Author Topic: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff  (Read 22002 times)

JGB

  • Stubble
  • **
  • Posts: 108
  • Age: 36
  • Location: Oklahoma
    • Grinderschool Poker Training
Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« on: July 28, 2014, 12:23:42 PM »
In the excellent thread Cheddar Stacker started to delve into the question of Investing vs Paying Off Debt, someone brought up the point of higher expenses in retirement resulting from holding a mortgage. My initial response was the simple "yes, but you have more assets and since you're earning more than the interest on the loan, you're better off." But then I decided to run the numbers, and I am intrigued...

Here's my situation:
I ran the numbers based on my family's current spending of $34k, which requires $850k to have a 4% SWR. My mortgage balance is $186k at 3.5%. I've done a number of long-term calculations which all say that we're ahead by over $50k (often the difference rises well over $100k) in the long run, if we keep the mortgage and invest the difference instead of paying it off.

Looking at the snapshot in time of when we will hit the $850k mark and could theoretically FIRE under the current timeline, we're about 3 years out based on our current savings rate and expenses. If, at that exact moment, we paid off the mortgage out of our $850k, we would be left with ~$680k (the mortgage will shrink a bit over the next 3 years as we make scheduled payments). The 4% number drops to $27k at this point, however our expenses (for principal and interest only -- tax will still need to be paid) drop by $10k. That means should only need $24k at this point. Suddenly we're at a withdrawal rate of 3.5% instead of 4%...

This seems to indicate that paying off a mortgage when pulling the trigger to FIRE may provide significant improvements to your success rate. Thoughts? Am I missing something? Are there arguments in favor of keeping the mortgage anyway?

« Last Edit: July 29, 2014, 08:43:06 AM by JGB »

Timmmy

  • Bristles
  • ***
  • Posts: 446
  • Age: 36
  • Location: Madison Heights, Michigan
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #1 on: July 28, 2014, 12:31:05 PM »
Isn't this based primarily on the difference between mortgage rate and SWR?

JGB

  • Stubble
  • **
  • Posts: 108
  • Age: 36
  • Location: Oklahoma
    • Grinderschool Poker Training
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #2 on: July 28, 2014, 12:37:01 PM »
Isn't this based primarily on the difference between mortgage rate and SWR?

I don't think so. My mortgage rate is already lower than the 4% SWR that I started with, so if that were the case, wouldn't the lower rate mean that I'd be better off keeping the mortgage?

Scandium

  • Handlebar Stache
  • *****
  • Posts: 2198
  • Location: EastCoast
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #3 on: July 28, 2014, 12:48:50 PM »
Another consideration is paying down the mortgage while working vs paying it off all at once before/once you FIRE, like you suggest. Now while we're working and taxed at 25%, the mortgage interest provides a nice tax deduction and give us more money to invest. Which presumably will earn more than the 4% rate (or even lower once the tax deduction is considered).

Once we FIRE however, if we have little or no regular income would the tax deduction not be less useful or even worthless? Maybe paying it off once in retirement is more valid, or at least not as sub-optimal as doing it while working?

But in your example you are still using $170K that could grow at 5%+ to pay off a 3.5% loan..

PeteD01

  • Bristles
  • ***
  • Posts: 413
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #4 on: July 28, 2014, 12:53:06 PM »
In the excellent thread Cheddar Stacker started to delve into the question of Investing vs Paying Off Debt, someone brought up the point of higher expenses in retirement resulting from holding a mortgage. My initial response was the simple "yes, but you have more assets and since you're earning more than the interest on the loan, you're better off." But then I decided to run the numbers, and I am intrigued...

Here's my situation:
I ran the numbers based on my family's current spending of $34k, which requires $850k to have a 4% SWR. My mortgage balance is $186k at 3.5%. I've done a number of long-term calculations which all say that we're ahead by over $50k (often the difference rises well over $100k) in the long run, if we keep the mortgage and invest the difference instead of paying it off.

Looking at the snapshot in time of when we will hit the $850k mark and could theoretically FIRE under the current timeline, we're about 3 years out based on our current savings rate and expenses. If, at that exact moment, we paid off the mortgage out of our $850k, we would be left with ~$680k (the mortgage will shrink a bit over the next 3 years as we make scheduled payments). The 4% number drops to $27k at this point, however our expenses (for principle and interest only -- tax will still need to be paid) drop by $10k. That means should only need $24k at this point. Suddenly we're at a withdrawal rate of 3.5% instead of 4%...

This seems to indicate that paying off a mortgage when pulling the trigger to FIRE may provide significant improvements to your success rate. Thoughts? Am I missing something? Are there arguments in favor of keeping the mortgage anyway?

I didn't check the numbers but I would be surprised if they turned out to be substantially incorrect. Your scenario is one of a fully funded retirement (not under- or overfunded) and the elimination of the liabilty of the mortgage loan favors not carrying the loan. Forget the emotional issue of "owning" your house or not - that is not relevant for the question of going into retirement with or without leveraged investments.

You can answer this question very simply: 1) underfunded: can't retire anyways and risk of failure to achieve a fully funded retirement is to be considered. With a sufficiently long time horizon and a good interest rate on your mortgage, leveraged investment may substantially increase your chances of eventually being fully funded. 2) overfunded = with a sufficiently long time horizon, leveraged investments may be advantageous as your liabilities are more than matched. 3) fully funded: your liabilities are matched and eliminating leverage decreases failure risk (illustrated in your case by the decrease in withdrawal rate from 4% to 3.5%).

Of course, the conditions, underfunded, fully funded, and overfunded, exist on a continuum and so should your leveraging strategy develop as well.
In any case, you are doing the correct analysis and this will enable you to exercise reasoned judgment when you make the necessary decisions.


frugalnacho

  • Magnum Stache
  • ******
  • Posts: 3334
  • Age: 36
  • Location: Madison Heights, Michigan
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #5 on: July 28, 2014, 01:18:08 PM »
So when you hit FIRE you will have 850k stash and 170k mortgage?  And your 170k mortgage is costing you 10k a year in principal and interest?  How is a 170k mortgage @3.5% costing you 10k/yr? 

What was your original mortgage amount?  I am having a hard time understanding the numbers.  A full 30 year mortgage with a starting point of 170k and 3.5% interest only requires about $9,200 a year, and that remains the same for all 30 years.   If you started with a significantly larger mortgage the yearly payments would be higher.

PeteD01

  • Bristles
  • ***
  • Posts: 413
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #6 on: July 28, 2014, 01:23:43 PM »
So when you hit FIRE you will have 850k stash and 170k mortgage?  And your 170k mortgage is costing you 10k a year in principal and interest?  How is a 170k mortgage @3.5% costing you 10k/yr? 

What was your original mortgage amount?  I am having a hard time understanding the numbers.  A full 30 year mortgage with a starting point of 170k and 3.5% interest only requires about $9,200 a year, and that remains the same for all 30 years.   If you started with a significantly larger mortgage the yearly payments would be higher.

It is at least a 186k mortgage.

frugalnacho

  • Magnum Stache
  • ******
  • Posts: 3334
  • Age: 36
  • Location: Madison Heights, Michigan
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #7 on: July 28, 2014, 01:36:21 PM »
Yea the op put that right in the post, I am retarded.  It makes sense.

tomsang

  • Handlebar Stache
  • *****
  • Posts: 1085
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #8 on: July 28, 2014, 01:51:55 PM »
Typically people forget that they can draw down on the principal of their investments to pay their monthly mortgage payment.  If you are retiring on a 4% SFR, then you are saying that your investments will be in excess of 4% over 30+ years.  The models probably indicate that you will be having investment returns in the 7%+.  I made an excel spreadsheet to show this for another topic.

