Mortgage - fixed versus interest only

I have taken out several interest-only mortgages and lines of credit and found that they worked well. If cash flow is an issue, they can significantly increase your monthly cash flow. This can be good if maintenance expenses are high and you intend to keep the property for a relatively short period and then sell. During the early years of a fixed term mortgage, there is not very much increase in equity over the short term, since most of the monthly payments go for interest anyway. But interest rates change. Interest only loans definitely are not a good idea if you intend to live in the house yourself for 30 years.

Go with a fixed rate. The interest only is like wetting the bed, at some point you will have to get up and change the sheets.

First, you realize, you are not saving $300/mo. You know this I hope. You are saving $507.50/yr in interest based on the rate difference.

Now I have a few questions for you: Can you not afford the extra $300/mo? Will something change in the next 5 years that you will either likely move or will have a higher income? What happens to your mortgage in 5 years? Does it turn into a 25 yr fixed? What is the maximum rate it can be? Can you afford that?

Knowing nothing else about you, I can only tell you this; we frequently revert to the 'how would you feel' questions here. Personally, I'd feel more comfortable with the fixed, fully amortizing mortgage. At some high rate, in a positive yield curve environment, I might consider a variable, but not here, for most of my adult life, that 6-3/8 looks like a sweet rate. Choose wisely. JOE

Thanks Joe, Here are all the answers. I can afford the extra $300.00 but wanted to increase my cash flow. I was planning to move in 3-4 years and might sell the house in that case. After 5 years, based on the rate then I have to take up a fixed rate. So worrying that at that time if the fixed rate is too high, I might be in trouble and would be forced to sell the house. If the market is down then, I might get a really bad deal.

My logic of going with the Fixed is that though interest only sits in well with my concept of selling in 3-4 years, these are the problems I see If the market is down, I would not really be able to sell it. In that case, if after 5 years I get a high interest, and the real estate are still down, i would really be under the gun to sell it and might not get the price I get.

With Fixed, though I will pay more, I am contributing towards the principal a bit. Also, there would be no pressure on me to sell it after 5 years.

Could you explain this part to me "First, you realize, you are not saving $300/mo. You know this I hope. You are saving $507.50/yr in interest based on the rate difference. "

Here were some calculations I made. Although, the fixed rate is about $300.00 higher, I am contributing about $120 towards principal. So at the end of 5 years, this extra sum is 12*5*300 = $18000

I can take some consolation with the fact that 12*5*120 =$7200.00 is towards the principal.

But based on your "feel" I think I will stick with the Fixed rate.

Thanks a lot

(snipped my original reply, as quoted below)

I looked at it this way; $203K x 6.125% = $12433.75 $203K x 6.375% = $12941.25

The difference is $507.50 for the first year in interest. Of course, your cash flow is improved, with no principal payments for 5 years, but as you state, the risk that the rates will be high, and market depressed when you need to sell, is there. It's all risk vs reward, if the dollars involved were higher, it would be different. I'd not take on this risk for only $2500 over 5 years. Enjoy the reduced principle, and I wish you a return to moderate price appreciation. JOE

Thanks guys!!! changed to Fixed rate!!!! Thank you all

I'm a little confused. A 1050/mo payment at 6.125% interest only would be a loan on closer to 206000 not 203000. Likewise 1350/mo at 6.375 would be on a loan for 216,000. Why the difference?

For 203,000 I get: interest only @ 6.125% = 1036/mo

30y fixed @ 6.375% = 1266/mo

Cost of loan after 5 years: interest only: 62200 Fixed : 62700

Is fixed rate really worth saving $100 per year?

I'm pretty new at this so can someone check my numbers?

Where did you get that monthly payment for the fixed-rate mortgage? According to my calculation, it should be $1266.46.

So far, all the posts have ignored the effect of paying down the principal. Since we're talking short-term, that's generally a pretty good approximation. However, since we're also talking about a potential savings of less than $1K, I think we have to take it in to account. So I drew up a spreadsheet for you. You can find it at:

As you can see from the spreadsheet, the MOST you will ever save from the interest-only mortgage is $858 (at 38/39 months). After that, your savings will start to decline, due to the principal pay-down. By month 76, you're in the red, meaning you've lost money on the interest- only mortgage.

