Paying off Mortgage Early – How bad is it for your FI Date?

[Faraday]

This is profoundly true. I still need to make that payment.

But I have money, because I wasn't simple-mindedly writing a check to pay down the mortgage from my paycheck: I was funding my mortgage pre-payments through short-term liquid investments. So I have a stepwise liquid cash reserve I use, much like an FU FUND, until I find the next job, get the loan re-cast, or use the cash reserve to do a refi.

So now your previous plan to prepay in lieu of investing has morphed into a plan to invest in cash in lieu of investing in stocks/bonds?  That is not addressing the "carry mortgage vs. pay off mortgage" decision (which is the subject of this thread), and is instead addressing the question of the advisability of engaging in market timing.  Even the person with no mortgage has to decide whether to deploy excess cash towards their bank account or their brokerage/mutual fund account.  And, consistent with the upthread attempts to stymie the off-topic detours into the matter of dividends, that's all I'm going to say on the subject.

Look: I'm not coming to you or anyone else to determine "my plan". I have a plan and I'm sticking to it based on my prior personal investment experience and the experience I've seen other people go through. If avoiding another Enron or GM bailout is "market timing", then I have to accept that you and I are never going to have a reasonable discussion about this question.

I'd truly be a lost soul to take guidance from people who've been so irrational in a public forum. You guys keep dreaming up ridiculous ideas in your head about what I'm really doing, simply because I contested a point and asked questions that so far none of you poop-flinging monkeys has an interest in answering.

Look: here's what's printed at the bottom of the Vanguard Investments website:

"All investing is subject to risk, including the possible loss of the money you invest. Diversification does not ensure a profit or protect against a loss. Investments in bonds are subject to interest rate, credit, and inflation risk."

[brooklynguy]

Not sure what questions you have asked that you believe haven't been answered, since answering your questions is all us "poop-flinging monkeys" have been exhaustively trying to do (which is what set this thread ablaze anew in the first place).

Also not sure why anyone would be considered a "lost soul" for taking advice from a public forum.  I myself am in that boat on this precise issue, as anyone can see by scrolling five pages back into this very thread.

[arebelspy]

We're not talking about your plan, mefla.  We don't even know what it is.  We're talking about paying off your mortgage versus investing.

[Eric]

I sense a new tagline:

The MMM forums:  Home of the most irrational people on the internet.

[matchewed]

I sense a new tagline:

The MMM forums:  The Internet: Home of the most irrational people on the internet.

FTFY ;)

[arebelspy]

I'd truly be a lost soul to take guidance from people who've been so irrational in a public forum ... you poop-flinging monkeys ...

I really did enjoy this though, it's almost sig-worthy.

[MDM]

... ideas in your head about what I'm really doing, simply because I contested a point and asked questions ...

Speaking of questions:

Mortgages are capitalized up-front, investments are capitalized "late".

Don't understand - could you elaborate?

arebelspy offered his opinion of what you were thinking, but it would be better to hear a first person response.  There has been a lot of back and forth since then, so I may have missed it, but what do you mean about the capitalization timing?

[arebelspy]

... ideas in your head about what I'm really doing, simply because I contested a point and asked questions ...

Speaking of questions:

Mortgages are capitalized up-front, investments are capitalized "late".

Don't understand - could you elaborate?

arebelspy offered his opinion of what you were thinking, but it would be better to hear a first person response.  There has been a lot of back and forth since then, so I may have missed it, but what do you mean about the capitalization timing?

I thought mefla agreed with my interpretation based on his response:

Hey arebelspy, it's not MDM that doesn't understand. MDM was quoting my comment, it's me saying "capitalized up front" so it's me who is not understanding.

But if not, feel free to chime in with what you did mean.

[BarkyardBQ]

Look: I'm not coming to you or anyone else to determine "my plan". I have a plan and I'm sticking to it based on my prior personal investment experience and the experience I've seen other people go through. If avoiding another Enron or GM bailout is "market timing", then I have to accept that you and I are never going to have a reasonable discussion about this question.

All investing is subject to risk, TRUE
including the possible loss of the money you invest. If you take out or sell your principal when it's down
Diversification does not ensure a profit or protect against a loss. sure... 1 goes up 1 goes down... usually more like 10 go up, 2 go down, 1 goes out
Investments in bonds are subject to interest rate, credit, and inflation risk. Cash in a savings account loses 2% when inflation is 3% on average and your savings account returns 1%... same for bonds if inflation goes above the average.

mefia, I'm not nearly as smart as some of the people trying to reply to you but I at least understand the math and especially risk tolerance and time horizon involved in weathering a devalued portfolio during a financial crisis... so what I'm going to offer are scenarios that may appeal to your emotional decision.

Scenario:
1) You have a 30 year mortgage with a 1k payment. The economy fumbles, you lose your job, you have no emergency fund, no investments, and you've simply made your normal monthly payments and burned all your remaining disposable income. You are screwed, you either go on subsidized funding and/or default on your mortgage.

2) You have a 30 year mortgage with a 1k payment. The economy fumbles, you lose your job, you have no emergency fund, no investments, and you've  doubled down on your mortgage payments for years. You own half your house, but you have no income, and nothing to liquidate. You will default on your next payment and you will not be able to buy groceries.

3) You have a 30 year mortgage with a 1k payment. The economy fumbles, you lose your job, you have no emergency fund, no investments, and you've doubled down on your mortgage payments for years. You own your house, you have no income, banks are in crisis mode and will not give you a HELOC. You cannot buy groceries.

4) You have a 30 year mortgage with a 1k payment. The economy fumbles, you lose your job, you have a 6 month emergency fund, no investments, and you've doubled down on your mortgage payments for years. [pick one], you have no income, you can chose to either default or buy groceries.

-or-

A) You have a 30 year mortgage with a 1k payment. The economy fumbles, you lose your job... You only ever made your minimum monthly mortgage payment, you built up a 6-12 month emergency fund (because you hate risk) which includes mortgage + expenses. You've invested 1k a month into the market, which is now worth 50% of your peak balance. Because you have 6 - 12 month emergency fund you can survive 6-12 months or you can cut back to beans and rice (you don't buy gas anymore cause you don't have a job) and you can stretch your fund +33% longer while you ride out the crisis or find a new job. You're portfolio is still worth something and can be utilized if this event depletes your emergency fund.


The point everyone else is trying to make is that you cannot predict the next crisis, but if it happens while you are prepaying your mortgage, you will have NO funds to weather the storm. If you are completely against investing in the market because you cannot handle the risk/time horizon for it to return again, then cash is your friend, it will become worth less over time and you should probably check out this thread http://forum.mrmoneymustache.com/welcome-to-the-forum/preppers/

The only way to avoid another Enron/GM, bailout, or crisis is to never invest in the market... 7 years later we have recovered quite well, read http://forum.mrmoneymustache.com/ask-a-mustachian/how-did-you-fare-during-the-20002008-stock-crashes-regale-this-stocks-noob/

I think this one from h2Ogal highlights it pretty well...

2008      151000 - left old job - options also underwater - lost them - downturn also
2009      103834 - at one point even lower like 70K
2010      161963

That's a speed bump... if you are driving around in one of those bottomed out Honda CRX's... enjoy driving through the parking lot like Frogger avoiding all the trouble.

[Faraday]

I thought mefla agreed with my interpretation based on his response:

Hey arebelspy, it's not MDM that doesn't understand. MDM was quoting my comment, it's me saying "capitalized up front" so it's me who is not understanding.

But if not, feel free to chime in with what you did mean.

I see now, I should have put "not understanding" in quotes. Those are your words, not mine. I was trying to make a peace offering to get you assholes off my back and engaged in meaningful communication.

Do you get it? It was a peace offering. it was me saying "stop, stop: I want to listen to your points...This crap where you guys make up stupid things I "allegedly" believe is driving me nutso and it serves no useful purpose except to make me think you are kids with no real-world investing experience.

Now, Here's the deal:

What I mean by "front capitalization" is pretty much what some of the alternative arguments are in the OTHER "mortgage thread" (that brooklynguy was trying to get me to shut up in, since others had come to my rescue there...).

That when you take out a mortgage, but you have no liquid investments, you are standing there in a fairly high state of risk with nothing to cover you in case of job failure. That you run the risk, with a 30 year mortgage, as many people have EXPERIENCED, of being upside down in the property with no way to restructure the debt.

This is the state in which all of you began the discussion, and properly assessed: without any investments you can draw on in the case of job loss, you're screwed. You're standing there owing however-many thousands of dollars. You've leveraged the house you live in and you MUST make the payments.

Day 1 of your mortgage, you and brooklynguy and all the other assholes who chimed in to kick me in the balls are right about. Day 1, you are "screwed".

At this point, the "screwee" is in what I call the "zone of risk". It's the state where job loss = home loss. Now, this is a WORSE state to be in than if you rented. All you assholes say this and I agree. Losing your job in either case is bad, but with a fresh 30 year mortgage, it has a far worse impact than being kicked out of your rental property.

So your first priority, above all - especially above prepaying the mortgage - is to build an "FU FUND". This is a common-enough mustachian principle that it's been made perfectly clear to me and I have done this personally myself. I have an FU fund of about 8 months right now and I've been building it to reach a 12 month level of security. And I've used the other principles that get commonly discussed, to determine what level that needs to be at, while simultaneously cutting my cost of living to fit down into a 4% SWR.

Now, what I did was I built multiple FU funds out of liquid investments. I got this idea from the "CD Ladder' concept. Realizing that I could reinforce those funds by making them investments, not just ramming that money straight into the mortgage or caching it in a  passbook savings account. (although, if that's all you have access to, it's better than nothing...)

And of course, because I'm not an idiot, I'm still investing. I'm contributing pre-tax to a 401k spread over a Vanguard indexed fund. This year, got HSA capability, so I'm pushing a teeny tiny bit to that also.

Fast forward to about two years ago. I've got multiple "tiered" FU funds, a 401k, some stock in an E*TRADE account and I'm restructuring my life to lower my cost of living to approach that magic 4% (or preferably 3%) SWR. Life is good.
 
Suddenly I realize something quite sucky: it's impossible for me to fit down into a 4% SWR while carrying the mortgage I have. I do some back-of-napkin estimates and I realize that I won't reach a high enough level of investment for at least 10 years, to be able to carry my mortgage + retire. I am over-leveraged in my mortgage, and it's REALLY not that big of a mortgage: $200,000.

Well screw that. I want to retire in 2020, five years from now. I can reach 4% SWR with no mortgage and the investment level I feel comfortable that I'll reach in five years.

I decide to accelerate my 15 year mortgage to a capitalization level low enough that I can then refi (remember, they will pay me) to a 10 year low-interest mortgage, with the plan to pay that off in 5 years.  And since I'm aiming for a lower capitalization, as long as I remain employed, I can still continue pre-paying the mortgage AND investing for FIRE.

And then I come across "Runge" and his simulator spreadsheet, and it damn near perfectly describes what I'm aiming for anyway:

HOWEVER, there's another option. What if I did scenario 2, and right before I FIREd, I paid off my mortgage in balance. This leaves me with the lower annual expenses of not having a mortgage while capitalizing on the higher returns of the market compared with the mortgage interest rate. I find that I can retire in year 23, a full 2 years earlier than scenario 1. I get the benefits of having liquid assets to live off of if I lose my job in the first decade over scenario 1 while putting that hard earned cash to work earlier. And this is assuming I never increase my savings over 30 years (and a pretty lame savings rate to boot).

I've attached my spreadsheet. Green cells mean I can FIRE on 4%. Did I miss anything crucial? Yes of course I'm assuming a steady 7% rate of return. Also, in scenario 2, after year 30, my SWR would drop well below 4% considering I'm going from 40k/year expenses to 30k/year expenses, so scenario 2 works out even better assuming I can get through the last 5 years.

So look: I'm not saying you are wrong about any math. I buy it, OK? What I'm saying is that your job, your employer, your personal history, investment options, mortgage terms, and even the design of the house one lives in, all dictate how you want to move money around among your various asset classes.

In my case, I've chosen to diversify by investing a little faster in my own home, given that, in my estimate, I'm over-leveraged in my mortgage and I don't want to sell this home to rent or buy another house. It has features in it that are very important and still somewhat unique that make 4% SWR much more feasible for me.

FIN

[TheNewNormal2015]

mefla - it might be best to just focus on winning in life rather than on this Internet forum. 

There do seem to be a few jerks, and I guess is it is human nature to state things confidently as black and white, when in reality there may be gray areas (I am guilty of this myself at times).  There are quite a few friendly + helpful ppl as well and they are really the ones who add value to the discussions.

I have gotten irritated at times myself, which has forced me to consider how much time I want to spend participating rather than just casually observing.  I do think it would ultimately be in everyone's interest to keep the tone friendly and welcoming, lest ppl who actually have something meaningful to contribute decide to disengage, but I leave that to others to police and manage, particularly the forum moderators and established participants.

Good luck in your endeavors

[Faraday]

mefla - it might be best to just focus on winning in life rather than on this Internet forum. 

There do seem to be a few jerks, and I guess is it is human nature to state things confidently as black and white, when in reality there may be gray areas (I am guilty of this myself at times).  There are quite a few friendly + helpful ppl as well and they are really the ones who add value to the discussions.

I have gotten irritated at times myself, which has forced me to consider how much time I want to spend participating rather than just casually observing.  I do think it would ultimately be in everyone's interest to keep the tone friendly and welcoming, lest ppl who actually have something meaningful to contribute decide to disengage, but I leave that to others to police and manage, particularly the forum moderators and established participants.

Good luck in your endeavors

I appreciate your comments. I got an answer that satisfies me from VikB in the other thread. I'm comfortable accepting your counsel and your considerations - I'm doing that now myself by winding down my postings.

What I didn't understand (and walked into blindly) is that there are people here getting their rocks off on making other people look stupid. To the point that they slice out small bits of comment, repeat what they said before, then make claims like "you don't understand the math".

I really wasn't prepared for that - I asked an honest question and the people who came at me the hardest have done their damndest not to answer it, preferring to extend the feeding frenzy on someone they want to think of a a weak-minded idiot.

The pathos of that overwhelms me, and yes, scares me into limiting my time on the forums.

But I'm happy, too. Two threads embraced the more complicated risk elements of the discussion, and one person actually gave me a real, honest-to-god answer, which I deeply appreciate. I've got something now I can act on, and that's a Good Thing.

As for arebelspy "enjoying the thread", more power to him. I sure as hell haven't. If he's supposed to be a moderator and trusted keeper of the forums, he damn well failed in that mission on this thread.

[Eric]

As for arebelspy "enjoying the thread", more power to him. I sure as hell haven't. If he's supposed to be a moderator and trusted keeper of the forums, he damn well failed in that mission on this thread.

LOL!  This from the person that just called everyone else assholes.

Most people take disagreement as just that -- disagreement.  You're the one the raised the ante with personal attacks.  So I can understand why you didn't enjoy the thread.  You're taking things that aren't personal and making them so.  The rest of us (arebelspy included I'm sure) enjoy the discussion even if there is disagreement.  That's how you learn.  Or at least that's how it's supposed to work unless you dig your heels in and consider any other points of view to be those of poop flinging monkeys.

[arebelspy]

I appreciate your comments. I got an answer that satisfies me from VikB in the other thread.

I took a look at that other thread.  No one ever answered that question because you never asked it.  If you want to talk about what to invest in, by all means, start a thread with your parameters.  Maybe I missed it.  Please quote where in this thread you asked what investments you should make.

Also VikB gave a quite shallow/quick answer (I don't disagree, but it's certainly not nuanced discussion - simply because no one knew what you were asking, IMO).  You should dig deeper.

