Printer Friendly

The marriage of mortgage lending & technology.

Technology has made many inroads in the business of mortgage lending. A brief review of the integration of several key technologies into the business shows the many ways technology investments have made a difference.

THE FIRST 30-YEAR, FIXED-RATE MORTGAGE instrument recorded by the Veterans Administration dates back to February 16, 1945. The loan was taken out by Elizabeth Lotts of Pittsburgh, Pennsylvania, who signed a purchase money deed in the amount of $7,000. Interestingly, enough, this was about the same time that the first electronic computer was being invented. This is, no doubt, one of those neat coincidences of history, but it also foreshadows the lengthy and fruitful relationship that would develop between emerging information technologies and mortgage banking.

Prior to World War II, homebuyers typically paid cash for their homes or took out a promissory note in which the entire principal balance was due at a specified date (similar to balloon mortgages offered today). Since then, however, there have been huge changes in the mortgage lending industry. Today's homebuyers literally, have thousands of ways to borrow money from a lender.

There are also many decisions that a lender must make once the loan has been originated. Dovetailing with these changes in the industry have been advances in the implementation of new technologies to support the new products and processes involved with the modern business of mortgage banking.

The business of mortgage lending is a very complicated one, indeed.

The many phases entailed in the production, sale and servicing of a mortgage create profound operational complexities and engender tons of paper that must somehow be managed by mortgage companies. Most lenders have turned to computer systems to help manage the mortgage process--from origination to servicing, and from pipeline management to secondary marketing. Others have used innovations in computer systems to gain competitive advantages in the marketplace.

More and more companies are starting to use computer technology to assist them in assessing risk and to help them better manage risk. In some cases, not having the necessary technology is a significant-enough handicap to damage a company's competitive position in the marketplace.

What mortgage companies were doing with technology

Although banks were a key commercial customer in the early era of computer development (1960s), mortgage companies typically limited their use of computers to accounting operations, such as loan servicing. The real rush to computerize began in 1964 with the introduction of the IBM System/360, according to Thomas Steiner and Diogo Teixeira, authors of a book called Technology in Banking. This was the first family of compatible computers spanning a wide range of performance. With the realization that their investment was protected and that more extensive automation would become possible, the entire financial community lined up at IBM's door.

This marked the first stage of automation, where affordable and practical computer technology became generally available to the mortgage industry. It was the core technology level that included hardware and systems software. Because this hardware and systems software could have been bought by anyone on the open market, it provided no competitive advantage, and thus acquisition of it simply became defensive in nature. Everyone had the technology or would soon get it, so your company better get it too. Even though this level of computerization provided little or no competitive advantage, its impact must not be underestimated. The next level of computerization underscores its importance.

What mortgage companies are doing today

Mortgage companies today support much of their operations and products with information technology. The heart of most mortgage companies' systems consists of a large mainframe computer, which tends to be oriented toward transaction-processing applications. These transactions basically consist of simple calculations, such as the amortization of loans, repeated over and over again on very large quantities of data. Huge data bases are used to store all of the data. The trend is toward decentralization, but economies of scale, security and control reasons have caused most companies to continue to centralize the bulk of their systems investments.

The second level of computerization consists of application programs. These programs support specific products and services. Thousands of programs support the operation of any major mortgage company. Although many vendors have invested millions of dollars to develop packages to carry out core loan-accounting functions, purchased software represents only 5 percent of over all systems expenses. Thus, most systems are proprietary.

This level of computerization is where most mortgage companies hope to compete. They are betting that their business ideas expressed through the application program infrastructure will get to the market first, capture it and maintain high market share. However, in reality, even the best technology will not make any competitive difference because there is nothing unique about the functions or products provided by this level of investment. Supporting a hundred different mortgage products with application software does not make a company a technological leader. What does matter is how technology is used by a company to be the first to meet a unique business need.

