Printer Friendly

The new thinking on prepayments.

Up until now, virtually all models to predict prepayments were built around interest rates. But this is about to change. The new research in serious circles is that equity may be what really drives prepayments.

The most recent surge in refinancing and the roller coaster of the last two years has created a sustained interest in prepayment forecasting. But the dramatic fall in interest rates appears to have misled our thinking about prepayments in general and, more importantly, the crucial role that equity appreciation is playing in decisions to pay off mortgages.

To put this in perspective, consider relatively recent experience in the housing market. Anyone who bought their house during, or since 1989 has had little or no equity appreciation. Thus, these homeowners are highly unlikely to resell their homes, regardless of where interest rates are. Furthermore, those owners who have seen their equity actually decrease, not only will be reluctant to sell, but may find it difficult or financially unattractive to refinance. This is rational and easily understandable behavior and it is clear that both homeowners and their mortgage lending counterparts generally behave quite rationally.

This observation is central to the idea that changes in equity are a principal cause of prepayments. Homeowners are consistently aware--either in the back of their minds or by careful calculation--of their current equity position and what it means in terms of their financial options. When most people stop and think about this, they quickly realize it's true in their own situation. Having asked this question of many people recently, the response has always been that they had a pretty good idea of their equity position. This regular "mark-to-market" behavior is entirely to be expected, because a home is generally the largest single investment most people ever make.

Equity's role in prepayments

Understanding future mortgage prepayments has always been the most critical factor in valuing loan portfolios, servicing and all forms of mortgage securities. Everyone in the industry knows what a difficult task this is. Anyone active in the mortgage markets today has seen how quickly and strongly unexpected prepayments can impact prices and yields. In spite of the significant efforts major firms have put into prepayment research, analysis and modelling, one might be tempted to conclude prepayments are simply too random to permit even adequate prediction.

Because of the huge dollar impacts of unanticipated prepayments, it is important to consider anything that might be a fundamental of prepayment behavior that has been overlooked by many modelers. Most prepayment models today are primarily based on interest rate movements--almost to the exclusion of any other major variables. However, the ability of that particular conceptual framework to adequately explain prepayments needs to be revisited.

This article maintains that equity change, as defined by loan-to-value (LTV) ratios may provide a more comprehensive and robust explanation of prepayment behavior, and that home prices and interest rates actually are the two secondary causal factors of prepayments. This theory is considered from an anecdotal, conceptual perspective that is the basis for ongoing research into the development of a new, empirically-driven mid-to-long term prepayment model.


Neither equity nor LTV ratios have been considered essential to the determination of prepayment behavior, though in recent years they have been closely linked to prepayments caused by defaults. The reason is that home prices have been more-or-less steadily rising over the past 20 years. This meant that homeowners' equity was constantly increasing, which allowed them to trade up or take out some of their equity profits. Of course, these steadily increasing resales and refinancings were occasionally, and sometimes dramatically, affected by changes in interest rates. This caused attention to be focused more on this one particular variable, that really only influenced, rather than caused the basic prepayment behavior.

Equity change-related prepayment behavior may be described by some existing models in terms of loan aging and turnover. This turnover was attributed to the fact that after so many years in one house, the typical young family grows in size and income to the point where they need a bigger and better house. The fact that this housing stock turnover rate was quickening was ascribed to the economy's growth and greater mobility of the baby boomers. A general review of the literature suggests that no attempt has been made to study the specific underlying causes of this turnover, though it may have been inherently understood that after you own a house for several years, your equity typically appreciates enough to allow you to buy a bigger or better house. This broad understanding also may have been held regarding home equity loans and equity take-out refinancings, since these are so directly related to equity appreciation.

But one of the problems with seriously investigating this topic is that the home has never been rigorously studied as an investment. There has been no statistical analysis of the point at which--in percent increase in equity--homeowners elect to take their profits by trading up or refinancing. Is there a threshold of equity increase over which the probability of prepayment rises dramatically? No one knows, because for many in the emerging field of prepayment research, up until now, it didn't seem relevant.