In this thread, I see one comment where you ask if you're missing any pre-tax investments and state what you're doing post-tax:

And what-the-hell are these awesome pre-tax investments I'm missing out on besides:
- 401k (Buying Vanguard but it's Fidelity. It has a contribution cap and .15% fees)
- HSA, which I dived into when it became available to me 3 months ago which I think has zero yields.

Post-tax I'm using Betterment.

And one where you ask if you should do a case study:

I'm serious about this, I want to get it right. What can we do here - start another thread with me as a case study?

And you're promptly told yes.  Did you do that?

As for arebelspy "enjoying the thread", more power to him. I sure as hell haven't. If he's supposed to be a moderator and trusted keeper of the forums, he damn well failed in that mission on this thread.

I enjoy pretty much everything.  I'm a regular person, moderator doesn't mean I represent anything in any way, it means I delete spam and deal with trolls.

If you're having trouble with multiple people in multiple threads, maybe you should look at how you're approaching the issue.

Good luck buddy.  :)

[arebelspy]

Most people take disagreement as just that -- disagreement ... You're taking things that aren't personal and making them so.  The rest of us (arebelspy included I'm sure) enjoy the discussion even if there is disagreement.

This.  Many of us have learned things and changed our minds about things based on discussions here.  We try to have intelligent, respectful discourse.  Unfortunately it doesn't always work out that way.

But disagreement does not necessarily have to be antagonistic.

[brooklynguy]

I do think it would ultimately be in everyone's interest to keep the tone friendly and welcoming, lest ppl who actually have something meaningful to contribute decide to disengage

I wholeheartedly agree with this sentiment, and would just note that it's sometimes difficult to communicate tone through internet postings so we should all try not to jump to conclusions.  I can see how mefla might have felt that everyone was ganging up on him (particularly given how aggressive some of the posters were getting in expressing their viewpoints in the other thread), but it's a two-way street and the guy who starts tossing around the asshole and poop-throwing monkey comments should probably develop a thicker skin.  As one of the designated assholes myself, I won't speak on my own behalf, but I found it funny that the other folks who were responding to and trying to help answer mefla's questions are some of the friendliest, nicest and most even-tempered folks we have in this forum, including arebelspy and MDM.

[MDM]

Getting back to the question posed in the thread title, I ran some calculations on a (I believe) realistic scenario. I assumed a 30 year mortgage with a beginning balance of $160,000 and an interest rate of 4.25%. This gives a monthly mortgage payment of $787. Annual expenses including the mortgage are $40,000, and excluding the mortgage equates to $30,500. I'm also assuming a 7% CAGR for investments (which is what the 4% rule is based off of).

Let's say I have an additional $1,200/mo to either:
*pay down my mortgage early and then invest $1,987/mo after the mortgage is gone
*invest the $1,200/mo.

What I find is that in scenario 1, I can pay down the house in the 7th year, and have zero investments during that time period. Fast forward to year 25 and I finally have enough invested that I can live off 4%/year.

In scenario 2, I reach FI in year 26, one year after scenario 1 and can live off my investments and continue paying my mortgage.

HOWEVER, there's another option. What if I did scenario 2, and right before I FIREd, I paid off my mortgage in balance. This leaves me with the lower annual expenses of not having a mortgage while capitalizing on the higher returns of the market compared with the mortgage interest rate. I find that I can retire in year 23, a full 2 years earlier than scenario 1. I get the benefits of having liquid assets to live off of if I lose my job in the first decade over scenario 1 while putting that hard earned cash to work earlier. And this is assuming I never increase my savings over 30 years (and a pretty lame savings rate to boot).

I've attached my spreadsheet. Green cells mean I can FIRE on 4%. Did I miss anything crucial? Yes of course I'm assuming a steady 7% rate of return. Also, in scenario 2, after year 30, my SWR would drop well below 4% considering I'm going from 40k/year expenses to 30k/year expenses, so scenario 2 works out even better assuming I can get through the last 5 years.

Nicely done, but there are some items that seem pertinent.
1) After retirement, presumably the extra $1200/mo (or $1987/mo in scenario 1) is not available.
2) In scenario 2, the monthly payments have to be deducted from the investment balance for calculating subsequent growth.
3) In scenario 3, the lump sum has to be deducted from the investment balance for calculating subsequent growth.

When those items are included (see attached), the investment balance after 30 years is
Scenario 2: Invest all savings and pay minimum on mortgage = $1,417K
Scenario 3: Invest all, then pay lump sum at retirement   = $1,409K
Scenario 1: Pay Down Mortgage before investing = $1,266K

...and retirement occurs at exactly the same time in both scenario 2 & 3.

The math remains the math: investing at 7% is better than prepaying at 4.25%.  Of course one can debate the riskiness of one vs. the other, but if one chooses to assume 7% investment vs. 4.25% mortgage then the answer is to pay the mortgage minimum.

[skyrefuge]

When I run a cfiresim simulation, I want to find out the historical success rate of withdrawing $20k per year, and not the historical success rate of withdrawing an amount of money sufficient to purchase $20k worth of beef as of the date I run the simulation.

Is that really what you want, or is that just a reasonable compromise between what you want and what is possible?

Here is what *I* want cFIREsim to tell me, where, if "A" is not implementable, then I'll settle for the next one on the list, and so on:

A) A visitor from the future to tell me that my retirement worked perfectly, where I never wanted for money, and died with $1 to my name.
B) A time machine that sends versions of me into the past, and tells me that "1871 skyrefuge" through "1970 skyrefuge" all would have made it successfully and happily through their retirements.
C) A time machine that sends a Generic Money-Spender into the past, and tells me that "1871 GMS" through "1970 GMS" don't run out of money.

Note that each step down comes at the cost of decreased accuracy in the prediction of my actual life path. "A" is the ultimate, because it removes all concern about whether the future is likely to resemble the past. "B" doesn't tell me anything about the future, but it tells me how *I* would have done in the past, with my desires and behaviors and flexibility in the face of differing economic conditions. "C", believe it or not, is what cFIREsim actually does ("whoa....")

And yes, until we can see the future, it will remain the best-of-the-worst retirement predictors, but that doesn't mean clairvoyance is the only way its class-leading predictive ability could be improved.

Take the hypothetical example of someone whose entire living expenditures does and will always consist of nothing but purchases of large quantities of pocket calculators.  When this person uses cfiresim to plan his retirement, should cfiresim take into account the actual historical cost of pocket calculators (which, I'm assuming, was extraordinarily higher in today's-dollars in 1965 than today)?

Yes, this is exactly what a B-Level cFIREsim would and should do. If there was actually such a person whose only desire was a fixed quantity of pocket calculators per year, then B-Level cFIREsim would look into its database for the historical calculator prices and show that cycles starting in 1980 were far more successful than cycles starting in 1950, as calculator-acquisition costs required much smaller WRs in later years. For the years before the invention of the pocket calculator, it would "know" that it's really calculation in any form that's important to our subject (rather than battery-powered electronic devices), and would substitute the cost of slide-rules for the cost of calculators, just as it would also know your meat-eating preferences and substitute chicken for beef when beef prices got too high for you.

Of course B-Level cFIREsim would be impossible to implement for a general, real-world user. First it would have to incorporate historical prices of everything into its database, though that would actually be the "easy" part. The hard part would be to incorporate the user's desires and behaviors and substitution preferences, which would take like a 20,000-line form even if all of that stuff was knowable to the user, which it isn't. So that's why a fully-functional B-Level cFIREsim would require genuine mind-reading and/or time-travel technology.

But for the special case of a calculator nut? A one-off B-Level cFIREsim is totally possible, because the "user" in that case is extremely easy to model since we already know everything in their mind.

And so similarly, a one-off B-Level cFIREsim for a leveraged-investing-via-mortgage nut would also be possible. If I was such a nut and had the choice of using a B-Level cFIREsim or a C-Level cFIREsim, of course I would choose the B-Level, because it would provide a more accurate prediction of my actual life path than C-Level. It would still be a terrible prediction, since, unlike an A-Level cFIREsim, it can't actually know the future. But that doesn't mean I'd say "pfft, since it's not as good a A-Level, I'm just gonna stick with C-Level".

So I can imagine a C-plus-Level cFIREsim, that sits between C-Level and B-Level, where there would be optional fields for "percentage of expenses directed to mortgage payments" and "percentage of expenses directed to pocket calculators". In fact, the variable spending method options are already a significant step in the direction of C-plus-Level, as they allow the user to incorporate some of their personal behaviors into the simulation, making Generic Money-Spender slightly less-generic.

Of course I'm not actually calling for such features to be added to cFIREsim, since they're silly and would confuse the heck out of things, even if they improved it for a few specific users/uses. It's really just a mental exercise to explore whether using historical mortgage rates should be considered an improvement.

But once we get these mind-reading and backward-time-travel things worked out, bo_knows should totally expect to hear from me!

[TheNewNormal2015]

Getting back to the question posed in the thread title, I ran some calculations on a (I believe) realistic scenario. I assumed a 30 year mortgage with a beginning balance of $160,000 and an interest rate of 4.25%. This gives a monthly mortgage payment of $787. Annual expenses including the mortgage are $40,000, and excluding the mortgage equates to $30,500. I'm also assuming a 7% CAGR for investments (which is what the 4% rule is based off of).

Let's say I have an additional $1,200/mo to either:
*pay down my mortgage early and then invest $1,987/mo after the mortgage is gone
*invest the $1,200/mo.

What I find is that in scenario 1, I can pay down the house in the 7th year, and have zero investments during that time period. Fast forward to year 25 and I finally have enough invested that I can live off 4%/year.

In scenario 2, I reach FI in year 26, one year after scenario 1 and can live off my investments and continue paying my mortgage.

HOWEVER, there's another option. What if I did scenario 2, and right before I FIREd, I paid off my mortgage in balance. This leaves me with the lower annual expenses of not having a mortgage while capitalizing on the higher returns of the market compared with the mortgage interest rate. I find that I can retire in year 23, a full 2 years earlier than scenario 1. I get the benefits of having liquid assets to live off of if I lose my job in the first decade over scenario 1 while putting that hard earned cash to work earlier. And this is assuming I never increase my savings over 30 years (and a pretty lame savings rate to boot).

I've attached my spreadsheet. Green cells mean I can FIRE on 4%. Did I miss anything crucial? Yes of course I'm assuming a steady 7% rate of return. Also, in scenario 2, after year 30, my SWR would drop well below 4% considering I'm going from 40k/year expenses to 30k/year expenses, so scenario 2 works out even better assuming I can get through the last 5 years.

Nicely done, but there are some items that seem pertinent.
1) After retirement, presumably the extra $1200/mo (or $1987/mo in scenario 1) is not available.
2) In scenario 2, the monthly payments have to be deducted from the investment balance for calculating subsequent growth.
3) In scenario 3, the lump sum has to be deducted from the investment balance for calculating subsequent growth.

When those items are included (see attached), the investment balance after 30 years is
Scenario 2: Invest all savings and pay minimum on mortgage = $1,417K
Scenario 3: Invest all, then pay lump sum at retirement   = $1,409K
Scenario 1: Pay Down Mortgage before investing = $1,266K

...and retirement occurs at exactly the same time in both scenario 2 & 3.

The math remains the math: investing at 7% is better than prepaying at 4.25%.  Of course one can debate the riskiness of one vs. the other, but if one chooses to assume 7% investment vs. 4.25% mortgage then the answer is to pay the mortgage minimum.

So is it logical to assume anyone who is following this strategy of paying the minimum mortgage payment has 0% allocation to bonds and is 100% stocks?

Why invest in bonds when you can earn 100s of bps more after tax and risk free by paying down your own mortgage?  And if you want to effectively "rebalance" you can simply extract equity that was paid down by selling bond portion to 0%.

And let's assume that for discussion purposes the risk of HELOC shutdown is de minimis (in reality the probability of a shutdown occurring is more than compensated by the extra spread you get by effectively owning your own MBS) and should not be a factor.

[brooklynguy]

So is it logical to assume anyone who is following this strategy of paying the minimum mortgage payment has 0% allocation to bonds and is 100% stocks?

No, but it is arguably logical to assume that they should do so, assuming they are capable of completely excising all emotion from their investing decisions (which most people (or all people?) aren't).  Some of us (including me) routinely argue that there is no logical reason to hold bonds, period, if you can be absolutely certain that your time horizon is sufficiently long (30 years passes that test, IMO), except to the extent the bonds operate as self-restraint tool to prevent your own irrational, emotionally-driven self to panic-sell during a market downturn (there are lots of threads on that topic, like this one).  But it is especially illogical to do so while simultaneously carrying a mortgage (which effectively operates as a bond holding and can be used as a substitute for the desired bond portion of your allocation).

[arebelspy]

So is it logical to assume anyone who is following this strategy of paying the minimum mortgage payment has 0% allocation to bonds and is 100% stocks?

No, but it is arguably logical to assume that they should do so, assuming they are capable of completely excising all emotion from their investing decisions (which most people (or all people?) aren't).  Some of us (including me) routinely argue that there is no logical reason to hold bonds, period, if you can be absolutely certain that your time horizon is sufficiently long (30 years passes that test, IMO), except to the extent the bonds operate as self-restraint tool to prevent your own irrational, emotionally-driven self to panic-sell during a market downturn (there are lots of threads on that topic, like this one).  But it is especially illogical to do so while simultaneously carrying a mortgage (which effectively operates as a bond holding and can be used as a substitute for the desired bond portion of your allocation).

Except for the whole rebalancing thing.

[brooklynguy]

Except for the whole rebalancing thing.

You mean as far as a reason that it makes sense to hold bonds, or as far as why you can't equate carrying a mortgage to holding a bond?

[arebelspy]

Except for the whole rebalancing thing.

You mean as far as a reason that it makes sense to hold bonds, or as far as why you can't equate carrying a mortgage to holding a bond?

Yes.

I believe it's significantly harder to access equity in a house when the market crashes and your AA is unbalanced than it is to sell bonds when that happens.

I think thinking of your house as a bond (and paying down your mortgage as holding bonds) makes a lot of sense.  But it's not a perfect comparison due to that issue.

[MDM]

...
The math remains the math: investing at 7% is better than prepaying at 4.25%.  Of course one can debate the riskiness of one vs. the other, but if one chooses to assume 7% investment vs. 4.25% mortgage then the answer is to pay the mortgage minimum.

So is it logical to assume anyone who is following this strategy of paying the minimum mortgage payment has 0% allocation to bonds and is 100% stocks?

The spreadsheet analysis makes no assumption about asset allocation.  It assumes only that one receives a certain (in the example, 7%) investment return, in order to demonstrate numerically that investing at a higher return is better than paying a mortgage with a lower interest rate.

[rpr]

Getting back to the question posed in the thread title, I ran some calculations on a (I believe) realistic scenario. I assumed a 30 year mortgage with a beginning balance of $160,000 and an interest rate of 4.25%. This gives a monthly mortgage payment of $787. Annual expenses including the mortgage are $40,000, and excluding the mortgage equates to $30,500. I'm also assuming a 7% CAGR for investments (which is what the 4% rule is based off of).

Let's say I have an additional $1,200/mo to either:
*pay down my mortgage early and then invest $1,987/mo after the mortgage is gone
*invest the $1,200/mo.

What I find is that in scenario 1, I can pay down the house in the 7th year, and have zero investments during that time period. Fast forward to year 25 and I finally have enough invested that I can live off 4%/year.

In scenario 2, I reach FI in year 26, one year after scenario 1 and can live off my investments and continue paying my mortgage.

