Stop worrying about the 4% rule

[Retire-Canada]

I should add that none of those comments add anything at all to the argument (which appears correct) that having a mortgage increases your SORR. If you are so inclined then go for it.

Sorry Steveo you really have not shown this ^^^ is true at all. The numbers I have posted [which you can verify with cFIREsim] show the mortgage account is less risky than the FIRE account at 4%WR over 30yrs. Instead of just insisting you are correct just because please demonstrate it. If it is true you'll be able to show it.

To Exflyboy's point there is nothing at all wrong with doing something that is not optimal according to the math for any number of reasons. If it makes you happy or more secure emotionally/psychologically in your retirement plans. Even if you just read a blog post and decided that sounds cool. We are not robots. We don't have to compute every decision.

That said I think it is important to dig down into these issues and see what's what so we can have the facts to support the various options.

[steveo]

I should add that none of those comments add anything at all to the argument (which appears correct) that having a mortgage increases your SORR. If you are so inclined then go for it.

Sorry Steveo you really have not shown this ^^^ is true at all. The numbers I have posted [which you can verify with cFIREsim] show the mortgage account is less risky than the FIRE account at 4%WR over 30yrs. Instead of just insisting you are correct just because please demonstrate it. If it is true you'll be able to show it.

To Exflyboy's point there is nothing at all wrong with doing something that is not optimal according to the math for any number of reasons. If it makes you happy or more secure emotionally/psychologically in your retirement plans. Even if you just read a blog post and decided that sounds cool. We are not robots. We don't have to compute every decision.

That said I think it is important to dig down into these issues and see what's what so we can have the facts to support the various options.

I agree with your comments about having to judge things rationally and factually. It's also okay to do something sub-optimal. I'm not convinced in relation to your analysis and I trust ERN's analysis more but he and myself could be wrong. I think I have shown that the analysis in relation to SORR is correct. I've used ERN's blog post (his analysis on the whole is the best I've seen on WR's) plus a logical definition of what could occur when it comes to holding a mortgage whereas I think your analysis is a little too simplistic however that doesn't mean you are wrong.

So we have different analysis stating that a mortgage is safer or less safe specifically when it comes to SORR. I suppose we have to leave it there until we know statistically the correct option based on this issue which may not occur. We also have to recognise that this is just betting with probabilities. The future may be different to the past.

We shouldn't though state that there is no impact in relation to SORR because the evidence says so. The evidence is at this point unclear however I'm erring on the side that a mortgage increases your SORR.

I agree it's good to dig into these issues and come up with good guidelines for people to use.

Personally there is no way in hell that I am leveraging into any market at this point in time but that is a subjective personal decision.

[boarder42]

The only evidence that it effects your safety is that you run out of money sooner with a mortgage in some cases you still fail fire in both situations. Doesn't really matter if the money lasts 20 years or 15 years you lost. The mortgage prevents failures of more historic scenarios specifically when related to inflation. 

[PizzaSteve]

SORR = sequence of returns risk.  It's a relatively new phrase in early retirement jargon.  It refers to the fact that if you have a bad few years at the beginning of retirement, that you can get into trouble with sustaining your retirement because the remaining investments can't recover from the double whammy of bad returns plus your withdrawals.

Thank you for the explanation!

Good explanation, though it is not a relatively new term.  It is an old concept. 

Basically, equity markets are volatile, so financial analysts sought to quantify the risks of the market.  Using old data (and assuming that old data has value as predictive of future events (a key assumption), the analysis showed that market volatility ranges could be quantified and then a range and probability of possible return outcomes could be simulated (fitting data to the bell curve of prior outcomes).

This type of simulation is the foundation behind papers like the Trinity Study and derived conclusions from it (like 4% is a pretty secure withdraw rate).  will the future fit the past?  No one knows, but a sequence of returns similar to the past is a fundamental assumption behind the models.  Note that I was doing this at Wharton on minicomputers using Minitab in 1984. 

Anyway, this all this boils down to using mathmatical models to help people plan.  Since you only have one life, it is up to you to decide if you want to work more, save more and be conservative or YOLO it, trust the data will apply to you and plan to live off better than 4% returns.

The thread is a good resource to explore various people's decisions.

[steveo]

The only evidence that it effects your safety is that you run out of money sooner with a mortgage in some cases you still fail fire in both situations. Doesn't really matter if the money lasts 20 years or 15 years you lost. The mortgage prevents failures of more historic scenarios specifically when related to inflation.

Where is your analysis that supports this ? I'd also suggest the time to run out of money does matter. You may be close to receiving social security or some other payments for instance.

It's probably a good idea to also read ERN's posting on this issue and even the whole WR question. There is lots of good stuff on his site.

[Classical_Liberal]

also suggest the time to run out of money does matter.

This is a very important point that seems to be lost on folks.  A failure is not just a failure.  Failure scenario "a" can be waaay different than scenario "b".  This is of particular importance to the crowd that promotes adding in extra income or decreasing spending to avoid SORR, or those with higher WR's who are OK with increased risk for those necessities at some point in a 50-60 retirement.  IOW a failure rate of 20%, but one in which a worst case is coming up 50K short, is totally different than a failure rate of 5% when the fewer in frequency failures are massive misses.  One implies minor corrections could put you at 100%.  While the other implies in 5% of the cases it's cat food and government housing, or some huge lifestyle changes at some point. 

I haven't run the numbers with mortgages and make no claim how one way or another works towards either end of such a failure spectrum.  I just think it's definitely something people should keep in mind when playing with these historical numbers and regarding AA.

[boarder42]

The only evidence that it effects your safety is that you run out of money sooner with a mortgage in some cases you still fail fire in both situations. Doesn't really matter if the money lasts 20 years or 15 years you lost. The mortgage prevents failures of more historic scenarios specifically when related to inflation.

Where is your analysis that supports this ? I'd also suggest the time to run out of money does matter. You may be close to receiving social security or some other payments for instance.

It's probably a good idea to also read ERN's posting on this issue and even the whole WR question. There is lots of good stuff on his site.

i've read his info ... as was said above he used fixed inflation - inflation was never fixed in our history.  Run the scenarios thru cFIREsim and you'll find the conclusion i've drawn and Retire canada drew. 

also see my analysis above showing perpetual mortgages and their great hedge against SORR.  you know based on historical data not some fixed conditions that have never occured.

[Retire-Canada]

This is a very important point that seems to be lost on folks.  A failure is not just a failure.  Failure scenario "a" can be waaay different than scenario "b".  This is of particular importance to the crowd that promotes adding in extra income or decreasing spending to avoid SORR, or those with higher WR's who are OK with increased risk for those necessities at some point in a 50-60 retirement.  IOW a failure rate of 20%, but one in which a worst case is coming up 50K short, is totally different than a failure rate of 5% when the fewer in frequency failures are massive misses.  One implies minor corrections could put you at 100%.  While the other implies in 5% of the cases it's cat food and government housing, or some huge lifestyle changes at some point. 

I haven't run the numbers with mortgages and make no claim how one way or another works towards either end of such a failure spectrum.  I just think it's definitely something people should keep in mind when playing with these historical numbers and regarding AA.

That's a good point. What do the failures look like and how hard is it to mitigate them?

I added in an $150K income event after 10yrs in my invest the mortgage simulation and pushed the 30yr success rate in cFIREsim to 100%. So in the example we've been playing with instead of taking out a $1M mortgage on your home and investing it all you could only invest $800K and keep $200K in some secure vehicle that matched inflation as closely as possible. Then deploy the $200K after 10yrs to get an investment plan that is far less risky than the standard 4%WR and definitely does not have higher SORR as Steveo claims.

The key with the mortgage as we've pointed out and ERN's analysis failed to take into account is the inflation protection. Once you get past the early years in FIRE inflation is the big risk and a mortgage is a great inflation hedge. Even excluding the equity in the property itself investing the mortgage as discussed in the paragraph above will make your overall FIRE less risky at a 4-5%WR since in some failure years for the 4-5%WR FIRE person their mortgage investment account has a surplus of funds they can draw on.

[Retire-Canada]

I agree with your comments about having to judge things rationally and factually. It's also okay to do something sub-optimal. I'm not convinced in relation to your analysis and I trust ERN's analysis more but he and myself could be wrong. I think I have shown that the analysis in relation to SORR is correct. I've used ERN's blog post (his analysis on the whole is the best I've seen on WR's) plus a logical definition of what could occur when it comes to holding a mortgage whereas I think your analysis is a little too simplistic however that doesn't mean you are wrong.

Rather than believing anyone Steveo about an important life decision like this I'd encourage you to do your own analysis. It's not that hard.

ERN fails to account for realistic inflation possibilities over the course of the 30yr mortgage. If you do use historical inflation values that lowers the risk of the plan considerably. He also fails to look at any of the obvious ways to take the invested mortgage risk to zero against historical market returns. For example only invest $800K of the $1M mortgage and drop the other $200K in after 10yrs or the SORR has played out.

[AdrianC]

"...SORR has played out."

When would you know that sequence of returns risk has played out?

[secondcor521]

SORR = sequence of returns risk.  It's a relatively new phrase in early retirement jargon.  It refers to the fact that if you have a bad few years at the beginning of retirement, that you can get into trouble with sustaining your retirement because the remaining investments can't recover from the double whammy of bad returns plus your withdrawals.

Thank you for the explanation!

Good explanation, though it is not a relatively new term.  It is an old concept. 

Basically, equity markets are volatile, so financial analysts sought to quantify the risks of the market.  Using old data (and assuming that old data has value as predictive of future events (a key assumption), the analysis showed that market volatility ranges could be quantified and then a range and probability of possible return outcomes could be simulated (fitting data to the bell curve of prior outcomes).

This type of simulation is the foundation behind papers like the Trinity Study and derived conclusions from it (like 4% is a pretty secure withdraw rate).  will the future fit the past?  No one knows, but a sequence of returns similar to the past is a fundamental assumption behind the models.  Note that I was doing this at Wharton on minicomputers using Minitab in 1984. 

Anyway, this all this boils down to using mathmatical models to help people plan.  Since you only have one life, it is up to you to decide if you want to work more, save more and be conservative or YOLO it, trust the data will apply to you and plan to live off better than 4% returns.

The thread is a good resource to explore various people's decisions.

Emphasis added by me.

As to the first bolded statement, as I said in my previous post that you quoted, it is a relatively new phrase - at least it's relatively new to me and I've been studying FIRE since the late 1990's.  I agree it's an old concept; in fact I think I've complained somewhere that recent bloggers think they've discovered something new which they haven't, and presenting it as such is just silly.

As for the second bolded statement, you are incorrect.  The Trinity study (at least the famous one that I assume you are referring to) was based on historical data, not parameterized simulations a la Monte Carlo.  There are no bell curves in the Trinity study as far as I know.

[EscapeVelocity2020]

As for the second bolded statement, you are incorrect.  The Trinity study (at least the famous one that I assume you are referring to) was based on historical data, not parameterized simulations a la Monte Carlo.  There are no bell curves in the Trinity study as far as I know.

I think you are missing the point of PizzaSteve's comment, if you read the entire thing.  Basically, by using the Trinity study, you assume that the rolling 30 year periods in the past capture then 30 rolling year period in the future.

FIRE simulators do the Trinity study one better by showing you all of the outcomes of 30 year periods for different asset allocations, income events, etc., but these still have limitations like we can't adjust the periods to start with a "post Fed bank bail-out, persistent low inflation, low unemployment, protectionist trade war, giant corporate tax cut, persistent high-PE, etc." year... 

What unworried 4%-ER'ers bet is that this is no worse than one of the 30 year rolling periods in the past.

[steveo]

I agree with your comments about having to judge things rationally and factually. It's also okay to do something sub-optimal. I'm not convinced in relation to your analysis and I trust ERN's analysis more but he and myself could be wrong. I think I have shown that the analysis in relation to SORR is correct. I've used ERN's blog post (his analysis on the whole is the best I've seen on WR's) plus a logical definition of what could occur when it comes to holding a mortgage whereas I think your analysis is a little too simplistic however that doesn't mean you are wrong.

Rather than believing anyone Steveo about an important life decision like this I'd encourage you to do your own analysis. It's not that hard.

ERN fails to account for realistic inflation possibilities over the course of the 30yr mortgage. If you do use historical inflation values that lowers the risk of the plan considerably. He also fails to look at any of the obvious ways to take the invested mortgage risk to zero against historical market returns. For example only invest $800K of the $1M mortgage and drop the other $200K in after 10yrs or the SORR has played out.

I definitely agree with you in relation to doing my own analysis. I don't agree with your points here though and I think you are underestimating the potential risk of taking on additional leverage. It's your call though. At the same time let's not try and make out that having a mortgage is always the right idea when that is definitely not clear.

Yes ERN's analysis might have some holes in it but I guess your analysis does as well.

Maybe we just have to accept that some things are not completely black and white and individual preference is going to play a role. A good example is asset allocation. There are many different asset allocations that have led to higher SWR's in the past but that does not mean they will retain that edge now and in the future.

At this point I have no interest in obtaining additional leverage to invest more into the stock market or whatever market I choose to invest in. I'm also extremely confident that you and no one else can prove categorically that having a mortgage is some easy path to increasing your WR with no potential negative side effects.

A good rule to remember is that there are no free lunches. So long as you aren't blind to potential downside risks and you accept those risks then make your own personal decision.

[Retire-Canada]

"...SORR has played out."

When would you know that sequence of returns risk has played out?

When you get through your initial phase of FIRE say after 10yrs and have had the pleasure of significant positive returns such that you feel it's unlikely your portfolio would fail due to a poor early sequence of returns.

If you started with a $40K/yr + inflation from $1M portfolio FIRE plan and at 10yrs you $1.8M after inflation I'd feel comfortable calling the early sequence of return risk to have not been an issue for your start year. If you re-started your FIRE at that point you'd be at a 2.2%WR and have 10yrs less duration to deal with.

In the example you were quoting from you could also just hold that $200K in some inflation mitigated vehicle and not add it to your main mortgage investment account. Unless you needed it and in that case you don't have to make any judgement call about SORR.

