Covering a mortgage without losing your ass(ets).

[Nords]

You live in an interesting market, too, UM.

I've spent some time in your area, hopefully the statute of limitations has expired by now...

I hear that New Orleans will be the American Venice of the 22nd century. And I can't imagine living on stilts without flood insurance!

 

[cute fuzzy bunny]

Re: Covering a mortgage without losing your ass(et

Sounds a little like a passive-agressive attack there, TH. Are you taking too much fish oil again?

I'm simply superbly pleased to have been able to have a discussion on the topic with different points of view where the other person didnt decide I was doing something "improper" or decide I must hate them for having a difference of opinion... :

I think a lot of people could actually LEARN something from THIS dialog...

By the way, I dont have a "passive aggressive" bone in my body.

Do you have something to add or are you just interested in trying to provoke me again today?

 

[sgeeeee]

Re: Covering a mortgage without losing your ass(et

Now that Nords has discussed the issue, I thought I might be able to present some data that is relevant to the mortgage/payoff decision.

Thank-you, Nords.

While none of us can say for sure whether playing the historical odds by not paying off a low interest mortgage will make money in the future or not, we can see whether it would have earned money in the past. A decision not to pay off a 30 year mortgage can reduce portfolio survival risk and increase probable terminal value of a reasonably allocated portfolio. The portfolio survival improvement realized by continuing to make payments rather than paying off the loan can be greater than the survival improvement realized by simply readjusting stock/bond allocation to a higher value.

Here's a case you can analyze with FIRECALC:

First consider the payoff case --
initial portfolio value: $1.0M
initial withdrawal rate: $50K
# of years: 30
Expense Ratio: 0.18
stock/bond ratio: 60/40

Probability of success: 87.9%
average terminal value: $1,943,681


Next consider a case with a 5.25% mortgage on $100K --
initial portfolio value: $1.1M
initial withdrawal rate: $50K + $6626.40 fixed mortgage
mortgage amount: $100K
mortgage rate: 5.25%
# of years: 30
Expense Ratio: 0.18
stock/bond ratio: 60/40

Probability of success: 90.2%
average terminal value: $2,137,980

Now, consider the payoff case with a higher stock/bond allocation --
initial portfolio value: $1.0M
initial withdrawal rate: $50K
# of years: 30
Expense Ratio: 0.18
stock/bond ratio: 66/34

Probability of success: 88.6%
average terminal value: $2,265,408

So, for these fairly reasonable values, keeping the mortgage would historically have resulted in lower risk (higher probability of portfolio survival) and higher average terminal value. Even if one were to consider a mortgage as equivalent to a bond in the portfolio (an assumption that cannot be completely justified), SWR risk is not reduced as much by this increased volatility risk as it is with the mortgage.

The reason for this result is fairly easy to understand. Since worst case survival occurs when poor real returns are realized in the early years of retirement, any strategy that helps to recover rapidly and solidly in the improved performance years that follow increases probability of survival. The mortgage option accomplishes this by maximizing the nest egg that is invested in the market when economic recovery occurs.

Not every situation will result in this kind of conclusion. As mortgage rates are increased, the SWR risk eventually increases rather than decreases. Similarly, as the ratio of loan value to initial portfolio value changes, so do the benefits of the loan.

Not considered in the analysis is the effect on taxes. Both the tax benefit of loan payments and the tax consequenses of higher required withdrawals have been ignored. These issues are more difficult to quantify as a general case, but are not difficult to estimate for your own personal situation. For many individuals the tax implications are negligible compared to the portfolio value.

If a FIRE candidate does not have other overwhelming feelings that dominate their thinking on the payoff decision, it might be worth the effort to run some simulations and work through the tax consequenses before they make their decision. . . like Nords did.

Of course . . . past performance is no guarantee of future results.

 

[preben]

Re: Covering a mortgage without losing your ass(et

I just paid cash!

 

[John Galt]

Re: Covering a mortgage without losing your ass(et

Over the past 3 years we bought our house, our condo, and a ranchette in Texas. Paid cash for all, and with no
number crunching at all to see if a mortgage was
advantageous, even though rates were very low over that whole period. Partly I am saving myself from
my tendancy to "slip" in my spending. I can still
access the money, but making it slightly inconvenient
works for me. For those of you thinking how easy it is
to get a HELOC, my tiny company owns all of the
real estate and most lenders don't want to give me a
HELOC on that basis. I do like to keep a lot of credit
available though, just in case.