On a more practical note, I think you should shop around more for mortgages. On BankRate.com, they say the national average for a 30- year fixed is currently 5.79%. In my specific area, I see offers for

5.25% with 1 point and 5.375% with no points. Of course, you can never actually get those rates, but they are sort of a barometer.

--Bill

You are right, mostly. I didn't calculate the payments, just interest delta. The fixed payment is 1266.46 as you say, and the interest only is

1036.15, so cash flow is improved by only $230, but interest saved is still only $500 or so the first year. What I thought most important here was the risk at the fifth year. With only $2500 in his "saved" kitty ($15,000 in total saved payments, if you wish) his payment will need to jump from the $1036 to $1323 to amortize the full loan in the remaining 25 years. 4 years to burn through 15K, but all the risk of the rate going up. OP didn't state the index or terms of the 6th year adjustment, but with what we know and how we replied, I believe he made the right choice. Your math and the 4th year crossover confirms how slight a benefit he'd even have in the 5 years he'd pay interest only. Good sheet. JOE

I hear you. My point with the spreadsheet is that the interest delta changes every month. By computing the amortization, you can clearly see where the delta switches from positive to negative and where the break-even point is. The spreadsheet also clearly shows the effect of building equity.

Good call. I completely forgot about the conversion of the loan. I just assumed it would stay interest-only at 6.125% forever. I suppose there's a chance the future interest rate could be lower. Of course, a snowball also has a chance in Hell...

--Bill

Amazing how a simple question/calculation can become so complex. Given the opportunity cost of money (i.e. you need to attribute a return on the monthly money saved) how about assuming OP makes the same payment on either mortgage. Both start at $203K, 6.375 on fixed, 6.125 on adjustable, but he pays the $1266.46 on both. After 5 years, the fixed is $189755, and adj, $186879. A delta of $2875. If we agree to ignore the slightly different tax implication, this would seem to be the better apple to apple comparison. The 'return' is $2875 vs the 'risk' of that rate change 5 years hence. (I'll read any reply, but I think I've beaten this to death in my own head) JOE

I am naive about the sales aspect, but didn't banks begin to push interest-only loans in a period when interest rates were low? If that is true, it is more likely that after 5 years the rate on one of these will be higher instead of lower.

No they could certainly be lower, if the US went into a recession.

Or rather when prices went soaring up.

At which point, the only way buyers could afford homes was 'interest only'.

The problem being that somehow they have to *repay* those loans. And housing prices might not rise as much in the future as they have in the past (almost certainly will not, in the next 5 years). And with low inflation, the amount you have to repay (in constant dollars) is much higher, because inflation isn't eating away the value of the principle (principal?).

Ther latter for sure!

I was suggesting that interest-only rates (and ARM rates) are more likely to increase than decrease, so that these types of loans are not good fors someone taking out a mortgage at the present time and gambling that the rates will stay the same or go down in future years. But it seems to me they could be OK for someone needing cash flow and intending to sell in a few years.

OK I misunderstood maybe. Given that the yield curve is normally upward sloping, a fix should cost you more than a floating rate.

so that these types of loans are not good fors

Buried under there, for most homeowners, is an assumption that prices go up, only. Which they do in the long run, but not necessarily on a

5 year view. I suspect most homeowners think that prices rise by an average of 10% a year, long run, whereas in fact in real terms they have averaged about 1% a year (adjusted for the qualilty of houses, which has improved eg a US home now is 60% larger than a US home in 1970). One reason being some 'high profile' markets (like English Bay in Vancouver, or New York/Boston/Los Angeles) have risen much, much faster than that, historically. But then there's Buffalo or Detroit, where housing prices now probably aren't much above where they were in 1990.

If one sells one's home every 5 years, and housing prices move up at their historic averages (rather than that of the last 10 years), and one has an interest only mortgage, then in 25 years one could wind up with no significant housing equity.

In the UK, where 'interest only' mortgages are a significant fraction of the market (perhaps as much as 40%?), this is going to be a real problem if we have a market downturn.

The banks and building societies insist that you have an 'alternate' plan for saving up the repayment amount (stock market savings scheme, etc.). Of course, what people do is cancel that plan, as soon as they get the mortgage money paid over.

