Pay off mortgage early ?

[renferme]

It's never rained money on me, but if it ever does, I'll have my bucket ready.

 

[Jarhead*]

JMortgage debt (especially if it is under 6%), puts you in the cat bird seat, if we get even close to the inflation of the late 60s though the early 80s.
I had a young family in 1966, and was 28 years old. For the next 16 years inflation and stagflation was the order of the day. (My parents were depression folks, and they went to their grave never trusting banks again).
Since 1982 or so, we have been in an almost cant miss environment, with bonds and stock returns making investment novices feel like they were in the class with Ted.
My own personal opinion, if I were in my earlly 50s, and was of the opinion that inflation was going to be a problem in the future, as long as I could service the debt with no problem, I dont think I would want to give away an excellent hedge.
In 1981 mortgages were running 16%. (CDs were paying about the same for 4 years.
Having real estate with a low interest rate was about the only way you could sell your property. The classified ads were picked clean every day, with real estate investors looking for assumable loans. If you had one, you were almost guaranteed a quick sale.
There are other ways of course, to hedge inflation, and this board (most posters) seems to like tips. I do too, and recently bought some.
The fact that your mortgage interest is deductible, (for most folks), makes to me, a pretty good case for holding on to a low interest mortgage.
Of course, if the current home you have is the last home you intend to live in, and have no intentions of ever selling, the piece of mind factor has to be figured in.
Jarhead

 

[haha]

Having real estate with a low interest rate was about the only way you could sell your property.  The classified ads were picked clean every day, with real estate investors looking for assumable loans.

Jarhead,

I think that assumable residential loans are about as comon as hen's teeth today. Still, that mortgage money is there to use in various money making ways, while you continue to live in the property.

Often though, the tax factors are negative. And, if your offsetting investment goes bad, your home is on the line. For some people, that can be a good enough reason to avoid mortgages.

Mikey

 

[sgeeeee]

And, if your offsetting investment goes bad, your home is on the line. For some people, that can be a good enough reason to avoid mortgages.

This argument keeps appearing, but it just does not make sense. Take a "typical" situation: investor has $1.1M at retirement and owes $100K on loan with 20 years remaining. The two options the investor has are to (1) pay off the loan and go into retirement with $1M or (2) start with $1.1M and keep making payments for 20 years (or pay off at any time in the future).

If the investor chooses (2) and ends up with their house on the line, then it means they would have had to have lost more than 90% (>$1M) of their nest egg in the first 20 years of retirement. Even if you assume that happens, then the poor guy who chose option (1) is in even worse shape unless he can sell his house. And if there were some reason why a rational person facing such an economic crises would prefer to own a house over eating, then all that a person who chose option 2 would have to have done is paid off the loan just before his balance dropped below the payoff amount.

If paying off your mortgage gives you a warm fuzzy that you just can't live without, then pay off your mortgage. But if you can afford to pay it off, then your house isn't on the line unless your entire lifestyle is on the line.

 

[haha]

You make a good point. Realistically, one's entire portfolio is carrying the mortgage, not just whatever offsetting investment was made with the proceeds of the loan (or the money that would have been earmarked for paying off the loan.)

Mikey

 

[Ted]

While I agree that paying off a mortgage isn't exactly a "no-brainer," one consideration that favors paying it off is this.

When a person's interest payments become fairly modest, as a mortgage matures, there may no longer be any tax advantage to holding the mortgage. This will happen if a person's itemized deductions fall below their standard deduction.

The most important consideration, though, is the interest rate on the mortgage. If it is above the after-tax return on secure investments such as a Treasury Note of equivalent maturity, it is best for most people to pay it off.

 

[renferme]

Ted says "When a person's interest payments become fairly modest, as a mortgage matures, there may no longer be any tax advantage to holding the mortgage. This will happen if a person's itemized deductions fall below their standard deduction.

The most important consideration, though, is the interest rate on the mortgage. If it is above the after-tax return on secure investments such as a Treasury Note of equivalent maturity, it is best for most people to pay it off"

A GREAT EXPLANATION. THANKS !

 

[sgeeeee]

. . . The most important consideration, though, is the interest rate on the mortgage. If it is above the after-tax return on secure investments such as a Treasury Note of equivalent maturity, it is best for most people to pay it off.

The converse of that is certainly true. . . If secure investments of equivalent maturity offer better after-tax return than the interest rate on the mortgage, it would be foolish to pay off the mortgage.

But this statement assumes much about the problem that is not neccesarily true. Run the numbers and see for your situation. Or don't run the numbers and decide based on your emotion, if that suits you better.

Example: (Case 1)
Investor has:
paid off home
$1M nest egg
75% equities
25% commercial paper
30 year horizon
0.18% expense ratio
Initial withdrawal rate = $40,000

FIRECALC probability of success = 100%
Average terminal value =$4,218,464
=============
Example: (Case 2)
Investor has:
remaining mortgage balance of $100K
@5%
for 10 more years
$1.1 M nest egg (original + $100K owed)
75% equities
25% commercial paper
30 year horizon
0.18% expense ratio
Initial withdrawal rate = $46,400 (case 1 + mortgage payments)
change in withdrawal by -$6440 after 10 years

FIRECALC probability of success = 100%
Average terminal value =$4,558,859

So case 2 provides equally high probability of success with a probability to increase withdrawal rate by more than 8% over case 1. Alternatively, that 8% can be used to offset items such as healthcare that increase at well over the average inflation rate.

The analysis ignores tax advantages of case 2.

When using retirement simulators like FIRECALC, we tend to focus on probability of success and safe withdrawal rate, but once acceptable SWR is achieved there is a lot of potential advantage to increasing the terminal value. That value is a relative indicator of how well you might be able to deal with unexpected inflation or other unexpected large expenses. Alternatively, it indicates a probability that you will be able to increase your SWR at some point in the future.

 

[Ted]

The converse of that is certainly true. . . If secure investments of equivalent maturity offer better after-tax return than the interest rate on the mortgage, it would be foolish to pay off the mortgage.  

But this statement assumes much about the problem that is not neccesarily true.  Run the numbers and see for your situation.  Or don't run the numbers and decide based on your emotion, if that suits you better.  

The only thing that this analysis illustrates is that if you can get a mortgage at a NEGATIVE 8.7% interest rate, you certainly will be better off taking it and investing the money.

Aside from the fact that the mortgage payment should be about $12,700 per year in this example, it should be recognized that taking the $100,000 from the mortgage and, in effect, investing it 75% in stocks and 25% in commercial paper is NOT investing it in "secure investments."  Running the analysis with FIRECalc using the correct numbers illustrates this -- especially if the planning period is less than 30 years and the withdrawal percentage is increased.  

Furthermore, per my previous comments about the tax deductibility of mortgage interest disappearing, there is not necessarily any tax advantage to keeping an "old" mortgage.

 

[Telly]

All the discussion here has been about an "all or nothing" payoff of a mortgage.

Another avenue is pre-paying months as-you-go.  This will not drain your ready cash, and you can feed it when you are comfortable with doing so.

Back when we did this, we had to pay $5 or so to have a loan payment table printed out for us from the lender.  It was $5 well spent.  Nothing like a complete list of payments for 30 years month by month.  Each line was a month, with Principal, Interest to be paid for that month. Principle to go, Interest etc. listed.

The usual monthly payment due is Principal + Interest.  So if you pay the month's payment, plus JUST the Principle for the next month, you have moved ahead two months.  Early on, almost all of a monthly payment is interest, so skipping ahead is easy.  As you get far into the loan, the extra Principle payments get heavier and heavier, but by then inflation and salary increases are helping out.  We did years at first, then chunks of years, down to just 2 mos or so when needed.  
I know it's not rocket science, but back then I was the only one I knew who did it.  Many figured that there was something "wrong" with it.  I felt there was something "wrong" with NOT doing it!

As inflation moves on, the amount of the mortgage that can be deducted from Fed tax decreases to zero, as mentioned before.  Unless one buys such an expensive house, and does a 95% loan and lives BEYOND ones means.

We kicked ourselves a bit for not looking into pre-paying earlier than we did.  That first year or so was almost ALL interest!  So the additional payment of principal would have been small change.  But we did catch on before too long.

Debt-free since '85.  Or maybe it was earlier than that  

 

[renferme]

we do something like that. Got a 30 year loan on our house 4 years ago, but began paying it off like it was a 15 year loan. Then a little over a year ago, when interest rates went way down, we refinanced, got a 15 year mortgage. Since we had already paid 3 years on the first mortgage, we now are paying the 15 year loan as if it were a 12 year loan. So now we have a little less than 11 years to go. That would take me to age 68 and I do believe that I would like this debt out of the way by the time I begin taking social security. That's kind of why I proposed this question in the first place.
Cheers, Ben

 

[earlyout]

In addition to the previous comments from everyone, some things to also consider: 1) if one needs to go into a nursing home under medicaid, I believe (someone correct if not right) your home is not considered an asset that must be used up before you qualify to go into a nursing home, leaving the other spouse with the house. Whereas stocks and other investments are capped and must be used up before entering nursing home (medicaid). In other words the stocks etc are not protected. 2) also I believe in certain states that your home is protected from certain liabilites/lawsuits and is an added form of protection of your asset (i.e. house).

Someone may have more details and info on these topics, please add and clarify.

Thanks

 

[John Galt]

I have heard of cases where people moved to a state
specifically because their home would be judgement proof, and then
tried to protect huge sums (invested in their residence)
from creditors. I assume there must be a statute of limitations involved, but it is surely one more factor to be
taken into account when considering where to reside.

John Galt

 

[sgeeeee]

The only thing that this analysis illustrates is that if you can get a mortgage at a NEGATIVE 8.7% interest rate, you certainly will be better off taking it and investing the money. . .

Which is more than any analysis others have presented proves because no other analysis has been presented.

I believe you'll find that the simulation is a valid simulation for a legitimate situation. It does overestimate the value of paying off the mortgage for a couple of reasons, but provides a worst case analysis for that situation. You may have simply misinterpreted the simulation.

The "secure investments" requirements is an arbitrarily constraint that I have not placed on the strategy. Since the point of the strategy is to realize gains above the interest rate of the mortgage, it would be foolish to place that money in low yielding bonds. If you could do that and come out ahead, then clearly you would be better off doing that. That situation may be available to people who re-finance today and face retirement in 10 years or so, but it would be a rare situation.

The risk issue is best dealt with at an overall portfolio level as opposed to trying to make a direct risk translation of your home mortgage to a specific investment in your portfolio. Requiring a direct trade of mortgage risk for a specific investment's risk would be a very narrow way to evaluate this strategy. Again, if you can make a direct risk translation from one mortgage to one investment, it makes the decision easy. But once you have achieved a risk level that is sufficient for you, you can begin to look at how to trade additional risk security for other factors.

I'm retired now, Ted. And this discussion reminds me of why I chose to get away from the office and retire early. I think you should pay off your mortgage if you haven't already. I hope it turns out to be a successful strategy for you.

 

[eridanus]

JohnGalt,

Isn't that what OJ did? He went to Florida, bought a huge house, and then took up residence there to avoid the civil lawsuits from his, uh, situation.

John Connally did that too in Texas. He was in fact the individual responsible for the limit placed on such homesteads because he bought a huge ranch right before the creditors came down on him. It's still a high limit, maybe unlimited for non-acreage, as I believe some of the Enron perps are using it too.

 

[cute fuzzy bunny]

Well, the Enron perps can hang photos of their nice estates on their prison cell walls.

Thankfully, I only owned Enron stock one time...when I bought it for about 50c just prior to bankruptcy and sold it a couple of days later for 70-something cents. Funny how drawing a paycheck allows for making riskier play-investing decisions...

I was sitting in the doctors office for a physical about 20-22 months or so ago. Of course they had the obligatory nearly year old magazines, including some financial rag that was published just before the bubble went pop. I laughed and laughed reading the stories about how you just "MUST OWN ENRON AND WORLDCOM, EVEN AT THESE PRICES...THESE BABIES STILL HAVE A LOT OF ROOM TO GO UP!!! BUY NOW!!!!". Not to mention thoughtful analysis on why internet stocks really do make sense and should be a core holding.

The house asset item is a real benefit and one I frankly hadnt thought of. While cash and investments can be seized or garnished for a variety of sins and ailments, the home is usually considered sacrosanct.

 

[Ted]

I believe you'll find that the simulation is a valid simulation for a legitimate situation.  It does overestimate the value of paying off the mortgage for a couple of reasons, but provides a worst case analysis for  that situation.  You may have simply misinterpreted the simulation.  

This isn't by any means the first time that I have seen a "proof" of something that was based on a completely unrealistic assumption, i.e., that a borrower can take out a mortgage for which the lender will pay THEM interest. A reasonable assumption is that the interest rate on the mortgage would be 5%, which means that the annual payment required to retire a $100,000 balance over 10 years would be $12,700 per year, and not $6,440 per year. (The latter is the annual payment required to amortize $100,000 over 30 years!)

Running the FIRECalc analysis using the correct annual mortgage payment demonstrates that a person's probablity of sustaining any particular withdrawal rate is higher if they pay off the mortgage than if they invest the $100,000 and continue making mortgage payments for 10 years.

 

[wabmester]

A reasonable assumption is that the interest rate on the mortgage would be 5%, which means that the annual payment required to retire a $100,000 balance over 10 years would be $12,700 per year, and not $6,440 per year.  (The latter is the annual payment required to amortize $100,000 over 30 years!)

Err, I believe that SG's reasonable assumption was that you had 10 years left on a 30-year mortgage, not that you got a 10-year loan that paid you interest

PLEASE, tell me it isn't intuitively obvious that unless the interest rate on the mortgage(edit: after tax?) is less than the SWR, then it's 'riskier' to have a mortgage than to pay it off.

Hmm? There's no correlation to the SWR and your mortgage rate. What you want to compare is your mortgage rate and expected long-term investment return. And you always have the option of paying off the mortgage if you feel the odds are against you.

Another way to play the game is to invest at your usual asset mix initially, but if/when "safe" interest rates rise above your mortgage rate, then go safe and just take the difference between new riskless return and your old mortgage which you were shrewd enough to get when loan rates were at historic lows.

 

[wabmester]

Oh, you're right about the mistaken payment amount in the example. However, I ran the numbers for my own payment and balance and FIREcalc gave me an 80% chance of investment returns outpacing mortgage payments.

SWR assumes inflation-adjusted withdrawls. Mortgage payments and rates are fixed, so you only need a nominal return greater than your mortgage, not a real return implicit in the SWR.

If you don't trust FIREcalc's historical return data to tell you if you can beat a 5% nominal return, why would you trust it to tell you that you can safely withdraw 4%?

In any case, I agree that you should do whatever your heart and the numbers tell you to do. No different than any other decision....

 

[sgeeeee]

Here is how you can simulate your own mortgage situation to see what your odds are of coming out ahead by paying off your mortgage. I took this example for a personal choice I had to make a few years ago. Should I buy my new home outright or get a loan. I was considering a 5.25% loan for $149,500 with 30 year duration. Below, I tabulate the inputs into FIRECALC for this situation and list and discuss the outputs. Your milage may vary.

FIRECALC Simulation to analyze risk/reward for paying off your loan: this example is specific to my situation. You can easily personalize for your own situation.

Withdrawals: $9907 (annual payments on 5.25%, 30 year loan)
Starting Portfolio: $149,500 (loan amount)
Lifespan of portfolio: 30 years (duration of loan)
percent of portfolio in stocks: 60% (specific to your portfolio)
where is the rest of investments: TIPS (specific to your portfolio)
Annual investment expenses: 0.2 (specific to your portfolio)
inflation estimate: None (since loan payments are fixed)
Check first year withdrawal box (loan payments begin immediately)

FIRECALC results:

We looked at the 101 possible 30 year periods from 1871 until 2002, and the 30 partial periods from 1972 until 2002, starting with a portfolio of $149,500 and taking out $9,907 the first year, and the same amount after adjustments for inflation (no inflation adjustment applied) each year .


Your Success Rate is 78.8%
==========================
The success rate represents your historical probability of coming out ahead by continuing to make regular payments on the loan rather than paying it off. In reality, this number may underestimate your odds since it does not account for the tax benefits of keeping the loan. The tax benefits can be significant in early years.

Details: The simulator shows that if you only consider data after 1894, your probability of success is better than 90%. Only twice since that time would the choice to keep making payments not have resulted in a financial gain over choosing the payoff -- Once in 1906 and a second time related to the great depression in the years 1926 - 1931. The 1906 case missed beating the payoff option by a total of $1813 (so with tax advantage, it too would probably have been a better choice).

 

[sgeeeee]

I found some additional analysis I did on this topic when I was making the loan vs payoff decision. I put together a spreadsheet that evaluated annual payments (principle and interest) then computed the tax advantage due to interest payments. The net actual cost was then subtracted from the unspent loan value and I applied my personal annual rate of return to that value. I looked at how I would have faired had I taken a 5.25% loan at the beginning of any of the last 10 years. I assumed a 20% tax rate. (Actually I played with that as a variable, but 20% is a number that probably underestimates my actual tax rate in Arizona in retirement by a small amount) Just for you guys, I have added the last couple of years to those spreadsheets so I could present a more complete story.

The worst case occured had I chosen to take this loan at the beginning of 2000. For that situation, I would now need to make an annual return on my investments of about 7% for the next 26 years in order to make the loan choice beat the payoff choice. That's the only case that I thought was marginal. Every other year looked like a no-brainer to me.

Next worse is to have taken the loan in 2001. To beat the payoff option, I would need to achieve an annual return on investment of 5.75% over the next 27 years.

Results of taking the 5.25% loan during other years:

YEAR REQUIRED ROR TO BEAT PAYOFF OPTION
2003 2.94%
2002 4.25%
2001 5.75%
2000 7.04%
1999 5.16%
1998 3.89%
1997 2.15%
1996 0.9%
1995 -1.49%
.
.
.

In case you are wondering, here is the personal rate of return I used which is calculated by Money. It represents what I earned on my overall portfolio for each year. Stock options and company investment plan constraints had a major effect on my performance, but I don't think these numbers are too far out of line for a more typical portfolio.

Year personal annual % return
1990 7.52%
1991 20.20%
1992 7.61%
1993 17.13%
1994 4.17%
1995 23.36%
1996 13.50%
1997 18.05%
1998 14.34%
1999 19.39%
2000 -4.63%
2001 -8.68%
2002 -9.70%
2003 22.19%

 

[Ted]

Here are the specific results of analyzing this example with FIRECalc, using Salaryguru's original assumption of a person with $1 million net worth deciding whether or not to pay off a mortgage at 5% interest with 10 years remaining.  Per Salaryguru's original assumption, the $1 million is invested 75% in stocks and 25% in commercial paper, and so is the $100,000 if it is not used to immediately pay off the mortgage.  But I will use a higher withdrawal rate that will not result both scenarios having a "100% probability of success," because that result masks the true differences in the riskiness of the respective strategies.

Case 1:  Pay off mortgage.  Invest $1 million.  Initial withdrawal rate $50,000 per year, increasing with inflation.  Probability of success: 88.6%  Mean remaning balance: $2.75 million.

Case 2:  Pay off mortgage over 10 years.  Invest $1.1 million.  Initial withdrawal rate $50,000 per year, increasing with inflation, plus $12,700 per year for 10 years to pay off mortgage.  Probability of success: 88.6%.  Mean remaining balance: $2.95 million.

(NOTE: To do this in FIRECalc is a bit tricky because the $50,000 increases with inflation, but the $12,700 doesn't. This post is revised to reflect that, and makes the outcome of Case 2 more attractive. However, the outcomes are essentially equal.)

 

[wabmester]

While not a major slam dunk, here's a conservative approach:

Take the money you would have used to payoff your mortgage and stick it in a federally insured 5.5% 7-year CD at PenFed. After seven years, look around for a better rate. If you can't find one, pay off your remaining balance.

I actually took this one step further. I got a HELOC at PenFed with a 1-year teaser rate of 1.9% and put that money into various safe short-term instruments that yield just shy of 3%. After one year, I plan to pay down the HELOC, but keep the credit line open as a disaster fund.

 

[sgeeeee]

A couple thoughts:  Just how many of the last 10/20/30 years would obtaining a fixed 5% or 5.25% mortgage (after expenses) been possible?  If one used the "real" expense adjusted mortgage rates for those years what would the results look like??  

Although that may be an interesting question, it is not the question at hand. I personnaly don't care what the odds of finding a low interst loan was at some time in the past, I care about whether I am likely to do well by keeping the loan that is possible for me to get today. The strategy of keeping your 13% loan from 1985 is probably a poor one, but the question is, if you hold a 5% or 5.25% loan, what is the best strategy.

As far as "tax advantage", since cap gains, dividends, and interest earned would all be taxed, it could be more tax advantaged to pay off the mortgage, if you're NOT able to deduct 100% of all mortgage interest.   For me Less than half the interest would be deductible and that would decrease as the std deduction continues to rise and annual mortgage interest decreases.   While nearly all the investment proceeds would be eventually taxed.

That may be for your case. That's why you should run your own numbers. Of course the FIRECALC analysis does not include any tax advantage or disadvantage. If you bother to run those simulations and look at the results, you can see that the tax difference is not likely to make a big affect on the conclusions.  It could in a few of the cases.

In the spreadsheet analysis I did originally, I eventually plugged in my actual tax estimates which included my working tax bracket in the first two years of the loan and then dropped to my retirement tax bracket estimates for the remainder. Since that was very specific to me, I did not include the results from those simulations in the post above. Because I was actually in a very high tax bracket for the first two years of the loan (when interest payments are high), it made the analysis even more favorable toward keeping the loan than the results I presented. But as I have said countless times, you need to run numbers specific to you yourself. The tax advantage will be small for most people, but not necesarily insignificant. And as you point out, there may be cases where it is a tax disadvantage -- certainly not for me.

I'm still NOT convinced there is any slam dunk here and the 80% chance of succes may well be based on some fairly faulty assumptions.  

I have not heard anyone suggest that it is a slam dunk. What I've said so many times that I can bearly stand to repeat is that people should run the numbers for their case. I've given instructions on how to do that fairly accurately and simply using FIRECALC (but neglecting tax advantage/disadvantage). And I've described some spreadsheet analysis that is not too difficult to do on your own that would allow you to put in specifics of your tax situation.

At this point, I certainly don't expect (or care) to convince anyone on this board that choosing not to pay off your mortgage is a good idea. Please pay it off and we will never discuss it again. I am more than $30,000 ahead today because of my own choice and I have significant probability to gain far more than that over then next 2-3 decades. That will be my reward for doing the analysis myself.

As far as faulty assumptions go, I haven't heard any specific faulty assumptions discussed as they relate to my analysis of my own situation.

 

[wabmester]

MOST HELOCs have some additional costs (appraisal & recording fees etc) which could eat into that 'massive' ~1% advantage--not to mention taxes. 

In my case, I opened the HELOC as part of a refi before I retired a couple years ago. There were no upfront fees, no annual fee, and no appraisal required if you had an appraisal that was less than 1-year old I believe.

I had already paid the HELOC off, and they sent me an offer with a 1-year 1.9% teaser if I started using it again. So, I figured why not -- if anybody wants to pay me $1000 for taking out a loan I don't need, I'll take the money. (And the interest is tax deductible to boot.)

Now, the 1% I'm making from arbitrage is actually on the low side. I could have bought another rung on a bond or CD ladder that would have paid approx 5%, and then paid off the loan when a lower rung on the ladder matured. All about as risk free as you can get.