|
|
|
|
| The Basics | Fixed vs. adjustable: Which mortgage wins?
|
The home-loan dilemma is particularly crucial as interest rates begin an almost certain rise. We take a look to see who came out best in years past.
By Liz Pulliam Weston
Alan Greenspan was right -- but perhaps not as right as he thought.
The Federal Reserve chairman has said that people who opted for adjustable-rate mortgages have been the winners in the recent low-rate environment. He opined that homeowners could have saved tens of thousands of dollars each had they abandoned fixed-rate loans for adjustables as rates skidded to generational lows.
No one can argue that adjustable-rate loans have saved borrowers money in recent years. But the savings might not have been as spectacular as Greenspan thought.
Furthermore, fixed-rate loans are usually the winners in rising-rate environments -- the kind that many economists predict well be facing soon.
Just how do they work out?
We at MSN Money wanted to see how borrowers would have fared with each type of loan -- fixed versus adjustable -- in each of three very different interest-rate environments:- 1977 to 1983, when rates rose to spectacular levels with the fixed rate topping out at over 18% in the fall of 1981.
- 1987 to 1993, when rates jumped up before falling back down.
- 1997 to 2003, when rates bounced around a bit before dropping sharply.
For any experiment like this, you have to make a number of assumptions which may or may not reflect an individual borrowers experience. I tried to make my assumptions as realistic as possible. If youre interested in the gory details, you can read about those assumptions below.
Another challenge: Adjustable-rate loans didnt really come into their own until the 1980s. Fannie Mae bought its first adjustable in 1981, and Freddie Mac only started tracking them in 1984. So rather than using Freddie Mac average rates, as I did for the other loans, I constructed a hypothetical loan rate with the help of Freddie Mac economist Michael Schoenbeck. (Essentially, we used one of the most commonly used benchmarks, the 1-year Treasury constant maturity rate, plus what Schoenbeck said was an average margin of 2.75 percentage points. Then, we knocked a percentage point off of that to determine the teaser rate.)
For both types of loans, I assumed the borrower took out a $200,000 mortgage in January and held it for seven years, or a total of 84 payments. I totaled all the payments made plus any refinance costs along the way. Then, I subtracted the equity built up over time to arrive at the total loan costs.
I didnt factor in tax effects, which would have varied widely, or an investment return for any savings the borrower might experience. (Besides ridiculously complicating the experiment, such a calculation may be moot, because Im not convinced many borrowers actually invest their savings. But youre welcome to give it a shot and tell me your results, if youd like.)
| Fixed-rate vs. adjustable-rate mortgages 1977-1983 | | 30-yr. fixed | Rate* | Payment | 1-year ARM | Rate* | Payment | Comment | | 1977 | 8.72% | $1,569 | 1977 | 7.64% | $1,283 | 6.64% teaser | | 1978 | 9.01% | $1,569 | 1978 | 9.71% | $1,552 | 8.64% cap | | 1979 | 10.39% | $1,569 | 1979 | 13.05% | $1,834 | 10.64% cap | | 1980 | 12.88% | $1,569 | 1980 | 14.73% | $2,126 | 12.64% cap | | 1981 | 14.90% | $1,569 | 1981 | 17.63% | $2,126 | 12.64% cap | | 1982 | 17.48% | $1,569 | 1982 | 15.60% | $2,126 | 12.64% cap | | 1983 | 13.25% | $1,569 | 1983 | 11.66% | $1,988 | | | Total loan cost | | $118,500 | Total loan cost | | $147,000 | |
|
*Prevailing rate in January of that year, per Freddie Mac
As you might expect, the fixed-rate loan was the clear winner. Interest-rate caps on the adjustable-rate loan prevented its payments from reflecting the full brunt of the rate increases between 1977 and 1982, but the borrower still paid a heavy price with annual payments that rose by more than two-thirds over the course of the loan.
Checking out the 1987-1993 period
Interest-rate caps werent triggered in our next example, despite rates that spiked in the early years -- before tumbling at the end:
| Fixed-rate vs. adjustable-rate mortgages 1987-1993 | | 30-yr. fixed | Rate | Payment | 1-year ARM | Rate | Payment | Comment | | 1987 | 9.20% | $1,638 | 1987 | 8.07% | $1,415 | 7.62% teaser | | 1988 | 10.38% | $1,638 | 1988 | 9.37% | $1,658 | | | 1989 | 10.73% | $1,638 | 1989 | 11.19% | $1,920 | | | 1990 | 9.90% | $1,638 | 1990 | 9.92% | $1,739 | | | 1991 | 9.64% | $1,638 | 1991 | 9.25% | $1,647 | | | 1992 | 8.43% | $1,638 | 1992 | 6.58% | $1,308 | | | 1993 | 7.99% | $1,638 | 1993 | 5.91% | $1,231 | | | Total loan cost | | $125,000 | Total loan cost | | $116,500 | |
|
Our fixed-rate borrower paid about $8,500 more for the security of an unchanging payment -- or about $101 a month.
