Printer Friendly

The physics of mortgage banking.

What would happen if you crossed physics with mortgage banking? People might start talking like this: "The HE/Index might be thought of as a measure of the refinance 'potential energy' in a market." Don't laugh, because that is an actual quote from a research report titled "Predicting & Understanding Regional Mortgage Prepayments" published by Reserve Financial Systems Corp., Miami.

And if there's any area of mortgage banking where rocket scientists might said to be at work, it's in the area of predicting prepayments--or attempting to more accurately predict prepayments, we should probably say. And if there was ever a year when mortgage bankers might appreciate the fruits of such labor, it would be 1992.

The HE/Index mentioned above is the product of the minds at work at Reserve Financial. The RF Home Equity Index is a "yardstick of average borrowers' equity" in all the major metropolitan housing markets across the country. The reason why such a measurement tool is important is because Reserve Financial has found that the amount of current equity in a loan closely correlates with probability of prepayment. They have tested out this correlation based on actual prepayments that occurred during March through June of this year on a group of more than 800,000 loans originated between 1986 and 1992.

To pinpoint current equity, Reserve has come up with a way to produce adjusted loan-to-value (LTV) ratios for portfolios based on where homes are located and the pattern of home price changes that have occurred in that particular area, as well as allowing for scheduled amortization. The real estate price data built into the RF Home Equity Database includes: "historical home price appreciation for 4 regions, 9 sub-regions, 50 states and more than 70 metropolitan housing markets for each origination year from 1981 to |first quarter~: 1992."

The Home Equity Index is calculated by using information about original LTVs, home appreciation by year of origination, and is weighted by the distribution of outstanding loans by year of origination. That together then yields some measure of how much built-up equity the average borrower in that market tends to have. The end result is the higher the Home Equity Index number for that market--the greater the likelihood of prepayments.

And if you want to get a sense of how loans from a particular state or metropolitan area have been paying off compared to a national norm, Reserve's quarterly reports include an extensive breakdown of states, regions and metro areas with a corresponding Prepayment Probability Index (PPI) based on the 800,000 loans in its data base. The reports also give the PPI for particular loan types and coupon rates by state. The PPI index is the ratio of the actual, observed prepayments in a category or tranche of loans divided by the national average for that same group.

So what does all this mean to you? Well, if you have a servicing portfolio concentrated heavily in Wisconsin and Michigan, you might want to know that you picked the two fastest prepaying states to build a portfolio around--at least based on prepayments observed in the second quarter 1992 in Reserve's group of 800,000 loans. The top five fastest prepaying states with their prepayment indices were: Wisconsin (1.77), Michigan (1.51), California (1.47), Illinois (1.46) and Oregon (1.41). An index of 1.00 would mean the area matched the prepayment performance that was the national norm, while a 1.77 means that unpaid principal balance-weighted prepayments were 177 percent of the national norm for that group.

The slowest five prepaying states were: Louisiana (.28), Alabama (.42), Tennessee (.46), Georgia (.49) and Oklahoma (.56).

We talked to some mortgage bankers to see how all these theories and forecasting tools might square with the real world. Bob Densmore, executive vice president, Source One Mortgage Services Corporation, Farmington Hills, Michigan, confirmed that the fastest prepaying region of the country for his portfolio was "by far the West." Source One's roughly 600,000 loan portfolio has approximately 19 percent of its loans based in California--the company's largest single-state concentration. The top four states where his servicing is most concentrated are California, Washington, Michigan and New York.

Densmore said that through June 30 of this year the payoff rate, on an annualized basis, of loans from California was "almost 40 percent." He added that both Washington and Colorado were "right up there" with high annualized payoff rates. Source One has the least prepayments in loans it services from the South, he said. That overall pattern squares with the Reserve Financial data, which shows the PPI for the West being the highest at 1.33 and for the South being the lowest at .64.

But Densmore observed that some of the states represented in his portfolio are showing annualized payoff rates of well under 20 percent. He noted that appreciation patterns in those states are not the only variable accounting for this relative slowness in prepayments. He added that "Theories are all good, but practicality is something else." He added that given the kind of low-rate market we are in, "you would think no state would be under 20 percent |in its annualized prepayment rate~." He said that his slowest prepaying state is New York at an annualized payoff rate of 14 percent. Part of the reason for this has nothing to do with local market appreciation. Apparently, New York state, because of regulatory requirements, is a terrible market to refinance in from a cost and procedural standpoint.

Another example that shows there are more variables than simply appreciation patterns determining prepayment activity is Source One's experience in Florida, where Densmore says the payoff rate is running at roughly 14 percent. The Source One executive says that quite likely the older borrower population in that state is the reason for the slower payoffs, because those borrowers are settled into their final retirement homes. Finally, also seemingly at odds with the theory of equity being the predominant factor determining prepayments is Source One's experience with loans from Massachusetts. Densmore says the annualized prepayment rate for such loans in his portfolio, as of mid-year 1992, was about 28 percent. But the well-documented recent depreciation in the Northeast would seem to work to slow prepayments in Massachusetts substantially, if the equity theory is correct.

But as Densmore explains, "There are a ton of variables" driving prepayments and his company has a whole refi unit dedicated to that aspect of the business. Source One stopped buying servicing toward the end of the second quarter, because "people were paying up unbelievably," Densmore said. Up until September, Source One had been able to keep its servicing portfolio at the same size by generating enough in new originations to cover the runoff from refinancings. But the runoff from Source One's servicing portfolio that hit in September finally exceeded the lender's production volume, after the company had managed to stay even in July and August, Densmore said.
COPYRIGHT 1992 Mortgage Bankers Association of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Boardroom View
Publication:Mortgage Banking
Date:Oct 1, 1992
Previous Article:Secondary market.
Next Article:The dynamics of overpriced markets.

Related Articles
Boardroom view.
Boardroom view.
Boardroom view.
Boardroom view.
Boardroom view.
Secondary market.
The mortgage banking food chain.
The new North American.
A mature market for offshore mortgage processing?
As Countrywide goes, so goes the industry.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters