 1.32% inflation, the amount withdrawn was $40,528, leaving $1,093,152 in the portfolio.

At the end of 1964, the portfolio's balance was $1,217,618, and with that year's 1.31% inflation, the amount withdrawn was $41,059, leaving $1,176,559 in the portfolio.

At the end of 1965, the portfolio's balance was $1,280,732.

After their inflation-adjusted withdrawals in 1963 and 1964, 1963 retirees had seen 28% nominal growth in their portfolio through the end of 1965. At the end of 1965, they could now restart the '4% rule' and withdraw $51,229, 4% of their current portfolio balance. Them withdrawing 4% of their current portfolio balance is no different at all from those who just retired in 1966 with $1,280,732 and started by withdrawing 4% of their portfolio balance, the same $51,229.
The Sensible Steward
User avatar
canadianbacon
Posts: 677
Joined: Sun Nov 10, 2019 9:04 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by canadianbacon »

Marseille07 wrote: Fri Jan 28, 2022 10:54 am
willthrill81 wrote: Fri Jan 28, 2022 10:19 am
marcopolo wrote: Thu Jan 27, 2022 10:51 am My main point is that soon as you get to a situation where the success rate is less that 100%, ratcheting up increases risk of failure.
Yes, ratcheting up as the OP has described it, which is just 'restarting' the '4% rule', is no less safe than the '4% rule' itself. It's actually safer because the retiree presumably has fewer than 30 years of retirement remaining. But if the '4% rule' is in danger, so is ratcheting up.
This is not true. Suppose 1966 was a year that the 4% rule failed. The problem with ratcheting up is that it would cause earlier years (1963~1965) to fail, when the 4% rule itself would have worked for the 1963~1965 retirees.
You seem to be persistently misreading Will’s posts in this thread. Maybe it’s easier to think of it another way. Take 3.8% as your base SWR because that is what worked for 1966. In any year that your portfolio appreciated from retirement start, you would be able to reset to 3.8% instead of only adjusting for inflation. This is because you’d be making the same decision as a fresh retiree and have fewer years to cover. But this is a mathematical/theoretical discussion because you never know your lifespan or the future market performance. In practice you must be flexible.
Bulls make money, bears make money, pigs get slaughtered.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

Seems like we were saying the same thing then. If a 30-year retirement *worked* then yeah, ratcheting up works just as well.

The problem is however, we only find out if your "4% rule" retirement worked after it's done. This is why I'm against the ratcheting-up idea, personally speaking.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

Marseille07 wrote: Fri Jan 28, 2022 11:39 am Seems like we were saying the same thing then. If a 30-year retirement *worked* then yeah, ratcheting up works just as well.

The problem is however, we only find out if your "4% rule" retirement worked after it's done. This is why I'm against the ratcheting-up idea, personally speaking.
Yes, we don't know what the 30 year SWR will be for 2022-2051. That can only be known in hindsight. And I agree that nobody should be rigidly following any SWR, including the '4% rule'.

My point all along has merely been that restarting something like the '4% rule' is no more dangerous in terms of risking premature portfolio depletion than is starting with the '4% rule'. If the 30 year SWR for 1966 retirees was 4% (4.1% according to the Simba spreadsheet for a 60/40), then it was perfectly fine for those who retired in earlier years to restart their withdrawals in 1966 at the very same withdrawal rate, 4%.
The Sensible Steward
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by marcopolo »

willthrill81 wrote: Fri Jan 28, 2022 11:31 am
Marseille07 wrote: Fri Jan 28, 2022 10:54 am
willthrill81 wrote: Fri Jan 28, 2022 10:19 am
marcopolo wrote: Thu Jan 27, 2022 10:51 am My main point is that soon as you get to a situation where the success rate is less that 100%, ratcheting up increases risk of failure.
Yes, ratcheting up as the OP has described it, which is just 'restarting' the '4% rule', is no less safe than the '4% rule' itself. It's actually safer because the retiree presumably has fewer than 30 years of retirement remaining. But if the '4% rule' is in danger, so is ratcheting up.
This is not true. Suppose 1966 was a year that the 4% rule failed. The problem with ratcheting up is that it would cause earlier years (1963~1965) to fail, when the 4% rule itself would have worked for the 1963~1965 retirees.
Please keep in mind what I said: restarting the '4% rule' is no more dangerous than starting with the '4% rule'. If the '4% rule' worked for 30 years for 1966 retirees, and the Simba backtesting spreadsheet indicates that it actually did, though barely, then it means that prior year retirees restarting their withdrawals at no more than 4% of their current portfolio balance were in no worse shape.

Let's look at some real numbers from the Simba backtesting spreadsheet to illustrate this.

Let's say that a 1963 retiree had $1m in a portfolio with 60% TSM and 40% TBM. At the end of that year, the portfolio was up to $1,133,680, and with that year's 1.32% inflation, the amount withdrawn was $40,528, leaving $1,093,152 in the portfolio.

At the end of 1964, the portfolio's balance was $1,217,618, and with that year's 1.31% inflation, the amount withdrawn was $41,059, leaving $1,176,559 in the portfolio.

At the end of 1965, the portfolio's balance was $1,280,732.

After their inflation-adjusted withdrawals in 1963 and 1964, 1963 retirees had seen 28% nominal growth in their portfolio through the end of 1965. At the end of 1965, they could now restart the '4% rule' and withdraw $51,229, 4% of their current portfolio balance. Them withdrawing 4% of their current portfolio balance is no different at all from those who just retired in 1966 with $1,280,732 and started by withdrawing 4% of their portfolio balance, the same $51,229.
Would you agree that a 3.33% WR is safer than 4%?

That is the scenario for the retiree that gets a 20% increase in their portfolio the first year of retirement.
If they do nothing, they are now at 3.33%WR, if they ratchet up, they are at 4%WR. The 4% WR rate might be pretty safe, but it is certainly riskier than 3.33%

The problem with what you are saying is the underlying assumption that 4% is 100% safe. Probability of failure is not a binary thing, the only reason it may appear to be 100% safe with some specific data sets is only a quirk due to nit having enough data. For example, running the cFIREsim backtest with the default parameters shows several (I believe 5 different starting years) that failed at 4%. The fact that small differences in AA can lead to multiple failures highlights the fact that the 100% success assumption is just due to lack of diversity in the data.

That is why more recent discussion refer to 95% success rate for the 4% "rule".

Once you allow for several failure years, it is quite trivial to show that ratcheting up significantly increases the number of starting years thatfail.. That would also be consistent with the obvious assertion that 4% WR is riskier than 3.33%.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

marcopolo wrote: Fri Jan 28, 2022 12:36 pm
willthrill81 wrote: Fri Jan 28, 2022 11:31 am
Marseille07 wrote: Fri Jan 28, 2022 10:54 am
willthrill81 wrote: Fri Jan 28, 2022 10:19 am
marcopolo wrote: Thu Jan 27, 2022 10:51 am My main point is that soon as you get to a situation where the success rate is less that 100%, ratcheting up increases risk of failure.
Yes, ratcheting up as the OP has described it, which is just 'restarting' the '4% rule', is no less safe than the '4% rule' itself. It's actually safer because the retiree presumably has fewer than 30 years of retirement remaining. But if the '4% rule' is in danger, so is ratcheting up.
This is not true. Suppose 1966 was a year that the 4% rule failed. The problem with ratcheting up is that it would cause earlier years (1963~1965) to fail, when the 4% rule itself would have worked for the 1963~1965 retirees.
Please keep in mind what I said: restarting the '4% rule' is no more dangerous than starting with the '4% rule'. If the '4% rule' worked for 30 years for 1966 retirees, and the Simba backtesting spreadsheet indicates that it actually did, though barely, then it means that prior year retirees restarting their withdrawals at no more than 4% of their current portfolio balance were in no worse shape.

Let's look at some real numbers from the Simba backtesting spreadsheet to illustrate this.

Let's say that a 1963 retiree had $1m in a portfolio with 60% TSM and 40% TBM. At the end of that year, the portfolio was up to $1,133,680, and with that year's 1.32% inflation, the amount withdrawn was $40,528, leaving $1,093,152 in the portfolio.

At the end of 1964, the portfolio's balance was $1,217,618, and with that year's 1.31% inflation, the amount withdrawn was $41,059, leaving $1,176,559 in the portfolio.

At the end of 1965, the portfolio's balance was $1,280,732.

After their inflation-adjusted withdrawals in 1963 and 1964, 1963 retirees had seen 28% nominal growth in their portfolio through the end of 1965. At the end of 1965, they could now restart the '4% rule' and withdraw $51,229, 4% of their current portfolio balance. Them withdrawing 4% of their current portfolio balance is no different at all from those who just retired in 1966 with $1,280,732 and started by withdrawing 4% of their portfolio balance, the same $51,229.
Would you agree that a 3.33% WR is safer than 4%?

That is the scenario for the retiree that gets a 20% increase in their portfolio the first year of retirement.
If they do nothing, they are now at 3.33%WR, if they ratchet up, they are at 4%WR. The 4% WR rate might be pretty safe, but it is certainly riskier than 3.33%

The problem with what you are saying is the underlying assumption that 4% is 100% safe. Probability of failure is not a binary thing, the only reason it may appear to be 100% safe with some specific data sets is only a quirk due to nit having enough data. For example, running the cFIREsim backtest with the default parameters shows several (I believe 5 different starting years) that failed at 4%. The fact that small differences in AA can lead to multiple failures highlights the fact that the 100% success assumption is just due to lack of diversity in the data.

That is why more recent discussion refer to 95% success rate for the 4% "rule".

Once you allow for several failure years, it is quite trivial to show that ratcheting up significantly increases the number of starting years thatfail.. That would also be consistent with the obvious assertion that 4% WR is riskier than 3.33%.
Of course a lower WR is safer than a higher one.

I've lost count of the number of times that I've said in this thread that the big assumption here is that the '4% rule' is safe going forward and that we don't know that it is.

That's never been my point. My point was merely that ratcheting up one's withdrawals to 4% is not inherently riskier than starting one's withdrawals at 4%. That's all.

Aside from all that, I personally put more faith in the quality of the Simba spreadsheet than cFIREsim, but that's just me. And as I've noted, if the '4% rule' works for 'only' 29.5 years vs. 31 isn't a big deal. As you note, the real world is not that binary.
The Sensible Steward
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by marcopolo »

willthrill81 wrote: Fri Jan 28, 2022 12:38 pm
marcopolo wrote: Fri Jan 28, 2022 12:36 pm
willthrill81 wrote: Fri Jan 28, 2022 11:31 am
Marseille07 wrote: Fri Jan 28, 2022 10:54 am
willthrill81 wrote: Fri Jan 28, 2022 10:19 am

Yes, ratcheting up as the OP has described it, which is just 'restarting' the '4% rule', is no less safe than the '4% rule' itself. It's actually safer because the retiree presumably has fewer than 30 years of retirement remaining. But if the '4% rule' is in danger, so is ratcheting up.
This is not true. Suppose 1966 was a year that the 4% rule failed. The problem with ratcheting up is that it would cause earlier years (1963~1965) to fail, when the 4% rule itself would have worked for the 1963~1965 retirees.
Please keep in mind what I said: restarting the '4% rule' is no more dangerous than starting with the '4% rule'. If the '4% rule' worked for 30 years for 1966 retirees, and the Simba backtesting spreadsheet indicates that it actually did, though barely, then it means that prior year retirees restarting their withdrawals at no more than 4% of their current portfolio balance were in no worse shape.

Let's look at some real numbers from the Simba backtesting spreadsheet to illustrate this.

Let's say that a 1963 retiree had $1m in a portfolio with 60% TSM and 40% TBM. At the end of that year, the portfolio was up to $1,133,680, and with that year's 1.32% inflation, the amount withdrawn was $40,528, leaving $1,093,152 in the portfolio.

At the end of 1964, the portfolio's balance was $1,217,618, and with that year's 1.31% inflation, the amount withdrawn was $41,059, leaving $1,176,559 in the portfolio.

At the end of 1965, the portfolio's balance was $1,280,732.

After their inflation-adjusted withdrawals in 1963 and 1964, 1963 retirees had seen 28% nominal growth in their portfolio through the end of 1965. At the end of 1965, they could now restart the '4% rule' and withdraw $51,229, 4% of their current portfolio balance. Them withdrawing 4% of their current portfolio balance is no different at all from those who just retired in 1966 with $1,280,732 and started by withdrawing 4% of their portfolio balance, the same $51,229.
Would you agree that a 3.33% WR is safer than 4%?

That is the scenario for the retiree that gets a 20% increase in their portfolio the first year of retirement.
If they do nothing, they are now at 3.33%WR, if they ratchet up, they are at 4%WR. The 4% WR rate might be pretty safe, but it is certainly riskier than 3.33%

The problem with what you are saying is the underlying assumption that 4% is 100% safe. Probability of failure is not a binary thing, the only reason it may appear to be 100% safe with some specific data sets is only a quirk due to nit having enough data. For example, running the cFIREsim backtest with the default parameters shows several (I believe 5 different starting years) that failed at 4%. The fact that small differences in AA can lead to multiple failures highlights the fact that the 100% success assumption is just due to lack of diversity in the data.

That is why more recent discussion refer to 95% success rate for the 4% "rule".

Once you allow for several failure years, it is quite trivial to show that ratcheting up significantly increases the number of starting years thatfail.. That would also be consistent with the obvious assertion that 4% WR is riskier than 3.33%.
Of course a lower WR is safer than a higher one.

I've lost count of the number of times that I've said in this thread that the big assumption here is that the '4% rule' is safe going forward and that we don't know that it is.

That's never been my point. My point was merely that ratcheting up one's withdrawals to 4% is not inherently riskier than starting one's withdrawals at 4%. That's all.

Aside from all that, I personally put more faith in the quality of the Simba spreadsheet than cFIREsim, but that's just me. And as I've noted, if the '4% rule' works for 'only' 29.5 years vs. 31 isn't a big deal. As you note, the real world is not that binary.
As you say, the world is not binary, so I don't understand why you keep using the word "safe", there is no such thing, and it is a terrible assumption to make. If that is your starting point, then the rest of your argument is built on a pretty weak foundation.

We should be using words like "safer", and "riskier", rather than absolutes like "safe"

You are very highly respected poster on this forum, I think you do a dis-service to less nuanced readers by suggesting that ratcheting up is just as "safe" as starting the 4% rule based on such a small number of observed outcomes that is not very robust.

I suspect small tweaks in AA, or adding a small expense ratio to Simba data would show similar failure scenarios.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

marcopolo wrote: Fri Jan 28, 2022 12:49 pm
willthrill81 wrote: Fri Jan 28, 2022 12:38 pm
marcopolo wrote: Fri Jan 28, 2022 12:36 pm
willthrill81 wrote: Fri Jan 28, 2022 11:31 am
Marseille07 wrote: Fri Jan 28, 2022 10:54 am

This is not true. Suppose 1966 was a year that the 4% rule failed. The problem with ratcheting up is that it would cause earlier years (1963~1965) to fail, when the 4% rule itself would have worked for the 1963~1965 retirees.
Please keep in mind what I said: restarting the '4% rule' is no more dangerous than starting with the '4% rule'. If the '4% rule' worked for 30 years for 1966 retirees, and the Simba backtesting spreadsheet indicates that it actually did, though barely, then it means that prior year retirees restarting their withdrawals at no more than 4% of their current portfolio balance were in no worse shape.

Let's look at some real numbers from the Simba backtesting spreadsheet to illustrate this.

Let's say that a 1963 retiree had $1m in a portfolio with 60% TSM and 40% TBM. At the end of that year, the portfolio was up to $1,133,680, and with that year's 1.32% inflation, the amount withdrawn was $40,528, leaving $1,093,152 in the portfolio.

At the end of 1964, the portfolio's balance was $1,217,618, and with that year's 1.31% inflation, the amount withdrawn was $41,059, leaving $1,176,559 in the portfolio.

At the end of 1965, the portfolio's balance was $1,280,732.

After their inflation-adjusted withdrawals in 1963 and 1964, 1963 retirees had seen 28% nominal growth in their portfolio through the end of 1965. At the end of 1965, they could now restart the '4% rule' and withdraw $51,229, 4% of their current portfolio balance. Them withdrawing 4% of their current portfolio balance is no different at all from those who just retired in 1966 with $1,280,732 and started by withdrawing 4% of their portfolio balance, the same $51,229.
Would you agree that a 3.33% WR is safer than 4%?

That is the scenario for the retiree that gets a 20% increase in their portfolio the first year of retirement.
If they do nothing, they are now at 3.33%WR, if they ratchet up, they are at 4%WR. The 4% WR rate might be pretty safe, but it is certainly riskier than 3.33%

The problem with what you are saying is the underlying assumption that 4% is 100% safe. Probability of failure is not a binary thing, the only reason it may appear to be 100% safe with some specific data sets is only a quirk due to nit having enough data. For example, running the cFIREsim backtest with the default parameters shows several (I believe 5 different starting years) that failed at 4%. The fact that small differences in AA can lead to multiple failures highlights the fact that the 100% success assumption is just due to lack of diversity in the data.

That is why more recent discussion refer to 95% success rate for the 4% "rule".

Once you allow for several failure years, it is quite trivial to show that ratcheting up significantly increases the number of starting years thatfail.. That would also be consistent with the obvious assertion that 4% WR is riskier than 3.33%.
Of course a lower WR is safer than a higher one.

I've lost count of the number of times that I've said in this thread that the big assumption here is that the '4% rule' is safe going forward and that we don't know that it is.

That's never been my point. My point was merely that ratcheting up one's withdrawals to 4% is not inherently riskier than starting one's withdrawals at 4%. That's all.

