The Day the 4% Rule Died

[randomguy]

Heck, 20% of American men aged 65 don't even survive to age 74 (according to the SSA).

If you are in that 20% cohort you are absolutely not going to be stressing about your portfolio at age 80...

How do you know if you are ahead of time?

At 64.5 go to the Doctor and they tell you that you stage 4 cancer and you have 9 months to live, you have a very good idea of what cohort you are in....

Seriously you can't know for sure but a lot of people have some pretty big clues. If you are 200lbs overweight, smoke 2packs/day, have diabetes and Cirrhosis, you probably don't need to plan out to 95. Or at the other extreme, if life expectancy of your peers is 95, the odds of you making it to 80 are very good.... Can you tell anything at 100%? Of course not. Well enough for reasonable planning purposes? Probably. Now for most of us it makes no difference. You can't cut off enough years (especially if you are a couple) to go hog wild and a few more years of life expectancy just don't move the numbers in terms of having to spend less.

[FactualFran]

Having a balance after 15 years equal to the initial balance adjusted for inflation was no guarantee of 30 years of withdrawals. With the 60/40 portfolio, a starting year of 1985, and an initial withdrawal rate that resulted in the balance after 15 years of inflation-adjusted withdrawal being equal to the initial balance adjusted for inflation, the portfolio would have supported only 8 more years of withdrawals.

So retirements starting in 1985 would have seen the 4% "rule" fail in year 23? I've never heard this before, anywhere. Or are you saying something different, regarding keeping the portfolio value the same in real, inflation adjusted, terms, by withdrawing less than 4%?

No, an initial withdrawal rate of 4% would not have failed in year 23 with 1985 as the staring year and a 60/40 portfolio of Large-Cap Stocks and Intermediate-Term Government bonds. The initial withdrawal rate that resulted in the balance after 15 years of inflation-adjusted withdrawals being equal to the initial balance adjusted for inflation was more than 4%. 4% is not the only possible initial withdrawal rate.

[McQ]

Contrary to randomguy, failure does not presume negative returns from stocks and bonds for a decade +. That is not how I set up the spreadsheet. In particular, a crash is not required. Bengen's rule had no trouble surmounting 1929: there was deflation not inflation, and bonds served as ballast.

In your spreadsheet you have negative CAGRs for both stocks and bonds for 10 years. I have you at a -3.3% CAGR for those first 10 years. Feel free to correct my math. And of course you have basically the worst crash in US history which gives you a horrible SOR. I am sure you can come up with some scheme where you have a positive CAGR over the first 10 years by really front loading losses. Make your draw down over 5 years instead of 2 and things get worse.

A 2022 retiree isn't betting a big bull market surge. They are betting these economic conditions don't turn into the worst 2 year period every and the economic recovery doesn't last a decade. It is only after they that decade shows up that you are betting on that bull market....

If you look at your example why did you fail?
a) Very Poor sequence of returns. Front loading the losses like you did really hurts
b) Very poor stock market returns over the whole time period. You have 20 years where the stocks return 7.3%. We have had a half a dozen of those since 1920 so they are possible but it is a bottom quartile result. But how many of those 20 year periods follow one of the worst 10 year periods for stocks in US history? It normally doesn't work that way.

There is always the chance this is the worst time ever. But we still need a lot of stuff to go wrong for this to be true.

Randomguy, I think our dispute is about semantics not math. When you said “crash,” I heard “a stock market decline significantly worse than minus 50%.” That’s not in my spreadsheet; I only have an ordinary bear, a little less than minus 50%. Compounded, however, with a significant bond market decline (the prospect of which makes 2022 thus far so interesting). And whatever the compounded growth rate in the first n years, once I plopped in a bull surge after 2028, all was fine.

And that’s the point of the speculation:
1. How many Bogleheads understand that the Bengen rule held because, after a downturn, US stocks have always surged to a new high sufficient to restore long-term returns to equal 1.0 Siegels?
2. And how many understand that after a downturn and partial recovery, an extended stretch of 7.3% nominal on stocks, 4.3% real, just isn’t good enough—even with a yield (and return) of 4.5% nominal on bonds?

[randomguy]

Heck, a 3% WR has been so conservative that over most periods, investors with global stock allocations in developed nations could have taken the greater of 3% of their current portfolio balance OR last year's withdrawal plus inflation and only rarely have seen their portfolio decline at all in inflation-adjusted value over the long-term (i.e., 20+ years).

The question is what do you do though when those rare cases show up? We all know that normally that SWR is up around 6% or you end up with like 2x your starting value or more. The problem is what on earth do you do about the bad times. Lets say 3% never fails. Great. But how will you feel 15 years in when your portfolio is off 65%.... At some point are you cutting to 2% so that you are only off 55%?

The OP sequence is sort of plausible case of how things fail. You have poor returns for the first 10 years. You have a big crash. And then you have slightly below average returns for 20 years. Maybe in the real world we get there by 10 poor years, 10 good years, 10 poor years.

Imagine our OP poster chart is right. Think about our 2000 60/40 guy who was looking really good last July with 800k and only 8.5 years left. You then get the -32% real, -25% real drop . We are now sitting at 320k with 6 years left. Things should be good but things have tightened up a lot. And it is going to be hard to make much of a recovery. You make 10% real and your portfolio drops in value after taking out 40k. You are going to limp to the finish line and really hope you don't need 35 years. And this was for a person who 22 years in was looking good...

[McQ]

When will the 4% rule fail? What is the combination of circumstances required? As willthrill81 pointed out, it has failed outside the US before (detailed case studies here: https://papers.ssrn.com/sol3/papers.cfm ... id=4001986). But why? How did the course of these markets differ from the US?

The primary failure cases are WWI, WWII, going communist and Nikkei crash.

Nigel_ht, I would not expect you to read in detail a paper when a preliminary scan indicated that it had nothing new to offer you.

Had you done so, you would have seen that there were no WW II examples in the paper (those were all ruled out as cheap shots, along with civil war and defeat in war and also revolutionary overthrow, e.g., going communist.)

The only WW I examples were of a victor (UK) and a neutral (Sweden). Retirees in each case would have done better if they had supplemented their domestic stock market with a world stock index investment.

An international stock index would have helped the US investor in 1965, not so much in 1910. YMMV.

The most interesting findings in the paper, to me, were the poor results for the post-1960 Italian and French retirees. No world war. No civil war. No revolution. Just high inflation plus weak stock and bond returns for a long time.

[Marseille07]

The question is what do you do though when those rare cases show up? We all know that normally that SWR is up around 6% or you end up with like 2x your starting value or more. The problem is what on earth do you do about the bad times. Lets say 3% never fails. Great. But how will you feel 15 years in when your portfolio is off 65%.... At some point are you cutting to 2% so that you are only off 55%?

The OP sequence is sort of plausible case of how things fail. You have poor returns for the first 10 years. You have a big crash. And then you have slightly below average returns for 20 years. Maybe in the real world we get there by 10 poor years, 10 good years, 10 poor years.

Imagine our OP poster chart is right. Think about our 2000 60/40 guy who was looking really good last July with 800k and only 8.5 years left. You then get the -32% real, -25% real drop . We are now sitting at 320k with 6 years left. Things should be good but things have tightened up a lot. And it is going to be hard to make much of a recovery. You make 10% real and your portfolio drops in value after taking out 40k. You are going to limp to the finish line and really hope you don't need 35 years. And this was for a person who 22 years in was looking good...

Use another withdrawal method.

SWR is not without merit but there is an obvious downside that is possibly running out of money. If you love the go-go years, VPW works really well in my opinion.

[randomguy]

The question is what do you do though when those rare cases show up? We all know that normally that SWR is up around 6% or you end up with like 2x your starting value or more. The problem is what on earth do you do about the bad times. Lets say 3% never fails. Great. But how will you feel 15 years in when your portfolio is off 65%.... At some point are you cutting to 2% so that you are only off 55%?

The OP sequence is sort of plausible case of how things fail. You have poor returns for the first 10 years. You have a big crash. And then you have slightly below average returns for 20 years. Maybe in the real world we get there by 10 poor years, 10 good years, 10 poor years.

Imagine our OP poster chart is right. Think about our 2000 60/40 guy who was looking really good last July with 800k and only 8.5 years left. You then get the -32% real, -25% real drop . We are now sitting at 320k with 6 years left. Things should be good but things have tightened up a lot. And it is going to be hard to make much of a recovery. You make 10% real and your portfolio drops in value after taking out 40k. You are going to limp to the finish line and really hope you don't need 35 years. And this was for a person who 22 years in was looking good...

Use another withdrawal method.

SWR is not without merit but there is an obvious downside that is possibly running out of money. If you love the go-go years, VPW works really well in my opinion.

Easy to say. harder to do. Look at that poor VPW guy. Barely spend any money for the first 15 years of retirement. Finally started spending 5 years ago. And now is back to spending no money. Sure they are taking out like 12% but the portfolio has shrunk enough so that instead of spending 80k/year you did for like 2 years, you are barely getting 40k... Constant percentage guy is even worse. They never spent money. They hoarded it and watched all their gains dissipated without every getting used productively.

The solution is going to be to derisk (i.e. they should have bough 10 years of tips in say 2020) to avoid the portfolio losses. But I am not sure you are going to be able to do that algorithmically.

[randomguy]

Pretty
Randomguy, I think our dispute is about semantics not math. When you said “crash,” I heard “a stock market decline significantly worse than minus 50%.” That’s not in my spreadsheet; I only have an ordinary bear, a little less than minus 50%. Compounded, however, with a significant bond market decline (the prospect of which makes 2022 thus far so interesting). And whatever the compounded growth rate in the first n years, once I plopped in a bull surge after 2028, all was fine.

And that’s the point of the speculation:
1. How many Bogleheads understand that the Bengen rule held because, after a downturn, US stocks have always surged to a new high sufficient to restore long-term returns to equal 1.0 Siegels?
2. And how many understand that after a downturn and partial recovery, an extended stretch of 7.3% nominal on stocks, 4.3% real, just isn’t good enough—even with a yield (and return) of 4.5% nominal on bonds?

Pretty much. You aren't talking about an average bear. You are talking about the worst 1 or 2 ever. The average bear is a lot more like 2020... If you want to call 1929 the only crash in US history you are free to do so. I think that is pretty restrictive. We have had like 20 bear markets in the last 100 years (peoples definition vary). 1929, 1973,2000,2007 were the worst. This one would either be the first or 2nd worst. And I am guessing that is the worst bond bear market every. And then you have a weak stock market recovery afterwards.

I am not sure people understand how bad the stock market returns you gave are. You are looking at something like 2.5% real for a 30 year period. That is about half of the worst time in post 1920 US history.

1. Isn't quite right. In the bad periods stocks didn't surge to get back to 1.0 Siegels. But you did get ~5% instead of the 2.5% you used. You need a lot of abnormally good years to get back to average after a 50% loss

2. That I don't know. I think a lot of people look at your numbers and go that was a nice recovery rather than stocks being off 35% or so after year 4. It isn't very intuitive how a 30% drop followed by a 30% gain leaves you far behind

In the end the debate will be is can we get returns like this. There have been plenty of papers talking about how 2 or 3% is the new 4%. They all use this basic logic of discounting historical stock market returns by 30-50%. And then when you get a bad sequence, you get this type of result. But it is very sensitive to assumptions. Take your example. Imagine stocks bounce back to 80% instead of 65%. Things will look a lot better when you are bleeding like 3k/year instead of 12k in your "steady state". It will take a lot longer to spiral down to the drain...

[Leesbro63]

Pretty
Randomguy, I think our dispute is about semantics not math. When you said “crash,” I heard “a stock market decline significantly worse than minus 50%.” That’s not in my spreadsheet; I only have an ordinary bear, a little less than minus 50%. Compounded, however, with a significant bond market decline (the prospect of which makes 2022 thus far so interesting). And whatever the compounded growth rate in the first n years, once I plopped in a bull surge after 2028, all was fine.

And that’s the point of the speculation:
1. How many Bogleheads understand that the Bengen rule held because, after a downturn, US stocks have always surged to a new high sufficient to restore long-term returns to equal 1.0 Siegels?
2. And how many understand that after a downturn and partial recovery, an extended stretch of 7.3% nominal on stocks, 4.3% real, just isn’t good enough—even with a yield (and return) of 4.5% nominal on bonds?

Pretty much. You aren't talking about an average bear. You are talking about the worst 1 or 2 ever. The average bear is a lot more like 2020... If you want to call 1929 the only crash in US history you are free to do so. I think that is pretty restrictive. We have had like 20 bear markets in the last 100 years (peoples definition vary). 1929, 1973,2000,2007 were the worst. This one would either be the first or 2nd worst. And I am guessing that is the worst bond bear market every. And then you have a weak stock market recovery afterwards.

I am not sure people understand how bad the stock market returns you gave are. You are looking at something like 2.5% real for a 30 year period. That is about half of the worst time in post 1920 US history.

1. Isn't quite right. In the bad periods stocks didn't surge to get back to 1.0 Siegels. But you did get ~5% instead of the 2.5% you used. You need a lot of abnormally good years to get back to average after a 50% loss

2. That I don't know. I think a lot of people look at your numbers and go that was a nice recovery rather than stocks being off 35% or so after year 4. It isn't very intuitive how a 30% drop followed by a 30% gain leaves you far behind

In the end the debate will be is can we get returns like this. There have been plenty of papers talking about how 2 or 3% is the new 4%. They all use this basic logic of discounting historical stock market returns by 30-50%. And then when you get a bad sequence, you get this type of result. But it is very sensitive to assumptions. Take your example. Imagine stocks bounce back to 80% instead of 65%. Things will look a lot better when you are bleeding like 3k/year instead of 12k in your "steady state". It will take a lot longer to spiral down to the drain...

What is a "Siegel"? I suppose someone defined it somewhere here, but I can't find it.

[willthrill81]

The question is what do you do though when those rare cases show up? We all know that normally that SWR is up around 6% or you end up with like 2x your starting value or more. The problem is what on earth do you do about the bad times. Lets say 3% never fails. Great. But how will you feel 15 years in when your portfolio is off 65%.... At some point are you cutting to 2% so that you are only off 55%?

The OP sequence is sort of plausible case of how things fail. You have poor returns for the first 10 years. You have a big crash. And then you have slightly below average returns for 20 years. Maybe in the real world we get there by 10 poor years, 10 good years, 10 poor years.

