Stop worrying about the 4% rule

[EscapeVelocity2020]

Early retirement blogger says: 25x Expenses Isn’t Enough for Early Retirement

http://retireby40.org/25x-expenses-isnt-enough-for-early-retirement/

Most of his points will be easily turned aside.

Yep.

"Using the 4% rule doesn't work if you refuse to follow the rule and instead spend more!"

As a counter-point, if you are already living a frugal life, you have no fat to cut when you get older.  And getting older typically entails having to hire younger people to do things you physically can no longer do (or are increasingly risky to push your limits to do) as well as other obvious medical costs.  I, personally, didn't mind his blog post and thought it was well intentioned.

Also, Chasefish, thanks for your thoughts.  Credit bubble is intentionally vague since it could mean a number of things.  I live my life generally thinking the next decade may suck (hence the justification for indexing over, say, individual stocks which have been soaring lately) so I'm usually pleasantly surprised.  But I am getting more and more tempted to put in the effort to prep.  I have reallocated toward a few hedges in depressed, strong dividend stocks, munis, and foreign currency.  None of these hedges have impacted the current standard of living, but I hope they defray any volatility going forward....

[TomTX]

Early retirement blogger says: 25x Expenses Isn’t Enough for Early Retirement

http://retireby40.org/25x-expenses-isnt-enough-for-early-retirement/

Most of his points will be easily turned aside.

Yep.

"Using the 4% rule doesn't work if you refuse to follow the rule and instead spend more!"

As a counter-point, if you are already living a frugal life, you have no fat to cut when you get older.  And getting older typically entails having to hire younger people to do things you physically can no longer do (or are increasingly risky to push your limits to do) as well as other obvious medical costs.  I, personally, didn't mind his blog post and thought it was well intentioned.

Also, Chasefish, thanks for your thoughts.  Credit bubble is intentionally vague since it could mean a number of things.  I live my life generally thinking the next decade may suck (hence the justification for indexing over, say, individual stocks which have been soaring lately) so I'm usually pleasantly surprised.  But I am getting more and more tempted to put in the effort to prep.  I have reallocated toward a few hedges in depressed, strong dividend stocks, munis, and foreign currency.  None of these hedges have impacted the current standard of living, but I hope they defray any volatility going forward....

By the time you get old enough to require hiring young people, your 4% withdrawal would generally be supplemented by Social Security.  Unless you get to the point of needing help to get to the bathroom, SS should be more than enough to cover a lawn guy, grocery delivery, uber rides and the occasional visit from a fixit guy around the house.

[Telecaster]

As a counter-point, if you are already living a frugal life, you have no fat to cut when you get older.  And getting older typically entails having to hire younger people to do things you physically can no longer do (or are increasingly risky to push your limits to do) as well as other obvious medical costs.  I, personally, didn't mind his blog post and thought it was well intentioned.

As a counter-counter-point, deciding you want to do more international travel is an example of adding fat. 

Here's the thing:  The headline "4% SWR Still Safe!" is boring.   The headline "I Found a Flaw with the 4% Rule!" is exciting and sexy!  It implies the author discovered new insights that no one else thought of.   "4% Rule Fails!" gets page clicks.   In this case, the author came up with the following bombshells:

1.  If you add discretionary spending to your budget, you need more money.

2.  If the future is worse than the past, the 4% rule fails.

Well, duh.  But worse, he also demonstrated a common noob misunderstanding of the 4% rule.  He's says (basically that based on higher than average P/E, it is a reasonable assumption that stock returns will be lower than average going forward.   No quibble from me on that point.  Based on that, he posits the 4% rule won't work in future.  But the 4% SWR isn't on average returns, it is the worst case scenario.   

So if the 4% rule fails--as he suggests could happen--it must be because we are now experiencing, or about to experience, worse market conditions than have ever existed in the past!   Yet he provided zero evidence this is the case, or even why it might be a likely scenario. 

TL;DR.  We can all stop worrying about the 4% rule. 

[Retire-Canada]

As a counter-point, if you are already living a frugal life, you have no fat to cut when you get older. 

Having read a metric shit ton of retirement stories/plans only a handful have been really bare bones where no cuts were possible. Add to that all the data I've seen points to a reduction in spending as people get older not an increase.

[deborah]

In Australia, we have superannuation, which is broadly similar to a 401k. A study has recently been published about long term withdrawal rates from superannuation in retirement - http://www.finsia.com/docs/default-source/jassa-current-issue/JASSA-2016-issue-2/superannnuation-drawdown-behaviour.pdf?sfvrsn=0

There are several interesting outcomes from the study. Firstly, the withdrawal amount doesn't increase as people age, although you would expect otherwise.

Secondly, people are withdrawing less than expected. In Australia there are two standardised retirement expenditures - modest (the expected requirements of a couple who live a modest lifestyle in retirement) and comfortable (those who have a more inflated lifestyle) - that were worked out by a bank by looking at the costs of various things. Couples on a modest lifestyle are supposed to require $34,216AUD, and $34,598AUD when they reach 85. However, the longitudinal study finds that people are actually living an even more modest lifestyle in retirement (the median expenditure) - about 20% less.

Thirdly, from the standardised retirement expenditures, people (on average) need about the same whether they are 65 or 85 - with the study noting that the 65 year olds tend to be spending more.

[arebelspy]

Early retirement blogger says: 25x Expenses Isn’t Enough for Early Retirement

http://retireby40.org/25x-expenses-isnt-enough-for-early-retirement/

Most of his points will be easily turned aside. He does talk about his own expense inflation, a very real risk for very early retirees on bare-bones budgets.

Well that was just.. bad.

He underestimated his ER expenses, and then complained that saving 25x (or 30, in his case) his underestimated expenses wasn't enough to pay for his new, higher expenses.

The problem there is not the 4% rule, it's in your poor ability to plan for the future.

His title "25x Expenses Isn’t Enough for Early Retirement" is totally wrong--it would be enough, if you actually had saved up 25x actual expenses, not a made up, lower number.  The poll on the bottom is ridiculous, too.  Almost half his respondents (43% as of this writing) say you need 40x or 50x expenses.  That's a 2% to 2.5% WR!  Crazy!  Of course, if you make up artificially low retirement budget numbers, and then save up 50x expenses, and then spend double in ER, guess what?  You're at a 4% real WR, and will do just fine, as long as that actually is your ER spending, and it doesn't keep rising and rising.  ;)

[Radagast]

I think retireby40 makes a good point that extreme early retirees may not have sufficient understanding of future expenses. There is still too much life ahead for a 29-year-old to know for certain about children, medical costs, divorces, modest increases in hedonism, etc. It is hard to extrapolate the next 60 years from the last 6 years with any accuracy,so an extra margin of safety is probably necessary.

[arebelspy]

I think retireby40 makes a good point that extreme early retirees may not have sufficient understanding of future expenses. There is still too much life ahead for a 29-year-old to know for certain about children, medical costs, divorces, modest increases in hedonism, etc. It is hard to extrapolate the next 60 years from the last 6 years with any accuracy,so an extra margin of safety is probably necessary.

If that's your point, say that.

Say that "here's what to watch out for when setting future expenses, and thus determining your number."  But "4% rule doesn't work" bullshit is just clickbait, and misguided.  In his example, he is the problem (way underestimating FIRE expenses), not the 4% rule.

He could have written a beautiful article with the premise you provide that almost everyone here would have agreed with.  Instead, he put that up. 

[AdrianC]

Well, duh.  But worse, he also demonstrated a common noob misunderstanding of the 4% rule.  He's says (basically that based on higher than average P/E, it is a reasonable assumption that stock returns will be lower than average going forward.   No quibble from me on that point.  Based on that, he posits the 4% rule won't work in future.  But the 4% SWR isn't on average returns, it is the worst case scenario.   

Quibble: 4% SWR isn't worst case scenario, is it?

Firecalc 30 years, 4%, 75/25
For our purposes, failure means the portfolio was depleted before the end of the 30 years. FIRECalc found that 6 cycles failed, for a success rate of 94.8%.
Zero failures at 3.6%.

Firecalc 40 years, 4%, 75/25
For our purposes, failure means the portfolio was depleted before the end of the 40 years. FIRECalc found that 15 cycles failed, for a success rate of 85.8%.
Zero failures at 3.3%.

[arebelspy]

Well, duh.  But worse, he also demonstrated a common noob misunderstanding of the 4% rule.  He's says (basically that based on higher than average P/E, it is a reasonable assumption that stock returns will be lower than average going forward.   No quibble from me on that point.  Based on that, he posits the 4% rule won't work in future.  But the 4% SWR isn't on average returns, it is the worst case scenario.   

Quibble: 4% SWR isn't worst case scenario, is it?

Firecalc 30 years, 4%, 75/25
For our purposes, failure means the portfolio was depleted before the end of the 30 years. FIRECalc found that 6 cycles failed, for a success rate of 94.8%.
Zero failures at 3.6%.

Firecalc 40 years, 4%, 75/25
For our purposes, failure means the portfolio was depleted before the end of the 40 years. FIRECalc found that 15 cycles failed, for a success rate of 85.8%.
Zero failures at 3.3%.

Probably a typo--it's missing a single letter.  It's the worst case scenarios.

;)

Yes, 4% WR was designed to have 95% success rate (being defined as having money left) after 30 years, historically, when withdrawing an initial 4% and then adjusting it up for inflation yearly.

[mathjak107]

actually the grand pappy of the 4% safe withdrawal rate bill bengen  found his   SAFEMAX  was actually  4.15% for   a 50/50 allocation for stocks and bonds and was 100% succesful  based on 1966 as  the worst case scenario .

it was only later on when the trinity study  used longer term corporate bonds rather  than bill's 5 year gov't bonds that they got poorer results so 4% dropped to only a 95% success rate . it took a 3.60% draw rate  or so to get a full 100% success rate under the trinity parameters  . bill also took his withdrawals the end of each year vs others which took it at the beginning .

so at worst there is about a 12% difference in pay checks between the two .

note: none count any fees at all in the equation .

[AdrianC]

I don't see why we keep harping on about Bengen and the Trinity studies. Sure, they did ground-breaking work. Now we can all do our own studies using cFIREsim or FIREcalc.

Unless someone thinks that these tools are inferior?

http://www.firecalc.com/

[mathjak107]

the math at the end of the day is what it is , at least for a 30 year time frame .  mathematically you need at least a 2% real return average over the  first 15 years of a 30 year retirement to support 4% inflation adjusted  .

going out longer i am not aware of anyone who actually broke out the math needed  so all you get is a statistic and nothing to really monitor going forward .

i know if 5 years in or so if i am not seeing at least a 2% real return i may want to cut back spending .  but if i had a 40 or 50 year retirement i don't know what the math is i would have to achieve  to have it hold . kitces only did the study on the math for a 30 year time frame .

[ender]

mathematically you need at least a 2% real return average over the  first 15 years of a 30 year retirement to support 4% inflation adjusted  .

Citation needed?

I can think of multiple scenarios where a 2% real return average for the first 15 years fails 30 year retirement. Or where a lower percentage succeeds.

[mathjak107]

here you go

https://www.kitces.com/blog/what-returns-are-safe-withdrawal-rates-really-based-upon/

[mathjak107]

mathematically you need at least a 2% real return average over the  first 15 years of a 30 year retirement to support 4% inflation adjusted  .

Citation needed?

I can think of multiple scenarios where a 2% real return average for the first 15 years fails 30 year retirement. Or where a lower percentage succeeds.

here are the actual numbers for the worst time frames we have had in history


suppose you were so unlucky to retire in one of those worst time framess ,what would your 30 year results look like :

1907 stocks returned 7.77% -- bonds 4.250-- rebalanced portfolio 7.02- - inflation 1.64--

1929 stocks 8.19% - - bonds 1.74%-- rebalanced portfolio 6.28-- inflation 1.69--

1937 stocks 10.12 - - bonds 2.13 - rebalanced portfolio -- 7.24 inflation-- 2.82

1966 stocks 10.23 - -bonds 7.85 -- rebalanced portfolio 9.56- - inflation 5.38

for comparison the 140 year average's were:

stocks 8.39--bonds 2.85%--rebalanced portfolio 6.17% inflation 2.23%

so what made those time frames the worst ? what made them the worst is the fact in every single retirement time frame the outcome of that 30 year period was determined not by what happened over the 30 years but the entire outcome was decided in the first 15 years.

so lets look at the first 15 years in those time frames determined to be the worst we ever had.

1907--- stocks minus 1.47%---- bonds minus .39%-- rebalanced minus .70% ---inflation 1.64%

1929---stocks 1.07%---bonds 1.79%---rebalanced 2.29%--inflation 1.69%

1937---stocks -- 3.45%---bonds minus 3.07%-- rebalanced 1.23%--inflation 2.82%

1966-stocks minus .13%--bonds 1.08%--rebalanced .64%-- inflation 5.38%

it is those 15 year horrible time frames that the 4% safe withdrawal rate was born out of since you had to reduce from what could have been 6.50% as a swr down to just 4% to get through those worst of times.

[AdrianC]

1966-stocks minus .13%--bonds 1.08%--rebalanced .64%-- inflation 5.38%

it is those 15 year horrible time frames that the 4% safe withdrawal rate was born out of since you had to reduce from what could have been 6.50% as a swr down to just 4% to get through those worst of times.

Sure, excepting 1966, of course. 4% did not work from 1964 through 1969 (according to cFIREsim).

[mathjak107]

depends on the portfolio . bill bengan's model passed 1966 at 4.16%  . the trinity did not . the types of bonds held matter . but the 2% real return gives you extra leeway

[mathjak107]

mathematically you need at least a 2% real return average over the  first 15 years of a 30 year retirement to support 4% inflation adjusted  .

Citation needed?

I can think of multiple scenarios where a 2% real return average for the first 15 years fails 30 year retirement. Or where a lower percentage succeeds.

as you see 2% real returns the first 15 years never failed . every worst case was less than 2% 

[AdrianC]

it is those 15 year horrible time frames that the 4% safe withdrawal rate was born out of since you had to reduce from what could have been 6.50% as a swr down to just 4% to get through those worst of times.

I tried 6.5% in FIREcalc just for fun:
50/50 FIRECalc found that 77 cycles failed, for a success rate of 33.6%.
75/25 FIRECalc found that 64 cycles failed, for a success rate of 44.8%.

6.1%:
75/25 FIRECalc found that 58 cycles failed, for a success rate of 50.0%.

[k9]

The poll on the bottom is ridiculous, too.  Almost half his respondents (43% as of this writing) say you need 40x or 50x expenses.  That's a 2% to 2.5% WR!  Crazy!  Of course, if you make up artificially low retirement budget numbers, and then save up 50x expenses, and then spend double in ER, guess what?  You're at a 4% real WR, and will do just fine, as long as that actually is your ER spending, and it doesn't keep rising and rising.  ;)

Yeah. When your remaining life expectancy (typically, for a male who "retires by 40", a little less than 40 years) is lower than your multiple of expenses invested, you know there's a problem. Get a life annuity and you'll automatically improve your WR (which will be the safest you can come up with, by definition: 100% success rate). And, once again, that's without considering ability to work part-time if needed, or without considering SS.

[ender]

as you see 2% real returns the first 15 years never failed . every worst case was less than 2%

That is a very different statement than:

mathematically you need at least a 2% real return average over the  first 15 years of a 30 year retirement to support 4% inflation adjusted  .

Mathematically you barely need over 0% real return to support a 4% withdrawal rate. If your portfolio keeps up with inflation and otherwise does exactly 0% return, you will still last 25 years with a 4% withdrawal. Mathematically you need your overall return to be only slightly over 0% real every year to support 30 years (or more).

[mathjak107]

25 years is not long enough and sequences of returns makes it even worse once you try to get enough return not to have any money left at all.

That draw down figures not having a penny left for a longer life or unexpected spending beyond the budget.

2% real returns is about the min i would consider safe enough for  a 30 year or longer retirement not zero

[arebelspy]

2% real returns, huh?

So if you assume 4% withdrawals, 3% inflation, and 5.06% nominal return for 10 years (which gives us a real return of 2% (Formula for real return = (1+nominal rate)/(1+inflation rate)-1)), ending portfolio after 10 years would be 6.675% higher than initial amount.

E.g. starting at 1MM, spending 40k/yr, increasing with inflation at 3%/yr, earning 5.06%/yr (this is, of course, assuming steady returns, which totally skews things, and is the whole point of the trinity study, but I'm ignoring that for the moment), you'll end with 1,066,750 after 10 years.

At what point along the way do you monitor, and with what triggers?  Saying "2% real after a decade" is nice, but I'd probably be looking at it before then, and looking at portfolio value, not just return amount.

[Frs1661]

2% real returns, huh?

So if you assume 4% withdrawals, 3% inflation, and 5.06% nominal return for 10 years (which gives us a real return of 2% (Formula for real return = (1+nominal rate)/(1+inflation rate)-1)), ending portfolio after 10 years would be 6.675% higher than initial amount.

E.g. starting at 1MM, spending 40k/yr, increasing with inflation at 3%/yr, earning 5.06%/yr (this is, of course, assuming steady returns, which totally skews things, and is the whole point of the trinity study, but I'm ignoring that for the moment), you'll end with 1,066,750 after 10 years.

At what point along the way do you monitor, and with what triggers?  Saying "2% real after a decade" is nice, but I'd probably be looking at it before then, and looking at portfolio value, not just return amount.

In real terms,  1MM after 10 years of inflation at 3% you would need 1.344MM to have the same buying power as your starting portfolio. So this investor is down by about 25% with 33%of their 30y retirement over... Seems OK. No reason to panic anyway.

[arebelspy]

2% real returns, huh?

So if you assume 4% withdrawals, 3% inflation, and 5.06% nominal return for 10 years (which gives us a real return of 2% (Formula for real return = (1+nominal rate)/(1+inflation rate)-1)), ending portfolio after 10 years would be 6.675% higher than initial amount.

E.g. starting at 1MM, spending 40k/yr, increasing with inflation at 3%/yr, earning 5.06%/yr (this is, of course, assuming steady returns, which totally skews things, and is the whole point of the trinity study, but I'm ignoring that for the moment), you'll end with 1,066,750 after 10 years.

At what point along the way do you monitor, and with what triggers?  Saying "2% real after a decade" is nice, but I'd probably be looking at it before then, and looking at portfolio value, not just return amount.

In real terms,  1MM after 10 years of inflation at 3% you would need 1.344MM to have the same buying power as your starting portfolio. So this investor is down by about 25% with 33%of their 30y retirement over... Seems OK. No reason to panic anyway.

Right.  And that was the minimum mathjak said he's looking for.  He's mentioned this in many, many threads.  I'm curious though what he's actually looking for.  I mean, unless you don't check until the 10 year mark, it's not a super helpful indicator.

[dougules]

2% real returns, huh?

So if you assume 4% withdrawals, 3% inflation, and 5.06% nominal return for 10 years (which gives us a real return of 2% (Formula for real return = (1+nominal rate)/(1+inflation rate)-1)), ending portfolio after 10 years would be 6.675% higher than initial amount.

E.g. starting at 1MM, spending 40k/yr, increasing with inflation at 3%/yr, earning 5.06%/yr (this is, of course, assuming steady returns, which totally skews things, and is the whole point of the trinity study, but I'm ignoring that for the moment), you'll end with 1,066,750 after 10 years.

At what point along the way do you monitor, and with what triggers?  Saying "2% real after a decade" is nice, but I'd probably be looking at it before then, and looking at portfolio value, not just return amount.

In real terms,  1MM after 10 years of inflation at 3% you would need 1.344MM to have the same buying power as your starting portfolio. So this investor is down by about 25% with 33%of their 30y retirement over... Seems OK. No reason to panic anyway.

Right.  And that was the minimum mathjak said he's looking for.  He's mentioned this in many, many threads.  I'm curious though what he's actually looking for.  I mean, unless you don't check until the 10 year mark, it's not a super helpful indicator.

I'm curious if any of these calculations are taking dividends into account.  A 2% WR would almost be covered by dividends alone at the moment, and dividend yields are really low compared to most of history. 

[ender]

I'm curious if any of these calculations are taking dividends into account.  A 2% WR would almost be covered by dividends alone at the moment, and dividend yields are really low compared to most of history.

Keep in mind if 2% dividends remain constant, it's 2% of market value.

So if you start with $1M, and 20k/year spend, but the market drops to $500k but still has 2% dividends, you are only getting $10k/year in dividends. Not a constant $20k.

This is a common mistake when considering dividend yield and SWRs.

[dougules]

I'm curious if any of these calculations are taking dividends into account.  A 2% WR would almost be covered by dividends alone at the moment, and dividend yields are really low compared to most of history.

Keep in mind if 2% dividends remain constant, it's 2% of market value.

So if you start with $1M, and 20k/year spend, but the market drops to $500k but still has 2% dividends, you are only getting $10k/year in dividends. Not a constant $20k.

This is a common mistake when considering dividend yield and SWRs.

The 2% was relative to the numbers as they stand now.  Dividends generally don't drop nearly as bad as prices when a crash happens, so if all your expenses are covered by dividends you should be golden.  Just a quick run through cfiresim shows that you might have a sequence of returns risk if there's 70s style inflation, but it wouldn't be that bad.  Anyway, my point is that a WR is pretty conservative if it is mostly covered by dividends when you start out. 

http://www.multpl.com/s-p-500-dividend-yield/

Notice the spikes that happened when the market went south. 

[ender]

Notice the spikes that happened when the market went south.

What spikes?

The only "spike" there was during the great depression. Everywhere else dividend yield stays fairly flat (2008/2009 is a good example, the market dropped much more than dividends increased).

[Classical_Liberal]

I'm curious if any of these calculations are taking dividends into account.  A 2% WR would almost be covered by dividends alone at the moment, and dividend yields are really low compared to most of history.

Keep in mind if 2% dividends remain constant, it's 2% of market value.

So if you start with $1M, and 20k/year spend, but the market drops to $500k but still has 2% dividends, you are only getting $10k/year in dividends. Not a constant $20k.

This is a common mistake when considering dividend yield and SWRs.

The 2% was relative to the numbers as they stand now.  Dividends generally don't drop nearly as bad as prices when a crash happens, so if all your expenses are covered by dividends you should be golden.  Just a quick run through cfiresim shows that you might have a sequence of returns risk if there's 70s style inflation, but it wouldn't be that bad.  Anyway, my point is that a WR is pretty conservative if it is mostly covered by dividends when you start out. 

http://www.multpl.com/s-p-500-dividend-yield/

Notice the spikes that happened when the market went south.

Doesn't matter, at a 2%WR you are immune to everything with the exception of collapse of western civilization.  Unless someone is concerned about a legacy, with a WR that low, the vast majority of assets can invested conservatively so that sequence risk goes away.

[mathjak107]

I'm curious if any of these calculations are taking dividends into account.  A 2% WR would almost be covered by dividends alone at the moment, and dividend yields are really low compared to most of history.

Keep in mind if 2% dividends remain constant, it's 2% of market value.

So if you start with $1M, and 20k/year spend, but the market drops to $500k but still has 2% dividends, you are only getting $10k/year in dividends. Not a constant $20k.