Aside from all that, I personally put more faith in the quality of the Simba spreadsheet than cFIREsim, but that's just me. And as I've noted, if the '4% rule' works for 'only' 29.5 years vs. 31 isn't a big deal. As you note, the real world is not that binary.
As you say, the world is not binary, so I don't understand why you keep using the word "safe", there is no such thing, and it is a terrible assumption to make. If that is your starting point, then the rest of your argument is built on a pretty weak foundation.

We should be using words like "safer", and "riskier", rather than absolutes like "safe"

You are very highly respected poster on this forum, I think you do a dis-service to less nuanced readers by suggesting that ratcheting up is just as "safe" as starting the 4% rule based on such a small number of observed outcomes that is not very robust.

I suspect small tweaks in AA, or adding a small expense ratio to Simba data would show similar failure scenarios.
Again, I've REPEATEDLY prefaced my statements that the assumption may not be safe going forward. I've never said that the '4% rule' is safe, and I've lost count of the number of times I've said that rigidly following it would be insane. It's not my fault if someone wantonly ignores all of that.
The Sensible Steward
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by marcopolo »

willthrill81 wrote: Fri Jan 28, 2022 12:53 pm
marcopolo wrote: Fri Jan 28, 2022 12:49 pm
willthrill81 wrote: Fri Jan 28, 2022 12:38 pm
marcopolo wrote: Fri Jan 28, 2022 12:36 pm
willthrill81 wrote: Fri Jan 28, 2022 11:31 am

Please keep in mind what I said: restarting the '4% rule' is no more dangerous than starting with the '4% rule'. If the '4% rule' worked for 30 years for 1966 retirees, and the Simba backtesting spreadsheet indicates that it actually did, though barely, then it means that prior year retirees restarting their withdrawals at no more than 4% of their current portfolio balance were in no worse shape.

Let's look at some real numbers from the Simba backtesting spreadsheet to illustrate this.

Let's say that a 1963 retiree had $1m in a portfolio with 60% TSM and 40% TBM. At the end of that year, the portfolio was up to $1,133,680, and with that year's 1.32% inflation, the amount withdrawn was $40,528, leaving $1,093,152 in the portfolio.

At the end of 1964, the portfolio's balance was $1,217,618, and with that year's 1.31% inflation, the amount withdrawn was $41,059, leaving $1,176,559 in the portfolio.

At the end of 1965, the portfolio's balance was $1,280,732.

After their inflation-adjusted withdrawals in 1963 and 1964, 1963 retirees had seen 28% nominal growth in their portfolio through the end of 1965. At the end of 1965, they could now restart the '4% rule' and withdraw $51,229, 4% of their current portfolio balance. Them withdrawing 4% of their current portfolio balance is no different at all from those who just retired in 1966 with $1,280,732 and started by withdrawing 4% of their portfolio balance, the same $51,229.
Would you agree that a 3.33% WR is safer than 4%?

That is the scenario for the retiree that gets a 20% increase in their portfolio the first year of retirement.
If they do nothing, they are now at 3.33%WR, if they ratchet up, they are at 4%WR. The 4% WR rate might be pretty safe, but it is certainly riskier than 3.33%

The problem with what you are saying is the underlying assumption that 4% is 100% safe. Probability of failure is not a binary thing, the only reason it may appear to be 100% safe with some specific data sets is only a quirk due to nit having enough data. For example, running the cFIREsim backtest with the default parameters shows several (I believe 5 different starting years) that failed at 4%. The fact that small differences in AA can lead to multiple failures highlights the fact that the 100% success assumption is just due to lack of diversity in the data.

That is why more recent discussion refer to 95% success rate for the 4% "rule".

Once you allow for several failure years, it is quite trivial to show that ratcheting up significantly increases the number of starting years thatfail.. That would also be consistent with the obvious assertion that 4% WR is riskier than 3.33%.
Of course a lower WR is safer than a higher one.

I've lost count of the number of times that I've said in this thread that the big assumption here is that the '4% rule' is safe going forward and that we don't know that it is.

That's never been my point. My point was merely that ratcheting up one's withdrawals to 4% is not inherently riskier than starting one's withdrawals at 4%. That's all.

Aside from all that, I personally put more faith in the quality of the Simba spreadsheet than cFIREsim, but that's just me. And as I've noted, if the '4% rule' works for 'only' 29.5 years vs. 31 isn't a big deal. As you note, the real world is not that binary.
As you say, the world is not binary, so I don't understand why you keep using the word "safe", there is no such thing, and it is a terrible assumption to make. If that is your starting point, then the rest of your argument is built on a pretty weak foundation.

We should be using words like "safer", and "riskier", rather than absolutes like "safe"

You are very highly respected poster on this forum, I think you do a dis-service to less nuanced readers by suggesting that ratcheting up is just as "safe" as starting the 4% rule based on such a small number of observed outcomes that is not very robust.

I suspect small tweaks in AA, or adding a small expense ratio to Simba data would show similar failure scenarios.
Again, I've REPEATEDLY prefaced my statements that the assumption may not be safe going forward. I've never said that the '4% rule' is safe, and I've lost count of the number of times I've said that rigidly following it would be insane. It's not my fault if someone wantonly ignores all of that.
can you point me to any such caveat in the post that I responded in this exchange?
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

marcopolo wrote: Fri Jan 28, 2022 12:55 pm
willthrill81 wrote: Fri Jan 28, 2022 12:53 pm
marcopolo wrote: Fri Jan 28, 2022 12:49 pm
willthrill81 wrote: Fri Jan 28, 2022 12:38 pm
marcopolo wrote: Fri Jan 28, 2022 12:36 pm

Would you agree that a 3.33% WR is safer than 4%?

That is the scenario for the retiree that gets a 20% increase in their portfolio the first year of retirement.
If they do nothing, they are now at 3.33%WR, if they ratchet up, they are at 4%WR. The 4% WR rate might be pretty safe, but it is certainly riskier than 3.33%

The problem with what you are saying is the underlying assumption that 4% is 100% safe. Probability of failure is not a binary thing, the only reason it may appear to be 100% safe with some specific data sets is only a quirk due to nit having enough data. For example, running the cFIREsim backtest with the default parameters shows several (I believe 5 different starting years) that failed at 4%. The fact that small differences in AA can lead to multiple failures highlights the fact that the 100% success assumption is just due to lack of diversity in the data.

That is why more recent discussion refer to 95% success rate for the 4% "rule".

Once you allow for several failure years, it is quite trivial to show that ratcheting up significantly increases the number of starting years thatfail.. That would also be consistent with the obvious assertion that 4% WR is riskier than 3.33%.
Of course a lower WR is safer than a higher one.

I've lost count of the number of times that I've said in this thread that the big assumption here is that the '4% rule' is safe going forward and that we don't know that it is.

That's never been my point. My point was merely that ratcheting up one's withdrawals to 4% is not inherently riskier than starting one's withdrawals at 4%. That's all.

Aside from all that, I personally put more faith in the quality of the Simba spreadsheet than cFIREsim, but that's just me. And as I've noted, if the '4% rule' works for 'only' 29.5 years vs. 31 isn't a big deal. As you note, the real world is not that binary.
As you say, the world is not binary, so I don't understand why you keep using the word "safe", there is no such thing, and it is a terrible assumption to make. If that is your starting point, then the rest of your argument is built on a pretty weak foundation.

We should be using words like "safer", and "riskier", rather than absolutes like "safe"

You are very highly respected poster on this forum, I think you do a dis-service to less nuanced readers by suggesting that ratcheting up is just as "safe" as starting the 4% rule based on such a small number of observed outcomes that is not very robust.

I suspect small tweaks in AA, or adding a small expense ratio to Simba data would show similar failure scenarios.
Again, I've REPEATEDLY prefaced my statements that the assumption may not be safe going forward. I've never said that the '4% rule' is safe, and I've lost count of the number of times I've said that rigidly following it would be insane. It's not my fault if someone wantonly ignores all of that.
can you point me to any such caveat in the post that I responded in this exchange?
Do you want disclaimers in every single post? :shock:

However, I did say "no more dangerous" in the post you quoted.
The Sensible Steward
randomguy
Posts: 11295
Joined: Wed Sep 17, 2014 9:00 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by randomguy »

willthrill81 wrote: Fri Jan 28, 2022 11:31 am After their inflation-adjusted withdrawals in 1963 and 1964, 1963 retirees had seen 28% nominal growth in their portfolio through the end of 1965. At the end of 1965, they could now restart the '4% rule' and withdraw $51,229, 4% of their current portfolio balance. Them withdrawing 4% of their current portfolio balance is no different at all from those who just retired in 1966 with $1,280,732 and started by withdrawing 4% of their portfolio balance, the same $51,229.
Yes and that means they share the fate of the 1966 retiree. That 1966 retire doesn't have a 5% chance of failure. They have a 100% chance (hand waving about picking the right starting month and AA to get a failure) . They just don't know it. The 1963 retiree who resets has gone from having aa 0% chance of failure to doing some math to see if the AA they picked lasts lasts 27 year or not....

It really comes down to how much risk that 1963 retiree wants. If they want 5%, they can't keep upping the withdrawal. The fact that they won the first 3 5% gambles they made doesn't mean they can make another roll of the dice and still only have a 5% chance of failure. Every time they up the withdrawal they are upping their risk. Now at a certain point the risk they are adding is pretty low. Reseting after year 1 might up your failure rate from like 5% to 8% (i.e. the 1965 and 1966 retirees fail instead of just 1966) but as you keep reducing the number of years to fund with a constant SWR, the risk of the resets drop. By year 10 of doing this your risk might be going from like 9% to 9.1% as the odds of the 4% rule failing over 20 years is getting pretty close to zero.
Patzer
Posts: 701
Joined: Wed Jun 10, 2015 10:56 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Patzer »

4% does sometimes fail, that is why we care about sequence of return risk.
You can't ratchet up 4%, without increasing your failure rate, because you are restarting the clock on your sequence of return risk every year.

If you go with a never fail number, then you could ratchet it up every year and it would continue to have a 0% chance of failure, but the never fail numbers are usually in the 3-4% range depending on your retirement time horizon, asset allocation, and percentage of total spend that comes from social security and how many years into retirement that social security starts.

A strategy might to figure out the no fail number, and start ratcheting up once you pass that.

I.e. you retire with a 4% SWR, accepting some chance of failure, and knowing that your no historical risk number is 3.5%.
In the first few years, you stick with 4%, but if you get a nice bull market and 3.5% coming out of your portfolio starts to exceed your old 4% number, then from that point on you can start ratcheting up based on 3.5% every year.
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by HomerJ »

randomguy wrote: Fri Jan 28, 2022 2:04 pm
willthrill81 wrote: Fri Jan 28, 2022 11:31 am After their inflation-adjusted withdrawals in 1963 and 1964, 1963 retirees had seen 28% nominal growth in their portfolio through the end of 1965. At the end of 1965, they could now restart the '4% rule' and withdraw $51,229, 4% of their current portfolio balance. Them withdrawing 4% of their current portfolio balance is no different at all from those who just retired in 1966 with $1,280,732 and started by withdrawing 4% of their portfolio balance, the same $51,229.
Yes and that means they share the fate of the 1966 retiree. That 1966 retire doesn't have a 5% chance of failure. They have a 100% chance (hand waving about picking the right starting month and AA to get a failure) . They just don't know it. The 1963 retiree who resets has gone from having aa 0% chance of failure to doing some math to see if the AA they picked lasts lasts 27 year or not....

It really comes down to how much risk that 1963 retiree wants. If they want 5%, they can't keep upping the withdrawal. The fact that they won the first 3 5% gambles they made doesn't mean they can make another roll of the dice and still only have a 5% chance of failure. Every time they up the withdrawal they are upping their risk. Now at a certain point the risk they are adding is pretty low. Reseting after year 1 might up your failure rate from like 5% to 8% (i.e. the 1965 and 1966 retirees fail instead of just 1966) but as you keep reducing the number of years to fund with a constant SWR, the risk of the resets drop. By year 10 of doing this your risk might be going from like 9% to 9.1% as the odds of the 4% rule failing over 20 years is getting pretty close to zero.
It's not quite that bad for the 1963 retiree... For one, they only need the money to last 27 years instead of 30 like the 1966 retiree.

Two, I am assuming that upping your spending for most people in this situation will be discretionary spending. An extra vacation, or splurging on a nicer car. Pretty easy to cut back to your original spending stuff I mean.

4% isn't set in stone. It's a "rule of thumb".

I probably wouldn't reset to 4% every single year the market goes up... But after 5 years, if I was up, I'd definitely increase my spending some. Nothing wrong with that.

There is no set rule for any of this. Just common sense, just like we all used when spending money while working.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

HomerJ wrote: Fri Jan 28, 2022 2:26 pm There is no set rule for any of this. Just common sense, just like we all used when spending money while working.
Sadly, common sense seems to be rather uncommon.
The Sensible Steward
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

willthrill81 wrote: Fri Jan 28, 2022 2:31 pm
HomerJ wrote: Fri Jan 28, 2022 2:26 pm There is no set rule for any of this. Just common sense, just like we all used when spending money while working.
Sadly, common sense seems to be rather uncommon.
It's not that people lack common sense. The WR discussions will be academic by nature, deliberately so.

I'd argue it's actually wrong to bring common sense into academic discussions because by that logic, anything *works*. If SHTF, cut back, common sense.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

Marseille07 wrote: Fri Jan 28, 2022 2:37 pm
willthrill81 wrote: Fri Jan 28, 2022 2:31 pm
HomerJ wrote: Fri Jan 28, 2022 2:26 pm There is no set rule for any of this. Just common sense, just like we all used when spending money while working.
Sadly, common sense seems to be rather uncommon.
It's not that people lack common sense. The WR discussions will be academic by nature, deliberately so.

I'd argue it's actually wrong to bring common sense into academic discussions because by that logic, anything *works*. If SHTF, cut back, common sense.
It drives me crazy for people to talk about 'failure rates' for Z withdrawal scheme, as though a person intelligent enough to build a relatively sizable portfolio will just blindly spend it down to zero without making any adjustments at all. I understand why people model such things 9 ways to Sunday, and I think that there can be real value in all that, but the risk of 'failure' seems to almost universally be overblown, IMHO.
The Sensible Steward
randomguy
Posts: 11295
Joined: Wed Sep 17, 2014 9:00 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by randomguy »

HomerJ wrote: Fri Jan 28, 2022 2:26 pm

It's not quite that bad for the 1963 retiree... For one, they only need the money to last 27 years instead of 30 like the 1966 retiree.

Two, I am assuming that upping your spending for most people in this situation will be discretionary spending. An extra vacation, or splurging on a nicer car. Pretty easy to cut back to your original spending stuff I mean.

4% isn't set in stone. It's a "rule of thumb".

I probably wouldn't reset to 4% every single year the market goes up... But after 5 years, if I was up, I'd definitely increase my spending some. Nothing wrong with that.

There is no set rule for any of this. Just common sense, just like we all used when spending money while working.
Pretty sure the 1966 retiree had failures in year 25 to 29 depending on the AA. So yes they will not fail every time like the 1966 retire will but there will be a chance above 0 depending on their AA. The 1965 retiree doing this is far more likely to fail.

Sure if you add the ability to cut spending, you can be more aggressive. If you don't want to cut spending, you need to be less aggressive. You can plot out all sorts of cases. If after 10 years you 1 million has turned into 2, I don't think anyone will complain about you upping your spend from 40k to 60k from a risk point of view. Upping to 100k because the 20 year SWR is 5% though might be more risk than you want.

The flexibility of spending is great but it is hard to talk about without specifics. A 6% SWR with flexible spending will work. You just better be ready for 50% cuts for a couple decades if times get rough.... Just saying well I will cut if things get bad is a bit vague. You need an idea of how much and for how long you will find acceptable.
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by marcopolo »

willthrill81 wrote: Fri Jan 28, 2022 2:31 pm
HomerJ wrote: Fri Jan 28, 2022 2:26 pm There is no set rule for any of this. Just common sense, just like we all used when spending money while working.
Sadly, common sense seems to be rather uncommon.
If we were discussing common sense, no one would be using the 4% rule.
This discussion was specifically about the 4% rule, and the question posed in the "Theory" section of the forum was whether "ratcheting increases risk or not".

The fact that no one follows the 4% rule, and if things get bad, people cut back spending, etc. is absolutely true, but does not impact the answer to that question. The fact is absolutely true that ratcheting up increases risk of running out of money. You are spending a higher percentage of your portfolio, how can that not be true?!? The fact that in the real world, people can fairly easily mitigate that risk does not mean that the increased risk is not there.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

marcopolo wrote: Fri Jan 28, 2022 2:42 pm
willthrill81 wrote: Fri Jan 28, 2022 2:31 pm
HomerJ wrote: Fri Jan 28, 2022 2:26 pm There is no set rule for any of this. Just common sense, just like we all used when spending money while working.
Sadly, common sense seems to be rather uncommon.
If we were discussing common sense, no one would be using the 4% rule.
This discussion was specifically about the 4% rule, and the question posed in the "Theory" section of the forum was whether "ratcheting increases risk or not".

The fact that no one follows the 4% rule, and if things get bad, people cut back spending, etc. is absolutely true, but does not impact the answer to that question. The fact is absolutely true that ratcheting up increases risk of running out of money. You are spending a higher percentage of your portfolio, how can that not be true?!? The fact that in the real world, people can fairly easily mitigate that risk does not mean that the increased risk is not there.
The fact that spending more increases one's risk of running out of funds later seems to be very common sensical to me.
The Sensible Steward
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

willthrill81 wrote: Fri Jan 28, 2022 2:39 pm It drives me crazy for people to talk about 'failure rates' for Z withdrawal scheme, as though a person intelligent enough to build a relatively sizable portfolio will just blindly spend it down to zero without making any adjustments at all. I understand why people model such things 9 ways to Sunday, and I think that there can be real value in all that, but the risk of 'failure' seems to almost universally be overblown, IMHO.
Practically, you're right that people will cut back. But when talking about X% WR, we have to assume they're withdrawing in full, every year, without making any adjustments, as unpractical as it is.

Otherwise, pretty much anything "works" because people do make adjustments to make it work.
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by marcopolo »

willthrill81 wrote: Fri Jan 28, 2022 2:44 pm
marcopolo wrote: Fri Jan 28, 2022 2:42 pm
willthrill81 wrote: Fri Jan 28, 2022 2:31 pm
HomerJ wrote: Fri Jan 28, 2022 2:26 pm There is no set rule for any of this. Just common sense, just like we all used when spending money while working.
Sadly, common sense seems to be rather uncommon.
If we were discussing common sense, no one would be using the 4% rule.
This discussion was specifically about the 4% rule, and the question posed in the "Theory" section of the forum was whether "ratcheting increases risk or not".

The fact that no one follows the 4% rule, and if things get bad, people cut back spending, etc. is absolutely true, but does not impact the answer to that question. The fact is absolutely true that ratcheting up increases risk of running out of money. You are spending a higher percentage of your portfolio, how can that not be true?!? The fact that in the real world, people can fairly easily mitigate that risk does not mean that the increased risk is not there.
The fact that spending more increases one's risk of running out of funds later seems to be very common sensical to me.
Yet, above, you kept insisting that it was "no more dangerous" :oops:
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

Marseille07 wrote: Fri Jan 28, 2022 2:45 pm
willthrill81 wrote: Fri Jan 28, 2022 2:39 pm It drives me crazy for people to talk about 'failure rates' for Z withdrawal scheme, as though a person intelligent enough to build a relatively sizable portfolio will just blindly spend it down to zero without making any adjustments at all. I understand why people model such things 9 ways to Sunday, and I think that there can be real value in all that, but the risk of 'failure' seems to almost universally be overblown, IMHO.
Practically, you're right that people will cut back. But when talking about X% WR, we have to assume they're withdrawing in full, every year, without making any adjustments, as unpractical as it is.

Otherwise, pretty much anything "works" because people do make adjustments to make it work.
I completely understand why Bengen used the SWR-style method that he did. Certainly retirees prefer stable over unstable retirement income. But we all know that all sane people will make some kind of adjustment along the way, especially if their portfolio is performing poorly. The practical question then becomes whether 4% is a reasonable starting point for someone anticipating a retirement lasting up to around 30 years, and I believe that it still is. Anything much more involved than that is likely to suffer from the Mike Tyson effect: "Everyone has a plan until they get punched in the mouth."
The Sensible Steward
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

marcopolo wrote: Fri Jan 28, 2022 2:47 pm
willthrill81 wrote: Fri Jan 28, 2022 2:44 pm
marcopolo wrote: Fri Jan 28, 2022 2:42 pm
willthrill81 wrote: Fri Jan 28, 2022 2:31 pm
HomerJ wrote: Fri Jan 28, 2022 2:26 pm There is no set rule for any of this. Just common sense, just like we all used when spending money while working.
Sadly, common sense seems to be rather uncommon.
If we were discussing common sense, no one would be using the 4% rule.
This discussion was specifically about the 4% rule, and the question posed in the "Theory" section of the forum was whether "ratcheting increases risk or not".

The fact that no one follows the 4% rule, and if things get bad, people cut back spending, etc. is absolutely true, but does not impact the answer to that question. The fact is absolutely true that ratcheting up increases risk of running out of money. You are spending a higher percentage of your portfolio, how can that not be true?!? The fact that in the real world, people can fairly easily mitigate that risk does not mean that the increased risk is not there.
The fact that spending more increases one's risk of running out of funds later seems to be very common sensical to me.
Yet, above, you kept insisting that it was "no more dangerous" :oops:
And withdrawing 4% of one's new portfolio balance isn't any more dangerous or any less safe than a new retiree starting by withdrawing 4% of their portfolio balance.

This is easy to understand.

C'mon dude, you're beating a horse that wasn't even there in the first place. Lay off already.
Last edited by willthrill81 on Fri Jan 28, 2022 2:52 pm, edited 1 time in total.
The Sensible Steward
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by marcopolo »

Marseille07 wrote: Fri Jan 28, 2022 2:45 pm
willthrill81 wrote: Fri Jan 28, 2022 2:39 pm It drives me crazy for people to talk about 'failure rates' for Z withdrawal scheme, as though a person intelligent enough to build a relatively sizable portfolio will just blindly spend it down to zero without making any adjustments at all. I understand why people model such things 9 ways to Sunday, and I think that there can be real value in all that, but the risk of 'failure' seems to almost universally be overblown, IMHO.
Practically, you're right that people will cut back. But when talking about X% WR, we have to assume they're withdrawing in full, every year, without making any adjustments, as unpractical as it is.

Otherwise, pretty much anything "works" because people do make adjustments to make it work.
Yeah, it seems very odd to me that we are discussing a specific withdrawal scheme, and then people keep changing it to some nebulous "well, then you just stay flexible and adjust to it". That is always an option with any methodology, but adds zero value when actually trying to assess the outcomes of specific methods.
Once in a while you get shown the light, in the strangest of places if you look at it right.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

marcopolo wrote: Fri Jan 28, 2022 2:50 pm Yeah, it seems very odd to me that we are discussing a specific withdrawal scheme, and then people keep changing it to some nebulous "well, then you just stay flexible and adjust to it". That is always an option with any methodology, but adds zero value when actually trying to assess the outcomes of specific methods.
Yes, the discussion would have to be somewhat impractical & academic; otherwise people start inserting their own set of assumptions. At that point the discussion is derailed and there's no common ground to discuss about.
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by marcopolo »

willthrill81 wrote: Fri Jan 28, 2022 2:50 pm
marcopolo wrote: Fri Jan 28, 2022 2:47 pm
willthrill81 wrote: Fri Jan 28, 2022 2:44 pm
marcopolo wrote: Fri Jan 28, 2022 2:42 pm
willthrill81 wrote: Fri Jan 28, 2022 2:31 pm

Sadly, common sense seems to be rather uncommon.
If we were discussing common sense, no one would be using the 4% rule.
This discussion was specifically about the 4% rule, and the question posed in the "Theory" section of the forum was whether "ratcheting increases risk or not".

The fact that no one follows the 4% rule, and if things get bad, people cut back spending, etc. is absolutely true, but does not impact the answer to that question. The fact is absolutely true that ratcheting up increases risk of running out of money. You are spending a higher percentage of your portfolio, how can that not be true?!? The fact that in the real world, people can fairly easily mitigate that risk does not mean that the increased risk is not there.
The fact that spending more increases one's risk of running out of funds later seems to be very common sensical to me.
Yet, above, you kept insisting that it was "no more dangerous" :oops:
And it isn't any more dangerous or any less safe.

C'mon dude, you're beating a horse that wasn't even there in the first place. Lay off already.
Just one more beating of the dead horse, I promise.

How can it be simultaneously "no more dangerous" AND "increases risk of running out of funds"?!?

I will let you have the last word
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

marcopolo wrote: Fri Jan 28, 2022 2:54 pm
willthrill81 wrote: Fri Jan 28, 2022 2:50 pm
marcopolo wrote: Fri Jan 28, 2022 2:47 pm
willthrill81 wrote: Fri Jan 28, 2022 2:44 pm
marcopolo wrote: Fri Jan 28, 2022 2:42 pm

If we were discussing common sense, no one would be using the 4% rule.
This discussion was specifically about the 4% rule, and the question posed in the "Theory" section of the forum was whether "ratcheting increases risk or not".

The fact that no one follows the 4% rule, and if things get bad, people cut back spending, etc. is absolutely true, but does not impact the answer to that question. The fact is absolutely true that ratcheting up increases risk of running out of money. You are spending a higher percentage of your portfolio, how can that not be true?!? The fact that in the real world, people can fairly easily mitigate that risk does not mean that the increased risk is not there.
The fact that spending more increases one's risk of running out of funds later seems to be very common sensical to me.
Yet, above, you kept insisting that it was "no more dangerous" :oops:
And it isn't any more dangerous or any less safe.

C'mon dude, you're beating a horse that wasn't even there in the first place. Lay off already.
Just one more beating of the dead horse, I promise.

How can it be simultaneously "no more dangerous" AND "increases risk of running out of funds"?!?

I will let you have the last word
The "no more dangerous" means that ratcheting one's withdrawals up to 4% of the current portfolio balance is "no more dangerous" than starting one's withdrawals at 4%.

If a retiree in the year 2015 started by withdrawing $40k from a $1m portfolio, and this person's portfolio grew to $1.2m by 2020, that person could withdraw 4% of the new portfolio balance (i.e., $48k) with no more danger of prematurely running out of funds than a new retiree who just retired with a $1.2m portfolio and started withdrawing $48k from their portfolio.