Lots of discussion in this thread.
http://forum.mrmoneymustache.com/investor-alley/paying-off-mortgage-early-how-bad-is-it-for-your-fi-date/

Cheddar Stacker

  • Magnum Stache
  • ******
  • Posts: 3714
  • Age: 41
  • Location: USA
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #9 on: July 28, 2014, 02:51:29 PM »
JGB, good question. I didn't run any numbers (I'm on my phone) but there are at least 2 reasons why I can think holding the mortgage will still be better:

1) inflation. Your portfolio can provide a 4% SWR because it returns 7-8% with 3-4% inflation. 3 years post fire that 34k will be 36k making it easier to make the mortgage payment.

2) eventual mortgage payoff would also clear 10k(ish) in expenses and will happen automatically. That will create a huge safety margin in 10 years when the mortgage matures. All the sudden your safety margin is 10k not 3k.

PeteD01

  • Bristles
  • ***
  • Posts: 413
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #10 on: July 28, 2014, 03:17:11 PM »
JGB, good question. I didn't run any numbers (I'm on my phone) but there are at least 2 reasons why I can think holding the mortgage will still be better:

1) inflation. Your portfolio can provide a 4% SWR because it returns 7-8% with 3-4% inflation. 3 years post fire that 34k will be 36k making it easier to make the mortgage payment.

2) eventual mortgage payoff would also clear 10k(ish) in expenses and will happen automatically. That will create a huge safety margin in 10 years when the mortgage matures. All the sudden your safety margin is 10k not 3k.

That sounds about right. However, the trade off is that the sequence of return risk is higher and the assumptions regarding inflation and ROI need to materialize. But this is what the OP has to weigh against the advantages of a lower withdrawal rate when making his decisions. The beauty of the lower withdrawal rate is that if ROI were actually 7-8% in the years after FIRE, his investments will do quite well.
« Last Edit: July 28, 2014, 04:25:32 PM by PeteD01 »

Tyler

  • Handlebar Stache
  • *****
  • Posts: 1141
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #11 on: July 28, 2014, 03:23:27 PM »
I think a lot of prospective early retirees get hung up on the difference between net worth and investable assets when talking about SWRs. In the accumulation phase, many people focus on net worth and ignore mortgage principal payments as a cost because they are net worth neutral (a perfectly reasonable thing to do). However, in retirement net worth is irrelevant and only investable assets matter from a SWR perspective. And your entire mortgage payment is an expense out of your investments.

Also, where you are in your current mortgage payoff makes a difference. Retiring in year 1 of a 30 year mortgage is a bit different than when you're 20 years into the mortgage (where your payments are a much higher percentage of the remaining balance).

For me, if I were to keep my mortgage in FIRE my withdrawal rate from my investments in year 1 would be 4%. If I paid it off, it's 3%. I plan to pay it off. IMHO, everyone should run their own numbers and not assume that a blanket solution makes most sense for them.
« Last Edit: July 28, 2014, 03:43:09 PM by Tyler »

defenestrate

  • 5 O'Clock Shadow
  • *
  • Posts: 45
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #12 on: July 28, 2014, 03:58:18 PM »


This seems to indicate that paying off a mortgage when pulling the trigger to FIRE may provide significant improvements to your success rate. Thoughts? Am I missing something? Are there arguments in favor of keeping the mortgage anyway?

On a theoretical level this makes sense. In ER, your mortgage is essentially leverage, and that leverage is not invested in your house, but typically in the stock market--this will increase your return by (market return-(1-tax rate)*mortgage rate) on the portion of your portfolio that would otherwise be used to pay off your debt. So lets say you have a mortgage rate of 4% and the market returns its long run average of 8%, and that your marginal tax rate on the interest expense is 25%. Your additional return would be 5%. So on your $170,000, you may expect an additional (pre-tax) return of $8500.

Now let's say you had no tax benefit on interest payments, the additional return would fall to $6,800. If you are only concerned with long run returns, and your interest rate is low enough, then the math says to invest in the market.

However, volatility is what is not taken into account. The distribution of returns in the stock market make it quite risky to use leverage while at the same time relying on income from the portfolio to fund living expenses, especially fixed expenses like a mortgage.

The standard deviation of returns on the stock market is ~20%. If you retire with a lot of fixed long-term costs, that you cannot adjust, this volatility can be more costly than the $6800-$8500 of excess return you may expect to reap every year. Let's say that you get hit with an 20% hit to the portfolio--that 170,000 becomes $136,000 and if you will need to take an additional 5% from this principal to cover the interest expense on the mortgage (probably double if you need to cover the principal payments as well).

If on the other hand you had payed off your mortgage, you reduce your exposure to volatility. Having a paid off mortgage is a hedge against bear markets. If you are working, or generating income, you have more flexibility to take advantage of the spread in returns. In early retirement, reducing your fixed costs is a hedge, and hedges are very valuable. This is especially true for those that FIRE at the 4% of portfolio mark.




PeteD01

  • Bristles
  • ***
  • Posts: 413
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #13 on: July 28, 2014, 05:02:15 PM »
Typically people forget that they can draw down on the principal of their investments to pay their monthly mortgage payment.  If you are retiring on a 4% SFR, then you are saying that your investments will be in excess of 4% over 30+ years.  The models probably indicate that you will be having investment returns in the 7%+.  I made an excel spreadsheet to show this for another topic.

Lots of discussion in this thread.
http://forum.mrmoneymustache.com/investor-alley/paying-off-mortgage-early-how-bad-is-it-for-your-fi-date/

What exactly is the principal of an investment? I hear this once in a while but nobody has ever been able to give me an explanation that made any sense beyond mental accounting. I'm serious, this concept can be extremely dangerous for an investor but is probably off topic in this thread.

PeteD01

  • Bristles
  • ***
  • Posts: 413
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #14 on: July 28, 2014, 05:33:40 PM »


This seems to indicate that paying off a mortgage when pulling the trigger to FIRE may provide significant improvements to your success rate. Thoughts? Am I missing something? Are there arguments in favor of keeping the mortgage anyway?

On a theoretical level this makes sense. In ER, your mortgage is essentially leverage, and that leverage is not invested in your house, but typically in the stock market--this will increase your return by (market return-(1-tax rate)*mortgage rate) on the portion of your portfolio that would otherwise be used to pay off your debt. So lets say you have a mortgage rate of 4% and the market returns its long run average of 8%, and that your marginal tax rate on the interest expense is 25%. Your additional return would be 5%. So on your $170,000, you may expect an additional (pre-tax) return of $8500.

Now let's say you had no tax benefit on interest payments, the additional return would fall to $6,800. If you are only concerned with long run returns, and your interest rate is low enough, then the math says to invest in the market.

However, volatility is what is not taken into account. The distribution of returns in the stock market make it quite risky to use leverage while at the same time relying on income from the portfolio to fund living expenses, especially fixed expenses like a mortgage.

