Printer Friendly

Wall Street's costly quest.

Wall Street is working on a new growth industry. Hedging prepayment risk has many of the street's best and brightest tinkering with a host of products with some very strange names. But the volatile world of prepayments taught some of the savviest firms on the street some very expensive lessons last year.

WHY BOTHER TO SHUT THE BARN DOOR AFTER THE horses are out? That old adage is heard often these days as mortgage bankers and Wall Street dealers in servicing hedges discuss the pros and cons of hedging servicing portfolios.

After all, interest rates are at 20-year lows. The big runoff already occurred, and it was a doozy.

Yet, paradoxically, more and more mortgage bankers are only now inquiring about, and actually buying hedges for servicing. "It's topic A," says Robert N. Husted, Jr., a managing director at Mortgage Industry Advisory Corporation (MIAC), a New York-based subsidiary of Treasury Bank, Ltd.

Is this just one of those illogical quirks of human nature, much the same as when homeowners rush out to buy flood insurance--after the flood?

To be sure, right now in the servicing business there is appropriate, heightened awareness of the losses that can occur after a calamity has already struck. There is a keen sense that the losses from prepayments can be far greater than previously imagined. "A lot more mortgage bankers are now aware of the prepay risk," says Carlos Perez, director of corporate development at Chase Home Mortgage Corporation in Tampa, Florida.

The pace of prepayments in the last year has come at such a speed that it has destroyed all previous assumptions. This year's unprecedented typhoon of mortgage prepayments, primarily from refinancings, has delivered a major blow to servicing cash flows and forced write-downs of acquired mortgage servicing and retained excess servicing.

Refinancings have driven the volume of mortgage banking business to its highest level ever, far surpassing 1987, a year when new originations, not refinancings, led a prepayment surge. Total mortgage business is estimated to be $840 billion to $850 billion for 1993, compared to an estimated $875 billion to $900 billion for all of 1992.

An increasing number of large and mid-sized mortgage banking firms are hedging acquired and excess originated servicing. At least six out of the top ten mortgage servicers hedge either internally or externally, according to sources at major Wall Street firms that sell the hedges to mortgage bankers.

While both mortgage bankers and Wall Street dealers like to keep a lid on exactly how much hedging goes on and by whom, key players in the market estimate that the notional balance of mortgages now covered by a variety of hedges devised on Wall Street is double the level it was at the beginning of 1992.

At the low end of the range, Hunter Wolcott, president and CEO of Reserve Financial of Miami, a firm specializing in servicing asset management, estimates that hedged servicing represents no more than $100 billion of all outstanding mortgages. At the high end, Husted at MIAC estimates that as much as $200 billion in servicing is now hedged.

MIAC has a joint venture with Hamilton, Carter, Smith & Co., headquartered in Los Angeles, to structure and evaluate hedges through the use of a sophisticated option-adjusted spread (OAS) model that plots 400 separate paths that interest rates might take during the term of a hedge. The joint venture, titled the Electronic Servicing Network Risk Management Service, has already advised Nomura Securities International, Inc. of New York on its interest rate-driven hedges.

Prepayments defying old norms

Mortgage bankers now must recalibrate their assumptions on prepayments. For one thing, it takes less of a drop in interest rates to provoke a new round of refinancings. At one time, a 150 to 200 basis point decline in interest rates was required to generate substantial refinancings. This year, however, prepayments have occurred at one and one-half to two times faster than their historical norms, according to Harley S. Bassman, the managing director responsible for mortgage derivative products at Merrill Lynch & Co., New York City.

Today, many homeowners are willing and able to refinance when there is only a 60 basis point decline in interest rates, Bassman says. "This has made mortgage servicing much riskier; mortgage bankers recognize that risk and are hedging now," he says. Wolcott says that a drop as little as 50 basis points can lead to refinancings.

"Prepayments have become more volatile," says Wolcott, noting that the mortgage lending industry has been largely responsible for this change by introducing no-cost, no-points refinancings. The increasing volatility of prepayments has alarmed mortgage bankers about potential future losses if competitive pressures increase and firms fight harder for market share. "People are afraid prepayment speeds could get uncontrollable as a result," says Bassman.

Some mortgage bankers now see a scenario where prepayments could rise even if interest rates go up in response to a surge in the economy. That's because many potential first-time and trade-up homebuyers are holding off buying a new home in the present uncertainty over the economy. Should consumer confidence return, there could be a surge of new originations that could send prepayments soaring.

Mortgage bankers may also be starting to rethink their assumptions that they can refinance everything that runs off, or that they are fully hedged by their own origination operations. "If it costs 50 basis points to refinance the runoff, that's expensive," says Husted. "If, on the other hand, you can hedge that runoff for between 5 and 10 basis points, that's a pretty good trade-off," he adds.

The argument in favor of hedging has proven even more attractive to those who bought prepayment hedges in the recent past and were partly protected from their losses in the past year. "Everything we expected the hedges to do, they have done and even more," says Sy Naqvi, president and chief operating officer of Sears Mortgage Corporation in Vernon Hills, Illinois. Sears purchased a prepayment-driven hedge tied to reference portfolios to cover the prepayment risk of a $6 billion portfolio acquired from Southeast Mortgage, a few years ago. The mortgage company has been very pleased with the hedge's performance.

Naqvi believes that "the time has come to hedge servicing whether it's purchased or originated because not doing so puts a lot of capital at risk." He admits, however, that Sears is not yet hedging internally originated mortgage servicing, "but we are talking about it."

Naqvi says that the mortgage banking industry may not yet fully appreciate the risk of not hedging originated mortgage servicing and may be too focused on the fact it involves an additional cost. "It's a new item on the expense line of the income statement, and it can be a big number," he says. From top management's point of view, "it only looks like a good idea in markets like 1991 and 1992. The rest of the time you ask, 'Why am I spending all this money?'"

By contrast, mortgage bankers have come to accept hedging the pipeline as a normal part of doing business. Naqvi believes that hedging servicing should also be a normal part of doing business. Naqvi believes that hedging servicing should also be a normal part of doing business and "not an issue" that has to be addressed from time to time. One impediment to advancing the practice of hedging servicing may be the fact that some mortgage bankers lack the capital markets expertise to view servicing as an investment that needs to be protected, Naqvi says.

Husted at MIAC, on the other hand, is slightly more optimistic. "The time is approaching when a significant percentage of the industry will have a strategy in place--they may not be executing it, but they will have a game plan."

Hedging strategy

While many mortgage bankers are mostly buzzing over hedging servicing but taking a wait-and-see attitude about actually buying hedges, mortgage banking firms owned by depository institutions appear to be moving more quickly to implement a servicing hedging strategy. These bank-affiliated firms may, in turn, be hedging more because of the increasing regulatory scrutiny of servicing that has occurred over the past three years. Bank-affiliated mortgage firms appear to be more focused than other mortgage bankers on the possibility of having to write down losses on excess servicing and acquired servicing.

For bank-affiliated mortgage firms, such write-downs pose a double threat: they reduce earnings, and they pose the danger of reducing the bank's risk-based capital. Under the new risk-based capital ratios, acquired servicing rights count 50 percent toward the Tier 1 capital for national banks.

Purchased mortgage servicing rights, or PMSR in the alphabet soup of the feds, has traditionally been considered an intangible asset by the regulators, who gave it little weight in risk-based capital computations. Bankers have long objected to this treatment of what they view as a solid asset. In April 1992, the Office of the Comptroller of the Currency (OCC) proposed an amendment to the risk-based capital rule for national banks that would resolve some of the servicing issues. The final rule, adopted in March 1993, increased from 25 percent to 50 percent the allowable limit of purchased servicing that can count toward Tier 1 capital. Offsetting this gain, the new rule requires banks to compute the value of servicing at 90 percent of fair market value or 100 percent of unamortized book value. Unamortized book value, in turn, is to be computed as the discounted value of estimated future net cash flows and reported via the monthly call reports banks file with the regulators and which resemble financial statements. Prior to the new rule, the book value could be based on estimated undiscounted future cash flows.

The treatment of excess servicing fees received (ESFR, in fedspeak) has not been under review by bank regulators. The OCC continues the policy of counting excess servicing, via monthly call reports from national banks, as a tangible asset, but only for servicing on mortgages for one- to four-family homes. According to the OCC, excess servicing is an upfront gain on the sale of the mortgages, which must be amortized over its life. It is not viewed as a revenue stream. As a capital gain, it qualifies as part of the Tier 1 capital base.

Banks are particularly sensitive about their capital levels these days. Under the Federal Deposit Insurance Corporation Improvement Act of 1991 (FDICIA), banks with lower capital levels face a higher level of regulation. FDICIA also requires banks to recalculate their capital ratios every quarter. Thus, the value of purchased servicing will be adjusted quarterly as the asset suffers impairment. This is an inducement for bank-affiliated mortgage bankers to hedge this asset, although income streams from a prepayment hedge will not be viewed as having prevented impairment of the asset in the eyes of federal bank regulators, according to the OCC.

While banks have insisted that their servicing is internally hedged by the ability of the institution to originate new mortgages, the OCC does not recognize banks' claims that they are internally hedged. "The federal bank regulators view servicing as a stand-alone risk," says Kevin T. Connors, senior vice president of derivative products at Lehman Brothers, New York City. Lehman has become a major player among Wall Street dealers in hedges in the last year by writing hedges for $30 billion of servicing. From Connors' point of view, the demand for hedges by mortgage bankers, responding to the concerns of banking regulators, has driven the growth in hedging that has occurred over the past year.

A source at the OCC concedes that queries on mortgage servicing assets by bank examiners have probably driven the surge in hedging that Connors has observed. According to the source, bank examiners ask bank managers if they can see the bank's program for managing prepayment risk on servicing assets. "We do not force them to hedge," the OCC source says, noting at the same time that "it is always prudent to hedge against catastrophic losses." The OCC's disclaimer notwithstanding, the bank examiner's query would likely lead a bank manager to believe he or she should put hedges on mortgage servicing assets to meet regulatory concerns. Otherwise, the bank manager would probably be unable to show that the bank had a risk-management program in place for prepayments.

Lehman Brothers has emerged as one of the leading innovators in hedge instruments, on the strength of its index-amortizing product, an interest rate-driven hedge introduced in May 1992. Lehman's first such hedge was placed on a $13 billion portfolio of Chase Home Mortgage using options tied to Treasury bonds. Perez, at Chase Home Mortgage, says he recommended tying the option to Treasuries. Although he had been happy with the principal caps he had purchased from other Wall Street firms, Perez says they had become too expensive.

The index-amortizing hedges, by contrast, have been far more reasonably priced, as low as 2 to 5 basis points by a variety of firms for selected products. These new index-amortizing floors are tied to the rates on either 7-year or 10-year Treasury bonds. They have, for the time being, largely displaced the prepayment-driven hedges as king of the hedging market.

The accounting issue

Prepayment-driven hedges were introduced four years ago largely to overcome accounting concerns that required that hedges correlate more exactly with the prepayment performance of mortgage servicing. Ironically, the first prepayment-driven hedge was not developed on Wall Street. It is widely acknowledged that First Interstate Bank in California sold the first such hedge in 1989 to Home Savings of America, Irwindale, California. (First Interstate is no longer in this business.)

The new generation of prepayment-driven hedges also offered other attractive features, including protection of the cash flow for the term of the hedge--often five years. Previous hedges, such as principal-only securities (POs) and supper POs, required more active management by the hedge purchaser and may be more properly considered a cash security than a derivative by some accountants. Cash securities do not qualify for hedge accounting, according to Halsey Bullen, project director on hedge accounting at the Financial Accounting Standard Board (FASB).

POs and super POs may also lack the required positive convexity needed to offset the negative convexity of mortgage servicing. If interest rates go up, the value of servicing does not rise in direct correlation. Instead, it lags behind the rise in interest rates. Conversely, if interest rates go down, servicing tends to lose value at a greater rate than the interest rate decline. Treasury bonds, by comparison, have positive convexity. Indeed, the spread between the value of Treasuries and mortgages narrows over time no matter which way interest rates turn--a central fact that lies at the heart of the interest rate-driven hedges.

The troubling hedge accounting requirement is derived from FASB's Statement 80, "Accounting for Futures Contracts," which requires a high degree of inverse correlation between the asset hedged and the hedge itself.

While prepayment-driven hedges approach 100 percent correlation, the interest rate-driven hedges lag behind at about 80 percent correlation. That's because some basis risk is kept by the mortgage banker. That basis risk was largely transferred to the dealer who sold the prepayment-driven hedges, which are tied to the prepayment performance of a reference portfolio similar date of origination, coupon, rate, type of mortgage and geographical location to the portfolio being hedged. In 1991, these prepayments hedges were reasonably priced--some as low as 4 or 5 basis points. Today, prepayment hedges are two to five times as expensive as interst rate--driven hedges, making them cost effective.

While dealers on Wall Street say that the price of prepayment hedges is higher because interest rates have fallen, other factors may have played a role. One dealer has confided, on the condition of anonymity, that Wall Street firms had miscalculated the cost of running the hedges when they priced them initially.

With the benefit of hindsight, the dealer says there are three major components of cost that are not germane to cash flows that the hedges provide. First, the hedges require the seller to load up his or her firm's balance sheet with assets to hedge the hedges. These include POs, super POs and other derivatives. The cost of carrying these hedges had to be included in the cost of the prepayment hedges. Secondly, the internal hedges are capital intensive, requiring Wall Street firms to reserve capital to cover them to meet the requirements of the Federal Reserve and the Securities and Exchange Commission. Finally, there are the mark-to-market costs and transactions costs associated with managing the internal hedges, the dealer source said. These, too, must be passed on in the price of the hedge for these products to be profitable.

The failure of some investment banking firms to actively manage their own hedges created what a number of observers claim to be sizable losses from prepayment-driven hedges at several Wall Street firms. Most of the dealers who sold prepayment-driven hedges, often called principal caps or revenue caps, have "toasted," confides one dealer still in the business.

The surge in prepayments hit the Wall Street dealers much the same way unexpected catastrophes hit insurance companies. It has driven down the number of major players in this market and driven up prices drastically, as those remaining are free to charge higher prices in the face of sustained demand. In fact, the prepayment-driven hedge almost disappeared at one point during the last year, according to Wall Street sources, just as umbrella policies for insurance liability have dried up in the wake of huge losses at Lloyd's of London and other insurers of the top end of the market.

Who lost how much?

One of the most closely guarded secrets on Wall Street these days is who lost money on prepayment hedges and how much. In spite of protestations to the contrary, some of the dealers in prepayment-driven hedges probably failed to hedge adequately for the hedges they sold. The basic problem for all the dealers is that "the prepayment models broke down in this interest rate cycle," says Randal Stoller, vice president of marketing on derivative products at Lehman Brothers. Lehman, a newcomer to the market, was fortunate enough to have entered the servicing hedge market as a major player after the horses had left the barn at other Wall Street firms.

One deal that reportedly led to a big loss for the hedge seller involved a multi-billion dollar portfolio of GNMA servicing covered by a prepayment-driven hedge sold by Goldman, Sachs. Goldman, Sachs has been a major player and innovator in this market since mortgage bankers began to hedge servicing. The servicing was owned by Barclays American/Mortgage Corporation, the Charlotte, North Carolina, mortgage banking subsidiary of Barclays Bank PLC, London.

According to sources on Wall Street and at other mortgage banking firms, Goldman, Sachs allowed Barclays to use its own portfolio as the reference portfolio for the hedge. Such arrangements, theoretically, pose a risk to the dealer because the purchaser of the hedge can speed up prepayments by soliciting refinancing from its own servicing portfolio.

Although Goldman, Sachs' agreement with Barclays prohibited the bank from soliciting new business from its clients, Goldman, Sachs had failed to consider that Barclays relies heavily on brokers for a significant portion of its originations. The brokers were free to solicit refinancings from Barclays mortgage holders and had every incentive to do so, because it generated new income for them. According to the sources, this drove prepayments above even the current accelerated levels, hitting Goldman, Sachs with a loss. Neither Goldman, Sachs nor Barclays would confirm or deny the reports, although a Goldman, Sachs source insists all hedges it sells are properly hedged. Indeed, Goldman, Sach is still active in the business and is offering a number of index-amortizing hedges tied to Treasuries.

Firms too slow

Another way that firms selling prepayment hedges may have lost money was to fail to respond quickly to fast-moving changes in the interest rate markets last year, when Federal Reserve Chairman Alan Greenspan engineered a long succession of drops in interest rates. The dealers were also plagued with the lack of an appropriate hedge for the prepayment driven hedges they were selling. The ideal hedge for such hedges is to sell short interest-only (IO) securities, according to Connors at Lehman Brothers. That's because servicing is analogous to having a long position in IOs. The problem for the Wall Street firms is that there is no market for short IOs, except those produced synthetically.

Events last year moved so quickly for those who dawdled that the price of internal hedges for prepayment hedges quickly became more expensive than the cost of the firm's exposure, according to Perez at Chase Home Mortgage. After failing to move quickly enough to buy appropriate hedges to cover the hedges they had sold, some dealers might have been tempted to then gamble by not buying any hedges at all or by failing to buy more hedge protection. Bassman at Merril Lynch unequivocally says his firm adequately covered its prepayment-driven hedges, in spite of some big payouts to such clients as JHM Mortgage Capital Corporation, McLean, Virginia. JHM Mortgage purchased "in the money" prepayment caps at a highly competitive rate from Merill Lynch. Indeed, Merrill Lynch remains a major player in this market.

Other firms may not have done as well in hedging their risks as Merrill Lynch's Bassman claims his firm has done. While there is no sure way of knowing who they are, Wall Street wags are almost unanimous in saying that both J. P. Morgan Securities, Inc. and Salomon Brothers suffered big losses on prepayment-driven hedges that might not have been adequately covered by internal hedges.

Formerly with J. P. Morgan, Austin Tilghman left last year to join UBS Securities, Inc. in New York City, where he introduced index-amortizing floors tied to Treasuries and made UBS a major player in the market. J. P. Morgan Securities, by contrast, seems to have faded from its once dominant position in the prepayment hedging market, according to some mortgage bankers.

Problems with prepayment hedges at J. P. Morgan and Salomon seem more likely in view of the fact that both firms acknowledged sustaining losses in their mortgage-backed securities business. J. P. Morgan reported in January that its trading revenues had fallen from $1.297 billion in 1991 to $959 million in 1992, a $338 million decline. The firm cited losses in trading in mortgage-backed securities as the primary cause of the decline in revenues. Analysts estimate that J. P. Morgan's losses in mortgage-backed securities for 1992 were between $150 million and $200 million.

Salomon Brothers has publicly cited $250 million in losses from trading in mortgage-backed securities in January and February of 1993. The Wall Street humor mill has been largely in agreement on the theory that the hefty "trading losses at both firms were due to losses tied to IOs. Alison A. Deans, a stock analyst with Smith Barney Harris Upham & Co., Inc., New York City, suggest that J. P. Morgan may have tried be a niche player in mortgage derivatives, but that the strategy had backfired. One credit analyst raised the possibility that Salomon may have sold amortizing swaps that required it to provide a revenue stream to the purchasers when interest rates declined.

If Salomon and J. P. Morgan had accumulated and inventory of IO strips in the belief that interest rates would rise, as the Wall Street sources surmise, it could account for much of the losses. As some mortgages prepaid, some IO tranches would have become worthless, while the value of surviving IO tranches would have fallen precipitously. If indeed J. P. Morgan and Salomon had made similar assumptions about prepayments and interest rates that led them to accumulate a big inventory of IOs, then it would suggest that they might also fail to internally hedge their prepayment hedges adequately. The losses from the prepayment-hedges, if, indeed, they were not hedged by the dealers, are nevertheless, probably only a small portion of the reported losses in mortgage-backed securities at both firms, according to most observers.

In the closely guarded world of Wall Street investment banking, sometimes what is not said can be more illuminating than what is said. In contrast to the firm assurances of profits from prepayment-hedging activities at Merill Lynch and Goldman, Sachs, neither J. P. Morgan nor Salomon Brother would categorically deny it had lost money on its hedging operations. Nor did they insist that their prepayment-hedging operations had been consistently profitable.

The new head of mortgage-backed securities at J. P. Morgan, Peter Bennett, says it is his firm's policy not to discuss the profitability of various operations. J. P. Morgan, however, is defended by its former hedge salesman, Austin Tilghman, who says that while at J. P. Morgan, the hedges he sold to mortgage bankers were very carefully and adequately hedged by J. P. Morgan.

Salomon Brothers was unable to provide anyone to discuss the firm's hedging activities for this article, in spite of repeated requests by Mortgage Banking over a period of six weeks. Brief, unofficial conversations with sources at Salomon, however, provided assurances that the firm is still actively in the business of selling prepayment hedges to mortgage bankers.

Future concern

In spite of the troubles that have forced prepayment-driven hedges to the sidelines, the future may not be all blue skies for the interest rate-driven hedges. The same troubling accounting issues that led to the creation of prepayment-driven hedges remain unresolved. FASB has not yet determined what degree of correlation can be counted as a high degree of inverse correlation. Thus, the fact that the interest rate-driven hedges provide about 80 percent correlation may leave them vulnerable to some future accounting ruling, even though accountants have, in practice, tended to agree that 80 percent represents an acceptable level of correlation.

The FASB staff is currently reviewing hedge accounting issues, but, according to the project director, Halsey Bullen, the staff has not yet been able to address the matter of what constitutes a high degree of inverse correlation. Furthermore, the OCC has not yet accepted the current version of Statement 80 as the basis for items entered onto the monthly call reports for banks. The OCC is following closely the hedging project at FASB but has no hedging projects of its own on the horizon, according to a source at OCC.

Bullen's Statement 80 project is prepared to make a recommendation that would defer both gains and losses on hedges used against mortgage servicing to the extent they are effective hedges. This would apply only to the hedges that are in place, not to future transactions. Such a move would remove some of the concerns mortgage bankers have about acquiring hedges.

FASB's changes in hedge accounting may be potentially less noteworthy than another FASB change that may be in the offing. The FASB staff is working on a proposal for the board to consider a project to weigh changes to Statement 65, "Accounting for Certain Mortgage Banking Activities." The changes in accounting for servicing assets being suggested for consideration could dramatically increase the universe of servicing assets placed on the books, according to Scott Woltemath, a practice fellow at FASB. One possible change to be considered is to create a new servicing asset consisting of retained servicing from originated loans that are sold.

The staff proposal is also likely to ask FASB to address the issue of how firms account for the impairment of all servicing assets over time. If FASB accepts the staff proposal, it would, by the end of 1993, set up a new project to address these mortgage banking issues, according to Woltemath. Should the board ultimately create a new servicing asset for originated loans with retained servicing, it would likely lead to a boom in the use of hedges that would dwarf the present scope of their use.

That prospect, no doubt, will keep Wall Street tinkerers working even later nights devising new instruments to meet this potential new demand.

Robert Stowe England is a freelance writer based in Arlington, Virginia.
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:hedging prepayment risks
Author:England, Robert Stowe
Publication:Mortgage Banking
Article Type:Cover Story
Date:Jun 1, 1993
Previous Article:Interest rates and charts.
Next Article:How do you measure up?

Related Articles
Perfecting the prepayment hedge.
Hedging runoff risk; severe runoff doesn't happen often, but often enough to teach dramatic lessons.
Selling protection.
Whistling past the graveyard.
A primer on hedging servicing.
Hedging servicing comes of age.
Hedging's high wire act.
Hedging's trial by turmoil.
The case against lowering the minimum servicing fee.

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