What mortgage companies will be doing

To address a specific business need, the appropriate hardware, systems software and application programs must be in place to support this final layer of technology within a business framework. This level of computerization is the set of technologies that support a business strategy. Those companies that have formulated a strategy, with its corresponding technology, to meet a unique business need are those that have been the most successful.

However, technological leadership, in and of itself, does not make a company successful. Bankers Trust Company, New York City, has gained the edge in optical imaging, not because it was the first to utilize the new technology, but because it used the technology to satisfy a business need--that of storing huge amounts of loan closing documentation, while providing instantaneous access to any field on any document. Citicorp, was actually the first company to use optical imaging--but it was used to track loan correspondence, not the actual closing documentation. Thus, Bankers Trust has been successful at capturing a large portion of the primary lending market.

Many mortgage companies have great incentive to move closer and closer to becoming "paperless" because mortgage loan applications require manual intervention and thus are costly to process. More and more employees are being displaced by technology, although the effects are slower in the mortgage industry because most transactions are still originated in paper form and a complete transformation to electronics cannot occur.

Industrywide, the escalation of investments in systems technology is encouraging mortgage companies to spread their financial and managerial investments in technological resources over a greater base, inevitably leading to mergers and consolidations. The newly consolidated companies should play an ever-increasing role in the industry. And, as they spend more and more on systems technology, these companies will dominate the industry's technological resources. (J. Reginald Campbell estimated in a March 1992 article in Banker's Monthly that technological investments in the mortgage industry were rising at an estimated rate of 12 percent a year.)

There is no clear dividing line between "new" and "old" technologies. Much of what is important is the constant progression of vendor upgrades in equipment and the ongoing introduction of new capabilities. These underlying capabilities, such as the capacity of on-line storage devices, help drive higher-level technology developments, such as image processing and artificial intelligence. Even in an area such as software development, some improvement is expected to occur.

Six categories of technology have been chosen for review in this article because of their potential for having major competitive impact on the mortgage banking business. They are: laptops, local area network (LAN) and wide area network (WAN), expert systems, open systems, imaging and information robotics.


Lomas Field Services (LFS) is a subsidiary of Lomas Mortgage USA, Dallas and specializes in providing servicing information to mortgage lenders. In a June 1990 article in Mortgage Banking called "High-Tech Servicing," Terry Deyoe explained how LFS had given each of its 98 field representatives laptops to conduct tens of thousands of physical inspections and valuations of properties in default. The laptops were also used to capture information gathered during personal interviews with borrowers, which is required by investors and mortgage insurers at various stages of default. The laptops also provided instant notification of vacant properties or cancellations of foreclosure actions from a servicer after receipt of a borrower's payment. The data captured on these laptops inputs data directly into the customer's on-line collection system or on-line foreclosure tracking system. The information that is provided is more timely and more accurate. Furthermore, the system also reduced the flow of paper between the mortgage servicers and LFS.

Laptops will increase in popularity as improvements in technology drive the price and weight down. For instance, increases in random access memory and disk storage capacity allows more and more information to be stored and accessed. The addition of internal expansion slots allows a LAN board to be connected to the laptop, which provides it with connectivity to the mainframe.

Advancements in speed are making desktop PCs and laptops similar in functionality. Toshiba has introduced a reduced instruction set computer (RISC)-based laptop powerful enough to compete with engineering work stations. Sony Corp. has introduced the PoqetPad, a pen-based portable PC that allows a user to write documents or commands on a seven-inch screen and have the information processed or recorded, without the use of a keyboard. This will be of particular use to the mortgage industry because the industry is extremely paper-intensive and will continue to have paper-driven work flows. Pen computing can replace paper, while at the same time retaining the way lenders carry on their business.

The key to the success of the laptop will be the software that runs on it. NEC has been a leader in developing programs for its laptops. To date, there are more than 12,800 software titles for NEC PCs. In 1993, NEC will announce a host of new applications available for its laptops, most of which are targeted to run with LAN software.

Thus, innovations in the laptop area will give lenders a competitive advantage in providing remote computing power to perform such tasks as credit verification, loan origination, creation of loan packages and preparation of closing documentation.


A LAN is a group of computers connected through cables or telephone lines that are capable of information transmission and communication. BancBoston Mortgage Corporation, Jacksonville, Florida, has used LAN and WAN technology to fully integrate its back-office operation. The main impetus for the implementation of a LAN a few years back was to support the company's QuickClose software, which was designed to process and approve conventional loans within 48 hours. (See Mortgage Banking, February 1990, "A Totally Integrated System," by Susan E. Budd.) Each of BancBoston Mortgage's loan originators was equipped with a Toshiba laptop computer, internal modem and printer, complete with a prequalification program and all the documents needed to process a loan application.

Expert systems

Artificial intelligence (AI) is a loose grouping of several related technologies, of which expert systems is the most relevant to mortgage banking. An expert system can account for missing or conflicting information and subjective processes and can incorporate experience. Instead of using fixed deterministic approaches, an expert system contains a knowledge base, which is a series of "rules" describing how a human expert might solve the problem. A rule can be viewed in simple terms as an if-then statement, although, in fact, the codification of the expert's knowledge is much more complicated.

Expert systems have been in the labs for years but only reached the stage of commercialization in the 1980s as the cost of central processing unit (CPU) cycles decreased. Expert systems can be very CPU intensive because they are nonprocedural. Because CPU costs should continue to decline, expert systems should continue to improve their cost effectiveness. It is estimated that investments in expert systems, in the mortgage banking industry may hit $200 million per year by the mid-1990s, according to Technology in Banking, by Steiner and Teixeira.

Back in December 1990, Institutional Investor ran an article on "Ten Technologies Ahead of the Curve" by Joe Kolman. In that article, Citicorp's underwriting expert system was considered so advanced that it made the top 10 list of financial technology. What set such software apart from others in mortgage lending was that it could give customers approval on the spot. Since Citicorp's entry into this area, other mortgage lenders have launched their own underwriting expert systems, and instant loan approvals have become a significant competitive arena for the future.

Under the expert system that won Citicorp distinction back in late 1990, mortgage brokers and others who were members of Citicorp's Mortgage Power Plus program could gain access to the system from remote PCs and fill out mortgage applications with their clients on-line. The system ran each application through a credit-decision model and checked credit bureau reports from various sources. The system was able to generate binding mortgage commitments, and it referred customers who did not get an immediate approval to a Citibank loan officer.

Open systems

"Open" systems enable any end user to communicate with another end user without being constrained by the equipment of a particular vendor. Several events have led to the development of open systems. The first is the development of a set of international standards that define a multilayer reference model called the open systems interconnection (OSI). The second is the signing of an agreement between AT&T and Retix Corporation for the development and marketing of a method of integrating OSI with AT&T's Unix System V. The results of the agreement make it easier for Unix and non-Unix systems to communicate. IBM announced in June 1992 its commitment to developing a Unix-based operating system for its mainframe products.

These developments are key to allowing more and more computer systems to communicate with one another, thereby opening up tremendous opportunities for the gathering and sharing of information between lenders and their customers.

Another mortgage industry player gained recognition in the December 1990 issue of Institutional Investor in its article on "Ten Technologies Ahead of the Curve." That article referenced how Greenwich Capital Markets has used the open system philosophy to gain a strategic advantage in the trading of mortgage-backed securities. Its trading and support system was built around a distributed data base powered by large Unix file servers that allowed the entire firm to share the same data. Numbers compiled by the mortgage-backed research group, for example, were immediately available to trading, sales and back-office pricing groups.

As a result, the firm's major trading areas were completely integrated, making it easy for managers to assess the firm's total risk exposure. While traders at other firms were often forced to do pre-trade analytics on stand-alone PCs, Greenwich's system integrated the analysis with the actual trade. A trader analyzed the effect of a trade on his or her position and his or her profit-and-loss statement, evaluated possible hedging strategies and calculated the effect on the firm's complete position. Then, if the trader liked the deal, he or she could book the trade instantly. The system allowed Greenwich to increase its transaction volume without increasing its staff, according to the article. The open architecture also meant the firm was able to extend its system to branches with a minimum of effort.

Another article from December 1990, "Interest Grows in Open Computer Systems," by M. William Friis, Banking Journal, documents how Mellon Bank, Pittsburgh, and Fleet/Norstar have integrated their Unix-based branch networks with the home office IBM mainframe using IBM's system network architecture (SNA). In this structure, customer service representatives can take advantage of Unix's ability to run several tasks simultaneously, while gaining access to the host computer's large repository of data. According to the December 1990 Mortgage Banking article, the system proved itself at Mellon.

Similarly, at Fannie Mae, highly complex economic forecasting is performed using high-powered Unix-based machines, which are connected to its mainframe. The communications between the two environments may appear transparent to the end user, but in reality, a complex configuration of special communication devices and network software is needed such that the process is not as seamless as a truly open system.

The advantages of open systems have far-reaching implications in that mortgage companies will be able to interact with their corporate customers, no matter what platform (environment) they use.


Imaging processing refers to the conversion of paper documents or their film copies to computer data so that they can be stored, retrieved, displayed or transmitted in digital form. Lenders use imaging to store and examine loan document images. These images are then compared to pertinent loan information on a host machine to answer such questions as: Are the signatures the same on the mortgage document and the bank note? Are the issue date of the loan and the date on the mortgage schedule the same?

In an article ("Imaging in a Multivendor Environment," by Clinton Swift) that appeared in Bank Management magazine in February 1990, it was noted that Bankers Trust purchased a $1 million imaging system from FileNet, Inc. The FileNet system used scanner devices for document entry, work stations for retrieval and display of images, and optical disk drives for storage, all linked by a LAN. This network approach was intended to prepare Bankers Trust to expand to remote document entry and retrieval in the branches.

The same article also noted that Banc One Mortgage Corporation in Indianapolis selected Kodak's KIMS imaging system. The imaging system consisted of a file server, two scanners, PC work stations with internal optical disks for storing images and a laser printer. IRMA boards (remote connectivity) make the work stations act like standard IBM terminals hooked to the mainframe that is at the other end of a fiber optic and telephone link to the main office in Columbus, Ohio.

According to Deyoe's June 1990 Mortgage Banking article, "High-Tech Servicing" Lomas Mortgage USA used laser optics to increase the volume and availability of information available to its servicing staff. Laser optics reduced the volume of paper documents and virtually eliminated the need to physically retrieve loan records. The June 1990 article noted that Lomas experienced an 84 percent productivity boost in the records management area and a 140 percent increase in the payoff group since the implementation of its laser optics data storage system.

Information robotics

Voice technologies hold great promise for mortgage companies because of their potential to reduce the cost of labor-intensive, telephone-based customer contact transactions. Automatic dialing is a simple form of the technology that has increased productivity by 20 percent to 40 percent in high-volume information-retrieval situations, according to Steiner and Teixeira in their 1990 book Technology in Banking. The economic savings are going to place a further emphasis on telephones and a further displacement of transactions away from branches, not to mention better overall levels of customer service.

Information robotics involve automated speech systems that are programmed by a complex language that can access recorded messages stored in a data base in digital form. Fannie Mae was one of the first companies in the mortgage industry to release such a system, according to InterVoice, Inc., in 1989 with its Pooltalk system, which allowed investors to dial in a pool number and retrieve key information on Fannie Mae pools. The advantage of the system was that the information it provided was more timely than the Bondbuyer book. The Bondbuyer book was sent via Federal Express to its customers. The information provided in the Bondbuyer book includes financial statistics on every pool traded in the secondary mortgage market, and it contains, consequently, hundreds of thousands of lines of information. The Pooltalk system eliminates the search process by allowing voice recordings of information requested on a specific pool. The system was so successful that Fannie Mae released a Japanese version for Japanese investors. Freddie Mac introduced a similar system in 1991 called "Freddie Answers."

Lomas Mortgage has used "Infobot" systems for years to ease the burden of servicing transfers and to provide general information about loans, such as tape-recorded explanations of ARM adjustments, escrow account analysis and how to obtain payoff information. Infobot systems use prerecorded messages so that incoming phone calls are routed to the appropriate place or directed to a customer service representative. In a February 1991 article in Mortgage Banking called "Servicing Smarter" by Brian Hershkowitz, Karen Sharp, systems administrator, was quoted as saying that Lomas' Infobot system provides a 30 percent completion rate (the percentage of time that the system answers the customer's inquiry) and that the logic built into it makes the system react just as an actual customer service representative would respond. In October 1992, an enhancement was added to the system to allow customers to order a copy of their payoff statement to a fax machine of their choice. The importance of voice simulation systems has grown so significantly that 70 percent of all banks in the United States with net assets of at least $5 billion have installed some form of automated customer service, according to Timothy Rector in his article "Maximizing Manpower," in Mortgage Banking, February 1991.

Mortgage companies of the future most likely will be substantially different from the mortgage companies of today. They will depend on much more technology, and that technology will look different from today's technology. In order to survive, mortgage bankers will need to incorporate all three levels of systems investment--infrastructure (level one), application software (level two) and business-need technology (level three). However, in order to succeed, they will need to be the first at implementing level-three technology.

The six areas reviewed in this article--laptops, LANs and WANs, expert systems, open systems, imaging and information robotics--were singled out because they probably will be widely adopted and they possess more potential for major competitive impact. Supporting these technologies are other developments that will have a profound influence on these technologies as well. They are advancements in the design and composition of the CPU leading to the millions of instructions per second (MIP) cost reduction; advancements in telecommunications improving connectivity between computers; growing popularity of object-based design (developed to optimize reusability of software) and graphical user interfaces (a noncharacter based software that eases user interaction with the computer); and the evolution of distributed computing environment (DCE) and distributed management environment (DME). DCE is the next step beyond sharing files across networks. It makes non-related or loosely connected systems appear as one system. Thus, it is transparent to the user that his files reside on different computers. DME is technology that provides a unified approach for efficiently managing systems, networks and user applications. It provides a consistent approach for the systems administration function.

The trend of substituting systems technology for employees has led to the shrinking of the employment base in the industry, with an anticipated loss of one-fifth of its employees by the year 2000. The aggregate size of the industry's data centers have been growing at rates of 25 percent to 40 percent per year. All told, the industry has roughly 30 times more systems power available to it than it did a decade ago, according to Steiner and Teixeira, in Technology in Banking. Thus, mortgage banks will need to improve productivity through the innovative use and adaptation of the proven technologies. As was wisely stated by the 17th century English philosopher Francis Bacon, "He that will not apply new remedies must expect new evils. For time is the greatest innovator." And only time will tell what a mortgage lending institution will look like in the next century.

Kim M. Cisney is senior data base consultant at the Federal National Mortgage Association, Washington, D.C.
COPYRIGHT 1993 Mortgage Bankers Association of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:mortgage banks' use of information technologies
Author:Cisney, Kim M.
Publication:Mortgage Banking
Date:Apr 1, 1993
Previous Article:Beyond mere compliance.
Next Article:Credit reports: does accuracy count?

Related Articles
The Mortgage Banking Sourcebook: A Reference Guide to Information Sources on the Mortgage Lending Industry.
Looking forward in tough times.
Northeast region-top 30 lenders. (Marketrac [R]).
Top 200 Lenders. (Rankings).
Top 200 Lenders: first-quarter 2003 (by $ amount). (Rankings).
Underperformance and oversight.
Family Lending Services taps Mortgage Cadence's ELS.

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