This is not saying that the '4% rule' is safe, dangerous, or anything else. The 'no more danger' is with regard to comparing the risk of the 2015 retiree who is now withdrawing $48k from a $1.2m portfolio prematurely depleting the portfolio and the risk of the 2020 retiree who is also withdrawing $48k from a $1.2m portfolio prematurely depleting that portfolio. The risks of premature portfolio depletion are equal because the withdrawal rates are equal.

If the 2015 retiree did not ratchet up the withdrawal rate to 4% of the new portfolio balance, then of course, that retiree has a less chance of premature portfolio depletion than does the 2020 retiree who is withdrawing 4% of that retiree's portfolio balance.
Last edited by willthrill81 on Fri Jan 28, 2022 3:00 pm, edited 1 time in total.
The Sensible Steward
aristotelian
Posts: 12277
Joined: Wed Jan 11, 2017 7:05 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by aristotelian »

4% rule has a 1-7% failure rate depending on your timeframe and allocation. It is not completely failsafe. If you "ratchet up," every time you do so increases the odds that your sequence will fall in the failure group. Do it enough times and eventually you actually have a pretty decent chance of failure.

I also would ask what is to be gained by taking the risk. If 4% was enough for you at time of retirement, why do you need more?

That said, I think you can use common sense and assume that if you have a decent sequence of returns you have a bit more flexibility. If your portfolio doubles in the first five years, it is probably safe to increase your withdrawal by 25% which should provide you with plenty of extra luxuries you weren't anticipating.
Last edited by aristotelian on Fri Jan 28, 2022 3:02 pm, edited 1 time in total.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

aristotelian wrote: Fri Jan 28, 2022 2:59 pm 4% rule has a 5% failure rate, it is not completely failsafe.
I agree that it's not guaranteed to work in the future, but whether it has ever failed in the past depends on which data are being used. The Simba backtesting spreadsheet indicates that the 30 year SWR for a 60/40 AA of TSM and TBM was 4.1%, indicating that the '4% rule' never failed.
The Sensible Steward
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

willthrill81 wrote: Fri Jan 28, 2022 2:49 pm I completely understand why Bengen used the SWR-style method that he did. Certainly retirees prefer stable over unstable retirement income. But we all know that all sane people will make some kind of adjustment along the way, especially if their portfolio is performing poorly. The practical question then becomes whether 4% is a reasonable starting point for someone anticipating a retirement lasting up to around 30 years, and I believe that it still is. Anything much more involved than that is likely to suffer from the Mike Tyson effect: "Everyone has a plan until they get punched in the mouth."
Well...I don't think anyone questions if 4% is a reasonable starting point. Imo it still is, too.

Now the question is the impact of ratcheting up, that's what we're interested in. And "if SHTF then make some kind of adjustment along the way" just won't cut it in my opinion because it doesn't say anything about ratcheting up in particular.
aristotelian
Posts: 12277
Joined: Wed Jan 11, 2017 7:05 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by aristotelian »

willthrill81 wrote: Fri Jan 28, 2022 3:01 pm
aristotelian wrote: Fri Jan 28, 2022 2:59 pm 4% rule has a 5% failure rate, it is not completely failsafe.
I agree that it's not guaranteed to work in the future, but whether it has ever failed in the past depends on which data are being used. The Simba backtesting spreadsheet indicates that the 30 year SWR for a 60/40 AA of TSM and TBM was 4.1%, indicating that the '4% rule' never failed.
I am not familiar with Simba. ERN seems to suggest higher failure rates. Regardless, even 1% chance repeated enough times adds up to an unacceptable risk IMO.

https://earlyretirementnow.com/2016/12/ ... t-1-intro/
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

aristotelian wrote: Fri Jan 28, 2022 3:03 pm
willthrill81 wrote: Fri Jan 28, 2022 3:01 pm
aristotelian wrote: Fri Jan 28, 2022 2:59 pm 4% rule has a 5% failure rate, it is not completely failsafe.
I agree that it's not guaranteed to work in the future, but whether it has ever failed in the past depends on which data are being used. The Simba backtesting spreadsheet indicates that the 30 year SWR for a 60/40 AA of TSM and TBM was 4.1%, indicating that the '4% rule' never failed.
I am not familiar with Simba. Regardless, even 1% chance repeated enough times adds up to an unacceptable risk IMO.

https://earlyretirementnow.com/2016/12/ ... t-1-intro/
The Simba backtesting spreadsheet is a very useful data set with some great tools. It was developed by BHs for BHs.

Again, the problem with reading much into a '1% chance of failure' is that it assumes that someone will follow the '4% rule' rigidly, which nobody ever does nor should.
The Sensible Steward
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by marcopolo »

willthrill81 wrote: Fri Jan 28, 2022 2:59 pm
marcopolo wrote: Fri Jan 28, 2022 2:54 pm
willthrill81 wrote: Fri Jan 28, 2022 2:50 pm
marcopolo wrote: Fri Jan 28, 2022 2:47 pm
willthrill81 wrote: Fri Jan 28, 2022 2:44 pm

The fact that spending more increases one's risk of running out of funds later seems to be very common sensical to me.
Yet, above, you kept insisting that it was "no more dangerous" :oops:
And it isn't any more dangerous or any less safe.

C'mon dude, you're beating a horse that wasn't even there in the first place. Lay off already.
Just one more beating of the dead horse, I promise.

How can it be simultaneously "no more dangerous" AND "increases risk of running out of funds"?!?

I will let you have the last word
The "no more dangerous" means that ratcheting one's withdrawals up to 4% of the current portfolio balance is "no more dangerous" than starting one's withdrawals at 4%.

If a retiree in the year 2015 started by withdrawing $40k from a $1m portfolio, and this person's portfolio grew to $1.2m by 2020, that person could withdraw 4% of the new portfolio balance (i.e., $48k) with no more danger of prematurely running out of funds than a new retiree who just retired with a $1.2m portfolio and started withdrawing $48k from their portfolio.

This is not saying that the '4% rule' is safe, dangerous, or anything else. The 'no more danger' is with regard to comparing the risk of the 2015 retiree who is now withdrawing $48k from a $1.2m portfolio prematurely depleting the portfolio and the risk of the 2020 retiree who is also withdrawing $48k from a $1.2m portfolio prematurely depleting that portfolio. The risks of premature portfolio depletion are equal because the withdrawal rates are equal.

If the 2015 retiree did not ratchet up the withdrawal rate to 4% of the new portfolio balance, then of course, that retiree has a less chance of premature portfolio depletion than does the 2020 retiree who is withdrawing 4% of that retiree's portfolio balance.
So, this is now a different horse. And i agree with you.
The 2015 retiree, by ratcheting up, assumes the same risk as the 2020 retiree starting.

I think that is a somewhat (very?) different question than if a retiree increases his own risk by ratcheting up. The 2015 retiree increase his own risk of failure by ratcheting up instead of staying with his predetermined withdrawal amount.

So, perhaps we are in agreement, but discussing somewhat different scenario comparisons.
Once in a while you get shown the light, in the strangest of places if you look at it right.
randomguy
Posts: 11295
Joined: Wed Sep 17, 2014 9:00 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by randomguy »

aristotelian wrote: Fri Jan 28, 2022 2:59 pm
I also would ask what is to be gained by taking the risk. If 4% was enough for you at time of retirement, why do you need more?
Because I wanted more money but the cost to get that money (i.e. working another 3 years) wasn't worth it. If I can get that money for "Free" why not? Now in general I do agree with your point to a large extent. If I retire spending 100k, what the heck benefit is there to a scheme that lets me spend 250k when I am 90+? Other than charity/grandkids, who the heck is ramping their spending up like that as they age? But let me spend 120k from 65-75 instead of the 100k I was planning on? I could sort of see that...
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

marcopolo wrote: Fri Jan 28, 2022 3:07 pm
willthrill81 wrote: Fri Jan 28, 2022 2:59 pm
marcopolo wrote: Fri Jan 28, 2022 2:54 pm
willthrill81 wrote: Fri Jan 28, 2022 2:50 pm
marcopolo wrote: Fri Jan 28, 2022 2:47 pm

Yet, above, you kept insisting that it was "no more dangerous" :oops:
And it isn't any more dangerous or any less safe.

C'mon dude, you're beating a horse that wasn't even there in the first place. Lay off already.
Just one more beating of the dead horse, I promise.

How can it be simultaneously "no more dangerous" AND "increases risk of running out of funds"?!?

I will let you have the last word
The "no more dangerous" means that ratcheting one's withdrawals up to 4% of the current portfolio balance is "no more dangerous" than starting one's withdrawals at 4%.

If a retiree in the year 2015 started by withdrawing $40k from a $1m portfolio, and this person's portfolio grew to $1.2m by 2020, that person could withdraw 4% of the new portfolio balance (i.e., $48k) with no more danger of prematurely running out of funds than a new retiree who just retired with a $1.2m portfolio and started withdrawing $48k from their portfolio.