This is a common mistake when considering dividend yield and SWRs.

unless dividends are cut or raised  the dividend is constant while the yield from share price changes with nav for new buyers only .

a dividend is declared as a dollar amount not a percentage of share price .

if i get a 1 dollar dividend it stays 1 dollar  regardless of share price . only those first buying in are getting a higher or lower yield percentage wise since they still get the same 1 dollar i do

[mathjak107]

I'm curious if any of these calculations are taking dividends into account.  A 2% WR would almost be covered by dividends alone at the moment, and dividend yields are really low compared to most of history.

Keep in mind if 2% dividends remain constant, it's 2% of market value.

So if you start with $1M, and 20k/year spend, but the market drops to $500k but still has 2% dividends, you are only getting $10k/year in dividends. Not a constant $20k.

This is a common mistake when considering dividend yield and SWRs.

The 2% was relative to the numbers as they stand now.  Dividends generally don't drop nearly as bad as prices when a crash happens, so if all your expenses are covered by dividends you should be golden.  Just a quick run through cfiresim shows that you might have a sequence of returns risk if there's 70s style inflation, but it wouldn't be that bad.  Anyway, my point is that a WR is pretty conservative if it is mostly covered by dividends when you start out. 

http://www.multpl.com/s-p-500-dividend-yield/

Notice the spikes that happened when the market went south.

sequence odf return risk can take those same average total returns and throw them off as much as a 15 year difference in how long that money lasts .

same exact total returns can see a huge difference

[arebelspy]

I'm curious if any of these calculations are taking dividends into account.  A 2% WR would almost be covered by dividends alone at the moment, and dividend yields are really low compared to most of history.

Keep in mind if 2% dividends remain constant, it's 2% of market value.

So if you start with $1M, and 20k/year spend, but the market drops to $500k but still has 2% dividends, you are only getting $10k/year in dividends. Not a constant $20k.

This is a common mistake when considering dividend yield and SWRs.

unless dividends are cut or raised  the dividend is constant while the yield from share price changes with nav for new buyers only .

a dividend is declared as a dollar amount not a percentage of share price .

if i get a 1 dollar dividend it stays 1 dollar  regardless of share price . only those first buying in are getting a higher or lower yield percentage wise since they still get the same 1 dollar i do

Right, he's saying they're often cut.  So if dividends remain 2% (which is a percentage), they'll be cut to match the new share price.  The graph shows they're much more aligned with market price than dividend chasers would have you believe.

[mathjak107]

got it , if that is the case .

[dougules]

I'm curious if any of these calculations are taking dividends into account.  A 2% WR would almost be covered by dividends alone at the moment, and dividend yields are really low compared to most of history.

Keep in mind if 2% dividends remain constant, it's 2% of market value.

So if you start with $1M, and 20k/year spend, but the market drops to $500k but still has 2% dividends, you are only getting $10k/year in dividends. Not a constant $20k.

This is a common mistake when considering dividend yield and SWRs.

unless dividends are cut or raised  the dividend is constant while the yield from share price changes with nav for new buyers only .

a dividend is declared as a dollar amount not a percentage of share price .

if i get a 1 dollar dividend it stays 1 dollar  regardless of share price . only those first buying in are getting a higher or lower yield percentage wise since they still get the same 1 dollar i do

Right, he's saying they're often cut.  So if dividends remain 2% (which is a percentage), they'll be cut to match the new share price.  The graph shows they're much more aligned with market price than dividend chasers would have you believe.

I'm not saying dividends don't go down in a recession.  They definitely do, but they're way less volatile than prices.  Spikes may not have been the best term for the graphic, but you'll notice that the dividend yield rose in most of the worst economic periods.  That illustrates that the dividends did a lot better than prices through the trouble. 

Anyway my point is that a WR that is purely dividends is very conservative.  Yes dividends can and do go down, but if they cover your expenses now it would take a pretty bad economy to make you dip into any principal.  Even then probably not much. 

[arebelspy]

[quote author=dougules link=topic=39064.msg1233781#msg1233781 date=1474321735. 
Anyway my point is that a WR that is purely dividends is very conservative.  Yes dividends can and do go down, but if they cover your expenses now it would take a pretty bad economy to make you dip into any principal.  Even then probably not much.
[/quote]

I agree with you, a dividend approach is much too conservative (and subpar for other reasons as well).

[mathjak107]

The term total return means your complete roi regardless of how the roi is achieved .

I am retired . I reinvest all dividends right now . I own dividend paying funds , bond funds and non dividend paying funds .

I have a withdrawal rate of 3.50% . I take that from cash instruments right now while letting all investments still grow . My total portfolio yield income wise is about 3% with total return about 5% right now ytd on the portfolio.