Imagine our OP poster chart is right. Think about our 2000 60/40 guy who was looking really good last July with 800k and only 8.5 years left. You then get the -32% real, -25% real drop . We are now sitting at 320k with 6 years left. Things should be good but things have tightened up a lot. And it is going to be hard to make much of a recovery. You make 10% real and your portfolio drops in value after taking out 40k. You are going to limp to the finish line and really hope you don't need 35 years. And this was for a person who 22 years in was looking good...

Use another withdrawal method.

SWR is not without merit but there is an obvious downside that is possibly running out of money. If you love the go-go years, VPW works really well in my opinion.

Easy to say. harder to do. Look at that poor VPW guy. Barely spend any money for the first 15 years of retirement. Finally started spending 5 years ago. And now is back to spending no money. Sure they are taking out like 12% but the portfolio has shrunk enough so that instead of spending 80k/year you did for like 2 years, you are barely getting 40k... Constant percentage guy is even worse. They never spent money. They hoarded it and watched all their gains dissipated without every getting used productively.

The solution is going to be to derisk (i.e. they should have bough 10 years of tips in say 2020) to avoid the portfolio losses. But I am not sure you are going to be able to do that algorithmically.

VPW is not the only reasonable withdrawal method with flexibility. The ABW method is even more flexible, and by incorporating expected forward returns into the method, it balances security with reduced withdrawal volatility (compared to VPW or fixed percentage-of-portfolio approaches) pretty well.

[Marseille07]

Easy to say. harder to do. Look at that poor VPW guy. Barely spend any money for the first 15 years of retirement. Finally started spending 5 years ago. And now is back to spending no money. Sure they are taking out like 12% but the portfolio has shrunk enough so that instead of spending 80k/year you did for like 2 years, you are barely getting 40k... Constant percentage guy is even worse. They never spent money. They hoarded it and watched all their gains dissipated without every getting used productively.

The solution is going to be to derisk (i.e. they should have bough 10 years of tips in say 2020) to avoid the portfolio losses. But I am not sure you are going to be able to do that algorithmically.

VPW starts off around 4.3% WR and goes higher (this depends on your N and AA; and obviously the actual dollar amount depends on the market). This is likely higher than your 4% SWR, at least initially. If this is "barely spend any money" then your expectation is quite off I must say.

[randomguy]

VPW starts off around 4.3% WR and goes higher (this depends on your N and AA; and obviously the actual dollar amount depends on the market). This is likely higher than your 4% SWR, at least initially. If this is "barely spend any money" then your expectation is quite off I must say.

You can't pay bills with WR. Focus on the number that matters which is the sub 30k/year that you are getting. And sure if you don't need the money that is fine. But most of us need our money and just didn't save to see numbers on a piece of paper.

[randomguy]

VPW is not the only reasonable withdrawal method with flexibility. The ABW method is even more flexible, and by incorporating expected forward returns into the method, it balances security with reduced withdrawal volatility (compared to VPW or fixed percentage-of-portfolio approaches) pretty well.

Plot up ABW for this case and you are going to get similar results. The problem is the totally crappy returns.... If you have 1 million dollars and you are going to lose 50% of it, there isn't a big difference between spending 40k of it and then losing 50% or just losing 50%.... That losing 50% is the killer.

[randomguy]

What is a "Siegel"? I suppose someone defined it somewhere here, but I can't find it.

The historical real stock market return (6.6%) according to the OP. I personally have never heard any one using it and have no clue if it is the average at the time the book was written or if it is an on going average.

[HomerJ]

And that’s the point of the speculation:
1. How many Bogleheads understand that the Bengen rule held because, after a downturn, US stocks have always surged to a new high sufficient to restore long-term returns to equal 1.0 Siegels?
2. And how many understand that after a downturn and partial recovery, an extended stretch of 7.3% nominal on stocks, 4.3% real, just isn’t good enough—even with a yield (and return) of 4.5% nominal on bonds?

Well, I've been telling people for years that, so far, the long-term average returns (1.0 Siegel - very weird term), INCLUDE the crashes.

That, every time, so far, after a crash, the U.S. stock market recovers strongly enough to reach new highs and give us good average returns even with the crash.

Yes, in your postulated scenario, if this time... we get a crash and then we don't get some really good years to make up for it later, that will cause 4% to fail.

All you are saying is... "Hey, you know if the next 30 years has a much lower average return than the worst 30 year-period on record, 4% will fail!"


Well, duh.

4% represents (roughly), the worst 30-year period in the past. That's a pretty conservative starting point.

If we've had 120 rolling 30-year periods, divided as below (made up numbers, but representative)

3 of those periods - you could have pulled 9% a year
12 of those periods - you could have pulled 8% a year
19 of those periods - you could have pulled 7% a year
41 of those periods - you could have pulled 6% a year
34 of those periods - you could have pulled 5% a year
11 of those periods - you could have pulled 4% a year

So going with 4% is a pretty conservative starting point.

Sure, if the next 30 years is worst in history, then 4% will fail.

But the stock market runs in cycles for a reason. Returns year-to-year (or at least 3-year-to-3-year periods or 5-year-to-5-year periods) are not fully independent events.

Low return periods change the variables so that high return periods are more likely to follow.

Is the future guaranteed to repeat the past? Absolutely not. We might see a big crash, and then just get low, average returns going forward from then on.

But so far, that's not how it has worked in the past.

[Marseille07]

You can't pay bills with WR. Focus on the number that matters which is the sub 30k/year that you are getting. And sure if you don't need the money that is fine. But most of us need our money and just didn't save to see numbers on a piece of paper.

You're all over the map.

You want to front-load, then when SWR dwindles 15 years later, you ask what the retiree should do?

Give me a break. You face the consequences then, whatever they are. Every method has tradeoffs whichever you end up choosing.

[McQ]

What is a "Siegel"? I suppose someone defined it somewhere here, but I can't find it.

The historical real stock market return (6.6%) according to the OP. I personally have never heard any one using it and have no clue if it is the average at the time the book was written or if it is an on going average.

My bad, Leesbro63, I’ve spent so many years with Jeremy Siegel’s book, Stocks for the Long Run, open on my desk that I thought his findings were part of the lingua franca among Bogleheads. The 5th edition has US stocks returning 6.6% real from 1802 to 2012; earlier editions using somewhat different historical data sets had it at 6.7%. Global returns have been considerably lower.

You may enjoy this Siegel thread someone started today: viewtopic.php?t=380293

[Admiral]

Wow seven pages on this.

I'd like to meet the retiree who continues blindly and blithely spending 4% per year when the stock market drops 50%, bonds tank, and there is no recovery. I find it very hard to believe such a person exists.

If your retirement budget and planning shows that you cannot cut back at all--that is, you've neglected to build in some discretionary expenses that cushion you in deep downturns--then your "plan" may fail.

Hope is not a plan. Take one less vacation or don't buy a new car. This does not seem complicated to me.

[59Gibson]

If we experience the worse 30yr period from here, folks will adapt.
Cut, move, un- retire -all the above. Living standards will naturally change. "necessities " will be different if we're in essentially 0% or negative growth for nearly 3 decades.

[Marseille07]

Wow seven pages on this.

I'd like to meet the retiree who continues blindly and blithely spending 4% per year when the stock market drops 50%, bonds tank, and there is no recovery. I find it very hard to believe such a person exists.

If your retirement budget and planning shows that you cannot cut back at all--that is, you've neglected to build in some discretionary expenses that cushion you in deep downturns--then your "plan" may fail.