This is not saying that the '4% rule' is safe, dangerous, or anything else. The 'no more danger' is with regard to comparing the risk of the 2015 retiree who is now withdrawing $48k from a $1.2m portfolio prematurely depleting the portfolio and the risk of the 2020 retiree who is also withdrawing $48k from a $1.2m portfolio prematurely depleting that portfolio. The risks of premature portfolio depletion are equal because the withdrawal rates are equal.

If the 2015 retiree did not ratchet up the withdrawal rate to 4% of the new portfolio balance, then of course, that retiree has a less chance of premature portfolio depletion than does the 2020 retiree who is withdrawing 4% of that retiree's portfolio balance.
So, this is now a different horse. And i agree with you.
The 2015 retiree, by ratcheting up, assumes the same risk as the 2020 retiree starting.

I think that is a somewhat (very?) different question than if a retiree increases his own risk by ratcheting up. The 2015 retiree increase his own risk of failure by ratcheting up instead of staying with his predetermined withdrawal amount.

So, perhaps we are in agreement, but discussing somewhat different scenario comparisons.
Yes, the 2015 retiree is increasing his risk by ratcheting up his withdrawals, but that risk is no more than the risk the 2020 retiree faces as their withdrawal rates are equal. That's been my point all along.
The Sensible Steward
randomguy
Posts: 11295
Joined: Wed Sep 17, 2014 9:00 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by randomguy »

aristotelian wrote: Fri Jan 28, 2022 3:03 pm
willthrill81 wrote: Fri Jan 28, 2022 3:01 pm
aristotelian wrote: Fri Jan 28, 2022 2:59 pm 4% rule has a 5% failure rate, it is not completely failsafe.
I agree that it's not guaranteed to work in the future, but whether it has ever failed in the past depends on which data are being used. The Simba backtesting spreadsheet indicates that the 30 year SWR for a 60/40 AA of TSM and TBM was 4.1%, indicating that the '4% rule' never failed.
I am not familiar with Simba. ERN seems to suggest higher failure rates. Regardless, even 1% chance repeated enough times adds up to an unacceptable risk IMO.

https://earlyretirementnow.com/2016/12/ ... t-1-intro/
ERN using monthly data. A lot of sources use yearly. ERN also often looks at 40+ year time frames and changes failure from going broke to having less than x% less. Not using monthly (stocks were up like 25% in 1929 before the crash) always feels like a big mistake but the others are pure preference.

People have different opinions on what taking a 5% risk means. Some people go that means I only want a 5% chance of failure. Other people go I am willing to take 5% risks repeatedly if I win the first couple of times I take the risk and I don't care if that means my failure chance is 10% instead of 5% since each individual risk is only 5%. I don't remotely understand the logic behind the second point of view but every thread about this had numerous people suggesting that is how they view the problem.
randomguy
Posts: 11295
Joined: Wed Sep 17, 2014 9:00 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by randomguy »

willthrill81 wrote: Fri Jan 28, 2022 3:12 pm Yes, the 2015 retiree is increasing his risk by ratcheting up his withdrawals, but that risk is no more than the risk the 2020 retiree faces as their withdrawal rates are equal. That's been my point all along.
The question is why if in 2015, the retiree only wanted to take on a 5% chance of failure, why are they willing to now take on a 7% chance of failure ( the 5% risk they already took and the new risk which is say 2% due to the shorter time frame)? The 2020 retiree on the other hand is only taking on a 5% chance. If the 2015 retiree was ok taking on a 7% risk, why didn't they start with a higher SWR back in 2015?

Now historically after 5 years, the risk probably isn't 2% but more like .5% but maybe you are a monte carlo fan and think the SWR for the next 25 years is 2.5% in which case you might be picking up a lot more than 2% risk....
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

randomguy wrote: Fri Jan 28, 2022 3:31 pm
willthrill81 wrote: Fri Jan 28, 2022 3:12 pm Yes, the 2015 retiree is increasing his risk by ratcheting up his withdrawals, but that risk is no more than the risk the 2020 retiree faces as their withdrawal rates are equal. That's been my point all along.
The question is why if in 2015, the retiree only wanted to take on a 5% chance of failure, why are they willing to now take on a 7% chance of failure ( the 5% risk they already took and the new risk which is say 2% due to the shorter time frame)? The 2020 retiree on the other hand is only taking on a 5% chance. If the 2015 retiree was ok taking on a 7% risk, why didn't they start with a higher SWR back in 2015?

Now historically after 5 years, the risk probably isn't 2% but more like .5% but maybe you are a monte carlo fan and think the SWR for the next 25 years is 2.5% in which case you might be picking up a lot more than 2% risk....
But the risk the 2015 retiree faces by restarting the '4% rule' in 2020 is not higher than the risk the 2020 retiree faces. Their withdrawal rates are identical. The risk the 2015 faces is actually less because the 2015 retiree only has 25 years to go, whereas the 2020 retiree has 30 years remaining.
The Sensible Steward
sailaway
Posts: 8215
Joined: Fri May 12, 2017 1:11 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by sailaway »

willthrill81 wrote: Fri Jan 28, 2022 3:33 pm
randomguy wrote: Fri Jan 28, 2022 3:31 pm
willthrill81 wrote: Fri Jan 28, 2022 3:12 pm Yes, the 2015 retiree is increasing his risk by ratcheting up his withdrawals, but that risk is no more than the risk the 2020 retiree faces as their withdrawal rates are equal. That's been my point all along.
The question is why if in 2015, the retiree only wanted to take on a 5% chance of failure, why are they willing to now take on a 7% chance of failure ( the 5% risk they already took and the new risk which is say 2% due to the shorter time frame)? The 2020 retiree on the other hand is only taking on a 5% chance. If the 2015 retiree was ok taking on a 7% risk, why didn't they start with a higher SWR back in 2015?

Now historically after 5 years, the risk probably isn't 2% but more like .5% but maybe you are a monte carlo fan and think the SWR for the next 25 years is 2.5% in which case you might be picking up a lot more than 2% risk....
But the risk the 2015 retiree faces by restarting the '4% rule' in 2020 is not higher than the risk the 2020 retiree faces. Their withdrawal rates are identical. The risk the 2015 faces is actually less because the 2015 retiree only has 25 years to go, whereas the 2020 retiree has 30 years remaining.
For better or for worse, none of us knows how long we have remaining.

If writing an IPS, consider something like klangfool's split the difference off into a sinking fund or reset the % to split the difference. Ie, reset to 3.8% once the portfolio has grown enough that 3.8% is greater than the original 4%+ inflation.

Or accept that there is increased risk...
randomguy
Posts: 11295
Joined: Wed Sep 17, 2014 9:00 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by randomguy »

willthrill81 wrote: Fri Jan 28, 2022 3:33 pm But the risk the 2015 retiree faces by restarting the '4% rule' in 2020 is not higher than the risk the 2020 retiree faces. Their withdrawal rates are identical. The risk the 2015 faces is actually less because the 2015 retiree only has 25 years to go, whereas the 2020 retiree has 30 years remaining.
Nobody disputes that. But why does the 2015 want to take on that added risk? Why do they want to take on the risk of either of the 2015 or 2020 periods failing instead of just taking on the risk of the 2015 period failing? If you are playing Russian Roulette and you don't die the first time you pull the trigger, do you take another pull just because you didn't die? After all the odds didn't change. If when you blow off your head, if your friend also dies are you better off? Misery might love company but you might prefer not to be miserable....

Now historically this scheme has pretty mild failures. You are going to up your failure rate to ~10% and you will be failing in year 29 or so of the retirement. If you were doing something like this over a 40-50 years retirement, you might find the added risk unacceptable. For 30 years? Historically it doesn't matter....
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

randomguy wrote: Fri Jan 28, 2022 4:10 pm
willthrill81 wrote: Fri Jan 28, 2022 3:33 pm But the risk the 2015 retiree faces by restarting the '4% rule' in 2020 is not higher than the risk the 2020 retiree faces. Their withdrawal rates are identical. The risk the 2015 faces is actually less because the 2015 retiree only has 25 years to go, whereas the 2020 retiree has 30 years remaining.
Nobody disputes that. But why does the 2015 want to take on that added risk? Why do they want to take on the risk of either of the 2015 or 2020 periods failing instead of just taking on the risk of the 2015 period failing? If you are playing Russian Roulette and you don't die the first time you pull the trigger, do you take another pull just because you didn't die? After all the odds didn't change. If when you blow off your head, if your friend also dies are you better off? Misery might love company but you might prefer not to be miserable....

Now historically this scheme has pretty mild failures. You are going to up your failure rate to ~10% and you will be failing in year 29 or so of the retirement. If you were doing something like this over a 40-50 years retirement, you might find the added risk unacceptable. For 30 years? Historically it doesn't matter....
Who are we to say how much risk retirees should be willing to take on? If an intelligent, informed person understands the risks and chooses to take them on, that's their choice.