HOWEVER, there's another option. What if I did scenario 2, and right before I FIREd, I paid off my mortgage in balance. This leaves me with the lower annual expenses of not having a mortgage while capitalizing on the higher returns of the market compared with the mortgage interest rate. I find that I can retire in year 23, a full 2 years earlier than scenario 1. I get the benefits of having liquid assets to live off of if I lose my job in the first decade over scenario 1 while putting that hard earned cash to work earlier. And this is assuming I never increase my savings over 30 years (and a pretty lame savings rate to boot).

I've attached my spreadsheet. Green cells mean I can FIRE on 4%. Did I miss anything crucial? Yes of course I'm assuming a steady 7% rate of return. Also, in scenario 2, after year 30, my SWR would drop well below 4% considering I'm going from 40k/year expenses to 30k/year expenses, so scenario 2 works out even better assuming I can get through the last 5 years.

Nicely done, but there are some items that seem pertinent.
1) After retirement, presumably the extra $1200/mo (or $1987/mo in scenario 1) is not available.
2) In scenario 2, the monthly payments have to be deducted from the investment balance for calculating subsequent growth.
3) In scenario 3, the lump sum has to be deducted from the investment balance for calculating subsequent growth.

When those items are included (see attached), the investment balance after 30 years is
Scenario 2: Invest all savings and pay minimum on mortgage = $1,417K
Scenario 3: Invest all, then pay lump sum at retirement   = $1,409K
Scenario 1: Pay Down Mortgage before investing = $1,266K

...and retirement occurs at exactly the same time in both scenario 2 & 3.

The math remains the math: investing at 7% is better than prepaying at 4.25%.  Of course one can debate the riskiness of one vs. the other, but if one chooses to assume 7% investment vs. 4.25% mortgage then the answer is to pay the mortgage minimum.

MDM and Runge -- Thank you.

[Runge]

@rpr, you're welcome!

Getting back to the question posed in the thread title, I ran some calculations on a (I believe) realistic scenario. I assumed a 30 year mortgage with a beginning balance of $160,000 and an interest rate of 4.25%. This gives a monthly mortgage payment of $787. Annual expenses including the mortgage are $40,000, and excluding the mortgage equates to $30,500. I'm also assuming a 7% CAGR for investments (which is what the 4% rule is based off of).

Let's say I have an additional $1,200/mo to either:
*pay down my mortgage early and then invest $1,987/mo after the mortgage is gone
*invest the $1,200/mo.

What I find is that in scenario 1, I can pay down the house in the 7th year, and have zero investments during that time period. Fast forward to year 25 and I finally have enough invested that I can live off 4%/year.

In scenario 2, I reach FI in year 26, one year after scenario 1 and can live off my investments and continue paying my mortgage.

HOWEVER, there's another option. What if I did scenario 2, and right before I FIREd, I paid off my mortgage in balance. This leaves me with the lower annual expenses of not having a mortgage while capitalizing on the higher returns of the market compared with the mortgage interest rate. I find that I can retire in year 23, a full 2 years earlier than scenario 1. I get the benefits of having liquid assets to live off of if I lose my job in the first decade over scenario 1 while putting that hard earned cash to work earlier. And this is assuming I never increase my savings over 30 years (and a pretty lame savings rate to boot).

I've attached my spreadsheet. Green cells mean I can FIRE on 4%. Did I miss anything crucial? Yes of course I'm assuming a steady 7% rate of return. Also, in scenario 2, after year 30, my SWR would drop well below 4% considering I'm going from 40k/year expenses to 30k/year expenses, so scenario 2 works out even better assuming I can get through the last 5 years.

Nicely done, but there are some items that seem pertinent.
1) After retirement, presumably the extra $1200/mo (or $1987/mo in scenario 1) is not available.
2) In scenario 2, the monthly payments have to be deducted from the investment balance for calculating subsequent growth.
3) In scenario 3, the lump sum has to be deducted from the investment balance for calculating subsequent growth.

When those items are included (see attached), the investment balance after 30 years is
Scenario 2: Invest all savings and pay minimum on mortgage = $1,417K
Scenario 3: Invest all, then pay lump sum at retirement   = $1,409K
Scenario 1: Pay Down Mortgage before investing = $1,266K

...and retirement occurs at exactly the same time in both scenario 2 & 3.

The math remains the math: investing at 7% is better than prepaying at 4.25%.  Of course one can debate the riskiness of one vs. the other, but if one chooses to assume 7% investment vs. 4.25% mortgage then the answer is to pay the mortgage minimum.

Thanks for the feedback. I don't agree with the change to scenario 2 in that you reach FIRE at year 23 in your attached spreadsheet. You're assuming that the annual expenses are ~30k, when you still need to be paying off your mortgage. This would require a higher portfolio value to make sure there is enough money to cover the mortgage payment. Otherwise you'd be pulling out 40k from a $760k balance which equates to a 5.26% withdrawal rate at the very beginning of FIRE. That doesn't sound like a recipe for success to me, cFIREsim calc notwithstanding. Your change in reflecting the drawdown of investments after FIRE looks correct to me, it's just starting too early.

As for my calculations not accurately calculating the post fire number correctly, I was lazy and didn't implement that. I just said, hey cool, green cells mean I'm eligible to FIRE. :D Thank's for taking the time to update it to reflect the FIRE choice.

...
The math remains the math: investing at 7% is better than prepaying at 4.25%.  Of course one can debate the riskiness of one vs. the other, but if one chooses to assume 7% investment vs. 4.25% mortgage then the answer is to pay the mortgage minimum.

So is it logical to assume anyone who is following this strategy of paying the minimum mortgage payment has 0% allocation to bonds and is 100% stocks?

The spreadsheet analysis makes no assumption about asset allocation.  It assumes only that one receives a certain (in the example, 7%) investment return, in order to demonstrate numerically that investing at a higher return is better than paying a mortgage with a lower interest rate.

This is correct. Ideally you'd want to somehow figure out a way to marry this with the discussion that brooklynguy and skyrefuge are having about cFIREsim. Only then can you include asset allocation in the calculations, and you'd also get an estimated success rate for the plan as a whole. That would be really neat to see.

[brooklynguy]

Here is what *I* want cFIREsim to tell me, where, if "A" is not implementable, then I'll settle for the next one on the list, and so on:

A) A visitor from the future to tell me that my retirement worked perfectly, where I never wanted for money, and died with $1 to my name.
B) A time machine that sends versions of me into the past, and tells me that "1871 skyrefuge" through "1970 skyrefuge" all would have made it successfully and happily through their retirements.
C) A time machine that sends a Generic Money-Spender into the past, and tells me that "1871 GMS" through "1970 GMS" don't run out of money.

Note that each step down comes at the cost of decreased accuracy in the prediction of my actual life path. "A" is the ultimate, because it removes all concern about whether the future is likely to resemble the past. "B" doesn't tell me anything about the future, but it tells me how *I* would have done in the past, with my desires and behaviors and flexibility in the face of differing economic conditions. "C", believe it or not, is what cFIREsim actually does ("whoa....")

I definitely would prefer version "A", but I'm not convinced that version "B" is better.  I think the version "C" that we actually have may be the more accurate predictor of future success, because it isn't biased by outdated information.  The success rate that B-Level cFIREsim spits out for our hypothetical pocket calculator nut will underestimate his actual chances of success, because it thinks pocket calculators are more expensive than they really are (or, more accurately, its success rate was determined on the basis of many cycles where pocket calculators were more expensive than they are today and will be in the future).

And so similarly, a one-off B-Level cFIREsim for a leveraged-investing-via-mortgage nut would also be possible. If I was such a nut and had the choice of using a B-Level cFIREsim or a C-Level cFIREsim, of course I would choose the B-Level, because it would provide a more accurate prediction of my actual life path than C-Level. It would still be a terrible prediction, since, unlike an A-Level cFIREsim, it can't actually know the future. But that doesn't mean I'd say "pfft, since it's not as good a A-Level, I'm just gonna stick with C-Level".

So I can imagine a C-plus-Level cFIREsim, that sits between C-Level and B-Level, where there would be optional fields for "percentage of expenses directed to mortgage payments" and "percentage of expenses directed to pocket calculators". In fact, the variable spending method options are already a significant step in the direction of C-plus-Level, as they allow the user to incorporate some of their personal behaviors into the simulation, making Generic Money-Spender slightly less-generic.

Ok, to take this back to where we started, I think we can agree that, assuming the leveraged-investing-via-mortgage version of C-plus-(or B-minus?)-Level cFIREsim does do a better job of predicting the success of leveraged-investing-via-mortgage, then that version of cFIREsim would also do a better job of predicting the success of any spending plan.  When you ask cFIREsim to tell you the likelihood that your investments will outperform your mortgage, you are really just asking it to tell you the chances of portfolio success for a specific spending plan that happens to match the amortization schedule of a certain mortgage that could be obtained today.  It doesn't matter if the dollars withdrawn from your portfolio in connection with that spending plan are being used to make mortgage payments, or to purchase beef, or to be doused in kerosene and set on fire.  So, if cFIREsim were tweaked to incorporate prevailing-mortgage-rate-information into the historical dataset it uses to calculate success rates, and that tweak improved its predictive ability regarding leveraged-investing-via-mortgage, then that tweak would also improve its predictive ability regarding sustainable withdrawal rates in general.  Which takes us back to my original point, that your hypothesis really has nothing to do with the mortgage issue, and has broader application to WR success rates generally.  In other words, if it's too optimistic to look at all historical periods to determine the success rate of a spending plan based on today's low mortgage rates (instead of looking only at those historical periods with similarly low-interest-rate-environments at the start date), then the same is true for any spending plan.

[MDM]

I don't agree with the change to scenario 2 in that you reach FIRE at year 23 in your attached spreadsheet. You're assuming that the annual expenses are ~30k, when you still need to be paying off your mortgage. This would require a higher portfolio value to make sure there is enough money to cover the mortgage payment. Otherwise you'd be pulling out 40k from a $760k balance which equates to a 5.26% withdrawal rate at the very beginning of FIRE. That doesn't sound like a recipe for success to me, cFIREsim calc notwithstanding. Your change in reflecting the drawdown of investments after FIRE looks correct to me, it's just starting too early.

Scenarios 2 & 3 are identical through year 23, correct?  Then, to pay the mortgage, you either withdraw a large amount at once (scenario 3) or smaller amounts every month (scenario 2).

In scenario 3 you have a withdrawal rate of (30000 + 57000) / 820000 = 10.6% in your first year due to the large lump sum payment.  But that's ok because it lasts only 1 year and then you drop to something <4% for years after that. 

Similarly, scenario 2 has the higher withdrawal rate you note, but only for seven years.  That's ok because after the mortgage is paid you drop to something <4% for years after that.

By not withdrawing the lump sum coincident with retirement, you take advantage of earning 7% on that amount while paying only 4.25% for the privilege.  So you really can retire at the same time either way - maybe even a little earlier if you don't pay the lump sum, but close enough.

[Runge]

Sweet swirling onion rings! You're right! It doesn't matter if you drag out your mortgage payment in scenario 2 or lump sum pay it in scenario 3 because the underlying assets for FIRE assuming a paid off mortgage are there already. I had to make some graphs to visualize it, but it'll roughly look like the attached pictures.

By not withdrawing the lump sum coincident with retirement, you take advantage of earning 7% on that amount while paying only 4.25% for the privilege.  So you really can retire at the same time either way - maybe even a little earlier if you don't pay the lump sum, but close enough.

Right, and the sooner one reaches fire, the more dramatic keeping that mortgage money in investments grows compared with paying off the mortgage with a lump sum at retirement. Doh! The math really is that clear.

[MDM]

Nice charts!

The discussion in http://forum.mrmoneymustache.com/welcome-to-the-forum/trying-to-get-a-better-understanding-of-that-%2725-times-annual-spending%27-rule/, particularly kaizen soze's post #10, helped clarify some things in my head.

[arebelspy]

Right, and the sooner one reaches fire, the more dramatic keeping that mortgage money in investments grows compared with paying off the mortgage with a lump sum at retirement. Doh! The math really is that clear.

Bingo!  :D   Do we have another convert to "keep the mortgage"?  ;)

It really does take some digging to convince onesself (as it should - question everything), but once you see it, you can't unsee it.

Also, I loved this:

Sweet swirling onion rings!

[TheNewNormal2015]

Right, and the sooner one reaches fire, the more dramatic keeping that mortgage money in investments grows compared with paying off the mortgage with a lump sum at retirement. Doh! The math really is that clear.

Bingo!  :D   Do we have another convert to "keep the mortgage"?  ;)

It really does take some digging to convince onesself (as it should - question everything), but once you see it, you can't unsee it.

Also, I loved this:

Sweet swirling onion rings!

I wish I could share some ppl's enthusiasm for a strategy that relies on assuming historical equity returns (despite the fact that there has been a massive bull move and with stretched valuations as a starting point) combined with once-in-a-lifetime low mortgage rates.

Only time will tell whether implementing the strategy now would work sufficiently well to reward the investor for the risk taken 30yrs from now.

But I think it would be more interesting and useful to see how it actually fared historically

And based on a few data points I looked at spending just a few minutes searching Google, the strategy hasn't worked that well historically (including the last 30yrs, despite the 1985 starting point which would have seemed like a no-brainer winner).  I haven't run all the numbers but just eyeballing it doesn't look like it would have paid to take what seems like a meaningful amount of risk

See for yourselves:
http://www.freddiemac.com/pmms/pmms30.htm

http://dqydj.net/sp-500-return-calculator/

[rpr]

Right, and the sooner one reaches fire, the more dramatic keeping that mortgage money in investments grows compared with paying off the mortgage with a lump sum at retirement. Doh! The math really is that clear.

Bingo!  :D   Do we have another convert to "keep the mortgage"?  ;)

It really does take some digging to convince onesself (as it should - question everything), but once you see it, you can't unsee it.

Also, I loved this:

Sweet swirling onion rings!

I wish I could share some ppl's enthusiasm for a strategy that relies on assuming historical equity returns (despite the fact that there has been a massive bull move and with stretched valuations as a starting point) combined with once-in-a-lifetime low mortgage rates.

Only time will tell whether implementing the strategy now would work sufficiently well to reward the investor for the risk taken 30yrs from now.

But I think it would be more interesting and useful to see how it actually fared historically

And based on a few data points I looked at spending just a few minutes searching Google, the strategy hasn't worked that well historically (including the last 30yrs, despite the 1985 starting point which would have seemed like a no-brainer winner).  I haven't run all the numbers but just eyeballing it doesn't look like it would have paid to take what seems like a meaningful amount of risk

See for yourselves:
http://www.freddiemac.com/pmms/pmms30.htm

http://dqydj.net/sp-500-return-calculator/

In 1985 the 30 year FRM was 12% while the annualized S&P500 TR CAGR was 11%. You would have been right in 1985 not to mortgage instead of investing.  We were coming off historical highs in 30 year rates. Big difference between that and the current 4%. All we are hoping for is that the next 30 years, we will get >4%.

[TheNewNormal2015]

Right, and the sooner one reaches fire, the more dramatic keeping that mortgage money in investments grows compared with paying off the mortgage with a lump sum at retirement. Doh! The math really is that clear.

Bingo!  :D   Do we have another convert to "keep the mortgage"?  ;)

It really does take some digging to convince onesself (as it should - question everything), but once you see it, you can't unsee it.

Also, I loved this:

Sweet swirling onion rings!

I wish I could share some ppl's enthusiasm for a strategy that relies on assuming historical equity returns (despite the fact that there has been a massive bull move and with stretched valuations as a starting point) combined with once-in-a-lifetime low mortgage rates.

Only time will tell whether implementing the strategy now would work sufficiently well to reward the investor for the risk taken 30yrs from now.