Except for the fact that i draw my money for living from a cash buffer my outcome in down markets is identical to if i spent the dividends directly and held less cash .

There is no diifference in down markets as to how my income is constructed ,whether apreciation or dividends or both .

The outcome in down markets is the same whether you use cash buckets , stocks and bonds or whether you spend dividends and don't reinvest them or even spend equally from all parts of the portfolio in a systematic withdrawal that preserves your allocation as you spend .

To think you are any different spending down dividends that come off the share price vs a whole portfolio in a down market or up market is nonsense.

Just ask yourself if you reinvested the dividend and took the same money out of the overall portfolio or out of a non dividend payer with the same return would a down market effect you any less . Of course not is the answer .you are just fooling yourself if you think there is any difference except for some small tax differences. Funds have no fee to sell

[dude]

Interesting article, related to the SafeMax discussion.  Back-tests the Kitces/Pfau "rising equity glide paths" strategy against other fixed allocation and declining glide path (the norm in financial planning) strategies.  Was surprised to see the 60% fixed allocation a pretty clear winner here, all things considered (esp. when considering its simplicity). Was equally surprised to see a 60/40 stocks/bills allocation generally superior to 60/40 stocks/bonds (probably good news for TSP participants with the G Fund).  It's long and detailed, but worth the read, IMHO:

https://www.onefpa.org/journal/Pages/MAR15-Retirement-Risk,-Rising-Equity-Glide-Paths,-and-Valuation-Based-Asset.aspx

[Classical_Liberal]

Interesting article, related to the SafeMax discussion.  Back-tests the Kitces/Pfau "rising equity glide paths" strategy against other fixed allocation and declining glide path (the norm in financial planning) strategies.  Was surprised to see the 60% fixed allocation a pretty clear winner here, all things considered (esp. when considering its simplicity). Was equally surprised to see a 60/40 stocks/bills allocation generally superior to 60/40 stocks/bonds (probably good news for TSP participants with the G Fund).  It's long and detailed, but worth the read, IMHO:

https://www.onefpa.org/journal/Pages/MAR15-Retirement-Risk,-Rising-Equity-Glide-Paths,-and-Valuation-Based-Asset.aspx

Thanks for the link, this is very interesting! 

Corporate issued bonds tend to have more correlation with stock prices. Both are driven by earnings.  In an earnings recession, the chance for default on corporate bonds increases which drives up rates and can hurt the face values of existing bonds.  Whereas the opposite is true for Tbills.  In a recessionary period treasury yields tend to decrease, both from the free market "safety" end and from a macroeconomic policy standpoint. This has the effect of increasing face values of previously owned treasuries.  The idea that Tbill allocation outperforms total bonds from a WR standpoint makes sense, theoretically anyway. 

If the point of a portfolio in the initial 10-15 years of drawdown is to minimize sequence risk, Tbills seem to be the best choice.  Decrease sequence risk= Minimized volatility = noncorrelated assets (even if total returns are lower).  The problem with this theory; Treasury yields are yet to recover from the last, rather extreme, economic cycle.  This back testing may not hold true for the near future.

[mathjak107]

the relationship between bonds and stock  is all over the place . above 3% they  correlate one way , under 3% they have correlated another and at the transition line flip a coin .

[index]

The value of the dividend comes right out of the stock price. That is why "dividend capture" trading schemes don't work. There is nothing magical about receiving a dividend. It is a forced taxable event and can be achieved in the exact same manner by selling shares.   

[Telecaster]

The value of the dividend comes right out of the stock price. That is why "dividend capture" trading schemes don't work. There is nothing magical about receiving a dividend. It is a forced taxable event and can be achieved in the exact same manner by selling shares.

^^ Thank Odin someone understands this! 

[mathjak107]

that is the hardest concept for folks to get in to their heads . it is so misunderstood .

[dougules]

The term total return means your complete roi regardless of how the roi is achieved .

I am retired . I reinvest all dividends right now . I own dividend paying funds , bond funds and non dividend paying funds .

I have a withdrawal rate of 3.50% . I take that from cash instruments right now while letting all investments still grow . My total portfolio yield income wise is about 3% with total return about 5% right now ytd on the portfolio.

Except for the fact that i draw my money for living from a cash buffer my outcome in down markets is identical to if i spent the dividends directly and held less cash .

There is no diifference in down markets as to how my income is constructed ,whether apreciation or dividends or both .

The outcome in down markets is the same whether you use cash buckets , stocks and bonds or whether you spend dividends and don't reinvest them or even spend equally from all parts of the portfolio in a systematic withdrawal that preserves your allocation as you spend .

To think you are any different spending down dividends that come off the share price vs a whole portfolio in a down market or up market is nonsense.