A look at 1997 to 2003
The price for security was a bit lower in our final example: The fixed-rate borrower paid $5,500 more in total, or about $65 extra a month. Adjustable-rate borrowers benefited from teaser rates that were unusually low compared with the fully indexed rate that would have otherwise applied, not to mention a dizzying drop in rates during the last years of the loan.
| Fixed-rate vs. adjustable-rate mortgages 1997-2003 | | 30-yr. fixed | Rate | Payment | Comment | 1-yr. ARM | Rate | Payment | Comment | | 1997 | 7.82% | $1,443 | | 1997 | 8.26% | $1,143 | 5.56% teaser | | 1998 | 6.99% | $1,443 | | 1998 | 8.32% | $1,401 | 7.56% cap | | 1999 | 6.79% | $1,279 | Refinanced | 1999 | 7.31% | $1,368 | | | 2000 | 8.21% | $1,279 | | 2000 | 8.63% | $1,540 | | | 2001 | 7.03% | $1,279 | | 2001 | 8.39% | $1,509 | | | 2002 | 7.00% | $1,279 | | 2002 | 5.01% | $1,119 | | | 2003 | 5.92% | $1,279 | | 2003 | 4.24% | $1,038 | | | Total loan cost | | $97,500 | | Total loan cost | | $92,000 | |
|
The savings the adjustable-rate borrower realized arent small potatoes, but theyre hardly the tens of thousands of dollars Greenspan predicted.
So, what can we learn from all this?
Well, I was vividly reminded how nasty interest rates can get -- and how handy caps on adjustable-rate mortgages can be. (Some mortgage professionals like to downplay those caps, as if interest rates could never soar enough to trigger them. Ha!)
The cost of avoiding uncertainty
I was also reminded that security often comes with a cost. The majority of buyers who opted for fixed-rate loans in recent years did pay a price -- but they may have thought the cost worthwhile to be able to avoid uncertainty.
Finally, no matter how many times we do this experiment, we wont be able to predict what will happen in the next seven years or the years after that. The past is no guarantee of the future. If youre willing to roll the dice on an adjustable, be my guest -- as long as youve done the math and know youll be able to swing the payments if rates climb.
Given the current interest-rate outlook, though, many borrowers will opt to lock in todays low rates, at least for as long as they expect to be in the house. Thats hardly a foolish choice, despite what Greenspan implied, and may, in fact, prove to be a pretty smart move.
The gory details
OK, now the fine print: Here are the assumptions used in creating the charts.
For the fixed-rate loan: I assumed the borrower paid the average rate reported in January of the originating year (1977, 1987 and 1997). I also assumed the borrower would refinance if the rate in any subsequent January was at least one percentage point below the borrowers current rate.
The only exception I made was for January 1993; while the rate was more than a point lower, the costs of refinancing couldnt have been recouped in the year the borrower had left on our assumed holding period.
I could have assumed the borrower would refinance anytime the rate dropped a point or more during the year, but I was trying to be realistic. Most people dont catch the exact bottom of any interest rate cycle. A borrower who did manage to catch the lowest rates each time, however, might have lowered the cost of the fixed-rate mortgage.
To determine refinancing costs: I used Freddie Macs average points and fees for the month in which the refinancing took place. These refinance fees were added to the annual payments to help determine the loans costs. Again, these might under- or overstate the costs an individual borrower might face.
Also, I had the borrower refinance to another 30-year mortgage, rather than to a shorter period, since typically most borrowers have done just that, particularly if theyre refinancing during the loans early years. Such a refinance lowers the monthly payments but slows the rate at which the borrower builds equity.
To determine the mortgages total costs, I subtracted the equity buildup from the sum of the loan payments and any refinance costs.
For the adjustable-rate loan: I assumed the mortgage would adjust once a year on the anniversary of its origination (in January, in other words).
Payments on the adjustable rate loans were determined by adding a margin to an underlying index rate. For this experiment, the index is the 1-year Treasury constant maturity rate for the prior December, plus the average margin reported by Freddie Mac for the month in which the loan was taken out. If no margin was available, as it wasnt for 1977, I used 2.75%.
Most adjustable-rate loans also come with caps that typically limit increases to 2 percentage points a year and no more than 6 percentage points over the life of the loan. Those assumptions are incorporated in the calculations.
Adjustable-rate loans, however, typically start with a teaser rate that can be one or two percentage points, or even more, lower than what the margin and index would otherwise equal. Freddie Mac reported average teaser rates for 1987 and 1997, which I used to determine the first years payments as well as the base rate for applying any caps that came into play.
As with the fixed-rate mortgage, I determined total cost by subtracting any equity growth from the total of all the annual payments made over the seven-year period.
Liz Pulliam Weston's column appears every Monday and Thursday, exclusively on MSN Money. She also answers reader questions in the Your Money message board.
|
|
|
|