The standard deviation of returns on the stock market is ~20%. If you retire with a lot of fixed long-term costs, that you cannot adjust, this volatility can be more costly than the $6800-$8500 of excess return you may expect to reap every year. Let's say that you get hit with an 20% hit to the portfolio--that 170,000 becomes $136,000 and if you will need to take an additional 5% from this principal to cover the interest expense on the mortgage (probably double if you need to cover the principal payments as well).

If on the other hand you had payed off your mortgage, you reduce your exposure to volatility. Having a paid off mortgage is a hedge against bear markets. If you are working, or generating income, you have more flexibility to take advantage of the spread in returns. In early retirement, reducing your fixed costs is a hedge, and hedges are very valuable. This is especially true for those that FIRE at the 4% of portfolio mark.

+1

Volatility is one risk that one can manage almost to the point that it doesn't qualify as a risk anymore.  If it weren't for that I wouldn't touch the stock market because volatility is a given. I like the idea of thinking of reducing fixed expenses as a hedge strategy because that's what it is and it is very much what this site is about.

dragoncar

  • Walrus Stache
  • *******
  • Posts: 8616
  • Registered member
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #15 on: July 28, 2014, 07:40:54 PM »
JGB, good question. I didn't run any numbers (I'm on my phone) but there are at least 2 reasons why I can think holding the mortgage will still be better:

1) inflation. Your portfolio can provide a 4% SWR because it returns 7-8% with 3-4% inflation. 3 years post fire that 34k will be 36k making it easier to make the mortgage payment.

2) eventual mortgage payoff would also clear 10k(ish) in expenses and will happen automatically. That will create a huge safety margin in 10 years when the mortgage matures. All the sudden your safety margin is 10k not 3k.

Yes, I ran these numbers the other day and I think they favored holding the mortgage.  Here's why:

Run firecalc to with a non-inflation adjusted spending model for 30 years (or whatever time remaining) on expenses of your mortgage payment.  Ask it to find the amount needed for x% success rate.  If the number is less than payoff amount, keep the mortgage.



I'm interested in seeing everyone's numbers

defenestrate

  • 5 O'Clock Shadow
  • *
  • Posts: 45
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #16 on: July 28, 2014, 09:25:54 PM »
Typically people forget that they can draw down on the principal of their investments to pay their monthly mortgage payment.  If you are retiring on a 4% SFR, then you are saying that your investments will be in excess of 4% over 30+ years.  The models probably indicate that you will be having investment returns in the 7%+.  I made an excel spreadsheet to show this for another topic.

Lots of discussion in this thread.
http://forum.mrmoneymustache.com/investor-alley/paying-off-mortgage-early-how-bad-is-it-for-your-fi-date/

What exactly is the principal of an investment? I hear this once in a while but nobody has ever been able to give me an explanation that made any sense beyond mental accounting. I'm serious, this concept can be extremely dangerous for an investor but is probably off topic in this thread.

My grandmother, who died at 100, in the last days of her life asked my father if he had "touched the principal" of her investment--to which my father smartly said "no." The idea of principal is only one held by individuals who have relatively low financial knowledge, and on the other spectrum, financial professionals trying to take advantage of the first group. Sadly, it is a concept that even places like vanguard use to describe financial performance. For us, our principal is whatever we have in the bank at any given time, it can rise and fall, but when we make decisions about our capital (principal) we should treat every dollar the same.

So perhaps "principal" can be seen as a discriminatory slur in finance. It treats the dollar originally invested different than the subsequent dollars earned on the investment. All dollars should be equally valued and embraced--irrespective of how they found their way into our portfolio.

I accept all dollars, and treat them all the same, as a principal...equal treatment of all dollars.

defenestrate

  • 5 O'Clock Shadow
  • *
  • Posts: 45
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #17 on: July 28, 2014, 09:37:00 PM »
So when you hit FIRE you will have 850k stash and 170k mortgage?  And your 170k mortgage is costing you 10k a year in principal and interest?  How is a 170k mortgage @3.5% costing you 10k/yr? 

What was your original mortgage amount?  I am having a hard time understanding the numbers.  A full 30 year mortgage with a starting point of 170k and 3.5% interest only requires about $9,200 a year, and that remains the same for all 30 years.   If you started with a significantly larger mortgage the yearly payments would be higher.

Original amount was 187,000 at 3.5%-- the mortgage will cost ~$840 a month or ~$10,100 per year

JGB

  • Stubble
  • **
  • Posts: 108
  • Age: 36
  • Location: Oklahoma
    • Grinderschool Poker Training
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #18 on: July 29, 2014, 05:22:54 AM »
Original amount was 187,000 at 3.5%-- the mortgage will cost ~$840 a month or ~$10,100 per year
$10k was an approximation. Actual initial balance was just under $190k. Monthly payments (excluding Tax/Insurance) are $850.46.

Scandium

  • Handlebar Stache
  • *****
  • Posts: 2198
  • Location: EastCoast
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #19 on: July 29, 2014, 10:37:04 AM »
Yes, I ran these numbers the other day and I think they favored holding the mortgage.  Here's why:

Run firecalc to with a non-inflation adjusted spending model for 30 years (or whatever time remaining) on expenses of your mortgage payment.  Ask it to find the amount needed for x% success rate.  If the number is less than payoff amount, keep the mortgage.


I'm interested in seeing everyone's numbers

Interesting. I did like you said and ran cFiresim with my mortgage. If I retire ~50 our mortgage would be $150k with 12 years left (we just refinanced), and costs us $25k/year.
From the model; 12 years of this spending with 95% success rate require $274,192.50. So based on this it would be better to take $150,000 from our stache and pay off the mortgage the day we FIRE?

aj_yooper

  • Handlebar Stache
  • *****
  • Posts: 1091
  • Age: 8
  • Location: Chicagoland
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #20 on: July 29, 2014, 10:52:48 AM »
Interesting thread.  JGB, I like how you posed the problem of mortgage expense in the context of the total stash. 

JCfire

  • Stubble
  • **
  • Posts: 145
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #21 on: July 29, 2014, 11:32:07 AM »
A few things change when you FIRE, and all of them make a mortgage less advantageous.  First, you will have less (if any) taxable income, so any tax benefit of mortgage interest becomes less valuable.  Second, your risk tolerance should get lower, because if something goes very wrong its harder to jump out of retirement into a new job than it would have been to keep your job in the first place -- you are more vulnerable. 

Third, and most importantly I think, you no longer have a regular cash inflow from your salary.  Instead, your cash outflow needs must be funded by investment income.  Some of that investment income may be in the form of dividends and interest payments, but likely some of it is in the form of selling assets.  Instead of dollar-cost-averaging into the market (which means you're buying more shares when they're cheaper which is good), you are dollar-cost-averaging OUT OF the market (selling more shares when they're cheaper which is bad). 

Your increased vulnerability (higher risk aversion) and the difference between dollar-cost-averaging IN versus OUT are two big reasons why the SWR is 4% and not 6%.

So your investment portfolio can fund a stream of cashflows equal to 4% of principal.  What can your mortgage fund?  A stream of cashflows equal to the future interest payments you make.  If your mortgage interest rate is above your SWR, you should prepay the mortgage when you FIRE.

If you choose not to do so, you will be better off ON AVERAGE!  This is counter-intuitive, why did I just recommend doing something that makes you worse off on average?  Because your retirement planning efforts don't focus on the average case, they focus on protecting you from the worst case.  In a worst-case scenario, the market crashes early in your retirement, and your investment portfolio is down 50%.  You have to sell twice as many shares to fund that mortgage payment because you are dollar-cost-averaging OUT, and when the market recovers back to its 7%-annual-appreciation trend, you don't fully participate in the recovery.  Your future income is impaired for life because of market volatility that would have made you wealthier, not poorer, if it happened before FIRE.  This risk is lower if you prepay your mortgage. 

Another way to demonstrate this is as follows -- if your mortgage rate is 5% on a $100,000 balance, and your SWR is 4%, you will need to save $125,000 in your investment account to give you enough reliable income to pay your mortgage interest.  If you instead use $100,000 of that money to prepay your mortgage, you'll have $25,000 and zero debt left, and that $25,000 will generate $1000 of reliable income throughout your retirement that you can use on other things.

dragoncar

  • Walrus Stache
  • *******
  • Posts: 8616
  • Registered member
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #22 on: July 29, 2014, 11:52:48 AM »
Yes, I ran these numbers the other day and I think they favored holding the mortgage.  Here's why:

Run firecalc to with a non-inflation adjusted spending model for 30 years (or whatever time remaining) on expenses of your mortgage payment.  Ask it to find the amount needed for x% success rate.  If the number is less than payoff amount, keep the mortgage.


I'm interested in seeing everyone's numbers

Interesting. I did like you said and ran cFiresim with my mortgage. If I retire ~50 our mortgage would be $150k with 12 years left (we just refinanced), and costs us $25k/year.
From the model; 12 years of this spending with 95% success rate require $274,192.50. So based on this it would be better to take $150,000 from our stache and pay off the mortgage the day we FIRE?

Yes, as between those two specific options, it seems like it would make sense to pay off the loan.  But I'm curious what your loan term/amount/rate are to get such a high monthly payment with such a low amount remaining with 12 years left. 

Remember, there are other options though -- different asset allocations and required success rates (basically risk tolerance) will affect the required stache to service the mortgage.

Scandium

  • Handlebar Stache
  • *****
  • Posts: 2198
  • Location: EastCoast
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #23 on: July 29, 2014, 12:19:54 PM »
Yes, I ran these numbers the other day and I think they favored holding the mortgage.  Here's why:

Run firecalc to with a non-inflation adjusted spending model for 30 years (or whatever time remaining) on expenses of your mortgage payment.  Ask it to find the amount needed for x% success rate.  If the number is less than payoff amount, keep the mortgage.


I'm interested in seeing everyone's numbers

Interesting. I did like you said and ran cFiresim with my mortgage. If I retire ~50 our mortgage would be $150k with 12 years left (we just refinanced), and costs us $25k/year.
From the model; 12 years of this spending with 95% success rate require $274,192.50. So based on this it would be better to take $150,000 from our stache and pay off the mortgage the day we FIRE?

Yes, as between those two specific options, it seems like it would make sense to pay off the loan.  But I'm curious what your loan term/amount/rate are to get such a high monthly payment with such a low amount remaining with 12 years left. 

Remember, there are other options though -- different asset allocations and required success rates (basically risk tolerance) will affect the required stache to service the mortgage.

Yes you're right. I'm an idiot. I just used our mortgage payment, which if course includes taxes etc! The loan payment is $18k/year. Which require $197k stach to pay for 12 years. So it's still better to pay off the $150k balance out of the portfolio, rather than continuing to pay.

(playing around with stock/bond ratio I got numbers between $188k - $215k)
ps: our loan is $300k/4.12%/30 years
« Last Edit: July 29, 2014, 12:28:17 PM by Scandium »

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2207
  • Age: 38
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #24 on: July 29, 2014, 01:48:58 PM »
For anyone with a sub-4% mortgage rate and a sufficiently long remaining period until maturity, the numbers strongly favor keeping the mortgage outstanding (as the firecalc test outlined by dragoncar will demonstrate).  The sequence of return risk is already reflected in the firecalc test (in other words, the results already include all historical periods where the investor was forced to sell assets during a crash early in the timeline, and they still returned an x% success rate).

frugaliknowit

  • Handlebar Stache
  • *****
  • Posts: 1662
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #25 on: July 29, 2014, 02:07:30 PM »
A few things change when you FIRE, and all of them make a mortgage less advantageous.  First, you will have less (if any) taxable income, so any tax benefit of mortgage interest becomes less valuable.  Second, your risk tolerance should get lower, because if something goes very wrong its harder to jump out of retirement into a new job than it would have been to keep your job in the first place -- you are more vulnerable. 

Third, and most importantly I think, you no longer have a regular cash inflow from your salary.  Instead, your cash outflow needs must be funded by investment income.  Some of that investment income may be in the form of dividends and interest payments, but likely some of it is in the form of selling assets.  Instead of dollar-cost-averaging into the market (which means you're buying more shares when they're cheaper which is good), you are dollar-cost-averaging OUT OF the market (selling more shares when they're cheaper which is bad). 

Your increased vulnerability (higher risk aversion) and the difference between dollar-cost-averaging IN versus OUT are two big reasons why the SWR is 4% and not 6%.

So your investment portfolio can fund a stream of cashflows equal to 4% of principal.  What can your mortgage fund?  A stream of cashflows equal to the future interest payments you make.  If your mortgage interest rate is above your SWR, you should prepay the mortgage when you FIRE.

If you choose not to do so, you will be better off ON AVERAGE!  This is counter-intuitive, why did I just recommend doing something that makes you worse off on average?  Because your retirement planning efforts don't focus on the average case, they focus on protecting you from the worst case.  In a worst-case scenario, the market crashes early in your retirement, and your investment portfolio is down 50%.  You have to sell twice as many shares to fund that mortgage payment because you are dollar-cost-averaging OUT, and when the market recovers back to its 7%-annual-appreciation trend, you don't fully participate in the recovery.  Your future income is impaired for life because of market volatility that would have made you wealthier, not poorer, if it happened before FIRE.  This risk is lower if you prepay your mortgage. 

Another way to demonstrate this is as follows -- if your mortgage rate is 5% on a $100,000 balance, and your SWR is 4%, you will need to save $125,000 in your investment account to give you enough reliable income to pay your mortgage interest.  If you instead use $100,000 of that money to prepay your mortgage, you'll have $25,000 and zero debt left, and that $25,000 will generate $1000 of reliable income throughout your retirement that you can use on other things.

This is an excellent analysis, Thank you, JC Fire!
One always needs to look at the worst case scenario and it's impact.

PeteD01

  • Bristles
  • ***
  • Posts: 413
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #26 on: July 29, 2014, 05:07:00 PM »
For anyone with a sub-4% mortgage rate and a sufficiently long remaining period until maturity, the numbers strongly favor keeping the mortgage outstanding (as the firecalc test outlined by dragoncar will demonstrate).  The sequence of return risk is already reflected in the firecalc test (in other words, the results already include all historical periods where the investor was forced to sell assets during a crash early in the timeline, and they still returned an x% success rate).