Linking the repayment of your house to the performance of the stockmarket also always struck me as a risky strategy. Endowment (with profits) mortgages were supposed to smooth out that risk, but in fact we had the selling scandal of all time (other than the privatisation of pensions) because endowments were sold on forecasts of 10 and 12% pa returns in the early 90s, and achieved returns have been as low as zero *after commissions*.

Funny how policy design never takes account of human incentives, particularly salesman's incentives ;-).

I've never thought of the housing "bull market" like this... that a major component of average price increases over the decades is that we're comparing today's bigger homes with more amenities to historically smaller homes.

Does anyone have any other thoughts on this - or any stats?

-HW "Skip" Weldon Columbia, SC

The effect is really only strong for really long time horizons. One of the best series available is for the Herengracht, one of the poshest canals in Amsterdam, since the 17th century. The houses haven' t really changed in size or structure (although they now have plumbing and heating) so they are quite comparable. And though Amsterdam has had its ups and downs, it's never done a Detroit, say. It remains an important Dutch city in a wealthy country.

But you could see how US housing prices would be 'rising' (on the basis of cost per home) but actually not going up anywhere as much (on the basis of price per square foot, or amenities like number of bathrooms etc).

A similar effect has taken place in cars: the standard car now has no- skid brakes, airbags etc. that would have been almost unknown in 1990, or only on the most luxurious of vehicles. Not to mention GPS etc. which are becoming standard equipment.

Rest assured, the movement in US housing prices since 1995 is very real, and upwards, and (perhaps unprecedentedly) strong across almost the whole country, even if you adjust for housing quality. If you look at 1. real income growth 2. population growth 3. fall in real interest rates (as measured by US TIPS) then it doesn't look like prices have risen ahead of fundamentals in a number of markets. In a number of other markets (California) prices are way ahead of fundamentals.

I remember a graphic (from Shiller Weiss, as below) which showed (as far as I can recall) that housing in Boise, Idaho, was the most undervalued, and the top overvalued markets were Los Angeles/S. California, Las Vegas and Washington. This was in the Atlantic magazine, I think in the spring of 2005.

Zoning has played a huge role. Edward Gleaser at Harvard has shown that in certain markets (mostly coastal ones eg Bay Area, Boston, NYC) housing has risen far, far faster than fundamentals such as income, due to a profound tightening of neighbourhood zoning laws since

1970*. The effect, in terms of discouraging industry and causing it to relocate, is actually quite tangible-- hence the superurban sprawl of the great northeastern cities.

• an example. In the 1920s boom, the brownstones on the upper East and West Side of NYC were knocked down and replaced with 10-15 story apartment buildings of the kind that line Central Park West. In the post 1990 period, they have not been replaced, Hong Kong of Shanghai like, by 40 story plus modern blocks. Manhattan would look like Hong Kong, if it were not for powerful political forces that prevent Manhattan expanding to meet the demand for apartments. Economic forces would do that, if they were allowed to.

pp 304-05 discusses the index composition a little bit

Shiller also discusses this in his intro to the SECOND edition of 'Irrational Exuberance'.

Skip, I wrote about this on my site,

to deconstruct the source of the 'boom'. Since much data is offered as 'mean' I found a study that offered me a mean home size for new construction in 2004 of 2330 sq. ft., vs 1400 sq ft in 1970. In just those three decades the home size grew 1.5%/yr.I found I didn't even need to adjust for size, that the drop in interest rates from 1981 to 2005 dropped the payment on the mortgage (despite a 4 fold price increase) so that the number of hours one needed to work was less to afford the mortgage in 05 than in 1981. The increase in size was just a bonus. The true cost* of houses actually fell during that period.

*True cost = the number of hours of labor needed to purchase an item. Even though the cost of the home increased faster than inflation, it did not increase faster than wages.

JOE

That is interesting, and I doubt if it is usually taken into consideration in making these comparisons. But I am wondering if size can be equated to quality. I have heard people say of current houses, "They don't build them like they used to," referring to the quality and thickness of the lumber, the amount of insulation, how long the roof is expected to last, etc. Yes, the modern houses have a lot of special features like marble counter tops,, superior lighting fixtures, built in internet access, double-glazed windows, what have you, and especially size and square footage. But I have seen houses and condos built 10 or 15 years ago that appeared strikingly modern at first and today look decrepit.

I my original speculation I was thinking more of investors than homowners relocating every five years.