John Galt

 

[Nords]

Thanks, SG!  One other improvement...

I wish I'd thought of running FIRECalc with the combined portfolio/mortgage numbers, but I'd mentally slotted them into two separate categories and I never re-examined that assumption. I'm glad you pointed this out.

th and I already bounced this around and it didn't make any difference to his FIRECalc runs, but inflation may affect the success rate when the mortgage is a sizeable fraction of the portfolio. I don't know how "sensitive" various portfolios are for changes to size of the portfolio, interest rates, expense ratios, and stock/bond mixes.

However for those who are thinking of running a couple hundred of their own scenarios, it's worth pointing out that the mortgage payment stays constant for the life of the mortgage and never rises with inflation. Other expenses will, but not the mortgage payment.

As long as FIRECalc's scenarios are limited to the life of the mortgage, the mortgage payment should be entered into the "Withdrawal Change 1" box as a positive number. (Clear the checkbox underneath because those dollars don't adjust for inflation.) Remaining (inflation-adjusted) expenses can stay in the original "Withdrawal" box.

For calculations longer than the period of the mortgage, the mortgage payment should also be recorded in "Withdrawal Change 2" box as a NEGATIVE number starting at the year when the mortgage ends. Again clear the checkbox underneath. This effectively cancels out the mortgage payment for the remainder of the period.

When we made our first decision, that was based strictly upon running the value of the mortgage through FIRECalc. So I ran the combined portfolio/mortgage scenario using our numbers:

"Withdrawals" & "Starting Portfolio": choose the first to be 4% of the second.
Lifespan: 80 years. (I'll be 124.)
Social Security: 0. (Actually about $12K/year total for the two of us if the system isn't bankrupted in the next 18 years.)
Withdrawal Change 1: Your mortgage payment as a positive number starting at year 0. Clear the checkbox.
Withdrawal Change 2: Your mortgage payment as a NEGATIVE number starting at year 30 (or whatever mortgage term you have). Clear the checkbox.
Withdrawal Change 3: If you're getting a pension then put it here. Check or clear the box as appropriate for a COLA pension. I chose a negative number equal to the mortgage payment, started it in year 0, and checked the box (no COLA). Since my pension is slightly larger than that and has a COLA, this seems most excessively conservative.
Portfolio: 100% stocks (we're actually ~95%, the rest is cash).
Annual expenses: 0.25%.
Inflation: PPI.
First year withdrawal: Check the box. Conservatively you'll need this money to pay the first year of the mortgage.

Those numbers gave 98.5% success. (Of course removing the mortgage from the calculation gave 100% since the SWR is 4%.) This decision was easy to make when the success rate was 74%. Anything over 90% is a slam dunk or, as Bernstein says, "irrelevant".

Upon further reflection, I guess our refinancing was a way to finagle a 30-year present-value lump-sum payment out of the govt in exchange for most of my pension cashflow. As long as that lump sum isn't frittered away foolishly, the odds are in our favor. Whether we're frittering or not remains to be seen.

Now I could spend 10 or 20 man-hours running FinancialEngines' Monte Carlo simulator with the two scenarios. Ouch... or maybe not.

BTW, that's why I put these questions on the board. It's a Darwinian critique process but it certainly exposes the flaws & improvements. Thanks again to all who took the time to think & post!

 

[cute fuzzy bunny]

Re: Covering a mortgage without losing your ass(et

Even if one were to consider a mortgage as equivalent to a bond in the portfolio (an assumption that cannot be completely justified), SWR risk is not reduced as much by this increased volatility risk as it is with the mortgage.

Actually my example showed a dramatic reduction in risk in my no-mortgage scenario, higher terminal results, and higher success rates with no mortgage...even if ones portfolio is substantially smaller. So this statement does not appear to be supported by the facts.

I eagerly await your explanation as to why such an assumption cannot be completely justified, and an explanation as to why you would retain the same stock:bond ratios both with and without a mortgage payout from your bond allocation.