[PizzaSteve]

"...SORR has played out."

When would you know that sequence of returns risk has played out?

When you get through your initial phase of FIRE say after 10yrs and have had the pleasure of significant positive returns such that you feel it's unlikely your portfolio would fail due to a poor early sequence of returns.

If you started with a $40K/yr + inflation from $1M portfolio FIRE plan and at 10yrs you $1.8M after inflation I'd feel comfortable calling the early sequence of return risk to have not been an issue for your start year. If you re-started your FIRE at that point you'd be at a 2.2%WR and have 10yrs less duration to deal with.

In the example you were quoting from you could also just hold that $200K in some inflation mitigated vehicle and not add it to your main mortgage investment account. Unless you needed it and in that case you don't have to make any judgement call about SORR.

Good point.

Another scenario is that your plan targets a certain age, for personal, benefit or professional accomplishment reasons (say 40 or 50), but your income and MMM inspired consumption levels has you saving in excess of your retirement needs.

Many high income savers could literally hold cash at 0% and never fail an @ 50 retirement.  There is nothing wrong with that person following a strategy of forgoing market returns for 100% safety because they wont ever need all the value from the money they earned.  It is not optimal, but it is fine.

Sure they could retire earlier with equities, but not without some SOR risk, and assuming they are an MD or something, enjoying performing heart surgeries to save lives or something rather than spending on a mercedes every year, they may want a zero stress, 100% win plan.  This is one reason why treasuries and the like sell so well.

In other words, a bond allocation shortens the SOR window in exchange for less top side wealth, with the extreme example being 100% cash, 0% SOR risk.

[sol]

with the extreme example being 100% cash, 0% SOR risk.

Also known as "100% inflation risk".

[pecunia]

How about these lower returns predicted by Vanguard?

https://www.cnbc.com/2017/11/20/jack-bogles-5-bold-investment-predictions-for-2018-and-beyond.html

Mr. Bogle says to expect about a 4% return.  If you FIRE, won't this shrink the stash and make you more susceptible to inflation problems?  He could be wrong too, but "Laissez les bon temps rouler" can't go on forever.

[EscapeVelocity2020]

with the extreme example being 100% cash, 0% SOR risk.

Also known as "100% inflation risk".

But if inflation is fixed at 2% (prices double in 36 years) then even cash can be OK for risk averse folks!  This is the Goldilocks scenario that we are currently making 30 year decisions from.

[sol]

But if inflation is fixed at 2%

I feel like assuming 2% long term inflation is like assuming 9% long term stock returns.  Sure, that could totally happen, but that's not the scenario we're trying to guard against when balancing the different types of portfolio failure risks.

[DreamFIRE]

How about these lower returns predicted by Vanguard?

https://www.cnbc.com/2017/11/20/jack-bogles-5-bold-investment-predictions-for-2018-and-beyond.html

Mr. Bogle says to expect about a 4% return.  If you FIRE, won't this shrink the stash and make you more susceptible to inflation problems?  He could be wrong too, but "Laissez les bon temps rouler" can't go on forever.

Yeah, he was predicting "nominal to zero" REAL returns three years ago:

https://forum.mrmoneymustache.com/investor-alley/bogle-projects-'nominal-to-zero'-real-returns-over-the-next-decade/

[Retire-Canada]

Many high income savers could literally hold cash at 0% and never fail an @ 50 retirement.  There is nothing wrong with that person following a strategy of forgoing market returns for 100% safety because they wont ever need all the value from the money they earned.  It is not optimal, but it is fine.

Just out of interest I ran some numbers in cFIREsim.

50yr FIRE @ 4%WR with 90/10 Portfolio - so $40K/yr off $1M = ~85% success

Holding 100% cash at 0% return for 50yrs spending $40K/yr you get the following success rates:

- $1M = 0%
- $2M = 7%
- $3M = 27%
- $4M = 49%
- $5M = 63%
- $6M = 69%
- $7M = 84%

Starting at $1M invested and adding $100K savings per year at 6% return after inflation it takes ~19.5yrs to get to $7M.
Starting at $1M invested and adding $200K savings per year at 6% return after inflation it takes ~14.5yrs to get to $7M.

Of course you can argue the person working on a cash FIRE will die sooner so they don't need to fund their retirement for the same duration. They might even die at their desk in the extra 14-19 years it takes to save up the required cash...that approach has a 0% chance of not running out of money that's independent of portfolio value.

[DreamFIRE]

If you want to be conservative, you can get cash equivalents that earn more than 0% nominal.  For example, a 3 year CD is yielding 3%.

[AdrianC]

"...SORR has played out."

When would you know that sequence of returns risk has played out?

When you get through your initial phase of FIRE say after 10yrs and have had the pleasure of significant positive returns such that you feel it's unlikely your portfolio would fail due to a poor early sequence of returns.

If you started with a $40K/yr + inflation from $1M portfolio FIRE plan and at 10yrs you $1.8M after inflation I'd feel comfortable calling the early sequence of return risk to have not been an issue for your start year. If you re-started your FIRE at that point you'd be at a 2.2%WR and have 10yrs less duration to deal with.

So really, we know that sequence of returns risk has played out when we are withdrawing substantially less than 4%. To quantify, would it be when we get a 100% success rate in cFIREsim?

[boarder42]

It could be 4% it also could be a shorter life expectancy than when you started. But yes when cfiresim says 100% and you don't have any borderline for a 40 year time line youve surpassed it basically if you get to a 3.5% wr after you've FIREd you're safe

[Exflyboy]

If you want to be conservative, you can get cash equivalents that earn more than 0% nominal.  For example, a 3 year CD is yielding 3%.

I have a relative who is 70 and he claims he is making about 5% just using CD products though Raymond James I believe.

Of course he got a bit cagey when I started asking him about fees and I pointed that interest rates are currently rising. Depending on how long the CD are locked in for and how fast rates rise he may get and effective sending growth that barely matches inflation.

Interesting concept for a "zero risk" portfolio though.

[Retire-Canada]

So really, we know that sequence of returns risk has played out when we are withdrawing substantially less than 4%. To quantify, would it be when we get a 100% success rate in cFIREsim?

If you accept the premise that the future will be no worse than the past a 100% success rate in cFIREsim means you can be pretty confident that you have mitigated both the SORR and inflation risk. So I'd be pretty happy with that.  You could select a different criteria that wasn't quite as high to decide if you wanted. 

But as I noted in the example we've been talking about if making the call is not something you are willing or comfortable doing you can take that requirement off the table and just invest 80% of the mortgage and keep 20% in reserve in an inflation mitigated vehicle. You lose out on some of the potential returns, but you push your success rate against historical data to 100% and your mortgage investment is less risky than the typical 4%WR FIRE plan and much less risky than a 5%WR FIRE plan.

[Retire-Canada]

At the same time let's not try and make out that having a mortgage is always the right idea when that is definitely not clear.

I would never suggest that ^^^ in fact I'll go further and say it's definitive that there are all sorts of situations we can conceive of where having a mortgage and investing the proceeds is not a good idea. My comments were only referring to the case we were examining which you presented along with the ERN blog post.

[MissNancyPryor]

Question for the group-  very early in the thread Nords is quoted as saying that anything over something like 80% on FIRECalc is just "meaningless precision".  Really?  How so? (Not a critical question, rather a genuine I-don't-get-it thing, sorry to be remedial here).  How does that precision thing work?   

My numbers are running in the 95% range on this and other calculators and still I sit a-quivering about my likelihood of success.  Supposed to be a 2019 cohort and my numbers are good but I am getting OMY sickness.   

[SwordGuy]

Question for the group-  very early in the thread Nords is quoted as saying that anything over something like 80% on FIRECalc is just "meaningless precision".  Really?  How so? (Not a critical question, rather a genuine I-don't-get-it thing, sorry to be remedial here).  How does that precision thing work?   

My numbers are running in the 95% range on this and other calculators and still I sit a-quivering about my likelihood of success.  Supposed to be a 2019 cohort and my numbers are good but I am getting OMY sickness.   

Well, first of all, FIRECalc is based on historical results, i.e., "the past".   You won't be living in the past, you'll be living in the future.   So, really, we don't know what will happen.     

What FIRECalc does is tell you that if the future is no worse than the past (from an FI perspective), then your odds of success are thus and so.  Obviously, the future could be worse than the past.      (It could also be much the same and even better, which would be more likely.) 

So, 95% success and 100% success are really about the same thing, given that the future could be different (and worse).


If Nords said 80% and 100% are really about the same thing (sounds familiar, but I didn't check), he probably meant the above plus he's willing to adapt to circumstances by cutting spending, working part time, etc.


So, if you quit your job, can you get another one similar in pay to it within a 12 month period of time?


Do you really need one at your old salary if things go bad for a couple of years?  Can you cut spending for those years?  Or just get a part-time job and cover the gap?


Can you cut spending to cover the gap?

If you answer any of these questions Yes, then you are FI ready (money-wise).   If not, you should consider a larger stash.


Best of luck.  I understand OMY syndrome, we did it.   Of course, we had a mentally handicapped daughter who would pay the price of our mistake, so we went for the extra safety, so I don't feel foolish for doing that.

[Exflyboy]

Honestly (unless you're in a job you HATE) OMY is not the end of the world. Heck AFTER I retired I took on some contracts paid hourly and it was the most fun (and profitable) work I have ever done.

Some would argue I have over saved just a smidge though..;)

[DreamFIRE]

95% is better than 80%, and there are good arguments that both are likely over estimated in today's environment:

https://forum.mrmoneymustache.com/welcome-to-the-forum/cfiresim-severely-overestimates-success-rates-for-mustachians/

https://forum.mrmoneymustache.com/investor-alley/cape-and-safe-withdrawal-rates/

https://forum.mrmoneymustache.com/investor-alley/start-worrying-about-the-4-rule/

And I think all of this was actually mentioned earlier in this thread.

[pecunia]

Swordguy - I like the way you posted this;

SNIP

Well, first of all, FIRECalc is based on historical results, i.e., "the past".   You won't be living in the past, you'll be living in the future.   So, really, we don't know what will happen.     

What FIRECalc does is tell you that if the future is no worse than the past (from an FI perspective), then your odds of success are thus and so.  Obviously, the future could be worse than the past.      (It could also be much the same and even better, which would be more likely.) 

So, 95% success and 100% success are really about the same thing, given that the future could be different (and worse).

SNIP

So, if you quit your job, can you get another one similar in pay to it within a 12 month period of time?


Do you really need one at your old salary if things go bad for a couple of years?  Can you cut spending for those years?  Or just get a part-time job and cover the gap?


Can you cut spending to cover the gap?

If you answer any of these questions Yes, then you are FI ready (money-wise).   If not, you should consider a larger stash.


Best of luck.  I understand OMY syndrome, we did it.   Of course, we had a mentally handicapped daughter who would pay the price of our mistake, so we went for the extra safety, so I don't feel foolish for doing that.

Great Questions - Firecalc is based on history and history is known to repeat itself so Firecalc should be given some level of trust.

  I'll apply this to myself: yes, yes and maybe.  Makes me feel good.

[TomTX]

I'm not saying that it's actionable for everyone (like if you are just getting started building a portfolio), and certainly not the same action for everyone, but when you are ahead of the game it is prudent to dial back exposure to risk and volatility.  Given the exceptional returns we have experienced since 2009, high stock exposure and 4% SWR is not my best bet.  Everyone here seems to think they have high risk tolerance, but I'll be interested to see how they feel in the middle of a bear market, especially if they are retired.  Fortunately I only need 2 - 3% WR currently, but that assumes inflation stays tame until I get to Medicare and SS.  I will probably ER in a year or two depending mostly on circumstances outside my finances (other than healthcare, I might still work to have access to my company plan).

Dude. You won. Seriously. If you like to keep working, that's fine. Otherwise, enjoy ER.

[TomTX]

The equity in your house does count as a part of your net worth [assets minus liabilities], but does not count as a part of your 'stache for 4% purposes. It may reduce your required expenses (if you own the house, hey, no mortgage! Somewhere to live!), but it doesn't generate income, so you can't use it for your 4% calculation. If you want to downsize and thus turn some home equity equity into extra 'stache, great!

I also consider home equity as a hedge against SORR. Before I retire, I intend to open a HELOC as a "just in case" measure. This will both allow a source of ready cash if the market takes a dump, and will keep variety in my credit report for longer.

[sol]

For anyone worried about SWRs of 3% or lower, and there have been several of you in this thread, please note that today the rate on 10 year US Treasuries went to 3.24%.  That's about as close to a risk-free guaranteed return as you can find, and it is higher than the SWR targeted by some folks here.

[AdrianC]

A half a percent real return isn’t going to do it.

“The current inflation rate for the United States is 2.7% for the 12 months ended August 2018, as published on September 13, 2018 by the U.S. Labor Department.“

[sol]

A half a percent real return isn’t going to do it.

“The current inflation rate for the United States is 2.7% for the 12 months ended August 2018, as published on September 13, 2018 by the U.S. Labor Department.“

You're right, if you were to put 100% of your portfolio into ultra-secure US treasury bonds today with real returns only half a percent above inflation, you would only be guaranteed 28 years of inflation adjusted withdrawals before you would have to find another source of income.  What's your life expectancy?  When can you draw social security?

If only there were some other asset class we could invest in to close that gap!

[EscapeVelocity2020]

A half a percent real return isn’t going to do it.

“The current inflation rate for the United States is 2.7% for the 12 months ended August 2018, as published on September 13, 2018 by the U.S. Labor Department.“

You're right, if you were to put 100% of your portfolio into ultra-secure US treasury bonds today with real returns only half a percent above inflation, you would only be guaranteed 28 years of inflation adjusted withdrawals before you would have to find another source of income.  What's your life expectancy?  When can you draw social security?

If only there were some other asset class we could invest in to close that gap!

At the risk of sounding like Suze Orman... If you're going to go a sarcastically suggested 'all bond' route, at least allocate some amount to something like VAIPX -a TIPS fund currently yielding 3.3%.

[sol]

At the risk of sounding like Suze Orman... If you're going to go a sarcastically suggested 'all bond' route, at least allocate some amount to something like VAIPX -a TIPS fund currently yielding 3.3%.

In this case, AdrianC reduced the current US treasury yield of 3.22% to 0.5% by subtracting off the recent inflation numbers of ~2.7%.  TIPS would generate the same thing.

And yet, for some reason, we still have people here arguing for a 3.0% SWR "just in case".

As a reminder, a 3.22% inflation-adjusted withdrawal rate of a mixed US stock/bond portfolio has never failed, at any point in history, for any length of retirement. 

[AdrianC]

You're right, if you were to put 100% of your portfolio into ultra-secure US treasury bonds today with real returns only half a percent above inflation, you would only be guaranteed 28 years of inflation adjusted withdrawals before you would have to find another source of income.  What's your life expectancy?  When can you draw social security?

If only there were some other asset class we could invest in to close that gap!

Actually, if you could guarantee a 0.5% real return and did a 3% SWR, your money would run out in about 37 years.

Personally, I want to do better than that - more years, higher SWR, so I'm invested in stocks, and I'm quite sure you are too.

We can't compare a nominal return to an SWR. It's not a meaningful comparison. Real returns are what counts.

[PizzaSteve]

At the risk of sounding like Suze Orman... If you're going to go a sarcastically suggested 'all bond' route, at least allocate some amount to something like VAIPX -a TIPS fund currently yielding 3.3%.

In this case, AdrianC reduced the current US treasury yield of 3.22% to 0.5% by subtracting off the recent inflation numbers of ~2.7%.  TIPS would generate the same thing.

And yet, for some reason, we still have people here arguing for a 3.0% SWR "just in case".

As a reminder, a 3.22% inflation-adjusted withdrawal rate of a mixed US stock/bond portfolio has never failed, at any point in history, for any length of retirement.

I think that is because of a combination of the 'shit happens' factor and because some folks are in a quite high income/low consumption situation, where hitting 3% is relatively easy by only putting in a few more years.  Why not stay in a challenging, hot job a few more years to be set beyond reach?  i gree this should not be the typical target, but shaming them for picking it is being mono visual.

It is not disrespectful to say that safety margin depends on circumstances and opportunity.  You are young and healthy. 

For example, I have a friend with a spouse with MS.  Their safety cushion needs are not yours.  They need a savings fund that is higher than usual becauee they know their future yearly spending will increase.   They know the costs of the disease will create issues for their later years, and 3% was sensible.  Also, if I was earning 500k/yr in a high tech job and committed to a 50k/yr consumption lifestyle, I would work OMY to hit 3% for sure (I have a friend who will earn an almost guaranteed 4-10M by working only two more years due to vested options at a tech winner.  Why would anyone throw away a job they love paying over 2M/yr, just to prove they can FIRE by a targeted minimum date?

My point is that 3% is the right number for some (including Suzi, perhaps, who is certainly far from typical 😉).

[Exflyboy]

Also a lot of FIREe's will end up with a ridiculously low WR simply because their portfolios will grow faster than they will spend it. Right now we are are running at somewhere slightly below 1.5% due to the fact that we have rental income and our pensions when they kick in will make our WR even more conservative.

Once that snowball starts rolling, unless you start spending boatloads of money then your net worth (and 1/your WR) will become ever larger as time goes by.

When I quit back in 2014 we had about $1.25M, Now we have about $2.5M simply because our spending rate hasn't risen that much.. Although I can tell you its easier to spend more in retirement than when you were working..:)

I suspect MMM himself with his $27k annual spend has a barley measurable WR!

[nereo]

Also a lot of FIREe's will end up with a ridiculously low WR simply because their portfolios will grow faster than they will spend it. Right now we are are running at somewhere slightly below 1.5% due to the fact that we have rental income and our pensions when they kick in will make our WR even more conservative.

Once that snowball starts rolling, unless you start spending boatloads of money then your net worth (and 1/your WR) will become ever larger as time goes by.

When I quit back in 2014 we had about $1.25M, Now we have about $2.5M simply because our spending rate hasn't risen that much.. Although I can tell you its easier to spend more in retirement than when you were working..:)

I suspect MMM himself with his $27k annual spend has a barley measurable WR!

By his own, seldom-reported accounts MMM is earning far more than they spend through his various semi-passive income streams.  He mentioned one year that htis blog generated $400k, and his rental properties have (at least in some years... he always seems to be buying and selling) covered all his spending.  He mentioned a while back that they've yet to even touch their original 'stache, and instead have been writing sizable checks to Betterment each year. 

with a low annual spend its pretty easy to meet that through a variety of independent measures. Earning enough to cover an $80k lifestyle can be tough without a FT job or several rental properties you have to manage-- earning $27k is easy.  One of the reasons we feel comfortable with you plan to use the 'glide-path' into FI/RE and go part time in our 40s. 

[matchewed]

Also a lot of FIREe's will end up with a ridiculously low WR simply because their portfolios will grow faster than they will spend it. Right now we are are running at somewhere slightly below 1.5% due to the fact that we have rental income and our pensions when they kick in will make our WR even more conservative.

Once that snowball starts rolling, unless you start spending boatloads of money then your net worth (and 1/your WR) will become ever larger as time goes by.

When I quit back in 2014 we had about $1.25M, Now we have about $2.5M simply because our spending rate hasn't risen that much.. Although I can tell you its easier to spend more in retirement than when you were working..:)

I suspect MMM himself with his $27k annual spend has a barley measurable WR!

By his own, seldom-reported accounts MMM is earning far more than they spend through his various semi-passive income streams.  He mentioned one year that htis blog generated $400k, and his rental properties have (at least in some years... he always seems to be buying and selling) covered all his spending.  He mentioned a while back that they've yet to even touch their original 'stache, and instead have been writing sizable checks to Betterment each year. 

with a low annual spend its pretty easy to meet that through a variety of independent measures. Earning enough to cover an $80k lifestyle can be tough without a FT job or several rental properties you have to manage-- earning $27k is easy.  One of the reasons we feel comfortable with you plan to use the 'glide-path' into FI/RE and go part time in our 40s.

My plan has changed a bit in a similar fashion. We have RE that after we move out of it can generate around $10k per year. I use only $7k for my calculations. An additional part time job for me and my SO will take us to only pulling a few thousand out of investments per year for a few years. After that we can ramp up the withdrawals and reduce the working. At some point we will sell the RE for a lump sum later in life and carry on with full income from investments.

It is another way of showing that all those issues that people have with the 4% rule can just easily be mitigated for a ridiculous success rate. And many of those things don't require OMY at some time/stress demanding job.

[PizzaSteve]

Agreed.  Thats why I get so annoyed about the mortgage debates.  It's really all about spending management, with investment returns really secondary.  Any decent investment strategy will do, ETFs, being debt free, individual stocks, rentals, even bonds or CDs are fine, assuming you live honestly and without that need to consume.

So much focus on x% withdraw rates misses the big picture. The models are just a tool/framework.  Lifestyle and savings are what matters, whether at a 2% or an 8% withdraw rate.  If you can manage yourself, track your status and be flexible, you will be fine.

Investment optimization threads are all fine, but secondary IMHO.

[matchewed]

Agreed.  Thats why I get so annoyed about the mortgage debates.  It's really all about spending management, with investment returns really secondary.  Any decent investment strategy will do, ETFs, being debt free, individual stocks, rentals, even bonds or CDs are fine, assuming you live honestly and without that need to consume.

So much focus on x% withdraw rates misses the big picture. The models are just a tool/framework.  Lifestyle and savings are what matters, whether at a 2% or an 8% withdraw rate.  If you can manage yourself, track your status and be flexible, you will be fine.

Investment optimization threads are all fine, but secondary IMHO.

Well the underlying what you're invested in still matters a great deal. There are probably more unsuccessful investment to SWR mixes than successful ones.

[Retire-Canada]

Well the underlying what you're invested in still matters a great deal. There are probably more unsuccessful investment to SWR mixes than successful ones.

Especially at 8%WR.The folks that FIRE at 2%WR at 65yrs with short lived family members in their gene pool probably can relax. ;-)

[TempusFugit]

Well the underlying what you're invested in still matters a great deal. There are probably more unsuccessful investment to SWR mixes than successful ones.

Especially at 8%WR.The folks that FIRE at 2%WR at 65yrs with short lived family members in their gene pool probably can relax. ;-)

Except for the spectre of approaching death. 

[Retire-Canada]

Except for the spectre of approaching death.

Let's face it people shooting for 2%WR or lower a secretly using the "if I die at my desk" strategy as a way to ensure they never have to worry about running out of money and perhaps even better "if I die at my desk I never have to face the scary possibility of actually having to stop working!" So I don't think death holds the same concern for them as it does for folks who are eager to retire and get off the hamster wheel. ;-)

[Retire-Canada]

The other item people forget to think about is the spending aspect of a WR. Spending is not a constant over 50 years.

When talking about %WRs spending is assumed to be appropriately budgeted. If you are at 2%WR and don't have a budget you can live with or you are at 8%WR and have 300% luxury spending built in than there is no point trying to even compare the two.

Before you bother working out a withdrawal rate you need to determine how much annual budget you need for your retirement. If you fail at that step nothing you do further down the planning process is going to be reliable.