I think everyone here understands this but the topic of the thread is the fact that deleveraging caused the withdrawal rate in the OP's example to drop. Obviously, if you keep the withdrawal rate constant, you will come to the conclusion that deleveraging is disadvantageous. But that is circular reasoning in this thread's context, with or without firecalc. On another note, the data firecalc simulations are based upon include the periods during crashes and we are most certainly not in a crash. This makes the success rates at least questionable. But that is a different topic

aj_yooper

  • Handlebar Stache
  • *****
  • Posts: 1091
  • Age: 8
  • Location: Chicagoland
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #27 on: July 29, 2014, 05:29:06 PM »
If there is a mortgage in retirement, following JGB's example, the stash 4% SWR may also need to pay taxes on the distribution.  In the example above, that would be $10,000/(1-tax rate).  With a 15% tax rate, the retiree would need to draw down $10,000/(1-.15) or $11,765, which leads to a bigger stash again.

dragoncar

  • Walrus Stache
  • *******
  • Posts: 8616
  • Registered member
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #28 on: July 29, 2014, 05:38:45 PM »
For anyone with a sub-4% mortgage rate and a sufficiently long remaining period until maturity, the numbers strongly favor keeping the mortgage outstanding (as the firecalc test outlined by dragoncar will demonstrate).  The sequence of return risk is already reflected in the firecalc test (in other words, the results already include all historical periods where the investor was forced to sell assets during a crash early in the timeline, and they still returned an x% success rate).

I think everyone here understands this but the topic of the thread is the fact that deleveraging caused the withdrawal rate in the OP's example to drop. Obviously, if you keep the withdrawal rate constant, you will come to the conclusion that deleveraging is disadvantageous. But that is circular reasoning in this thread's context, with or without firecalc. On another note, the data firecalc simulations are based upon include the periods during crashes and we are most certainly not in a crash. This makes the success rates at least questionable. But that is a different topic

The data firecalc simulations are based upon include Thursdays and we are most certainly not in a Thursday.

The Happy Philosopher

  • Bristles
  • ***
  • Posts: 343
    • thehappyphilosopher
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #29 on: July 29, 2014, 05:43:07 PM »
Just thinking off the cuff here but a potential blind spot in any comparison is that the mortgage will eventually end, causing expenses to drop at some point and effectively lower the swr at some point in the future (ie: if there are 10 years left on the note, in 10 years your swr will change due to lowered expenses).  Paying off a mortgage the day before retirement changes the swr forever.  In other words it's difficult to really compare the two without taking that into account.  Sorry if someone already mentioned this, I read the thread quickly.

PeteD01

  • Bristles
  • ***
  • Posts: 413
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #30 on: July 29, 2014, 06:02:26 PM »
For anyone with a sub-4% mortgage rate and a sufficiently long remaining period until maturity, the numbers strongly favor keeping the mortgage outstanding (as the firecalc test outlined by dragoncar will demonstrate).  The sequence of return risk is already reflected in the firecalc test (in other words, the results already include all historical periods where the investor was forced to sell assets during a crash early in the timeline, and they still returned an x% success rate).

I think everyone here understands this but the topic of the thread is the fact that deleveraging caused the withdrawal rate in the OP's example to drop. Obviously, if you keep the withdrawal rate constant, you will come to the conclusion that deleveraging is disadvantageous. But that is circular reasoning in this thread's context, with or without firecalc. On another note, the data firecalc simulations are based upon include the periods during crashes and we are most certainly not in a crash. This makes the success rates at least questionable. But that is a different topic

The data firecalc simulations are based upon include Thursdays and we are most certainly not in a Thursday.

I don't quite understand what Thursdays have to do with it but I would really like a function allowing me to adjust the denominator according to the derivative of market performance. Not that I put much stock into backtesting, but it would be interesting to play around with different intervals and such. Fortunately, one can simulate a crash by subjecting one's assets to a precipitous decline of any desired magnitude and then plug that number in. It's funny how that works out.
« Last Edit: July 29, 2014, 06:55:21 PM by PeteD01 »

dragoncar

  • Walrus Stache
  • *******
  • Posts: 8616
  • Registered member
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #31 on: July 29, 2014, 06:31:43 PM »
Just thinking off the cuff here but a potential blind spot in any comparison is that the mortgage will eventually end, causing expenses to drop at some point and effectively lower the swr at some point in the future (ie: if there are 10 years left on the note, in 10 years your swr will change due to lowered expenses).  Paying off a mortgage the day before retirement changes the swr forever.  In other words it's difficult to really compare the two without taking that into account.  Sorry if someone already mentioned this, I read the thread quickly.

You can factor this in by considering only interest payments for SWR purposes (allowing yourself to draw down your liquid stache, but recognizing that as a transfer into equity... Thus it doesn't matter if you end up with a few hundred thousand fewer liquid assets at the end because the same result would have occurred if you paid off equity in a lump sum)

Or run firecalc for your mortgage term, not your life expectancy.

Or firecalc could model fixed expenses that end at a point in the future. I think there's already some capacity for this, but I don't think you can separate inflation indexed expenses from truly fixed expenses.

Tyler

  • Handlebar Stache
  • *****
  • Posts: 1141
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #32 on: July 29, 2014, 06:36:57 PM »
Just thinking off the cuff here but a potential blind spot in any comparison is that the mortgage will eventually end, causing expenses to drop at some point and effectively lower the swr at some point in the future (ie: if there are 10 years left on the note, in 10 years your swr will change due to lowered expenses).  Paying off a mortgage the day before retirement changes the swr forever.  In other words it's difficult to really compare the two without taking that into account.  Sorry if someone already mentioned this, I read the thread quickly.

This is absolutely true. It is also true that the withdrawal rates & returns in the first few years of retirement have the largest affect on long term success, so one can still find value in talking about them.

I do like Dragoncar's suggestion of treating the mortgage (and associated investment balance equal to the mortgage amount) separately from other expenses from a Firecalc perspective.

tomsang

  • Handlebar Stache
  • *****
  • Posts: 1085
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #33 on: July 29, 2014, 06:48:25 PM »
Typically people forget that they can draw down on the principal of their investments to pay their monthly mortgage payment.  If you are retiring on a 4% SFR, then you are saying that your investments will be in excess of 4% over 30+ years.  The models probably indicate that you will be having investment returns in the 7%+.  I made an excel spreadsheet to show this for another topic.

Lots of discussion in this thread.
http://forum.mrmoneymustache.com/investor-alley/paying-off-mortgage-early-how-bad-is-it-for-your-fi-date/

What exactly is the principal of an investment? I hear this once in a while but nobody has ever been able to give me an explanation that made any sense beyond mental accounting. I'm serious, this concept can be extremely dangerous for an investor but is probably off topic in this thread.

If you are not making extra principal payments on your mortgage, then you are expected to place those payments in your investment portfolio.  When you retire or if you lose your job, you then draw down your portfolio to continue to make your mortgage payment. When the mortgage is paid off at term you will most likely have a sizable amount left in your portfolio that is extra money. The logic is that your investment portfolio will out perform 3.5% over 30 years. If you don't think that your portfolio will achieve greater than 3.5% over the longterm then you should not retire until your SWR is sub 1% as all the models are based on the past and we have never had a 30 year period of time that had returns less than 6%+.

All of this and the math is in the link that was provided. A few hang ups are that people consider there mortgage payment a permanent expense even if it ends in 15 or so years. I also believe that you are safer investing your extra payment vs. paydown as you are creating liquidity risk while taking on poor returns. So you are getting the worst of both worlds.


brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2207
  • Age: 38
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #34 on: July 29, 2014, 06:51:22 PM »
Just thinking off the cuff here but a potential blind spot in any comparison is that the mortgage will eventually end, causing expenses to drop at some point and effectively lower the swr at some point in the future (ie: if there are 10 years left on the note, in 10 years your swr will change due to lowered expenses).  Paying off a mortgage the day before retirement changes the swr forever.  In other words it's difficult to really compare the two without taking that into account.  Sorry if someone already mentioned this, I read the thread quickly.

You can factor this in by considering only interest payments for SWR purposes (allowing yourself to draw down your liquid stache, but recognizing that as a transfer into equity... Thus it doesn't matter if you end up with a few hundred thousand fewer liquid assets at the end because the same result would have occurred if you paid off equity in a lump sum)

Or run firecalc for your mortgage term, not your life expectancy.

Or firecalc could model fixed expenses that end at a point in the future. I think there's already some capacity for this, but I don't think you can separate inflation indexed expenses from truly fixed expenses.

I think the easiest way to conceptualize it and model it with firecalc simulations is to mentally break down your stash into two stashes:  one that will service your mortgage principal and interest (firesim inputs for this one should end at maturity and have non-inflation adjusted spending) and one for all other expenses.

PeteD01

  • Bristles
  • ***
  • Posts: 413
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #35 on: July 29, 2014, 08:07:45 PM »
Typically people forget that they can draw down on the principal of their investments to pay their monthly mortgage payment.  If you are retiring on a 4% SFR, then you are saying that your investments will be in excess of 4% over 30+ years.  The models probably indicate that you will be having investment returns in the 7%+.  I made an excel spreadsheet to show this for another topic.

Lots of discussion in this thread.
http://forum.mrmoneymustache.com/investor-alley/paying-off-mortgage-early-how-bad-is-it-for-your-fi-date/

What exactly is the principal of an investment? I hear this once in a while but nobody has ever been able to give me an explanation that made any sense beyond mental accounting. I'm serious, this concept can be extremely dangerous for an investor but is probably off topic in this thread.

If you are not making extra principal payments on your mortgage, then you are expected to place those payments in your investment portfolio.  When you retire or if you lose your job, you then draw down your portfolio to continue to make your mortgage payment. When the mortgage is paid off at term you will most likely have a sizable amount left in your portfolio that is extra money. The logic is that your investment portfolio will out perform 3.5% over 30 years. If you don't think that your portfolio will achieve greater than 3.5% over the longterm then you should not retire until your SWR is sub 1% as all the models are based on the past and we have never had a 30 year period of time that had returns less than 6%+.

All of this and the math is in the link that was provided. A few hang ups are that people consider there mortgage payment a permanent expense even if it ends in 15 or so years. I also believe that you are safer investing your extra payment vs. paydown as you are creating liquidity risk while taking on poor returns. So you are getting the worst of both worlds.

Is this in any way related to my question or what you were proposing in your original post?
You wrote that one can draw down on the principal of one's investments and my question is what that exactly means. This is not a trivial question because your statement implies a major financial misunderstanding. The issue is volatility and how to manage it after FIRE.
Don't worry about the math, no issues here.
« Last Edit: July 29, 2014, 08:09:54 PM by PeteD01 »

tomsang

  • Handlebar Stache
  • *****
  • Posts: 1085
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #36 on: July 29, 2014, 08:35:09 PM »
Typically people forget that they can draw down on the principal of their investments to pay their monthly mortgage payment.  If you are retiring on a 4% SFR, then you are saying that your investments will be in excess of 4% over 30+ years.  The models probably indicate that you will be having investment returns in the 7%+.  I made an excel spreadsheet to show this for another topic.

Lots of discussion in this thread.
http://forum.mrmoneymustache.com/investor-alley/paying-off-mortgage-early-how-bad-is-it-for-your-fi-date/

What exactly is the principal of an investment? I hear this once in a while but nobody has ever been able to give me an explanation that made any sense beyond mental accounting. I'm serious, this concept can be extremely dangerous for an investor but is probably off topic in this thread.

If you are not making extra principal payments on your mortgage, then you are expected to place those payments in your investment portfolio.  When you retire or if you lose your job, you then draw down your portfolio to continue to make your mortgage payment. When the mortgage is paid off at term you will most likely have a sizable amount left in your portfolio that is extra money. The logic is that your investment portfolio will out perform 3.5% over 30 years. If you don't think that your portfolio will achieve greater than 3.5% over the longterm then you should not retire until your SWR is sub 1% as all the models are based on the past and we have never had a 30 year period of time that had returns less than 6%+.

All of this and the math is in the link that was provided. A few hang ups are that people consider there mortgage payment a permanent expense even if it ends in 15 or so years. I also believe that you are safer investing your extra payment vs. paydown as you are creating liquidity risk while taking on poor returns. So you are getting the worst of both worlds.

Is this in any way related to my question or what you were proposing in your original post?
You wrote that one can draw down on the principal of one's investments and my question is what that exactly means. This is not a trivial question because your statement implies a major financial misunderstanding. The issue is volatility and how to manage it after FIRE.
Don't worry about the math, no issues here.

Drawing down the principal is how I describe selling shares to pay for your mortgage or other payments.  Each month, quarter or year you are balancing your portfolio to ensure that you have enough liquidity to pay all of your bills.  It is expected that you will be liquidating share of the bucket where you invested all of the payments that you were going to pay down your mortgage.

 Cfiresim, firecalc, and the 4% rules/tools already take into account volatility.  If you understand the math, then you understand how these tools work and you also understand that earning 7%+ on your investments while paying out 3.5% on your loan is going to create a significant cushion over the years. Paying your mortgage off early by monthly contributions is creating liquidity risks, while locking a sub optimal 30 year return. If you are assuming that your investment return is less than 3.5% over 30 years, you are projecting out a swr of less than 1%.

defenestrate

  • 5 O'Clock Shadow
  • *
  • Posts: 45
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #37 on: July 29, 2014, 09:08:43 PM »
  If you understand the math, then you understand how these tools work and you also understand that earning 7%+ on your investments while paying out 3.5% on your loan is going to create a significant cushion over the years. Paying your mortgage off early by monthly contributions is creating liquidity risks, while locking a sub optimal 30 year return. If you are assuming that your investment return is less than 3.5% over 30 years, you are projecting out a swr of less than 1%.

This is true, but only if there is great discipline in following the strategy. Lets say that you had a $190,000 mortgage in its 17th year in the year 2000 with a 3.5% rate. At this point the loan balance is ~$113,000. From 2000-2013 you live through two bear markets (using the S&P 500) from 2000-2002 and 2008.