Just ask yourself if you reinvested the dividend and took the same money out of the overall portfolio or out of a non dividend payer with the same return would a down market effect you any less . Of course not is the answer .you are just fooling yourself if you think there is any difference except for some small tax differences. Funds have no fee to sell

The value of the dividend comes right out of the stock price. That is why "dividend capture" trading schemes don't work. There is nothing magical about receiving a dividend. It is a forced taxable event and can be achieved in the exact same manner by selling shares.

^^ Thank Odin someone understands this!

I completely agree.  Earnings are earnings whether they're retained or distributed as dividends.  I was more just trying to make the point that a WR is really conservative when it's covered by dividends.  You're not even taking retained earnings into account at all except as a SHTF back-up. 

that is the hardest concept for folks to get in to their heads . it is so misunderstood .

Honestly it took me a while to wrap my head around this.  It does take some thought to understand the fact there is almost always some way to reinvest to earn more or cut costs. 

[dougules]

I am probably wasting my time writing this, but the incorrect statements keep coming.

RECEIVING A DIVIDEND IS NOT THE SAME AS SELLING A SHARE OF STOCK!

I agree as a practical note that an investor should focus on total return as apposed to dividends or capital appreciation at the exclusion of the other. However, dividends are different then capital appreciation. These are viewed differently by the IRS, SEC, and GAAP.

Dividends and capital appreciation affect the ownership of the company differently. If I own several shares of Target and sell several shares Target has the exact same market capitalization but I own a smaller piece of it as I sold some of my ownership to another investor (Transaction between another investor and me). If Target pays a dividend my ownership % stays the same but the value of what I own decreases(Transaction between Target and me).

There is a theory that when a dividend is received an investor will have the same net worth, ignoring taxes; but it is practically impossible to view this in the real world as the market price is affected by so many variables simultaneously. Further this theory leaves out the fact that the markets valuation of a company is based on many different factors such as future growth and not liquidation value of the companies assets. Why does this matter? I think an example will shed light on this.

LinkedIn has a  market cap of $26 billion and a cash balance of $546 million. If it issued a dividend totaling $540 million do you think the market cap would decrease to $25.46 billion after it was paid? It would likely decrease much more because they would be cash starved. Dividends affect the financial statements of entities and are an indicator of organizational health, investment opportunities, and the growth stage of the organization.

Well, yeah, dividends vs reinvested retained earnings do pan out differently in the real world.  I personally like it when companies retain earnings right now since I'm still working.  Dividends are going to get taxed at the high pre-FIRE rate for me instead of the nice post-FIRE capital gains rate for selling shares.  Plus reinvestment is the high risk, high reward option which is good when you're not relying on your stash.  I'm not going to get involved in the companies' business, so it's all the same to me owning fewer shares of a bigger company vs more shares of a smaller company (autoreinvest dividends). 

I think my least favorite is companies sitting on piles of cash.  It seems to be popular right now (like your LinkedIn example). 

At the end of the day, though, earnings are still earnings.  I'm trusting that the market is more efficient than me at valuing what companies do with them. 

[mathjak107]

stocks pay dividends even when they lose money . dividends are just an amount of money paid out decided by the board . right now there are quite a few major stocks that are paying out more than they even earned .

[k9]

EDIT: read too fast.

[markbike528CBX]

  Dividends are going to get taxed at the high pre-FIRE rate for me instead of the nice post-FIRE capital gains rate for selling shares.

for 10 and 15% tax brackets, BOTH dividends and cap gains are taxed at 0%. 

[mathjak107]

the issue ,especially with funds is what is a qualified dividend .there are many years a lot of  our dividends are not qualified .

dividends are not based on a 1 year holding period . you have special holding requirements personally and then the fund has it's own set of requirements .

Holding periods

Although the holding period requirement is the same whether you received a dividend for shares you hold directly or in a mutual fund during the tax year, how you determine the holding period may vary, as outlined below.
Note: When counting the number of days the fund was held, include the day the fund was disposed of, but not the day it was acquired.

Mutual funds
All of the following requirements must be met:
The fund must have held the security unhedged for at least 61 days out of the 121-day period that began 60 days before the security’s ex-dividend date. (The ex-dividend date is the date after the dividend has been paid and processed and any new buyers would be eligible for future dividends.)
For certain preferred stock, the security must be held for 91 days out of the 181-day period, beginning 90 days before the ex-dividend date. The amount received by the fund from that dividend-generating security must have been subsequently distributed to you.
You must have held the applicable share of the fund for at least 61 days out of the 121-day period that began 60 days before the fund’s ex-dividend date.
Stock
You must have held those shares of stock unhedged for at least 61 days out of the 121-day period that began 60 days before the ex-dividend date.
For certain preferred stock, the security must be held for 91 days out of the 181-day period beginning 90 days before the ex-dividend date.