Another recession or new era?

Today, however, times have changed. In 1989, home prices appeared to have peaked. Since then, price changes have been flat-to-down. Many who owned a home before then lost some increased equity. And many homebuyers purchasing after 1989, lost some of their original equity.

This development has disrupted significantly the housing turnover pattern of the last 20 years. There is simply little increased equity to fuel resales or equity take-outs and, as a result, we have seen fewer transactions or prepayments related to these events. This can be observed in the very slow pace of prepayments at the end of 1990 and in the summer of 1991. Without the strong rate-related refinancing activity that began in the fall of 1991, prepayments would have continued at that slow pace.

The turndown in home prices in 1989 may have coincided with or led the start of the recession. So, it just may be that resales and home prices will edge back upward after the recession ultimately runs its course. But, this recession isn't behaving quite as others have. In spite of winning a war and the deepest rate cuts in memory, there is persistently slow growth. The National Association of Realtors' Housing Affordability Index is at an all-time high, yet there is slow-or-no increase in home sales. Even if, or when home sales notably increase, there is an enormous inventory of homes for sale that must be worked through, before prices can again start to edge back up. An informal poll of local real estate boards recently showed inventories of from one to two years at current sales rates.

A larger concern of some is whether the slowdown in home sales is related to the end of the baby boom generation moving through the housing cycle. Certainly, the maturing of the population is having an impact on the total number of home sales. Figure 1 shows the baby boomers' bulge moving through the nation's population. As baby boomer families settle into their long-term residences, there are fewer first-time and trade-up buyers coming of age in the generations that follow. Even with immigration, population growth has declined significantly.

To the degree these considerations are valid, it becomes increasingly uncertain as to when, or if home sales and prices will rise again. It is not terribly risky to speculate that we may not see the appreciation rates of the mid-1980s before the end of the 1990s. All this taken together suggests it may be some time before home prices again move up in any broad-based way, and that brings us back to the central issue of equity--or LTV--as a principal cause of prepayments.

Four prepayment options

If we can accept, as fact, the notion that homeowners--and lenders--are aware of their current equity positions in specific properties and behave rationally based on that, then we can consider, from their point of view, the four major prepayment event options they face each month. Those four options are:

* Should I sell my house and buy another one?

* Should I refinance my current mortgage?

* Should I make a partial prepayment on my mortgage?

* Should I default on my mortgage?

These four options define the universe of mortgage prepayments. It is enlightening to recreate the decisionmaking process a typical homeowner goes through in regard to each of these options. These hypothetical scenarios will demonstrate how the causal factors under discussion--equity (LTV) and rates--interrelate and will suggest which factor appears more frequently to control the outcome.

Resale or trade

A homeowner may think about buying a new house for many reasons. His or her family may have outgrown their present home. The owner may be doing better financially. He or she may be relocating. Or, lower rates could cause the owner to think about the bigger mortgage their current payment could cover in the new, more favorable, rate environment.

Whatever the reason might be to sell, the next consideration most likely will be how much house can the owner afford. Almost as quickly will come an estimate of current equity and a quick calculation of how expensive a house can be bought using the homeowner's current equity as a down payment. If that's attractive enough, the owner will continue the process. But, if the new house the sellers can afford doesn't represent enough of an improvement, the consideration probably will end there. However, other forces might still cause the owner to move, such as the need of a young family to flee a deteriorating urban neighborhood for the suburbs, or the desire to relocate to a different region of the country where economic conditions appear more robust. But even in these later instances, the issue of equity and down payment constraints will be weighed.

The conclusion being offered here is that equity is a primary factor driving this prepayment decision and rates modify the basic consideration and outcome. Lower rates may push the owner to trade sooner, while higher rates could delay the trade.


The two types of refinancings--rate and take-out--should be considered separately, since they occur for quite different reasons.

Rate Refinancing--Whenever interest rate cuts cause mortgage rates to drop, traditionally homeowners with mortgage rates that were more than 200-250 basis points higher than the new rate would take steps to calculate whether it would make financial sense to refinance. (In the recent refinancing boom, the spread at which homeowners consider refinancing has narrowed to less than 200 basis points in many cases.) Nevertheless, the rough analyses performed by would-be refinancing customers are of varying degrees of sophistication.

If the refinancing appears to make financial sense, then the homeowner and the lender have to determine if the proposed new loan results in an LTV of 80 percent or less. If the owner's equity is less than 20 percent, then the homeowner must either pay down the mortgage or pay for private mortgage insurance. The increased costs of doing so could change the financial incentive to refinance sufficiently that some owners will not go through with the transaction. Therefore, in the case of a rate-driven refinancing, the primary cause is rate, but LTV oftentimes determines if the loan can be made.

Take-Out Refinancing--Equity take-out refinancings are more like resales in terms of why they occur. The homeowner realizes that his or her equity position has increased enough that a trade-up is feasible, based on the equity providing a larger down payment. However, if the homeowner doesn't want to trade-up, there is the option of simply taking up to 80 percent of the profits out in cash through an equity take-out refinancing. If the prevailing mortgage rates are equal to or less than the borrower's current rate, he or she will readily refinance. If rates are higher, the homeowner may think about it and may hold off.

So, equity take-outs are primarily driven by an increase in equity. Mortgage rates are secondary. Only if they have moved much higher since the homeowner's first mortgage, might rates play a role in convincing the homeowner to wait until they drop to initiate the refinancing.


Partial prepayments are the least understood type of prepayment event, though they appear to account for 5 percent to 15 percent of prepayments, depending on the coupon and remaining term on the mortgage. Generally, partial prepayments occur for one of three reasons. Those reasons are because most homeowners:

* Want to own their homes sooner;

* Use curtailment as a form of savings account;

* Want to reduce their costs and effective rate by "investing" at their mortgage rate.

Because so little is known about curtailment, which is the term for unscheduled, voluntary, additional mortgage payments by borrowers--it's awkward to speculate about its causes. However, certain reasonable assumptions can be made:

* If interest rates move down and homeowners' limited equity or other circumstances make refinancing difficult, investing in the existing higher mortgage rate is reasonable.

* As time goes on, curtailment has an increasing effect on hastening relative equity growth.


Default is the smallest component of total prepayments, though the incidence of it has been increasing over the past five years and this is of great concern to the industry. Default has been studied more than any other type of prepayment event, since default is an unscheduled prepayment that someone other than the homeowner has to make.

In terms of the two principal variables under consideration--equity and rates--defaults are solely related to the former. Of course, they are primarily caused by financial distress of the homeowner. When under such pressures, the homeowner makes some very pragmatic assessments. Typically, the homeowner looks at the monthly cost of the mortgage. If it's much greater than alternative housing costs, the owner evaluates his or her current equity. If it's little or nothing, it may be easy to walk away from the obligation. If it is substantial, the owner will find a way, if at all possible, to keep the asset by making the payments.

These are straightforward, well-understood and accepted facts. Pestre, Richardson and Webster's "Default Probabilities and Credit-Adjusted Spreads for Non-Agency Mortgage Securities," Journal of Fixed Income, June 1992, concluded that "the most important variable by far in explaining default is home price appreciation. |Their~ analysis shows that default activity is very negatively correlated with the robustness of the local housing market. Defaults are low during periods of strong housing markets and rise sharply when home price appreciation is sluggish or negative." Figure 2 illustrates the negative relationship between default and home appreciation for selected metropolitan statistical areas (MSAs). Interest rate movement plays little or no role in the decision. Because the decisionmaking process that a TABULAR DATA OMITTED homeowner facing default goes through is so readily understood and accepted, it would seem to support the idea that equity consideration may play an equally major role in the other types of prepayment decisions.

Is equity dominant?

Overall, in considering the relative importance of home equity and interest rates in causing each of the four types of prepayments, it becomes increasingly clear that equity position is a dominant factor. Interest rates are the primary cause only of rate-related refinancings and these can only occur when there are rate cuts. Thus, it would seem that an important variable in prepayment modelling should be equity position, yet most models are primarily driven by interest rates. Therefore, unless it can be shown that equity and rates co-vary inversely, even models that could dependably and accurately forecast interest rates, could not forecast prepayments effectively.

What is more interesting from a conceptual perspective is the continuous relationship of equity changes to prepayments. As equity increases, so do prepayments due to rate or take-out refinancings or resales. As equity decreases, so do prepayments due to defaults. This suggests that there is an orderly structure to prepayment behavior that can be discerned and used as the framework for forecasting future prepayments. Chart 1 illustrates the basic structure of prepayments related to equity, which could alternately be viewed in terms of LTV.

Recent trends

The current situation in the housing market illustrates this equity-driven behavior very well. Interest rates have been falling since December 1990 and yet home sales have fallen over the same period. Prepayments have sporadically been up fairly significantly due to refinancings, but the trend over the period has looked like a roller coaster. Each time rates moved down, refinancing occurred, temporarily boosting prepayments. When those who could refinance did, prepayments dropped until the next rate cut. The pattern went like this: up in the spring of 1991, down in the summer, up in the fall, down in the winter, up in the spring of 1992, down in early summer.

But prepayment speeds are now heading up again based on the most recent rate cut, but they will surely be back down by the end of fall, unless home sales and prices pick up.

During the last year and a half there is no evidence that home equity and rates show inverse covariance. Rates are down, but equity is not up. This is because home sales and prices have not moved up, which obviously would have caused equity to increase.

Perhaps, one might argue that sales and prices lag interest rates. If so, it seems they are currently lagging by such a long period that the relationship is unusable. Further, it is far more likely that home sales depend much more on other variables than rates.

Earlier, we raised the question of whether reaching the end of the baby boom generation's main homebuying years might have altered the supply versus demand equation. Whatever the cause, it is clear that the supply of houses at prices that would give their sellers a meaningful profit, currently exceeds the demand. This means that at some point in the future when home prices and equity have increased sufficiently, there will be a significant, longer term increase in prepayments because of resales and equity take-outs. At that point, this will happen pretty much regardless of where interest rates are, though rates will dampen or accelerate the trend.

This article has attempted to show that homeowners' natural responses to changes in their equity positions might provide a sound behavioral framework for forecasting prepayments. Home prices and interest rates would be the two variables that drove the model, with the former having a more dominant longer-term effect.

A great deal of empirical research needs to be done to validate these theories and to understand the specific patterns of behavior toward changes in equity. Because studies have already shown that defaults are negatively correlated to home appreciation, there are two critical hypotheses to validate:

* That annualized changes in equity for existing homes that were sold were greater than for those unsold.

* That for those borrowers with equal refinancing incentives, more of those with lower LTVs (less than or equal to 80 percent) refinanced than those with higher LTVs (greater than 80 percent).

The imminent reasonableness of the theory that changes in equity are a major cause of prepayments offers new and promising ground for researchers and analysts to investigate to provide improved prepayment forecasts. The timing for such developments could not be better. The unlikely confluence of low rates, reduced home prices, a stagnant economy and an aging population bulge make the near term highly unpredictable. Strategists in every aspect of the mortgage industry are in for the time of their lives.

Gordon Monsen is a mortgage banking consultant located in Riegelsville, Pennsylvania. He developed the Advance Factor Service for Telerate Systems, Inc. and served as president of its mortgage securities division.
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:prepayment of mortgages
Author:Monsen, Gordon
Publication:Mortgage Banking
Article Type:Cover Story
Date:Oct 1, 1992
Previous Article:The only game in town.
Next Article:The brave new world of artificial intelligence.

Related Articles
Mortgage prepayments slow to a crawl; steadily declining prepayments are likely to continue at a snail's pace as a fall off in first-time homebuyers...
Dodging the prepayment bullet.
Using private information: a pricing strategy.
Clocking the pace of prepayments.
The physics of mortgage banking.
Managing a slippery asset.
Managing prepayment float.
Building more profitable portfolios.
Differentiating prepayment risk.
Battling prepayment risks.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters