Printer Friendly

Secondary market.

Wall Street dealers seem to have a fixation with purity, at least when it comes to their good delivery guidelines for settlement of to be announced (TBA) trades for the GNMA I pool program. These guidelines, which mirror, but in some instances are more restrictive than GNMA's requirements, prohibit GNMA issuers from pooling standard buydown loans with regular single-family loans; limit eligible loan maturities; and constrain the number and size of loans that may be pooled. However, conventional mortgage-backed securities issued by Fannie Mae and Freddie Mac are not bound by these, and certain other, pooling limitations.

MBA believes that a system that perpetuates two sets of good delivery guidelines-one for GNMA I securities and another for conventional securities - is no longer warranted and represents yet another factor contributing to the diminishing attractiveness of issuing GNMA I securities.

Over the years MBA has worked with GNMA and Wall Street dealers through their trade association, the Public Securities Association (PSA), to relax onerous pooling requirements. As a matter of course, GNMA will not make a change unless MBA and PSA agree. Even though MBA and PSA meet regularly to discuss issues of importance and although the meetings are constructive, they often conclude without clear results. Consequently, a number of unnecessary differences still exist between government and Fannie Mae and Freddie Mac securities.

These differences and their potential consequences merit serious consideration. A key unanswered question arises at least partly due to the more onerous pooling requirements for GNMA securities: Will the share of conventional pools continue to grow at the expense of government pools?

According to PSA good delivery guidelines, there are no restrictions for the inclusion of standard buydown loans (i.e., 2-1 buydown) in Fannie Mae and Freddie Mac securities. Non-standard buydown loans (buydowns with interest rates "bought down" more than 2 percent or with buydown periods of more than two years) are limited to 10 percent of the pool principal balance. However, GNMA, in keeping with PSA's guidelines, does not permit the inclusion of buydown loans in any pools other than those pools specifically designated as buydown, under the current GNMA program.

MBA has strongly urged the commingling of standard buydown loans with other single-family loans in GNMA I pools. The basis of our arguments is the lack of conclusive evidence that prepayment speeds for 2-1 buydowns differ from those of standard, fixed-rate products. Furthermore, we believe the Wall Street bias against including buydown loans in GNMA I pools results less from actual loan performance than from lingering misperceptions created by the misuse and abuse of the FHA operative builder program offered in 1980.

In April 1991, MBA wrote GNMA asking the agency to amend its guide to provide for the unlimited inclusion of standard buydowns in GNMA I securities. Unfortunately, MBA's efforts were thwarted by PSA's analysis. PSA maintains the pricing of securities with an unlimited amount of 2-1 buydowns could be adversely affected. In support of its position, PSA cited that GNMA II securities, which contain buydowns, trade 6 to 10 basis points cheaper than GNMA I single-family securities. PSA reasoned that, if an unlimited amount of 2-1 buydowns were to be included in GNMA I pools, then GNMA I securities would trade closer to the GNMA II securities. To GNMA, this scenario translates into unfavorable pricing to benefit a few buydown borrowers at the expense of non-buydown borrowers.

But MBA pointed out the fundamental flaws in PSA's reasoning on this. As everyone is well aware, there are more differences between the GNMA I and II securities to account for the pricing differential, than simply the presence of buydown loans. For example, the wider range of coupons, the delayed principal and interest payment date to investors, and,the lack of liquidity in these securities, all contribute to altering the yield of the GNMA II security and are reflected in the price difference between GNMA I and II securities.

In a recent PSA newsletter, PSA clarified that under the 30-year good delivery guidelines for Fannie Mae and Freddie Mac, mortgage pools with maturities of 15 years and one month but less than 30 years will be considered good delivery for 30-year TBA trades. No such guideline exists for government securities. Instead, for 30-year TBA GNMA transactions, PSA guidelines require the final maturity of the security to be at least 28 years in length from the date of issuance. All other GNMA securities should be traded on a specified basis.

Not only is the PSA rule for government securities more restrictive than the conventional rule, it contradicts GNMA's own guidelines. Section 3-3a of the GNMA I Mortgage-backed Securities Gui states that at least 80 percent of the pool must be in mortgages with maturities within 30 months of the latest loan maturity and at least 90 percent with maturities of 20 years or more.

PSA has even gone so far as to urge GNMA to tighten up its rules on this issue. MBA believes that GNMA's pooling rules are stringent enough to provide investors with protection against shorter than expected maturities.

Over the years, MBA and PSA have debated the pros and cons of liberalizing the GNMA pool composition rules. MBA favors liberalization due to the increase in average loan size as a result of home price inflation and because of the statutory increase in the maximum FHA and VA loan amounts. PSA's opposition stems from fears of altered prepayment expectations and variability.

Nevertheless, in 1989 and 1990 MBA approached GNMA with requests to reduce the minimum number of loans in GNMA pools and increase the percentage that a single loan could represent in a pool. GNMA was willing to make the changes, assuming both MBA and PSA agreed upon them. In 1990, MBA and PSA compromised, and GNMA subsequently issued an All Participants Memo lowering the minimum number of loans to 10 and allowing a single loan to represent 15 percent of a pool. At that time, PSA wanted to double the minimum pool size, but chose not to do so to demonstrate its desire to work with MBA.

Interestingly, when developing its MBS program in late 1981, Fannie Mae used the Ginnie Mae security as its basic blueprint. Fannie Mae then added some features and streamlined procedures of its own. The success of the Fannie Mae MBS program indicates that these changes make the Fannie Mae MBS very attractive to lenders. Fannie Mae and Freddie Mac specify a minimum pool amount and eligible types of mortgages but are silent on composition structure. Yet, even with this added flexibility built into the conventional MBS programs, PSA's member firms have no problem making the market for conventional securities.

On the basis of this, MBA questions why PSA deems it necessary to have a different set of good delivery guidelines for Fannie Mae and Freddie Mac securities than for GNMA securities. PSA's recurring answer seems to be it's due to the lack of loan-level information provided by GNMA to the market. Admittedly, Fannie Mae and Freddie Mac make available a wealth of information that is both quick and easy to access and that increases investors' certainty as to the performance of their investments. GNMA currently does not have this capability but is working hard to provide more information regarding characteristics of new and seasoned pools, including the weighted average maturity (WAM).

MBA believes issuers should not suffer while GNMA develops its investor information system because there is little likelihood that our suggested changes would affect prepayment speeds, as PSA claims. Last November, GNMA echoed a similar sentiment in a letter to PSA. "GNMA feels strongly that, while GNMA is moving forward in developing additional information, such as WAM, the issuer community and the GNMA market should not be denied the ability to compete with the products of the other secondary mortgage market participants."

So why haven't these changes been made? Now more than ever it is important that GNMA redesign and implement program changes. For the first time in the history of the GNMA program, pool runoff exceeds pool issuance. Against this backdrop, GNMA must do some self-examination and make reasonable changes that may stimulate the program, while maintaining its integrity. Any roadblocks erected by constituent groups in housing finance, whether by PSA through its all-consuming quest for homogeneity or, FHA with its move to overhaul lending practices, must be overcome by GNMA. GNMA has always been a leader not a follower. It must continue to lead the industry through innovation and workable programs.
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:trading activities
Author:Allen-Malzahn, Deborah
Publication:Mortgage Banking
Date:Jul 1, 1992
Words:1415
Previous Article:Technology.
Next Article:Boardroom view.
Topics:


Related Articles
Secondary markets expect record year.
Proposed and final regulations affecting PTPs.
Transition Planning: Will your child be ready for life after high school?
Web-Enabling the CMBS World.
Disaster recovery: regulatory issues.
Mortgages looking up.
Canada's prescription drugs an issue in U.S. elections.
Tracking bonds' record: a new bond index from Swiss Re provides needed transparency to CAT bond market, perhaps more.
The appeal of second-tier markets: smaller, second-tier cities are hot markets for commercial real estate investors.

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