 

[sgeeeee]

Re: Covering a mortgage without losing your ass(et

Actually my example showed a dramatic reduction in risk in my no-mortgage scenario, higher terminal results, and higher success rates with no mortgage...even if ones portfolio is substantially smaller.  So this statement does not appear to be supported by the facts.

Here we go again . . .

Well . . . the statement is not supported by the facts only if you quote me out of context. If you go back to my original post, I begin the paragraph in the sentence just prior to your quote with "So, for these fairly reasonable values . . ."

You see, I was only talking about one case. The reason I was only speaking about one case is because I have always supported the idea of examining each situation on a case by case basis. As I've posted many times (and again in my previous post) the historical record will show that payoff was the most advantageous step financially for many situations while keeping the mortgage was financially advantageous in other cases. Since most people on this board seem to appreciate the first point, I offered one case that showed the second point.

I hoped I was making it explicitly clear that different situations would result in different conclusions in my prior post when I wrote:

"Not every situation will result in this kind of conclusion. As mortgage rates are increased, the SWR risk eventually increases rather than decreases. Similarly, as the ratio of loan value to initial portfolio value changes, so do the benefits of the loan."

So, I hope you can find that satisfying. I do not claim that keeping a mortgage has always been financially advantageous over the option of payoff. It has not.

I eagerly await your explanation as to why such an assumption cannot be completely justified, . . .

Are you really "eagerly awaiting"?

An assumption of equivalence is clearly disproven with a single example where the results using one approach does not align exactly with the other approach. I've presented such and example and you indicate that you have run simulations with different results too.

Dozens of posts and countless words have already been posted on the various early retirement boards discussing questions like: Should I include the value of my home in my net worth?, Should I include real estate as part of my equity allocation or as part of my bond allocation? . . . There are several different opinions on these subjects and all of them can be partially justified. Your house is not equivalent to either your bonds or your equity positions. Simlarly the mortgage behind your house is not equivalent to either equity or bonds.

More specifically, a mortgage is not equivalent to a bond because it involves the place where you live while bonds do not. A mortgage is not equivalent to a bond because local property values underly the mortgage while that is not true of bonds. . . . I could list these things all day. Clearly there are liquidity, volatility and versatility differences between your mortgage and a bond.

I'm not saying that a mortgage does not have some similarities to a bond, simply that it is not equivalent. In my opinion, the differences can be significant. You and others may disagree with this opinion, but I would hope that you don't disagree with the fact that they are not exactly equivalent.

. . . and an explanation as to why you would retain the same stock:bond ratios both with and without a mortgage payout from your bond allocation.

I didn't make any reccommendations about stock/bond allocations. I presented two separate payoff cases: 1) with an equivalent stock bond allocation and 2) with an increased equity allocation to account for the mortgage as if it were a bond. I have run whole families of runs examining this issue for a number of cases. Higher stock allocations with a payoff tend to increase average terminal value beyond that achieved by keeping the mortgage, but do not always offer as much survival safety. Of course when stock allocation is increased, the achievement of increased terminal value comes at the expense of increased short term volatility.

 

[db]

Re: Covering a mortgage without losing your ass(et

I'd add a 6th reason for not paying off a mortgage (or 7th if webmester's 6th is accepted) by liquidating portfolio:

If all, or most, of a portfolio is held in tax deferred status so that taking the funds to payoff a mortage would invoke substantial tax liability.

db

 

[cute fuzzy bunny]

Re: Covering a mortgage without losing your ass(et

Jeepers Creepers! Did Salaryguru just say (in 10,000 words) that my point of view actually was valid and had merit?

You mean my flim-flammery, misdirection and financial sleight of hand has gotten so good that even he is now buying it?!?

Holy cow. Drinks for everyone on me!

Can you take off the "He Hate Me" shirt now?

Well...while we're almost having a legitimate conversation...

Yes, bonds are not == a mortgage. Clearly there are a number of differences.

But if I was holding 400-600k in the usual index fund bucket o' bonds and getting 3%-3.5% on them while paying 5% on a mortgage, paying the bond i've sold to the bank from that stash bears some serious consideration. Given the way financial markets are right now, removing money from any of the usual stock/bond/cd/commodities/reit/etc buckets into something that pays me more feels like a good idea. One of the better benefits of a house vs a bond is if the value of both drops significantly, I can still live in the house. I guess I could live in the bonds or wallpaper with them, but its not quite the same.