I have never in this thread argued that retirees should ratchet up their spending. My only point all along has been that the retirees who restart a SWR are not taking on more risk than a new retiree starting at the same WR. That's all.
The Sensible Steward
seajay
Posts: 1656
Joined: Sat May 01, 2021 3:26 pm
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by seajay »

willthrill81 wrote: Fri Jan 28, 2022 4:14 pmretirees who restart a SWR are not taking on more risk than a new retiree starting at the same WR.
Consider a retiree who opts for a bucket approach. Starts with 25/75 stock/bonds, spends bonds first, doesn't rebalance ... that for a 25 year 3%/year SWR transitions to being 100/0 after 25 years. Averages 62.5/37.5 up to the point when all of bonds had been spent. Another retiree who instead prefers constant weighted 62.5/37.5 (yearly rebalanced) broadly has the same probabilities as the bucket investor, however their actual individual outcomes will tend to differ according to the sequence of returns.

If valuations are high at the start then the bucket approach perhaps has the better prospects, whilst if valuations are low at the start then constant weighted is more likely to have the better prospects. Absent of any firm view of valuations and 50/50 of the two will tend to be better than the worse case alone. As might stopping and restarting with the same SWR but where you flipped from one choice/style to the other according to perceived valuations - started with constant weighted, did well and valuations soared, so stopped and restarted but using buckets for instance.

PS Inspect historic worst cases for each of bucket and constant weighted and in some (many?) cases the worst start years will differ, a factor of having different degrees of (weightings to) risk exposure and sequence of returns.

I see no harm in stop/restarts, which might also include revision of how much $$$ are being drawn (same SWR % applied to a higher base capital value a.k.a. ratchet).
randomguy
Posts: 11295
Joined: Wed Sep 17, 2014 9:00 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by randomguy »

willthrill81 wrote: Fri Jan 28, 2022 4:14 pm Who are we to say how much risk retirees should be willing to take on? If an intelligent, informed person understands the risks and chooses to take them on, that's their choice.
Definitely. The question is why on day 1 of retirement did they want a 5% risk and then later on did they decide to increase the risk? If they were happy with a 7% failure rate in year 5, why weren't they happy with that failure rate in year 1? There might be reasonable reasons why you are ok upping your risk (maybe the reward is better. You get to spend 50k instead of say 43k. Maybe the failure cases are better like running out of money in year 28 instead of 26). But if you are choosing to add risk to your retirement you should be aware of it and not try to justify it by saying the person next to me will also fail.

Again we are talking about marginal failures with the historical data. Maybe something like a 90% stock guy in Japan in 1985 would end up with crazy different results by ramping up until the crash but in US history with a balanced portfolio, the failure cases are all pretty mellow to start with and reducing the time to failure is only going to make them more palatable.
lostdog
Posts: 5368
Joined: Thu Feb 04, 2016 1:15 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by lostdog »

We'll be doing this based on annual portfolio balance with flexibility.
Stocks-80% || Bonds-20% || Taxable-VTI/VXUS || IRA-VT/BNDW
Kenneth Almquist
Posts: 65
Joined: Mon Sep 26, 2016 1:59 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Kenneth Almquist »

foursix wrote: Wed Jan 26, 2022 6:02 pm I've been looking at various retirement withdrawal methods and have a question regarding the standard 4% method. If your portfolio value grows in retirement such that a new 4% calculation yields a larger withdrawal than you're currently taking, is it OK to reset your WR to the new value?

It seems to me that's the same as basically restarting your retirement, for withdrawal purposes. Assuming the 4% rule never fails, it shouldn't fail with the new amount either, right?
That's basically my strategy, but with a 3.25% withdrawal rate rather that a 4% withdrawal rate. The assumption that a 4% withdrawal rate never fails (or more precisely, that the risk of failure is too small to worry about) is not one that I would make.
StillGoing
Posts: 360
Joined: Mon Nov 04, 2019 3:43 am
Location: U.K.

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by StillGoing »

willthrill81 wrote: Fri Jan 28, 2022 12:38 pm

Aside from all that, I personally put more faith in the quality of the Simba spreadsheet than cFIREsim, but that's just me. And as I've noted, if the '4% rule' works for 'only' 29.5 years vs. 31 isn't a big deal. As you note, the real world is not that binary.
I'm not entirely sure exactly what is in the Simba spreadsheet (I've not yet used it in my modelling), but the likely errors in SWR's may be much larger than is anticipated.

1) For example: Going from monthly returns to annual returns using a different start month can change the SAFEMAX value by 20 basis points (70 basis points for an asset allocation of 100%) [using Shiller returns] - I was puzzled by why I was getting different answers to cfiresim (as far as I can tell, that uses January when I was using December).

2) Changing the bond duration (e.g. by using a mix of both bills and bonds or synthesising the returns of a bond fund as people on this forum have done) can change the SAFEMAX value by 50 basis points (this is strongly dependent on country - for the USA and UK, using bills instead of bonds tends to increase the SWR for poor retirements, while this does not happen for Switzerland and Japan).

3) Inflation model. While the US appears to have a single official set of historical inflation values, the UK has at least 4 models that I have found (for pre-1950) which lead to a differences in SAFEMAX of up to about 20 bps.

4) Shiller uses monthly average prices rather than end of month prices, and the SWR does differ if you use the latter (I'm sorry, I don't currently have the difference to hand).

cheers
StillGoing
milktoast
Posts: 626
Joined: Wed Jul 10, 2019 8:17 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by milktoast »

I agree that ratcheting up each year is equally safe to starting a 4% SWR that year. Math says that is true.

But here is more math. If we assume that each year has a 5% uncorrelated of failure, then each time you reset it’s a fresh roll of a 20 sided die. If you ever roll a 1, you fail.

I think from that, you can see the flaw of resetting. Start with 4% you have 95% chance of success. Reroll once and now you are at 0.95*0.95. Only a 90% chance of success.

In reality, it’s not uncorrelated. So the math is hard. But think about it, you will always be at 4% of your highest ever balance. That is much more likely to fail because you’ll alway be withdrawing as if you retired at the peak. Where without ratchet you had to get unlucky to retire at the peak.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

milktoast wrote: Sat Jan 29, 2022 1:30 pm I agree that ratcheting up each year is equally safe to starting a 4% SWR that year. Math says that is true.

But here is more math. If we assume that each year has a 5% uncorrelated of failure, then each time you reset it’s a fresh roll of a 20 sided die. If you ever roll a 1, you fail.

I think from that, you can see the flaw of resetting. Start with 4% you have 95% chance of success. Reroll once and now you are at 0.95*0.95. Only a 90% chance of success.

In reality, it’s not uncorrelated. So the math is hard. But think about it, you will always be at 4% of your highest ever balance. That is much more likely to fail because you’ll alway be withdrawing as if you retired at the peak. Where without ratchet you had to get unlucky to retire at the peak.
Restarting 4% withdrawals is not equivalent to 'rerolling the dice'. The risk of premature portfolio depletion in fewer than 30 years for someone who has been retired and is withdrawing 4% of his current portfolio balance is identical in every way to a new retiree who starts with the same 4% withdrawal rate.

Remember that it's necessary for the retiree's portfolio to have increased in value in order for the ratcheting up to occur. As such, it's wrong to view ratcheting up as another roll of the dice.

In virtually all of the historic scenarios where the maximum WR was 4% or close, the portfolio dropped in value very shortly after the individual retired, which pushed the effective amount being withdrawn to well above 4%.

What many have correctly pointed out is that if a retiree resets his withdrawals to 4% of his current portfolio balance, he is increasing his risk of premature portfolio depletion compared to not resetting the withdrawals. But that does not mean that the retiree is taking on more risk than a new retiree who is starting his withdrawals at a 4% rate.
The Sensible Steward
milktoast
Posts: 626
Joined: Wed Jul 10, 2019 8:17 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by milktoast »

willthrill81 wrote: Sat Jan 29, 2022 1:35 pm But that does not mean that the retiree is taking on more risk than a new retiree who is starting his withdrawals at a 4% rate.
I agree. They are taking a slightly reduced risk (shorter horizon) than those retiring that year.

However, there is extra cumulative risk from a reset strategy.

If everyone who starts in 1966 fail, but those who start earlier or later do not. The ratchet guarantees that the 1965 and 1964 join their 1966 cohort in failure.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

milktoast wrote: Sat Jan 29, 2022 2:05 pm
willthrill81 wrote: Sat Jan 29, 2022 1:35 pm But that does not mean that the retiree is taking on more risk than a new retiree who is starting his withdrawals at a 4% rate.
I agree. They are taking a slightly reduced risk (shorter horizon) than those retiring that year.

However, there is extra cumulative risk from a reset strategy.

If everyone who starts in 1966 fail, but those who start earlier or later do not. The ratchet guarantees that the 1965 and 1964 join their 1966 cohort in failure.
That's not a cumulative risk though. By resetting their withdrawal rate, the 1965 retirees were simply 'hitching their wagon' to the fate of the 1966 retirees.
The Sensible Steward
Post Reply