But I think it would be more interesting and useful to see how it actually fared historically

And based on a few data points I looked at spending just a few minutes searching Google, the strategy hasn't worked that well historically (including the last 30yrs, despite the 1985 starting point which would have seemed like a no-brainer winner).  I haven't run all the numbers but just eyeballing it doesn't look like it would have paid to take what seems like a meaningful amount of risk

See for yourselves:
http://www.freddiemac.com/pmms/pmms30.htm

http://dqydj.net/sp-500-return-calculator/

In 1985 the 30 year FRM was 12% while the annualized S&P500 TR CAGR was 11%. You would have been right in 1985 not to mortgage instead of investing.  We were coming off historical highs in 30 year rates. Big difference between that and the current 4%. All we are hoping for is that the next 30 years, we will get >4%.

The high in rates was actually 1980 (it didn't work starting then for 30yrs either).

Look at the data and come to your own conclusions

It seems a bit illogical to use historical data to assume future equity returns to calculate perceived safe withdrawal rates and other investment strategies but not use the data staring you in the face for mortgage vs equity returns.

[TheNewNormal2015]

Right, and the sooner one reaches fire, the more dramatic keeping that mortgage money in investments grows compared with paying off the mortgage with a lump sum at retirement. Doh! The math really is that clear.

Bingo!  :D   Do we have another convert to "keep the mortgage"?  ;)

It really does take some digging to convince onesself (as it should - question everything), but once you see it, you can't unsee it.

Also, I loved this:

Sweet swirling onion rings!

I wish I could share some ppl's enthusiasm for a strategy that relies on assuming historical equity returns (despite the fact that there has been a massive bull move and with stretched valuations as a starting point) combined with once-in-a-lifetime low mortgage rates.

Only time will tell whether implementing the strategy now would work sufficiently well to reward the investor for the risk taken 30yrs from now.

But I think it would be more interesting and useful to see how it actually fared historically

And based on a few data points I looked at spending just a few minutes searching Google, the strategy hasn't worked that well historically (including the last 30yrs, despite the 1985 starting point which would have seemed like a no-brainer winner).  I haven't run all the numbers but just eyeballing it doesn't look like it would have paid to take what seems like a meaningful amount of risk

See for yourselves:
http://www.freddiemac.com/pmms/pmms30.htm

http://dqydj.net/sp-500-return-calculator/

In 1985 the 30 year FRM was 12% while the annualized S&P500 TR CAGR was 11%. You would have been right in 1985 not to mortgage instead of investing.  We were coming off historical highs in 30 year rates. Big difference between that and the current 4%. All we are hoping for is that the next 30 years, we will get >4%.

Low real rates = low equity returns

http://www.economist.com/blogs/buttonwood/2013/02/investing

[TheNewNormal2015]

Right, and the sooner one reaches fire, the more dramatic keeping that mortgage money in investments grows compared with paying off the mortgage with a lump sum at retirement. Doh! The math really is that clear.

Bingo!  :D   Do we have another convert to "keep the mortgage"?  ;)

It really does take some digging to convince onesself (as it should - question everything), but once you see it, you can't unsee it.

Also, I loved this:

Sweet swirling onion rings!

I wish I could share some ppl's enthusiasm for a strategy that relies on assuming historical equity returns (despite the fact that there has been a massive bull move and with stretched valuations as a starting point) combined with once-in-a-lifetime low mortgage rates.

Only time will tell whether implementing the strategy now would work sufficiently well to reward the investor for the risk taken 30yrs from now.

But I think it would be more interesting and useful to see how it actually fared historically

And based on a few data points I looked at spending just a few minutes searching Google, the strategy hasn't worked that well historically (including the last 30yrs, despite the 1985 starting point which would have seemed like a no-brainer winner).  I haven't run all the numbers but just eyeballing it doesn't look like it would have paid to take what seems like a meaningful amount of risk

See for yourselves:
http://www.freddiemac.com/pmms/pmms30.htm

http://dqydj.net/sp-500-return-calculator/

In 1985 the 30 year FRM was 12% while the annualized S&P500 TR CAGR was 11%. You would have been right in 1985 not to mortgage instead of investing.  We were coming off historical highs in 30 year rates. Big difference between that and the current 4%. All we are hoping for is that the next 30 years, we will get >4%.

Low real rates = low equity returns

http://www.economist.com/blogs/buttonwood/2013/02/investing

With data going back to 1926:
http://www.economist.com/news/finance-and-economics/21564845-low-real-interest-rates-are-usually-bad-news-equity-markets

[Cressida]

Do we have another convert to "keep the mortgage"?  ;)

Probably. After all the talk on this topic over the past few days, I have certainly revised my opinion. I'm not going to prepay my mortgage as it stands now (we recently refinanced to a 15-year and put some money in to get the loan below jumbo level; I'm not sorry for that part, although I'm sure some will disagree). Instead, I've added the mortgage amount to my stache threshold for FI. I haven't decided if I would pay off the mortgage at that point - so I guess you can color me not ENTIRELY converted), but I'm certainly more comfortable with the idea of investing the cash rather than paying down the mortgage.

[rpr]

Low real rates = low equity returns

http://www.economist.com/blogs/buttonwood/2013/02/investing

This only shows correlation between rates and subsequent 5 year equity returns. Is there a table showing 30 year equity returns?

With data going back to 1926:
http://www.economist.com/news/finance-and-economics/21564845-low-real-interest-rates-are-usually-bad-news-equity-markets

Thanks, but this requires a registration/subscription which I don't have.

[rpr]

Do we have another convert to "keep the mortgage"?  ;)

Probably. After all the talk on this topic over the past few days, I have certainly revised my opinion. I'm not going to prepay my mortgage as it stands now (we recently refinanced to a 15-year and put some money in to get the loan below jumbo level; I'm not sorry for that part, although I'm sure some will disagree). Instead, I've added the mortgage amount to my stache threshold for FI. I haven't decided if I would pay off the mortgage at that point - so I guess you can color me not ENTIRELY converted), but I'm certainly more comfortable with the idea of investing the cash rather than paying down the mortgage.

cressida -- The 15 year is a great way to go especially, if you got a sub 3% rate. But don't prepay anymore. For those that commit to a shorter term, they benefit from the lower interest rate.

http://thefinancebuff.com/borrow-30-year-and-invest-the-difference.html

From the linked article:

Note this is not a typical “should I prepay my mortgage or invest” situation. Merely prepaying your mortgage does not lower the rate on the entire outstanding balance. You get the lower rate on the whole balance only when you make a firm commitment to make a higher monthly payment each and every month. It’s a gift to the committed.

[TheNewNormal2015]

Low real rates = low equity returns

http://www.economist.com/blogs/buttonwood/2013/02/investing

This only shows correlation between rates and subsequent 5 year equity returns. Is there a table showing 30 year equity returns?

With data going back to 1926:
http://www.economist.com/news/finance-and-economics/21564845-low-real-interest-rates-are-usually-bad-news-equity-markets

Thanks, but this requires a registration/subscription which I don't have.

That's unfortunate

I don't feel comfortable copy/pasting the data if they want to keep it to subscribers only

I guess it is up to each individual to decide how he wants to interpret the data provided.  I have a feeling many will just want to continue to believe what they want and keep their head in the sand.

I wonder whether that type of cognitive bias would be categorized under "Ostrich Effect"

[rpr]

TheNewNormal2015 -- there are many roads to Dublin as they say. Pick whichever works for you.

I was able to read that article. Thanks for the link. Unfortunately, that article does not say very much about long term returns of the market following a period of low interest rates. All it says, is that there is a strong relationship between the interest rate and the return from securities for that year.

There are a couple of possibilities.

I. Interest rates remain depressed for quite a while, say the next thirty years. Could happen, e.g. Japan. In this case stock returns would also be muted over this entire duration. In such a case everything is bad, stocks, bonds, etc.