If you kept the $113,000 invested, and paid the mortgage from this principal each year, you would end up with a balance of $21,000 in 2013. Is $21,000 a lot? Sure it is, but is it worth holding onto the balance to reap that benefit? Remember, this is a future value of $21,000--the present value of that money (using a 3% inflation rate) is ~$15,000. Is $15,000 really worth carrying the leverage? Probably not.

   loan   113000      
   Mortgage Payment   ($6,650.03)   
                         Portfolio Balance   Additions from portfolio to principal
17   2000   -9.03%   $96,144.02    ($6,382.03)
18   2001   -11.85%   $78,101.15    ($6,605.40)
19   2002   -21.97%   $54,295.39    ($6,836.59)
20   2003   28.36%   $63,041.25    ($7,075.87)
21   2004   10.74%   $63,163.61    ($7,323.52)
22   2005   4.83%   $59,567.21    ($7,579.85)
23   2006   15.61%   $62,217.15    ($7,845.14)
24   2007   5.48%   $58,979.56    ($8,119.72)
25   2008   -36.55%   $30,771.12    ($8,403.91)
26   2009   25.94%   $32,101.66    ($8,698.05)
27   2010   14.82%   $30,209.45    ($9,002.48)
28   2011   2.10%   $24,193.33    ($9,317.57)
29   2012   15.89%   $21,387.76    ($9,643.68)
30   2013   32.15%   $21,612.85    ($9,981.21)
                                    ($112,815.01)



dragoncar

  • Walrus Stache
  • *******
  • Posts: 8616
  • Registered member
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #38 on: July 29, 2014, 09:29:32 PM »


This is true, but only if there is great discipline in following the strategy. Lets say that you had a $190,000 mortgage in its 17th year in the year 2000 with a 3.5% rate. At this point the loan balance is ~$113,000. From 2000-2013 you live through two bear markets (using the S&P 500) from 2000-2002 and 2008.

If you kept the $113,000 invested, and paid the mortgage from this principal each year, you would end up with a balance of $21,000 in 2013. Is $21,000 a lot? Sure it is, but is it worth holding onto the balance to reap that benefit? Remember, this is a future value of $21,000--the present value of that money (using a 3% inflation rate) is ~$15,000. Is $15,000 really worth carrying the leverage? Probably not.

   loan   113000      
   Mortgage Payment   ($6,650.03)   
                         Portfolio Balance   Additions from portfolio to principal
17   2000   -9.03%   $96,144.02    ($6,382.03)
18   2001   -11.85%   $78,101.15    ($6,605.40)
19   2002   -21.97%   $54,295.39    ($6,836.59)
20   2003   28.36%   $63,041.25    ($7,075.87)
21   2004   10.74%   $63,163.61    ($7,323.52)
22   2005   4.83%   $59,567.21    ($7,579.85)
23   2006   15.61%   $62,217.15    ($7,845.14)
24   2007   5.48%   $58,979.56    ($8,119.72)
25   2008   -36.55%   $30,771.12    ($8,403.91)
26   2009   25.94%   $32,101.66    ($8,698.05)
27   2010   14.82%   $30,209.45    ($9,002.48)
28   2011   2.10%   $24,193.33    ($9,317.57)
29   2012   15.89%   $21,387.76    ($9,643.68)
30   2013   32.15%   $21,612.85    ($9,981.21)
                                    ($112,815.01)

I don't get it.  You're saying in probably one of the worst market scenarios you can devise, you still come out $15,000  ahead?  $15,000 is more than zero, and you got all that liquid money for 13 years absolutely free on top of that?  And that's a bad thing?

Cheddar Stacker

  • Magnum Stache
  • ******
  • Posts: 3714
  • Age: 41
  • Location: USA
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #39 on: July 29, 2014, 10:09:05 PM »
The data firecalc simulations are based upon include Thursdays and we are most certainly not in a Thursday.

Of course we are not in a Thursday. Are you crazy. We are in a Tuesday. You need to run the firecalc simulations in a Tuesday or the results are useless. Amatuer hour.

I really have enjoyed this thread, but I think it really comes down to the same question: invest or paydown debts? If you like the safe boring play (bonds) pay down your mortgage. If you are willing to take on more risk (stocks) keep the mortgage and invest the difference.

The math will work itself out. If you get different results from firecalc or a spreadsheet your did something wrong. 8% > 4% every time.

Tyler

  • Handlebar Stache
  • *****
  • Posts: 1141
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #40 on: July 29, 2014, 10:33:40 PM »
The math will work itself out. If you get different results from firecalc or a spreadsheet your did something wrong. 8% > 4% every time.

It's more complicated than that. Each year you have a decision to make on whether keeping the mortgage is worth it over simply paying it off.

A $160k mortgage at a 4% interest rate has an annual principal + interest payment of $9156. Plugging that $160k stache into Firecalc with annual spending of $9156, no inflation, and a 30-year horizon gives a success rate of just over 95%. The mortgage makes a lot of sense.

With 15 years left on the same mortgage, the balance is $103k. Plug that in with the same spending and a remaining 15-year duration, and the success rate drops to 82%. Still not bad.

With 5 years left, the balance is $40k and the success rate is 53%. Still worth it?

So unless you continually refinance to a 30-year mortgage, the risk of the investments lagging the mortgage increases over time. One can still argue the odds are in your favor, but at some point (especially for someone who is already FI) the risk just seems unnecessary. That breaking point is up to you.

So 8% average (with a large SD) > 4% fixed only some of the time, with the odds reducing as the mortgage payoff progresses.
« Last Edit: July 29, 2014, 10:37:30 PM by Tyler »

Cheddar Stacker

  • Magnum Stache
  • ******
  • Posts: 3714
  • Age: 41
  • Location: USA
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #41 on: July 30, 2014, 06:10:18 AM »
With 5 years left, the balance is $40k and the success rate is 53%. Still worth it?

Very well explained Tyler but...

I would still take those odds. Its better than a coin flip and that $40k will be a drop in the bucket of my net worth. Don't forget this isn't the success rate of your retirement, its the chance you will end up with more money under this one area or your planning.

Even at that point the first 25 years of ROI should dwarf any losses you might incur over the last 5 years.

So again, do you carry a portfolio of 25% stocks and 75% bonds? Aggressive or conservative?

JCfire

  • Stubble
  • **
  • Posts: 145
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #42 on: July 30, 2014, 09:29:37 AM »
The math will work itself out. If you get different results from firecalc or a spreadsheet your did something wrong. 8% > 4% every time.

It's more complicated than that. Each year you have a decision to make on whether keeping the mortgage is worth it over simply paying it off.

A $160k mortgage at a 4% interest rate has an annual principal + interest payment of $9156. Plugging that $160k stache into Firecalc with annual spending of $9156, no inflation, and a 30-year horizon gives a success rate of just over 95%. The mortgage makes a lot of sense.

With 15 years left on the same mortgage, the balance is $103k. Plug that in with the same spending and a remaining 15-year duration, and the success rate drops to 82%. Still not bad.

With 5 years left, the balance is $40k and the success rate is 53%. Still worth it?

So unless you continually refinance to a 30-year mortgage, the risk of the investments lagging the mortgage increases over time. One can still argue the odds are in your favor, but at some point (especially for someone who is already FI) the risk just seems unnecessary. That breaking point is up to you.

So 8% average (with a large SD) > 4% fixed only some of the time, with the odds reducing as the mortgage payoff progresses.

Principal part of the payment is irrelevant -- it's just moving your wealth from one pocket (i.e. stocks) to another (real estate equity investment).

Tyler

  • Handlebar Stache
  • *****
  • Posts: 1141
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #43 on: July 30, 2014, 10:04:16 AM »

Principal part of the payment is irrelevant -- it's just moving your wealth from one pocket (i.e. stocks) to another (real estate equity investment).