Hope is not a plan. Take one less vacation or don't buy a new car. This does not seem complicated to me.

Randomguy on this thread is one such person, because when anyone suggests variable withdrawals, they're so concerned that variable withdrawals might not provide enough income for them.

They also seem to love the front-loading idea, which pretty much requires steady withdrawals early in retirement (otherwise you aren't front-loading).

[willthrill81]

VPW is not the only reasonable withdrawal method with flexibility. The ABW method is even more flexible, and by incorporating expected forward returns into the method, it balances security with reduced withdrawal volatility (compared to VPW or fixed percentage-of-portfolio approaches) pretty well.

Plot up ABW for this case and you are going to get similar results. The problem is the totally crappy returns.... If you have 1 million dollars and you are going to lose 50% of it, there isn't a big difference between spending 40k of it and then losing 50% or just losing 50%.... That losing 50% is the killer.

No withdrawal method can provide 100% security and 0% withdrawal volatility. The two are incompatible, and most withdrawal methods other than a strict SWR and a percentage-of-portfolio approach are somewhere between those extremes. That said, using a combination of something like 1/CAPE and current TIPS' yields in proportion to one's AA has been a pretty good balance between smoothing returns while also guaranteeing that you won't completely deplete your portfolio prematurely (as long as the underlying assets don't go to zero, of course).

That said, if you look at something at drastic as the drawdown from 2008-2009, using the above approach would have resulted in remarkably little reduction in one's withdrawals because expected returns went up in almost direct proportion to stocks' drop in price.

[marcopolo]

VPW is not the only reasonable withdrawal method with flexibility. The ABW method is even more flexible, and by incorporating expected forward returns into the method, it balances security with reduced withdrawal volatility (compared to VPW or fixed percentage-of-portfolio approaches) pretty well.

Plot up ABW for this case and you are going to get similar results. The problem is the totally crappy returns.... If you have 1 million dollars and you are going to lose 50% of it, there isn't a big difference between spending 40k of it and then losing 50% or just losing 50%.... That losing 50% is the killer.

No withdrawal method can provide 100% security and 0% withdrawal volatility. The two are incompatible, and most withdrawal methods other than a strict SWR and a percentage-of-portfolio approach are somewhere between those extremes. That said, using a combination of something like 1/CAPE and current TIPS' yields in proportion to one's AA has been a pretty good balance between smoothing returns while also guaranteeing that you won't completely deplete your portfolio prematurely (as long as the underlying assets don't go to zero, of course).

That said, if you look at something at drastic as the drawdown from 2008-2009, using the above approach would have resulted in remarkably little reduction in one's withdrawals because expected returns went up in almost direct proportion to stocks' drop in price.

My observation is that valuations are only used as a hand-wringing argument to drive WR ever lower. I don't see much talk of increasing withdrawal rates (and thus spending) while portfolios are crashing.

[willthrill81]

VPW is not the only reasonable withdrawal method with flexibility. The ABW method is even more flexible, and by incorporating expected forward returns into the method, it balances security with reduced withdrawal volatility (compared to VPW or fixed percentage-of-portfolio approaches) pretty well.

Plot up ABW for this case and you are going to get similar results. The problem is the totally crappy returns.... If you have 1 million dollars and you are going to lose 50% of it, there isn't a big difference between spending 40k of it and then losing 50% or just losing 50%.... That losing 50% is the killer.

No withdrawal method can provide 100% security and 0% withdrawal volatility. The two are incompatible, and most withdrawal methods other than a strict SWR and a percentage-of-portfolio approach are somewhere between those extremes. That said, using a combination of something like 1/CAPE and current TIPS' yields in proportion to one's AA has been a pretty good balance between smoothing returns while also guaranteeing that you won't completely deplete your portfolio prematurely (as long as the underlying assets don't go to zero, of course).

That said, if you look at something at drastic as the drawdown from 2008-2009, using the above approach would have resulted in remarkably little reduction in one's withdrawals because expected returns went up in almost direct proportion to stocks' drop in price.

My observation is that valuations are only used as a hand-wringing argument to drive WR ever lower. I don't see much talk of increasing withdrawal rates (and thus spending) while portfolios are crashing.

Sadly, some do just that. But using the method I describe above, the current stock downturn might have little to no impact on the withdrawal amount because 1/CAPE has increased in almost exactly the same proportion that stock prices (at least the S&P 500) have fallen.

[marcopolo]

VPW is not the only reasonable withdrawal method with flexibility. The ABW method is even more flexible, and by incorporating expected forward returns into the method, it balances security with reduced withdrawal volatility (compared to VPW or fixed percentage-of-portfolio approaches) pretty well.

Plot up ABW for this case and you are going to get similar results. The problem is the totally crappy returns.... If you have 1 million dollars and you are going to lose 50% of it, there isn't a big difference between spending 40k of it and then losing 50% or just losing 50%.... That losing 50% is the killer.

No withdrawal method can provide 100% security and 0% withdrawal volatility. The two are incompatible, and most withdrawal methods other than a strict SWR and a percentage-of-portfolio approach are somewhere between those extremes. That said, using a combination of something like 1/CAPE and current TIPS' yields in proportion to one's AA has been a pretty good balance between smoothing returns while also guaranteeing that you won't completely deplete your portfolio prematurely (as long as the underlying assets don't go to zero, of course).

That said, if you look at something at drastic as the drawdown from 2008-2009, using the above approach would have resulted in remarkably little reduction in one's withdrawals because expected returns went up in almost direct proportion to stocks' drop in price.

My observation is that valuations are only used as a hand-wringing argument to drive WR ever lower. I don't see much talk of increasing withdrawal rates (and thus spending) while portfolios are crashing.

Sadly, some do just that. But using the method I describe above, the current stock downturn might have little to no impact on the withdrawal amount because 1/CAPE has increased in almost exactly the same proportion that stock prices (at least the S&P 500) have fallen.

Which is why i generally ignore all the hand-wringing about valuations.
Either it is not a good predictor of future performance in which case it is just noise, or it is somewhat predictive but it does not affect what i need to do because my smaller portfolio now has higher expected returns.

[HomerJ]

Wow seven pages on this.

I'd like to meet the retiree who continues blindly and blithely spending 4% per year when the stock market drops 50%, bonds tank, and there is no recovery. I find it very hard to believe such a person exists.

If your retirement budget and planning shows that you cannot cut back at all--that is, you've neglected to build in some discretionary expenses that cushion you in deep downturns--then your "plan" may fail.

Hope is not a plan. Take one less vacation or don't buy a new car. This does not seem complicated to me.

Exactly.

[randomguy]

Wow seven pages on this.

I'd like to meet the retiree who continues blindly and blithely spending 4% per year when the stock market drops 50%, bonds tank, and there is no recovery. I find it very hard to believe such a person exists.

If your retirement budget and planning shows that you cannot cut back at all--that is, you've neglected to build in some discretionary expenses that cushion you in deep downturns--then your "plan" may fail.

Hope is not a plan. Take one less vacation or don't buy a new car. This does not seem complicated to me.

It doesn't seem complicated when you don't give a useful solution. Let's say our OP skips one 10k vacation. The rough math will be he loses 50% during the drop. It then rallies 50%. So he has an extra 7.5k more. So know instead of 614k, he has 622k. Do you think his problem was solved? He needs to be taking out 50k in a portfolio making 6.2% or like 39k. He isn't going to make it.

The spending cuts need are more like a 30 years of skipped vacations. And in the OP case I am not sure even that is enough. Those returns are grim...

[HomerJ]

Wow seven pages on this.

I'd like to meet the retiree who continues blindly and blithely spending 4% per year when the stock market drops 50%, bonds tank, and there is no recovery. I find it very hard to believe such a person exists.

If your retirement budget and planning shows that you cannot cut back at all--that is, you've neglected to build in some discretionary expenses that cushion you in deep downturns--then your "plan" may fail.

Hope is not a plan. Take one less vacation or don't buy a new car. This does not seem complicated to me.

It doesn't seem complicated when you don't give a useful solution. Let's say our OP skips one 10k vacation. The rough math will be he loses 50% during the drop. It then rallies 50%. So he has an extra 7.5k more. So know instead of 614k, he has 622k. Do you think his problem was solved? He needs to be taking out 50k in a portfolio making 6.2% or like 39k. He isn't going to make it.

The spending cuts need are more like a 30 years of skipped vacations. And in the OP case I am not sure even that is enough. Those returns are grim...

Nope.

The OP case are made up numbers.

[nigel_ht]

When will the 4% rule fail? What is the combination of circumstances required? As willthrill81 pointed out, it has failed outside the US before (detailed case studies here: https://papers.ssrn.com/sol3/papers.cfm ... id=4001986). But why? How did the course of these markets differ from the US?

The primary failure cases are WWI, WWII, going communist and Nikkei crash.

Nigel_ht, I would not expect you to read in detail a paper when a preliminary scan indicated that it had nothing new to offer you.

Had you done so, you would have seen that there were no WW II examples in the paper (those were all ruled out as cheap shots, along with civil war and defeat in war and also revolutionary overthrow, e.g., going communist.)

The only WW I examples were of a victor (UK) and a neutral (Sweden). Retirees in each case would have done better if they had supplemented their domestic stock market with a world stock index investment.

An international stock index would have helped the US investor in 1965, not so much in 1910. YMMV.

The most interesting findings in the paper, to me, were the poor results for the post-1960 Italian and French retirees. No world war. No civil war. No revolution. Just high inflation plus weak stock and bond returns for a long time.

Part of my job is peer reviewing journals. I read pretty fast although, yes, I didn't read all 60 pages in detail. Only the parts relevant to your comment: "As willthrill81 pointed out, it has failed outside the US before (detailed case studies here: ... But why? How did the course of these markets differ from the US?". The RMD section at the beginning I perused quickly and then went back to re-read pertinent sections when necessary.

Yes, WWII was not in the scope of your paper but by using 1910 world portfolio it includes the European market in WWI which would include both victors and losers including those that ceased to exist after the war (Austro-Hungarian Empire, Ottoman Empire) except as reduced successor states.

Given your comment about the 1910 not being a particularly fair test on page 23...why did you spend 3 pages presenting it?

The next example is the UK...in 1910.

While the UK didn't lose the war it lost 887,858 troops. 744,000 from the British Isles (43m total population in 1911). There were an additional 1.6M wounded. The empire lost a total of about 949K to 1.1M troops. Deaths as a % of the population was around 1.7-2%. Total losses in human capital when you count the wounded would be higher.

The high inflation you noted was due to spending about 25% of their GDP on the war.

Now the US spent 40% of GDP in WWII but again...if we enter a World War, barely win and suffer 6M dead then yes, all bets are off. Taiwan isn't likely to be that kind of war.

It is really hard for 2022 US to suffer an existential crisis that isn't either self-inflicted or involves nuclear weapons.. I suppose that asteroidal impact or other large natural disasters also apply. Anyone attempting to mitigate for the Yellowstone super volcano exploding?

Sweden in 1912. Still within that WWI timeframe.

Sweden's economy was deeply impacted by WWI when German subs targeted neutral shipping in 1917 (280 Swedish merchant ships were lost in the war) and by unrestricted Allied blockade starting in 1915 and allied embargoes of neutrals to keep Germany from getting food and war material. Sweden went from having an export surplus to dramatically less trade overall (both import and export). Trade as a percentage of GDP was halved after 1915 becoming more regional and less global. Given that about 30% of government revenue was from taxes on import/export, which cratered, and increased military spending (because there's a global war going on not too far away) and its not surprising that even neutral countries struggled. Poor harvests in 1917 and Allied blockades lead to food shortages in Sweden and there were large food riots. Toward the end of the war coal could no longer be imported from Germany so there were also fuel shortages on top of that.

So your 77 year old Swedish retiree was hungry and cold in 1917 because of the war and, if hale enough, was possibly protesting in the streets.

To recap...a country that depended a lot on foreign imports lost a sizable portion of their merchant marine (17% of their tonnage) to war, suffered from diminished trade because of blockade (from war) and had a famine/poor harvest.

Yep...that leads to a bad scenario including high sustained inflation. Neutrality mitigated a portion of repercussions of the war but WWI still had very large detrimental effects on Sweden. This isn't normal SORR...and not very applicable to the US.

Post 1960s France was when the British and French realized they were no longer tier 1 powers after the Suez and from losing their colonies. Empires that lose their colonies tend to do poorly for a while. France lost Lebanon in 1943 and Syria in 1945. Cambodia, Laos and Vietnam by 1954, Morocco and Tunisia in 1956, French West Africa and French Equatorial Africa in 1960, Algeria in 1962 (after 8 years of war). Not a particularly stellar time for France.

And the Algerian war was significant...400,000 French troops were in Algeria by 1956 and 1.5m mobilized on the French side. It was a significant cause for the fall of the Fourth Republic in 1958. In April 1961 there was the coup attempt against de Gaulle. Not entirely "no civil war" even it mostly happened in Algeria...for France the Algerian War was the final fall of their empire after 15 years of rearguard wars and decline.

And arguably the cause of the fall of the French Empire was WWII. They started losing colonies during the war and never stopped until their empire was gone.

The equivalent scenario for the US is to start losing states and heading we're back down to our original 13 states. Those are our colonial holdings...yes, if this starts happening things will go poorly.

As a note...our original 13 colonies were about 865K sq miles. In comparison, France is about 211K sq miles. Folks tend to forget how big we are in comparison to European countries...another aspect of how we differ.

Italy...well...it's had a troubled economic history.

So back to the context:

"As willthrill81 pointed out, it has failed outside the US before (detailed case studies here: ... But why? How did the course of these markets differ from the US?".

My answer is:

They differ a lot from the US.
The primary failure cases remain WWI, WWII, going communist and Nikkei crash.

Your paper adds:

Being Italy. Which I would claim is comes under the heading of "differs a lot from the US."

When Bretton Woods and Suez Crisis happens to us, all bets are off. I say that pretty often and probably even in this thread...and it COULD happen to the US fairly rapidly if we have 2 world wars and a Great Depression. Decolonization is unlikely without a civil war.

So will we see 4% fail in the future without the collapse of the American Empire? Perhaps...but not by very much because large failure likely means we have already lost reserve currency and superpower status. The latter is a huge trump card...one we don't ever really use as a trump card because we don't face external existential crises...but if we ever did...

[dknightd]

It was never meant as rule. Just useful guidance. I still consider it useful

[willthrill81]

When will the 4% rule fail? What is the combination of circumstances required? As willthrill81 pointed out, it has failed outside the US before (detailed case studies here: https://papers.ssrn.com/sol3/papers.cfm ... id=4001986). But why? How did the course of these markets differ from the US?

The primary failure cases are WWI, WWII, going communist and Nikkei crash.

Nigel_ht, I would not expect you to read in detail a paper when a preliminary scan indicated that it had nothing new to offer you.

Had you done so, you would have seen that there were no WW II examples in the paper (those were all ruled out as cheap shots, along with civil war and defeat in war and also revolutionary overthrow, e.g., going communist.)

The only WW I examples were of a victor (UK) and a neutral (Sweden). Retirees in each case would have done better if they had supplemented their domestic stock market with a world stock index investment.

An international stock index would have helped the US investor in 1965, not so much in 1910. YMMV.

The most interesting findings in the paper, to me, were the poor results for the post-1960 Italian and French retirees. No world war. No civil war. No revolution. Just high inflation plus weak stock and bond returns for a long time.

Part of my job is peer reviewing journals. I read pretty fast although, yes, I didn't read all 60 pages in detail. Only the parts relevant to your comment: "As willthrill81 pointed out, it has failed outside the US before (detailed case studies here: ... But why? How did the course of these markets differ from the US?". The RMD section at the beginning I perused quickly and then went back to re-read pertinent sections when necessary.

Yes, WWII was not in the scope of your paper but by using 1910 world portfolio it includes the European market in WWI which would include both victors and losers including those that ceased to exist after the war (Austro-Hungarian Empire, Ottoman Empire) except as reduced successor states.

Given your comment about the 1910 not being a particularly fair test on page 23...why did you spend 3 pages presenting it?

The next example is the UK...in 1910.

While the UK didn't lose the war it lost 887,858 troops. 744,000 from the British Isles (43m total population in 1911). There were an additional 1.6M wounded. The empire lost a total of about 949K to 1.1M troops. Deaths as a % of the population was around 1.7-2%. Total losses in human capital when you count the wounded would be higher.

The high inflation you noted was due to spending about 25% of their GDP on the war.

Now the US spent 40% of GDP in WWII but again...if we enter a World War, barely win and suffer 6M dead then yes, all bets are off. Taiwan isn't likely to be that kind of war.

It is really hard for 2022 US to suffer an existential crisis that isn't either self-inflicted or involves nuclear weapons.. I suppose that asteroidal impact or other large natural disasters also apply. Anyone attempting to mitigate for the Yellowstone super volcano exploding?

Sweden in 1912. Still within that WWI timeframe.

Sweden's economy was deeply impacted by WWI when German subs targeted neutral shipping in 1917 (280 Swedish merchant ships were lost in the war) and by unrestricted Allied blockade starting in 1915 and allied embargoes of neutrals to keep Germany from getting food and war material. Sweden went from having an export surplus to dramatically less trade overall (both import and export). Trade as a percentage of GDP was halved after 1915 becoming more regional and less global. Given that about 30% of government revenue was from taxes on import/export, which cratered, and increased military spending (because there's a global war going on not too far away) and its not surprising that even neutral countries struggled. Poor harvests in 1917 and Allied blockades lead to food shortages in Sweden and there were large food riots. Toward the end of the war coal could no longer be imported from Germany so there were also fuel shortages on top of that.

So your 77 year old Swedish retiree was hungry and cold in 1917 because of the war and, if hale enough, was possibly protesting in the streets.

To recap...a country that depended a lot on foreign imports lost a sizable portion of their merchant marine (17% of their tonnage) to war, suffered from diminished trade because of blockade (from war) and had a famine/poor harvest.

Yep...that leads to a bad scenario including high sustained inflation. Neutrality mitigated a portion of repercussions of the war but WWI still had very large detrimental effects on Sweden. This isn't normal SORR...and not very applicable to the US.

Post 1960s France was when the British and French realized they were no longer tier 1 powers after the Suez and from losing their colonies. Empires that lose their colonies tend to do poorly for a while. France lost Lebanon in 1943 and Syria in 1945. Cambodia, Laos and Vietnam by 1954, Morocco and Tunisia in 1956, French West Africa and French Equatorial Africa in 1960, Algeria in 1962 (after 8 years of war). Not a particularly stellar time for France.

And the Algerian war was significant...400,000 French troops were in Algeria by 1956 and 1.5m mobilized on the French side. It was a significant cause for the fall of the Fourth Republic in 1958. In April 1961 there was the coup attempt against de Gaulle. Not entirely "no civil war" even it mostly happened in Algeria...for France the Algerian War was the final fall of their empire after 15 years of rearguard wars and decline.

And arguably the cause of the fall of the French Empire was WWII. They started losing colonies during the war and never stopped until their empire was gone.

The equivalent scenario for the US is to start losing states and heading we're back down to our original 13 states. Those are our colonial holdings...yes, if this starts happening things will go poorly.

As a note...our original 13 colonies were about 865K sq miles. In comparison, France is about 211K sq miles. Folks tend to forget how big we are in comparison to European countries...another aspect of how we differ.

Italy...well...it's had a troubled economic history.

So back to the context:

"As willthrill81 pointed out, it has failed outside the US before (detailed case studies here: ... But why? How did the course of these markets differ from the US?".

My answer is:

They differ a lot from the US.
The primary failure cases remain WWI, WWII, going communist and Nikkei crash.

Your paper adds:

Being Italy. Which I would claim is comes under the heading of "differs a lot from the US."

When Bretton Woods and Suez Crisis happens to us, all bets are off. I say that pretty often and probably even in this thread...and it COULD happen to the US fairly rapidly if we have 2 world wars and a Great Depression. Decolonization is unlikely without a civil war.

So will we see 4% fail in the future without the collapse of the American Empire? Perhaps...but not by very much because large failure likely means we have already lost reserve currency and superpower status. The latter is a huge trump card...one we don't ever really use as a trump card because we don't face external existential crises...but if we ever did...

In all fairness, you don't have to go back to WWII or prior to see failures of the '4% rule' in developed nations. Portfolio Charts uses data only going back to 1970 but indicates that Japan, Australia, and Spain, IIRC (I don't have good Internet access at the moment so this is only off the top of my head), are examples, even if half their residents' stock holdings were outside their home country.

[nigel_ht]

I am not sure people understand how bad the stock market returns you gave are. You are looking at something like 2.5% real for a 30 year period. That is about half of the worst time in post 1920 US history.

That's not so terrible. In his paper the black line in Figure 1-3 is 2.5%/0% real for 30 years. The red lines are 4.5%/2% real (Figures 1-4).

Yep that will ruin your day real quick and not too surprisingly a constant 0% real makes Bengen fail.

But hey, at least the gray line of constant 0%/-2.5% real was removed as unrealistic after Figure 1...

[nigel_ht]

"As willthrill81 pointed out, it has failed outside the US before (detailed case studies here: ... But why? How did the course of these markets differ from the US?".

My answer is:

They differ a lot from the US.
The primary failure cases remain WWI, WWII, going communist and Nikkei crash.

Your paper adds:

Being Italy. Which I would claim is comes under the heading of "differs a lot from the US."

When Bretton Woods and Suez Crisis happens to us, all bets are off. I say that pretty often and probably even in this thread...and it COULD happen to the US fairly rapidly if we have 2 world wars and a Great Depression. Decolonization is unlikely without a civil war.

So will we see 4% fail in the future without the collapse of the American Empire? Perhaps...but not by very much because large failure likely means we have already lost reserve currency and superpower status. The latter is a huge trump card...one we don't ever really use as a trump card because we don't face external existential crises...but if we ever did...

In all fairness, you don't have to go back to WWII or prior to see failures of the '4% rule' in developed nations. Portfolio Charts uses data only going back to 1970 but indicates that Japan, Australia, and Spain, IIRC (I don't have good Internet access at the moment so this is only off the top of my head), are examples, even if half their residents' stock holdings were outside their home country.

https://portfoliocharts.com/2017/06/09/ ... awal-rate/

This one? It's for a 40 year SWR...

"These numbers are calculated since 1970, do not account for fees, and look at a 40-year retirement period (more conservative than the traditional 30-year period found in most studies to compensate for the less conservative smaller data set — read this for more methodology info)."

That doesn't include Spain though so maybe different from the one you're thinking of.

[Glenn]

Wow seven pages on this.

I'd like to meet the retiree who continues blindly and blithely spending 4% per year when the stock market drops 50%, bonds tank, and there is no recovery. I find it very hard to believe such a person exists.

If your retirement budget and planning shows that you cannot cut back at all--that is, you've neglected to build in some discretionary expenses that cushion you in deep downturns--then your "plan" may fail.

Hope is not a plan. Take one less vacation or don't buy a new car. This does not seem complicated to me.

Amen

[randomguy]

Nope.

The OP case are made up numbers.

Definitely. But do you really think spending 10k less in 1966 would have made much of a difference for our 1966 retiree?

[randomguy]

I am not sure people understand how bad the stock market returns you gave are. You are looking at something like 2.5% real for a 30 year period. That is about half of the worst time in post 1920 US history.

That's not so terrible. In his paper the black line in Figure 1-3 is 2.5%/0% real for 30 years. The red lines are 4.5%/2% real (Figures 1-4).

Yep that will ruin your day real quick and not too surprisingly a constant 0% real makes Bengen fail.

But hey, at least the gray line of constant 0%/-2.5% real was removed as unrealistic after Figure 1...

There is a vast difference in outcomes between a steady 2.5% and one that starts off with a 60% market decline and slowly creeps back up to 2.5% over 30 years when you are taking money out. And of course we are talking stocks only. Throw in the bonds and I doubt the portfolio breaks 2%.....

[HomerJ]

Nope.

The OP case are made up numbers.

Definitely. But do you really think spending 10k less in 1966 would have made much of a difference for our 1966 retiree?

Yes.

[randomguy]

Nope.

The OP case are made up numbers.

Definitely. But do you really think spending 10k less in 1966 would have made much of a difference for our 1966 retiree?

Yes.

The I suggest you run the math on it. You might find it educational...

[nigel_ht]

Nope.

The OP case are made up numbers.

Definitely. But do you really think spending 10k less in 1966 would have made much of a difference for our 1966 retiree?

Yes.

The I suggest you run the math on it. You might find it educational...

Which numbers? The OP's numbers? Meh, don't have the spreadsheet so it's annoying to do.

Add in SS income at age 70 and want to bet it still works? Not too worried about 4% working even if I've decided to switch to 3% constant percentage with a floor.

[Admiral]

Nope.

The OP case are made up numbers.

Definitely. But do you really think spending 10k less in 1966 would have made much of a difference for our 1966 retiree?

Yes.

The I suggest you run the math on it. You might find it educational...

In 1966, the median family income was $7400. So yes, I think someone who could afford to “spend $10k less” would have been fine. Because they were already rich.

[GAAP]

Wow, way too much discussion of (or argument about) this:

PS: if your takeaway was something like, “Forget the quest for an SWR , I am going with a VWR framework,” I think that’s a very appropriate conclusion to draw.

When the far more useful reminder occurred immediately previously:

Now you know what it takes for the 4% rule to prevail, as it always has in post-1926 US history thus far, because in that history:
1. Stocks and bonds seldom tumble together
2. Stocks always roar back after taking a tumble
3. US stocks generally maintain a real return of 1.0 Siegels or better

Regardless of your individual choice of method, it is critical to understand what assumptions you're using. Bad assumptions = bad plan.

I use a variable method to calculate a number -- but it's essentially a budgetary target, not an absolute withdrawal, let alone actual spending. My guess is that most people on this site using an SWR approach effectively do the same. Those who only read the "financial news" may be much more rigid in their approach until it is too late.

[aristotelian]

Using $686,400 as the starting value and $40,000 as the withdrawal rate, Portfoliovisualizer Monte Carlo simulation leaves you out of money only in the lowest 10% of outcomes. So you'd still have a 90% success rate even changing nothing in your withdrawal rate.

For 20 year timeframe, you still have money 100% of the time.

Earlyretirementnow.com and others have already found something like 95% success rate for 4% constant dollar withdrawal, so it is well known that there is risk of failure. Certainly the risk increases starting from a bad first year.

[nigel_ht]

Wow, way too much discussion of (or argument about) this:

PS: if your takeaway was something like, “Forget the quest for an SWR , I am going with a VWR framework,” I think that’s a very appropriate conclusion to draw.

There is discussion because while VWR strategies are useful (I just switched) SWR still works. If you want a large remaining balance while not risking not getting enough income then constant percentage with a floor works better.

It’s not an appropriate conclusion because the data is fiction. If given a bad enough sequence all withdrawal strategies fail either by running out of money or not providing enough income to meet needs.

[Marseille07]

It’s not an appropriate conclusion because the data is fiction. If given a bad enough sequence all withdrawal strategies fail either by running out of money or not providing enough income to meet needs.

This depends on how you look at it.

Percentage methods do not fail. Now, you can argue they might not provide enough income...but often this is self-imposed. For example, I might be doing 3% WR...market crashes and 3% gets tight. If I do 4% then, is it the end of the world? Probably not. Spending floor works similarly, since when you hit the spending floor you're necessarily going over the WR you initially set, it's just that the amount is fixed instead of a percentage.

[rockstar]

I think, the reality is that you need a buffer in your retirement savings. Maybe be able to draw down 3% but have a good enough buffer that when your portfolio shrinks you can withdraw 4% instead. It's really hard for folks to cut back when the essentials are experiencing a lot of inflation right now. This is something to think about when the spend less approach doesn't work if you retired with bare minimum. And it doesn't help when the pool of money you're withdrawing from is shrinking whether it be bonds or equities.

[Marseille07]

I think, the reality is that you need a buffer in your retirement savings. Maybe be able to draw down 3% but have a good enough buffer that when your portfolio shrinks you can withdraw 4% instead. It's really hard for folks to cut back when the essentials are experiencing a lot of inflation right now. This is something to think about when the spend less approach doesn't work if you retired with bare minimum. And it doesn't help when the pool of money you're withdrawing from is shrinking whether it be bonds or equities.

Kind of, yes. I mean, if you start off with 3% and SHTF, 4% would generate the same amount of income 3% used to generate...spending floor works similarly.

[rockstar]

I think, the reality is that you need a buffer in your retirement savings. Maybe be able to draw down 3% but have a good enough buffer that when your portfolio shrinks you can withdraw 4% instead. It's really hard for folks to cut back when the essentials are experiencing a lot of inflation right now. This is something to think about when the spend less approach doesn't work if you retired with bare minimum. And it doesn't help when the pool of money you're withdrawing from is shrinking whether it be bonds or equities.

Kind of, yes. I mean, if you start off with 3% and SHTF, 4% would generate the same amount of income 3% used to generate...spending floor works similarly.

But you need stuff to cut to go the budget route. Sure, if you have allocated for travel, you can cut that out. But eventually, if costs continue to increase, you'll run out of places to cut. So you might end up back working at least part time. Remember Archie Bunker drove a cab to help make ends meet. I guess, you could go the Uber or Lyft route today. I see lots of old people working at the grocery store. I guess there has to be a plan, and you have to have good enough health to execute it.

[Marseille07]

But you need stuff to cut to go the budget route. Sure, if you have allocated for travel, you can cut that out. But eventually, if costs continue to increase, you'll run out of places to cut. So you might end up back working at least part time. Remember Archie Bunker drove a cab to help make ends meet. I guess, you could go the Uber or Lyft route today. I see lots of old people working at the grocery store. I guess there has to be a plan, and you have to have good enough health to execute it.

Shrug...having to go back to work is much better than plowing through 4% SWR without adjustments and run out of money 20 years later. I'd certainly make that tradeoff should it be necessary.

[McQ]

Wow, way too much discussion of (or argument about) this:

PS: if your takeaway was something like, “Forget the quest for an SWR , I am going with a VWR framework,” I think that’s a very appropriate conclusion to draw.

When the far more useful reminder occurred immediately previously:

Now you know what it takes for the 4% rule to prevail, as it always has in post-1926 US history thus far, because in that history:
1. Stocks and bonds seldom tumble together
2. Stocks always roar back after taking a tumble
3. US stocks generally maintain a real return of 1.0 Siegels or better

Regardless of your individual choice of method, it is critical to understand what assumptions you're using. Bad assumptions = bad plan.

I use a variable method to calculate a number -- but it's essentially a budgetary target, not an absolute withdrawal, let alone actual spending. My guess is that most people on this site using an SWR approach effectively do the same. Those who only read the "financial news" may be much more rigid in their approach until it is too late.

Thank you, GAAP. Your comment captures what I hoped to accomplish in the thread.

[McQ]

...

Part of my job is peer reviewing journals. ...

Sweden in 1912. Still within that WWI timeframe.

Sweden's economy was deeply impacted by WWI when German subs targeted neutral shipping in 1917 (280 Swedish merchant ships were lost in the war) and by unrestricted Allied blockade starting in 1915 and allied embargoes of neutrals to keep Germany from getting food and war material. Sweden went from having an export surplus to dramatically less trade overall (both import and export). Trade as a percentage of GDP was halved after 1915 becoming more regional and less global. Given that about 30% of government revenue was from taxes on import/export, which cratered, and increased military spending (because there's a global war going on not too far away) and its not surprising that even neutral countries struggled. Poor harvests in 1917 and Allied blockades lead to food shortages in Sweden and there were large food riots. Toward the end of the war coal could no longer be imported from Germany so there were also fuel shortages on top of that.

So your 77 year old Swedish retiree was hungry and cold in 1917 because of the war and, if hale enough, was possibly protesting in the streets.

To recap...a country that depended a lot on foreign imports lost a sizable portion of their merchant marine (17% of their tonnage) to war, suffered from diminished trade because of blockade (from war) and had a famine/poor harvest.

Yep...that leads to a bad scenario including high sustained inflation. Neutrality mitigated a portion of repercussions of the war but WWI still had very large detrimental effects on Sweden. This isn't normal SORR...and not very applicable to the US.

Sorry again that the paper did not repay your effort to read.

But I found value in the detail you added about the vicissitudes suffered by Sweden around WW I. Do you have a source that I can cite? I may want to add some of those details if I revise the paper.

In the meantime, Sweden presents a particularly interesting case for this thread. Based on the Credit Suisse Global Investment Returns Yearbook 2022, for the balanced fund investor (50-50), Sweden recorded almost the very best returns since 1900 for any market in the world, a hair behind South Africa and measurably ahead of the US. [4.555 v 4.5% v. 4.35% annualized real].

And yet, the Swede who retired in 1912 with a 60-40 or 30-70 balanced mix ran out of money in 16 or 20 years respectively. (For the paper I applied the RMD schedule, 3.65% initially, adjusted for inflation—not 4%). It was the worst outcome of any of the edge cases I considered.

Moral: winning the race over 122 years does not imply a successful withdrawal course at every juncture. Were something like the projected returns in my spreadsheet to transpire in the US (I am not saying they will), then the US might one day be the inverse of Sweden 1900-2021, with all the great returns coming up front instead of later; and those historical returns not guaranteed of continuance, just as Sweden after 1922 was not condemned to limp along at the rear of the pack forever.

Pity none of us has a 122 year investing horizon.

[HomerJ]

...

Part of my job is peer reviewing journals. ...

Sweden in 1912. Still within that WWI timeframe.

Sweden's economy was deeply impacted by WWI when German subs targeted neutral shipping in 1917 (280 Swedish merchant ships were lost in the war) and by unrestricted Allied blockade starting in 1915 and allied embargoes of neutrals to keep Germany from getting food and war material. Sweden went from having an export surplus to dramatically less trade overall (both import and export). Trade as a percentage of GDP was halved after 1915 becoming more regional and less global. Given that about 30% of government revenue was from taxes on import/export, which cratered, and increased military spending (because there's a global war going on not too far away) and its not surprising that even neutral countries struggled. Poor harvests in 1917 and Allied blockades lead to food shortages in Sweden and there were large food riots. Toward the end of the war coal could no longer be imported from Germany so there were also fuel shortages on top of that.

So your 77 year old Swedish retiree was hungry and cold in 1917 because of the war and, if hale enough, was possibly protesting in the streets.

To recap...a country that depended a lot on foreign imports lost a sizable portion of their merchant marine (17% of their tonnage) to war, suffered from diminished trade because of blockade (from war) and had a famine/poor harvest.

Yep...that leads to a bad scenario including high sustained inflation. Neutrality mitigated a portion of repercussions of the war but WWI still had very large detrimental effects on Sweden. This isn't normal SORR...and not very applicable to the US.

Sorry again that the paper did not repay your effort to read.

But I found value in the detail you added about the vicissitudes suffered by Sweden around WW I. Do you have a source that I can cite? I may want to add some of those details if I revise the paper.

In the meantime, Sweden presents a particularly interesting case for this thread. Based on the Credit Suisse Global Investment Returns Yearbook 2022, for the balanced fund investor (50-50), Sweden recorded almost the very best returns since 1900 for any market in the world, a hair behind South Africa and measurably ahead of the US. [4.555 v 4.5% v. 4.35% annualized real].

And yet, the Swede who retired in 1912 with a 60-40 or 30-70 balanced mix ran out of money in 16 or 20 years respectively. (For the paper I applied the RMD schedule, 3.65% initially, adjusted for inflation—not 4%). It was the worst outcome of any of the edge cases I considered.

Moral: winning the race over 122 years does not imply a successful withdrawal course at every juncture. Were something like the projected returns in my spreadsheet to transpire in the US (I am not saying they will), then the US might one day be the inverse of Sweden 1900-2021, with all the great returns coming up front instead of later; and those historical returns not guaranteed of continuance, just as Sweden after 1922 was not condemned to limp along at the rear of the pack forever.

Pity none of us has a 122 year investing horizon.

Moral: World War on your doorstep means all bets are off.

Thanks for the insight.