II. Loose monetary policies eventually lead to higher inflation. And in periods of higher inflation, stocks have a lower real return. But I would assume that interest rates would rise and people with investable assets could presumably get CDs which would definitely pay more than the lower interest rate that are locked with low mortgages from now. 





Sent from my iPad using Tapatalk

[MDM]

I guess it is up to each individual to decide how he wants to interpret the data provided.  I have a feeling many will just want to continue to believe what they want and keep their head in the sand.

I wonder whether that type of cognitive bias would be categorized under "Ostrich Effect"

Data from http://people.stern.nyu.edu/adamodar/New_Home_Page/datafile/histretSP.html.  X-axis is the beginning year of the 30 year period.


From http://www.dailywealth.com/2248/housing-bust-is-over.  This was the only chart I found (after an admittedly short search) with data pre-1971.  Note that it ends ~2012, and rates have continued low (or lower) since then.


I leave it to someone else to put everything on one graph if interested. 

[brooklynguy]

It seems a bit illogical to use historical data to assume future equity returns to calculate perceived safe withdrawal rates and other investment strategies but not use the data staring you in the face for mortgage vs equity returns.

In my view, it's exactly the opposite.  It seems illogical to use historical data to assume future equity returns for the purpose of determining perceived safe withdrawal rates but not use the same historical data to assume future equity returns for the purpose of determining the advisability of leveraged-investing-via-a-mortgage-having-today's-low-rate.  This is exactly the debate skyrefuge and I have been going back and forth about.  I completely agree that there may be reasons to think that current indicators could possibly be signs that future returns will be lower than historical returns (even though I'm not sure if that's actually true and don't share the level of pessimism some people, like Pfau, have on that question), but once you make an assumption about the sub-performance of future returns, that affects your perceived SWR just the same as your perceived advisability of leveraged-investing-via-mortgage.  If the next 30 years don't produce returns high enough to outperform today's low rate mortgages, then anyone retiring today on anything higher than an extremely-low WR is in a boatload of trouble.

[boarder42]

i was in the wrong post. I like this post much better gonna follow here ... though i'm already a convert by basically digging up all this info my self.   

[TheNewNormal2015]

TheNewNormal2015 -- there are many roads to Dublin as they say. Pick whichever works for you.

I was able to read that article. Thanks for the link. Unfortunately, that article does not say very much about long term returns of the market following a period of low interest rates. All it says, is that there is a strong relationship between the interest rate and the return from securities for that year.

There are a couple of possibilities.

I. Interest rates remain depressed for quite a while, say the next thirty years. Could happen, e.g. Japan. In this case stock returns would also be muted over this entire duration. In such a case everything is bad, stocks, bonds, etc.

II. Loose monetary policies eventually lead to higher inflation. And in periods of higher inflation, stocks have a lower real return. But I would assume that interest rates would rise and people with investable assets could presumably get CDs which would definitely pay more than the lower interest rate that are locked with low mortgages from now. 





Sent from my iPad using Tapatalk

Scenario II seems to require market timing and dramatic asset allocation shifting to work.  How does one determine when to move to CDs?

[Retire-Canada]

When I first started out I wanted a 30yr mortgage because the payments were low.

As I became for money conscious I was horrified by how much interest I was paying and I wanted to hammer that down as fast as possible. Because debt and interest was bad.

Now that I am deep into FIRE planning I'm back to wanting a long mortgage because I realize I can probably make more money investing my savings than to pay down my mortgage. And that cash flow is more important than killing debt if the interest on the debt is low.


My situation:

- I live in a high COL area and just bought a house so I would need 10yrs of focused effort to pay my mortgage off.

- I want to start downshifting in 2016 and be working less than half-time by 2020 so paying off the mortgage and saving for FI are not compatible.

- I'm in Canada with a prime - 0.75% variable $350K mortgage starting a new 5yr term. I can lock into a 5yr fixed rate at anytime I want, but until then I am not protected against rate increases.

- I can pay down $75k/yr extra without penalty and in 2020 I can pay the balance from my investments if I want to.

- If I killed my mortgage my COL drops by ~25% and my FI target drops by $250K.

-- Vik

[tomsang]

When everyone is digging up market returns you need to be grabbing nominal market returns with dividends reinvested as a 30 year fixed rate mortgage is a perfect hedge against inflation and dividends are part of the return.  Typically the times that the market did not perform well were the times that the US had high inflation. If you keep your mortgage you actually are better off with high inflation as your payment is locked in. Cfiresim does this already, but it appears that others are doing it themselves which I think is great. The more people understand how their investments fair with inflation, how keeping low cost fixed debt helps their success, how equity appreciation vs. dividend return doesn't matter, how to structure their portfolio for longterm success, and other knowledge about their portfolio management the better off they will be.

[TheNewNormal2015]

It seems a bit illogical to use historical data to assume future equity returns to calculate perceived safe withdrawal rates and other investment strategies but not use the data staring you in the face for mortgage vs equity returns.

In my view, it's exactly the opposite.  It seems illogical to use historical data to assume future equity returns for the purpose of determining perceived safe withdrawal rates but not use the same historical data to assume future equity returns for the purpose of determining the advisability of leveraged-investing-via-a-mortgage-having-today's-low-rate.  This is exactly the debate skyrefuge and I have been going back and forth about.  I completely agree that there may be reasons to think that current indicators could possibly be signs that future returns will be lower than historical returns (even though I'm not sure if that's actually true and don't share the level of pessimism some people, like Pfau, have on that question), but once you make an assumption about the sub-performance of future returns, that affects your perceived SWR just the same as your perceived advisability of leveraged-investing-via-mortgage.  If the next 30 years don't produce returns high enough to outperform today's low rate mortgages, then anyone retiring today on anything higher than an extremely-low WR is in a boatload of trouble.

The discussion in this thread has mainly revolved around whether mortgage rates beat equity returns for long holding periods, not what a perceived SWR actually should be.

It seems that proponents of using a mortgage as leverage are cherry picking today's record low interest rate without considering whether the interest rate strongly influences or determines future equity returns (which they seem to do, both theoretically - discounting cash flows? - as well as in practice).

If 4% withdrawal appears to have been safe historically, it seems like it would make sense to look at what the range and average historical yields and valuations of assets were and compare vs where we are today to give you a sense of where we lie on the spectrum

Your last sentence is precisely what the investing greats and legends of finance are saying today: expect much lower asset returns going forward.  For purposes of this discussion that leads me to be cautious in assuming the unknown part of the equation (future equity returns)

[tomsang]

- If I killed my mortgage my COL drops by ~25% and my FI target drops by $250K.

-- Vik

You are Canadian so you are different....  Well at least your access to a 30 year fixed rate mortgage is non existent, but your comment is consistent with a lot of people. The problems with this comment is it fails to take into acount the asset side of the equation. Your FI does not drop by $250k. As you are building up your asset you are getting closer to FI. If your assets are returning more than your mortgage interest rate then you are getting to the finishline quicker by keeping your mortgage.

The second fallacy is that your mortgage payment includes principal payments. This should not be listed as an expense. You are moving money from one pocket to the other. Ie taking money out of investments to pay down a debt. The downside is the money that could have been invested could earn more than the mortgage payment. On a longterm scale this has happened every time over the past 144 years on a 30 year time frame. With your short timeframe is gets more complicated, but it is still something to consider.

The third area is liquidity, which if it all hits the fan having investments vs a partially paid off house provides a provides the funds to live off of while the world is coming back online.

[TheNewNormal2015]

When everyone is digging up market returns you need to be grabbing nominal market returns with dividends reinvested as a 30 year fixed rate mortgage is a perfect hedge against inflation and dividends are part of the return. 

The link I provided above does calculate total returns on the S&P with dividends reinvested.

Typically the times that the market did not perform well were the times that the US had high inflation.

I am not sure this statement is correct for pre-war data, and is mixed at best for post-war.  The two largest selloffs in market history from peak to trough were during periods of low or negative inflation.