And if I was buying a new home or selling a mcmansion and trading down, paying cash for your new home can bring advantage. In my case the home I bought was empty, the kids were selling grammas old house after they moved her in with them. Fast, cheap close was worth money off the asking price, and put me ahead of two buyers that offered more, but would take 2-3 months to close rather than a week and a half.

I guess the only other discussion point is the volatility increase brought about by a higher stock allocation. My feeling is that with a reduced withdrawal rate and more control over that withdrawal rate, the volatility becomes less of an issue. You say you've seen a lower survival percentage with a payoff...I didnt...I saw a higher survival percentage even if I reduced the size of the payoff scenario's portfolio. In my above example, a 700k payoff port was where survivability dropped below 100%, whereas I needed $1M with a 200k mortgage to stay above 100%...which meant to me that I could retire on slightly less than the mortgage difference without a mortgage than I could with one.

I almost hate to broach the scenario in which someone moves after the usual 7 or 10 years rather than fulfilling the full 30 year payoff. In that case, you've primarily paid interest and very little principal. In the event you arent getting a huge tax break off of the mortgage interest I can imagine a scenario showing three 10 year or four 7 year mortgages in a row against paying cash would really tip the advantage to paying cash.

 

[unclemick]

Re: Covering a mortgage without losing your ass(et

Hmmmmm - In our case the 400-600k in bonds are in Trad IRA's - sooo the trade off would be rent vs low down mortgage (?wouldn't it?) rather than getting hosed on taxes.
I've never really run the numbers - just assumed - should living in Hurricane Alley roll the dice - that would be the back up - rent first and then assess the housing market. Assume the existing house as a throw away item to Mother Nature (no insurance).

 

[Nords]

Hey, waitaminnit.

The home's sale price has nothing to do with the ability to qualify for a mortgage or paying cash. In your example you equated that with being able to close faster, but there are other ways to satisfy the seller.

I've bought & sold plenty of homes over the years, and I've never had a seller say "Well, you're offering us more money, but you're getting a mortgage and we want cash..." The seller always gets cash when the escrow has paid out. (Unless they're underwater on their own mortgage!)

We've always included financing & appraisal contingencies in our offer. If we've been asked to remove them (e.g., seller has another all-cash offer, or claims to) then we remove them-- or we don't.

The point was that the seller accepted our offer because we were there first with the best price or willing to pay rent after a certain date or some other incentive-- not because "cash" is inherently better than "mortgage". That's strictly the seller's perception, and if they can't see past cash vs mortgage then they need a more creative perspective.

If you're in an area where cash speaks louder than mortgage financing, then a number of local banks would be getting their financing heads handed to them by an aggressive mortgage broker.

But I do agree that having a bigger portfolio with a lower withdrawal rate makes volatility much more survivable. And we're all smarter than FIRECalc, which doesn't include the "cutting-back-on-expenses" option for each year's returns.

But the smartest choice of all may be to stop arguing among various plans & stock/bond/cash mixes when all of them have success rates above 80%. As Bernstein says, anything above that is claiming that nothing bad will happen for 30-40 years. Given the history of the last seven-eight decades, that claim is more improbable than a 100-year bear market!

 

[wabmester]

Re: Hey, waitaminnit.

The home's sale price has nothing to do with the ability to qualify for a mortgage or paying cash. In your example you equated that with being able to close faster, but there are other ways to satisfy the seller.

I've experienced the same as TH. In a competitive bidding scenario, I've had sellers go with me simply because I waived the financing contigency and offered a fast close, even when my bid was lower.

Nobody likes uncertainty, and offering cash removes a lot of uncertainty for the seller.

 

[John Galt]

Re: Covering a mortgage without losing your ass(et

I have sold a lot of real estate and many times I would have taken less if I had received a "non-contingent" offer.
I am seller in a small deal now. We have had a contract since last summer, but the buyer is financing through
a state agency providing home financing for vets.
As of today (I'm guessing 3 months+ after acceptance)
I still don't know if the seller is approved and no closing date is set. Fortunately for me it's small deal. Knowing what I know now, I would have countered the offer
and required cash. As it is, I am still uncertain if we
will close. A final note. I told my realtor to keep the property on the market until the financing was in place.
I suspect that everyone kind of forgets about you once
a contract is pending though.