Irrelevant to net worth, but highly relevant to these calculations because the pocket matters. See a previous post.
« Last Edit: July 30, 2014, 10:22:20 AM by Tyler »

Tyler

  • Handlebar Stache
  • *****
  • Posts: 1141
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #44 on: July 30, 2014, 11:42:24 AM »
I would still take those odds. Its better than a coin flip and that $40k will be a drop in the bucket of my net worth. Don't forget this isn't the success rate of your retirement, its the chance you will end up with more money under this one area or your planning.

Even at that point the first 25 years of ROI should dwarf any losses you might incur over the last 5 years.

Sunk profits are just like sunk costs - they're in the past and shouldn't influence your decision making looking forward.  I personally consider a 50% chance of profit on an investment to be gambling so it's not in my nature. But as I said, that is up for each person to decide.

tomsang

  • Handlebar Stache
  • *****
  • Posts: 1085
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #45 on: July 30, 2014, 12:18:41 PM »
The math will work itself out. If you get different results from firecalc or a spreadsheet your did something wrong. 8% > 4% every time.

It's more complicated than that. Each year you have a decision to make on whether keeping the mortgage is worth it over simply paying it off.

A $160k mortgage at a 4% interest rate has an annual principal + interest payment of $9156. Plugging that $160k stache into Firecalc with annual spending of $9156, no inflation, and a 30-year horizon gives a success rate of just over 95%. The mortgage makes a lot of sense.

With 15 years left on the same mortgage, the balance is $103k. Plug that in with the same spending and a remaining 15-year duration, and the success rate drops to 82%. Still not bad.

With 5 years left, the balance is $40k and the success rate is 53%. Still worth it?

So unless you continually refinance to a 30-year mortgage, the risk of the investments lagging the mortgage increases over time. One can still argue the odds are in your favor, but at some point (especially for someone who is already FI) the risk just seems unnecessary. That breaking point is up to you.

So 8% average (with a large SD) > 4% fixed only some of the time, with the odds reducing as the mortgage payoff progresses.

I think you are missing the crucial point that at the end of a mortgage most of the monthly payment is going to principal. It is not an expense. A better question is can you beat 3.5% over the term of the mortgage or more appropriately over your expected life span as your investment portfolio should be designed to withstand 40+ years. When doing cfiresim, firecalc, etc. you have to bifurcate your mortgage payment into fixed vs. variable. With the Principal and interest fixed and insurance and taxes variable. The P&I should be set up as ending on the date of the last payment.

If the answer is that 3.5% is your best investment opportunity then ou should pay off your mortgage. You should also use a sub 1% SWR as all the models are based on 7%+ 30 year returns based on the past.

aj_yooper

  • Handlebar Stache
  • *****
  • Posts: 1091
  • Age: 8
  • Location: Chicagoland
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #46 on: July 30, 2014, 12:31:06 PM »
Tyler, you state your points very well.  MMM has a paid for house in retirement so you would be in good company. 

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2207
  • Age: 38
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #47 on: July 30, 2014, 12:45:54 PM »
I would still take those odds. Its better than a coin flip and that $40k will be a drop in the bucket of my net worth. Don't forget this isn't the success rate of your retirement, its the chance you will end up with more money under this one area or your planning.

Even at that point the first 25 years of ROI should dwarf any losses you might incur over the last 5 years.

Sunk profits are just like sunk costs - they're in the past and shouldn't influence your decision making looking forward.  I personally consider a 50% chance of profit on an investment to be gambling so it's not in my nature. But as I said, that is up for each person to decide.

Keep in mind that the yearly decision on whether or not to pay off the mortgage applies only to the (dwindling) remaining balance of the mortgage, while the original stash that started out equal to the mortgage balance will have grown over time (with a sufficiently long horizon).

In other words, in your example, the 95% success rate for investments outperforming mortgages with a 30-year horizon means that (historically) 95 out of 100 times you would have been better off keeping the mortgage outstanding for the entire 30 year period.  It's not as if those 95 success cases suddenly got cut in half when year 25 rolls around.

So your analysis explains why someone with a long horizon remaining on their mortgage is probably better off keeping the mortgage outstanding.  And it also explains why it may make sense for that same person to eventually pay off the mortgage prior to maturity, at a time when he or she is sitting on a stash that started out equal to the mortgage balance but now vastly exceeds the remaining balance.

I think you articulated this clearly in your thoughtful explanation, but just wanted to throw this out there to prevent anyone from drawing the wrong conclusion from your analysis.

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2207
  • Age: 38
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #48 on: July 30, 2014, 12:51:55 PM »
I think you are missing the crucial point that at the end of a mortgage most of the monthly payment is going to principal. It is not an expense. A better question is can you beat 3.5% over the term of the mortgage or more appropriately over your expected life span as your investment portfolio should be designed to withstand 40+ years. When doing cfiresim, firecalc, etc. you have to bifurcate your mortgage payment into fixed vs. variable. With the Principal and interest fixed and insurance and taxes variable. The P&I should be set up as ending on the date of the last payment.

If the answer is that 3.5% is your best investment opportunity then ou should pay off your mortgage. You should also use a sub 1% SWR as all the models are based on 7%+ 30 year returns based on the past.

The fact that most of the payment is allocated to principal does not change this analysis.  You are still paying x% (whatever your mortgage rate happens to be) on the outstanding principal balance.  Tyler's point is that at any given time, you need to analyze whether keeping the remaining balance outstanding makes sense.  25 years into paying the mortgage (i.e., with a 5 year horizon left), the answer may be no.  But, as I said in my post immediately above, this still argues for opting to keep the mortgage outstanding during the years when you have a long horizon to maturity.

Bob W

  • Magnum Stache
  • ******
  • Posts: 2947
  • Age: 60
  • Location: Missouri
  • Live on minimum wage, earn on maximum
Re: Invest vs Payoff Debts: SWR Effects of Mortgage Payoff
« Reply #49 on: July 30, 2014, 01:16:49 PM »
Great job managing your money and a very smart question!

The money smart long term strategy in your situation is to invest at 9% and pay at 3%.   The numbers work very nicely over 30 year periods.   Over 60 years,  which I consider a "lifetime" you would have an extra 5 million dollars assuming you kept the mortgage around 150K.  If you just pay off the home you might have 1 million in home value.   So the question might be,  would you rather have 5 million or 1 million to pass on to your heirs.   Good stewardship takes many, many years but it is the seed of the small acorn that grows into the giant oak tree.  (I know cliché)

Shorter term consider these options.   1.  Switch to an interest only mortgage with 20% equity.  That changes your scenario significantly.   (not a variable but a 7-15 year interest only)  2.  Your home is too big and too expensive.  (many of us on this site are included in this including MMM) So consider downsizing or renting something smaller.   3.  Post FI you may have the opportunity to move to an area like mine where 100K buys a nice home on rolling wooded acres.   Or you may decide to travel for a couple of years.  So you may consider listing your home now at a value 15% above similar homes.  You might get lucky.   

If you don't want to move, because let's face it we are habit creatures,  I would seriously consider option 1.    If you choose it,  remember you can always change your mind down the road.   

That said you may have a problem that many of the FIs have in that you didn't mention long term care insurance, LTCI.    If you don't have this you should consider it a need and not a want.   It is expensive.  But if you don't have LTCI then you should  reserve in excess of 1 million per person for this.   That wouldn't cover a long term nursing home fully at a 4% return but would minimize the risk.   Otherwise you may be fooling yourself with the "independent" term as you would be relying on Medicaid to cover nursing home care.

Submitted respectfully and just MHO.