Printer Friendly

Servicing's great divide.

SERVICING'S GREAT DIVIDE

In keeping with Leo Tolstoy's observation about life more than 100 years ago, KPMG Peat Marwick's 1989 Mortgage Servicing Performance Study (MSPS) found the same principles applying to mortgage servicing. The lower cost servicers were low cost for very similar reasons; the higher cost servicers experienced higher costs for very different reasons.

Furthermore, the mixture of results from the study still indicates that opportunities abound when servicers view both cost and profitability factors simultaneously.

For its fourth annual MSPS, KPMG Peat Marwick expanded the study's scope to include the key components of profitability as well as the components of servicing cost. Data collection was expanded in some areas, including portfolio characteristics, investor reporting, collections, foreclosures, and real estate owned (REO) costs. The number of participants grew, resulting in a database comprised of more than two million loans or $116 billion in loans serviced.

Servicing industry leaders from across the country submitted critical 1989 unit-cost and productivity, operating, profitability and balance-sheet data in early March. Stringent validation techniques were applied, and 19 servicing functions were analyzed, along with key portfolio, compensation and profitability statistics. Analytical results were provided to the participants in mid-May. Senior executives of participating firms met in early June to jointly review the results and to profile operational efficiencies. The information shared during the two days of discussion produced some enlightening insights on the results of the study.

Overall cost results

Core servicing cost, defined as the direct costs tied to each functional area within servicing, averaged $94 per loan. Of the 12 percent increase in the 1988 average core cost, nearly half is attributed to the change in capturing core costs in the REO area to more aptly reflect true costs incurred by servicers who also manage foreclosed property. The lower-cost servicers averaged $57 in core costs per loan, while the higher-cost servicers averaged $140 per loan. The relative distribution of core costs is reflected in Chart 1.

The total servicing cost per loan includes core costs, collateral losses, interest and other expenses resulting from bad loans plus allocated overhead. In contrast to the increase in core costs, the results for total servicing cost remained reasonably stable - averaging $144 versus 1988's average of $149. The contribution to total servicing cost from core costs increased from 56 percent to 65 percent; however, this increase was more than offset by the significant decrease in bad loan losses from 34 percent to 20 percent. A distribution of the different elements contributing to overall servicing costs is depicted in Chart 2.

Cost trends

The study's primary analytical focus was on the 19 servicing functional areas. Some overall, as well as some specific characteristics and patterns have emerged from the results of the last four years of the study. The following are some specific insights into the mortgage servicing operation derived from the current year's study.

* Once again, loans serviced per full-time equivalent (FTE) proved to be a general predictor of core servicing cost performance. For the lower-cost study participants, loans serviced per FTE averaged 919; that is well above the average of 672 loans per FTE for all the participants in the study. Higher-cost servicers averaged 472 loans per FTE. The majority of repeat participants showed an increase in productivity, averaging 13 percent above their 1988 average.

* Consistent with prior years' studies, the overall size of each participant's portfolio was not a predictor of core servicing costs. The average portfolio size for lower-cost servicers (257,000 loans) did exceed the average portfolio size for all participants in the study in 175,000 loans. However, the higher-cost servicers averaged 166,000 loans in their loan portfolio.

* The most significant cost increase came in the area of collections and foreclosures, which includes bankruptcy and real estate owned functions. These costs rose 29 percent from the 1988 average. Interestingly, the lower-cost participants managed the expenses of this key functional area up-front by spending twice as much, and using twice the FTEs in the collections area as in the foreclosure area. Higher-cost servicers spent twice as much on the foreclosure function as on the collection function.

* Overall turnover proved to be a reasonably good predictor of core cost performance. Surprisingly, a very low turnover rate had an adverse effect on core costs. Turnover by functional area was not a good predictor for functional core cost performance.

* While industry analysts have been predicting a growing trend toward servicing consolidation, a minimal amount of consolidation was detected in the 1989 study.

Lower cost vs. higher cost

As noted, lower-cost servicers share some similarities, including:

* a higher concentration of GNMA loans in their portfolios, with GNMA concentrations averaging 46 percent versus 20 percent for the remaining participants;

* a higher concentration of fixed-rate loans than the remaining participants, 94 percent versus 79 percent, respectively;

* in general, a larger amount of net growth within each portfolio, whereby the higher-cost servicers experienced little or negative growth; and

* a lower salary cost per FTE. Salary and other personnel costs had a significant impact on overall costs as 40 percent of core costs were personnel costs. (This is depicted in Chart 3.)

Like Tolstoy's unhappy families, the study's higher cost servicers shared few characteristics. However, certain features did emerge as contributing to the higher costs experienced by these servicers. They included:

* a higher concentration of loans in one or two states that did not provide lower cost opportunities for these servicers;

* use of either monthly billings or coupon books, or both (while lower-cost servicers used coupon books almost exclusively);

* portfolios dominated by conventional fixed-rate, ARM, and GNMA product, respectively for the top three high-cost servicers;

* no confirmed pattern of above-average prepayments, delinquencies or foreclosures;

* somewhat higher salaries per FTE, but more FTEs in general.

The three highest-cost servicers had their portfolios serviced by a service bureau, a modified in-house service bureau, and a custom in-house system indicating no connection that could be automatically linked to higher servicing costs.

One characteristic shared by most higher-cost servicers was a substantial parent company overhead allocation, either in the form of data processing costs or management fees. A larger expense allocation for data processing increased the core servicing cost, while the higher management fees increased the total servicing cost.

Profitability trends

The study for 1989 showed that total servicing revenues averaged 48 basis points on an average loan balance of $64,000. As expected, servicing fees dominated the revenue area, accounting for 79 percent of gross revenues. Ancillary income, including late charges, insurance fees, assumption fees, and the like, contributed 17 percent of gross revenues, or $49 per loan. Chart 4 provides a breakout of these servicing-related revenue sources.

There was, however, no clear relationship between revenues generated from servicing and net income. There was also no relationship between gross revenues and core cost performance. Because this was the first year profitability data was collected, we were unable to discern whether this was a continuing trend or not.

Emerging issues

This study is a peer analysis that bases its results on cost comparisons to a comprehensive mortgage servicing performance analysis. It is still evolving, and we encourage management involvement at all levels and areas. The study will continue to focus on current industry concerns including emerging tax issues, and accounting issues involving purchased and excess servicing.

As our understanding of servicing's cost components grows, so does our list of questions. While maintaining the consistency of data is an ongoing challenge, study participants are to be congratulated for their diligence in assisting us in molding their data into comparable form. Perhaps the more difficult part of the study is asking enough questions, particularly in those areas experiencing rising costs or declining profitability. For example, does the increase in the collections and foreclosure cost area complement the decline in industry delinquency and foreclosure rates? Or, does it serve as a predictor of future rising delinquency and foreclosure rates?

In Part II of this article (scheduled to appear in the October issue), we will discuss additional analyses beyond cost performance in an attempt to address some of these issues. Statistically significant relationship, derived from four year's worth of collected data, will help promote a further understanding of servicing performance.

Study participants have used the results of these annual analyses to conduct yearly internal management reviews of their servicing operation and to get an updated view of their servicing department's performance relative to the best in the industry. In the future, study participants will be provided updated information and ongoing analyses to better position their servicing department to take advantage of emerging trends and issues. This data should help other servicing managers share in the solutions and further cultivate ideas that are available from the most successful servicers in the industry.
COPYRIGHT 1990 Mortgage Bankers Association of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:mortgage servicing, part 1
Author:Simmons, Linda C.
Publication:Mortgage Banking
Article Type:Cover Story
Date:Sep 1, 1990
Words:1453
Previous Article:Boardroom view.
Next Article:The road to zero inflation.
Topics:


Related Articles
Where are the mermaids?
Virtual Speculation.
Stuart Bruck of Time Equities, Inc.
Saving Homes.
Meridian Capital Group.
Meridian Capital Group.
Automating the mortgage supply chain. (eMortgage).
Survey-happy.
Balancing act: SunTrust Mortgage Inc., Richmond, Virginia, keeps most of the servicing it originates. The mortgage company's president and CEO,...

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