John Galt

 

[cute fuzzy bunny]

Re: Covering a mortgage without losing your ass(et

Yeah Nords, I have to disagree that cash isnt king when buying. The seller had a buyer that farted around for 2 months on financing before it became clear that they werent going to get it and they had to find a new buyer. They decided that cash in hand and a 10 day close beat the heck out of waiting a while. A similar home a few lots over that "sold" at the same time sat with the realtors 'sold' sign hanging out front for almost 3 months before we saw moving trucks out front.

Heck, when I sold my mcmansion I had a "preapproved buyer" who after signing the paperwork ended up doing some crazy 100% financing thing with two different banks in different parts of the country and our "quick close" dragged out to 10 weeks. While a half million of my assets was sitting on its foundation waiting release, we had a nice run-up in the markets that I couldnt apply that money to.


Further, in this part of california its traditional for the buyer and seller to split all closing costs 50/50. With cash, our closing costs were a few hundred bucks instead of a few thousand.

No contingencies, getting the cash right now, no uncertainties and lower costs. Its an advantage. Or I'm one heck of a negotiator, because a dozen homes sold in my subdivision within a month or two and the price on mine sticks out like a sore thumb.

 

[sgeeeee]

Re: Covering a mortgage without losing your ass(et

. . . Holy cow.  Drinks for everyone on me!

. . .

I've spent the last 3 days on a camping trip and am not about to read everything that has been posted during that time. But I did catch this line.

Please send my drink to me ASAP.

 

[John Galt]

Re: Covering a mortgage without losing your ass(et

I bought a new winter coat yesterday (on sale - 50%
off) and before selecting it I got into a conversation with another shopper (I talk to everyone. My wife thinks it is
quite interesting). Anyway, this guy is 57 years old, has
multiple credit cards, owned a couple of houses and claims he has never paid any interest to anyone in his
life. I congratulated him and opined that he was
pretty unusual. Man, if that was the norm our economy would grind to a halt.

John Galt

 

[sgeeeee]

Re: Covering a mortgage without losing your ass(et

I bought a new winter coat yesterday (on sale - 50%
off) and before selecting it I got into a conversation with another shopper (I talk to everyone.  My wife thinks it is
quite interesting).  Anyway, this guy is 57 years old, has
multiple credit cards, owned a couple of houses and claims he has never paid any interest to anyone in his
life.  I congratulated him and opined that he was
pretty unusual.  Man, if that was the norm our economy would grind to a halt.

John Galt

Did the man own a kayak? folding or otherwise?

 

[cute fuzzy bunny]

Re: Covering a mortgage without losing your ass(et

Please send my drink to me ASAP.

You and GDER need to re-read my original post.

"Drinks for everyone are on me".

GDER's drink is on my knee and will be consumed shortly.

You can guess where YOUR drink is...

 

[sgeeeee]

Re: Covering a mortgage without losing your ass(et

. . .
You can guess where YOUR drink is...

In your kayak

 

[cute fuzzy bunny]

Re: Covering a mortgage without losing your ass(et

In your kayak

Noooo..but I guess you could call it that as a euphemism...

 

[Nords]

Ass(ets) redux.

We refinanced again this month-- barely nine months after the last refinance (which has already paid for itself).

This time we went from NFCU's 5.5% to 5.375%. Zero points, less than 80% LTV, no cash out, applying online, no documentation or appraisal, and haggling with the title company brought the total refinancing cost down to $1151. Our payments dropped by $44 so this time it'll take 27 months to pay back the refinancing costs.

Not that I'm in any hurry to do this again. One eighth of a point was almost not worth the hassle but the zero-points rates have already "zoomed" back up to 5.625% after hovering at/below 5.5% for the last year. (So I'm calling the bottom on the best rates in the last 40 years.) We could have paid points for a lower rate but I think that points are priced for the bank and not the borrower. (Saluki, Brewer, am I right or wrong on that?)

Back in Oct 2004 (the beginning of this thread) we moved a large chunk of our portfolio, including the refinancing proceeds, into the S&P600/Barra Small-cap Value ETF (IJS). I'm tracking this on a spreadsheet with the assumption that we subtract 15% from dividends (to pay taxes) and reinvest the remainder (which Fidelity brokerage does for free). The ETF is up 18% since then and reinvesting dividends has boosted the return to 18.7%. Sure it looks smart now, but let's give this another 29 years before making a judgment call. We could look equally stupid next spring but we'll keep reinvesting those after-tax dividends.

By the time this mortgage is paid off I'll be nearly 75 years old.

Here's an interesting commentary on financial fiduciaries. Two weeks after the Oct 2004 refinancing the NFCU cold-called to offer us a HELOC at zero cost (based on the paperwork from the refinancing closing). We graciously accepted (it's only about 10% of the home's equity) since it's just one more source of emergency cash that we don't intend to use. NFCU even paid a notary to come out to our back lanai for the paperwork.

Two weeks after this latest refinancing we got another NFCU cold call offering us another HELOC. Spouse graciously accepted and the caller got busy on the paperwork. About two-thirds of the way through the script my spouse said "How does this work-- do we use our current checkbook or do you send us another one?" The caller asked "What do you mean by your 'current checkbook'?" Spouse told her we still had the first HELOC and the caller said "I'm sorry, you must be mistaken, we don't show that on our records here." Spouse said "Sure, whatever you say! When do we sign?" The caller said "Uh, we'll call you back..." and we haven't heard from them since.

 

[Patrick]

Re: Ass(ets) redux.

We refinanced again this month-- barely nine months after the last refinance (which has already paid for itself).

This time we went from NFCU's 5.5% to 5.375%.  Zero points, less than 80% LTV, no cash out, applying online, no documentation or appraisal, and haggling with the title company brought the total refinancing cost down to $1151.  Our payments dropped by $44 so this time it'll take 27 months to pay back the refinancing costs.

It looks like 0.5% less for the 15-year mortgage.  What would your payments have been with that vs. the 30-year?

 

[brewer12345]

I think that points are usually priced such that you would have to stay in the loan for seeveral years to break even. I don't pay points as a rule because I know that the likelihood that I will keep any given loan long enough to come out ahead (positive NPV) is pretty slim.

Nords, I actually got my sister to lock in a refi about 4 weeks ago right before rates shot up again. She is going to refi a higher rate mortgage, plus the house has appreciated enough that she will be able to roll the pigggyback 10% HELOC into the first mortgage and stay at 80% LTV.

 

[Nords]

It looks like 0.5% less for the 15-year mortgage. What would your payments have been with that vs. the 30-year?

"Extremely painful." They would have risen by over 40% and destroyed cash flow.

While a 15-year-mortgage does a great job of reducing the loan's total interest, the shorter term is a little tight on the odds of the stock market returning more over that period. I agree that it'd have to be a heckuva bear market, but it could also be a miserable prolonged sideways market... kinda like Japan or the one we're in now.

But if we can keep getting cheap money to play the stock market's long-term (30 year) performance, while handling the cash flow within the confines of my pension payments, then I'll never pay off a mortgage. If I had to dip into the investment to make the payments then I think the whole project would become much more risky and subject to loss of capital.

With mortgages at their lowest rates since the early 1960s, this may have been a once-in-a-lifetime arbitrage. But if I can still do the math in another 40 years then I'll keep an eye on it.

I think that points are usually priced such that you would have to stay in the loan for seeveral years to break even.  I don't pay points as a rule because I know that the likelihood that I will keep any given loan long enough to come out ahead (positive NPV) is pretty slim.

Nords, I actually got my sister to lock in a refi about 4 weeks ago right before rates shot up again.  She is going to refi a higher rate mortgage, plus the house has appreciated enough that she will be able to roll the pigggyback 10% HELOC into the first mortgage and stay at 80% LTV.

Our breakeven on points was usually 8-9 years. For a small drop like this (less than 0.5%) it might have been decades.

Excellent news for your sister, now all she has to do is stay in the house for a while and defy the national averages!

NFCU is pretty slow to raise rates and very quick to drop them, so I wonder if they're a harbinger of the industry. A rise of this magnitude this quickly from NFCU may signal that long-term rates are going up again and that the yield curve isn't going to invert after all. Which means that the next decade could include a situation where our 5.375% mortgage is less than the after-tax return on long-term CDs. Life is good!