Printer Friendly


This continues a series of articles aimed at helping readers to make critical software choices. Each of the articles concentrates on a mortgage banking business area, highlighting the strategic and management issues associated with selecting appropriate software. The articles describe the range of functions the reader can expect to find in software marketed to support the business area. In each article, we present products that are representative of those found on the market We discuss how well the products support each business function, point out prominent or unique features and review the experience of users. The product reviews rely on information obtained from product literature, interviews with vendor representatives, talks with product users and the authors' own experience with mortgage systems. We do not endorse or recommend any specific products.

In earlier articles we looked at software to support the loan origination business function. This time we look at support for a concurrent function: risk management. For the purposes of this article, we have chosen to include those activities that manage the loan during the entire time the mortgage banker is at risk for it, from the day it locks until the day it is sold. These functions include: * measuring your exposure; * deciding what and how to sell; * making trades; * allocating loans to cash and security commitments; * allocating securities to trades; * analyzing performance.

Measuring your exposure is an analytical function and reporting function that enables you to decide what your company's risk exposure should be. Of course, this means the system must take into account the organization's guidelines and risk management paradigm. Features typically include three reports: a position report to measure exposure, a mark-to-market report to measure gains or losses and a sensitivity analysis that estimates what your position and mark would be if the market were to trade up or down. As part of estimating exposure, the system must calculate the risk for locked loans in the pipeline. We are going to look at systems that support two ways of doing this: traditional fallout calculations and pipeline delta hedging (explained later in this article).

Deciding what and how to sell is a tactical decision based on the loans in the pipeline and current market prices. Features might include a report that stratifies your pipeline, summarizing loans into possible pools and comparing the pools to your forward sales; a best-execution model to compare cash sales and pool sales, as well as the value of excess service fees to help with buy-up/buy-down fees; and an ARM-pool model. A well-estimated production forecast will help alleviate the need for pair-offs.

Making trades involves the actual sale of loans and pools as well as trading in hedging instruments. Trading itself is not a system function, but the system must record those trades. The system should handle forward sales, futures, futures options and cash options. It should allow easy entry of a hedge ratio and a delta for trades and calculate (on the position report) the amount of coverage provided by the trades using these figures. Additionally, it should monitor futures and options for expiration. Finally, the system needs to provide a mechanism for trades to be paired off, including one-trade-to-many pair-offs and vice versa.

Allocating loans is the operational function of selecting loans into cash sale commitments or into pools being formed. Features typically include sophisticated selection of loans by pre-defined or ad-hoc criteria and an interface to Mornet/Midanet (or other investors' equivalent) to ship loans for sale or swap.

Allocating securities to trades prepares for notice day and settlement day. The system should include features to match securities trades with pools and print-delivery schedules.

Analyzing performance involves analyzing historical data to predict future market performance. The system needs to have an accessible database, a range of analytical reports and an ad-hoc reporting tool. For example, you should be able to analyze fallout by any of a number of factors-product, refinanced versus original loan, branch office, loan officer, location, source (e.g., retail or wholesale), process status and time until closing.

Now, back to the two methods of estimating exposure. The traditional method is to use fallout calculations. The second method is to treat the locked loans as options, a technique known as "hedging the pipeline delta."

The fallout method analyzes historical data to derive a fallout function. This function might look like a spreadsheet, with factors such as source of origination, loan purpose, processing status and time remaining to close as row titles and price differential (between loan and market) as column titles. The number in each intersecting cell would then represent the probability that a loan meeting the cell's criteria would close (100 percent minus fallout percent). Applying this function to the pipeline on a loan-by-loan basis at the current market rate produces an estimate of each loan's probability of closing. Further, by applying this probability to the loan amount, an adjusted loan amount may be calculated. By grouping these adjusted amounts by your pooling rules, you would know what securities you are likely to be able to produce and thus be able to sell forward the estimated production. By applying the fallout function multiple times, each time varying the market rate to a "predicted" rate, you could measure the sensitivity of your pipeline to market rate changes.

Hedging the pipeline delta uses an option-pricing model rather than historical data. This model analyzes many of the same variables as the fallout calculation, but instead of estimating the pull-through percentage (the probability that the loan will close), it estimates the probability that the loan will close at a loss. As an example, consider the case of a 9 percent approved original loan with only three days until closing and the market at 8 percent. Fallout would estimate a high probability of the loan closing because of the short time remaining, while delta hedging would estimate a near-zero probability of the loan closing at a loss. Both numbers are true, but they are used differently. Under delta hedging, loss probability percentages are applied to pipeline loan amounts to set the amount of forward selling needed for an adequate hedge, letting the remaining pipeline ride. Advocates of delta hedging disdain fallout forecasting as no better than forecasting the interest rate direction, while those who favor the fallout method maintain that delta hedging leads to too much trading and that, in any case, fallout calculations are required to estimate production and recognize what pool types to sell forward.

In addition to supporting these functions, a risk management system may offer several features common to all of these functions. For example, a realtime pricing interface would allow you to go out to a rate service (for example, Knight-Ridder) for price updates and recalculate mark-to-market and position reports during the day at the touch of a key. Another convenience would be the ability to schedule a batch of reports to be run at set times, for example, an overnight run based on the previous day's closing prices. There should be a good deal of flexibility in the system, because even though the functions required in risk management systems may be fairly standardized, the implementation and end products, such as reports, are often subject to highly individualized taste and business style.

We have chosen to describe four systems that are representative of products that support the risk management functions. It is our intention not to recommend these products, but rather to use them as a means of demonstrating the types of products available. When selecting software, we recommend that your research go beyond this short list. The packages to be discussed include: * The Advanced Secondary Marketing

System (ASMS) from MortgageFlex

Systems, Jacksonville, Florida - ASMS

is an example of a product that is integrated

with a loan origination system,

and it supports all of the risk management

functions just described. * The Risk Management Analysis System

(RMAS) from Baldwin Financial

Corporation, Parker, Colorado - This

product is an example of a

stand-alone risk management system,

and it also supports all of the

risk management functions. * The Risk Management System

(RMS) from Quantitative Risk Management

Group (QRM), Chicago - RMS

RMS is an example of a stand-alone

system that is more analytically oriented

to support the strategic and

tactical functions. * Tuttle & Company, Mill

Valley, California - This

is a service bureau that

is also analytically-oriented

and is the

strongest advocate of

the pipeline delta hedging

method to estimate


The remainder of this month's article addresses the MortgageFlex software; we will describe the others next month.

ASMS from MortgageFlex

ASMS is an example of secondary marketing software that supports all of the aforementioned risk management functions and is fully integrated with a companion loan origination system. Risk measurement is based on the fallout method. ASMS runs on a variety of hardware platforms, but to date, has been sold only with MortgageFlex's Residential Lending System, their own loan origination software.

ASMS has combined various levels of summaries and pipeline stratifications into its position report, providing all of the exposure information in one report. For mark-to-market calculations, the system has a real-time pricing interface to Knight-Ridder's Spreadsheet Connection. Additionally, ASMS allows you to enter estimated pull-through percentages calculate in the loan origination system. ASMS also allows you to enter the calculations to use for gain/loss on various entities.

We were able to see a pre-release version of ASMS'S best-execution module. For fixed-rate loans, this new module considers remittance options, excess servicing fees and multiple prepayment assumptions. The stratification of the pipeline is accomplished as part of the position report.

ASMS can record any type of trade. It also records the hedge ratio for futures and the hedge ratio and delta- for options. On the position report, the amount of coverage that futures and options trades provide are adjusted for these numbers.

Allocating loans to commitments and pools is fully integrated with loan origination - a benefit because more data elements are available to the selection process. ASMS allows you to enter restrictions on commitments, and it monitors the loans selected to those against the restrictions. This is true for both commitments used to sell loans and commitments used to acquire loans.

Allocating securities to trades is done on an ongoing basis, using the position report and an outstanding trades register report. In fact, the system allows you to be quite specific about the inventory for which a trade provides coverage.

In addition to extensive standard reports, ASMS has an optional report writer that you can use to generate ad-hoc reports against any data in the system.

The system handles large volumes well. Mark James, corporate development manager, Directors Mortgage Loan Corporation in Riverside, California, which closes more than $600 million per month, says that ASMS has "all the building blocks in place for a company our size."

Next month, we will review the three remaining systems previously listed.

Bill Lehman and Ray Roberson are senior consultants at CC Pace Systems, Inc., an information systems consulting firm in Fairfax, Virginia that specializes in real estate finance systems. To provide ideas or feedback for this department, call them directly at (703) 631-6600.
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:software for risk management
Author:Lehman, Bill; Roberson, Roy
Publication:Mortgage Banking
Date:Jun 1, 1992
Previous Article:'Buy and Hold Real Estate Strategy.'
Next Article:Technology.

Related Articles
Software Risk Assessment.
Open debate on open source.
Handling twin takes of ERM: many corporations look to focus on compliance and corporate governance needs today, yet maintain the opportunity to grow...
Software Sizing, Estimation, and Risk Management: When Performance Is Measured Performance Improves.

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