A- Breaks Away From The Subprime Pack
Loans to “just-miss” borrowers become one
of the hottest niches of 2000.
By Jess Lederman, Tom Millon, Stacey Ferguson, Cedric Lewis
During the mid-1990s, subprime lending emerged as one of the fastest-growing
sectors of the residential mortgage market. This year, subprime production
is expected to total about $120 billion - down from a peak of $150 billion
achieved during the refi boom two years ago - but still 12% of all residential
mortgage originations.
The real story for 2000 is not in the subprime, or B/C market, it’s
the emergence of a new niche: A- lending, that is changing the world of
mortgage finance.
A- is the upper credit tier of subprime. But it is far from a narrow, limited
niche opportunity. Actually the exact opposite is true, for A- loans are
estimated to account for as much as 60% of all subprime production.
A- is a $70 billion opportunity, a fluid market where the pricing, product
characteristics and the players themselves are changing every day.
Characteristics of A- loans
While many factors contribute to the overall credit quality of any loan,
a reasonable definition of A- can be made by focusing on credit score, mortgage
history, bankruptcies, and debt ratio.
A PaperA-B/C
Minimum Credit Score620575540
12-Month Mortgage History0x302x304x30
No Bankruptcies in Last7 years3 years1 year
Maximum Debt-to-Income40%50%>50%
It should be noted that there can be many exceptions to these general guidelines.
A borrower with exceptional reserves and a suitable explanation for a 600
credit score might still qualify for an A quality loan. Conversely, a no-income
verifier to a borrower with a 650 credit score and modest reserves might
be considered an A- loan.
AU, newbies transform subprime
One of the most important factors contributing to the rise of A- lending
was the liquidity crisis that occurred during the fall of 1998, which led
to the demise of many of the pure B/C lenders who had previously dominated
the subprime market.
This created an opportunity for some of the blue-chip, well-capitalized
A paper shops - the major conduits and wholesalers - to enter and ultimately
emerge as the new leaders in subprime lending.
Many of the pure B/C lenders had found it highly profitable to price all
subprime production at the widest possible spreads to A paper. They had
no economic incentive to think of a smooth continuum of credit, and therefore
of pricing, from A paper to D. The traditional A paper players, on the other
hand, found their greatest comfort in the upper tier of subprime - indeed,
for some, their initial entry into subprime would be restricted entirely
to the A- sector.
To these new entrants, the pricing strategies of the pure B/C shops suggested
a tremendous opportunity to offer more attractive rates to over half of
all potential subprime borrowers. Their marketing pitch was simple and powerful:
better pricing and a more stable source of capital, all in one package.
1998 was not only the year of the great subprime shakeout, it was also the
year in which automated underwriting came of age. AU enabled the two most
important blue-chip players - Fannie Mae and Freddie Mac - to enter the
A- market. Their entry was perhaps the ultimate affirmation of the legitimacy
of A- lending, and has done the most to broaden originator participation
and narrow the pricing differential between A and A- loans.
Automated underwriting also has been important in enabling personnel who
had previously focused entirely on A paper to feel comfortable about handling
subprime loans.
Until recently, loan officers who typically made prime quality conventional
loans would hand their declines (or obviously subprime referrals) over to
an LO who specialized in B/C.
With the market contraction that began in 1999, however, and the proliferation
of both agency and private conduit AU systems for A-, this began to change.
LOs had a compelling economic incentive to seek out any and all loans, and
the AU system eliminated any guesswork about whether loans would receive
investor approval.
Insurers enter A-
Mortgage insurers first became involved in the A- market around 1997. Stretching
their credit criteria to enter the upper tier of subprime seemed a logical
way to expand volume. Their impact, however, would be profound.
The pure B/C shops that dominated subprime through 1998 had never been interested
in working with the MI companies, and traditionally even high LTV subprime
loans did not require insurance.
But many A paper lenders were fundamentally uncomfortable with this aspect
of the subprime market. This was particularly true for banks and thrifts,
the most important portfolio lenders, who are subject to close regulatory
scrutiny and to risk-based capital requirements that penalize uninsured
loans over 80% LTV.
The availability of mortgage insurance has both allowed new investors to
enter the market, and allowed existing investors to expand their product
guidelines. By requiring MI, some A- lenders who had previously capped their
LTVs at 90% have stretched to 95%, increased their maximum loan amounts,
allowed lower credit scores at higher LTVs, or have begun accepting investment
properties.
Mortgage insurance for A- loans is, in many cases, only marginally more
expensive than for A paper. As shown in the table below (based on premiums
quoted by MGIC, the nation’s largest MI), annual premiums for the upper
tier of A- are only five to ten basis points higher than for traditional
mortgage insurance. Only LTVs above 90% that have credit scores below 601
feature a material increase in premium level.
Servicing deteriorates
One of the conundrums of the subprime market is that the borrowers who are
best at improving their credit are also the first to refinance. This means
that, without an extraordinary retention program, the credit quality of
a subprime servicing portfolio may continually deteriorate.
Fannie Mae has come up with a highly innovative way to combat this phenomenon.
Their Timely Payment Rewards program gives borrowers the opportunity to
reduce their mortgage interest rate by 1% (thereby dramatically reducing
refinance risk) if they make 24 consecutive on-time payments following the
loan closing (i.e., improve their credit).
The program is available for certain loans that fall within Fannie Mae’s
highest acceptable A- risk level. These loans are of course priced at the
greatest yield spread to Fannie Mae’s generic A paper pricing - precisely
the loans where borrowers would have the greatest incentive to refinance
if their credit rating improved.
In addition to assisting with portfolio retention, potential rate reduction
is also a highly effective marketing point in selling to subprime borrowers.
While Fannie Mae is currently the leader in offering a rate reduction program
for A- loans, there is a clear trend in the private market to emulate this
feature, and we expect that similar programs will become standard in the
marketplace.
A- pricing, hedging, execution
Pricing:Fixed-rate A- pricing is commonly based on a spread to the U.S.
Treasury curve, rather than on a spread to the mortgage-backed security
(MBS) market. Market convention has been to calculate duration based on
a 21% CPR (constant annual prepayment rate), and to price the product at
a spread to the yield of an interpolated Treasury note of the same duration.
Most A- ARM product is priced at a spread over LIBOR. However, most A- ARM
product is originated by portfolio lenders, who are generally not concerned
about hedging their ARM product. Their primary concerns are credit risk,
duration risk, extension risk and spread.
The portfolio lender originates product that is an acceptable credit risk
for its portfolio, has a duration closely matched to the duration of its
liabilities, will not extend duration far beyond that of its liabilities,
and earns an acceptable risk-adjusted yield spread over its cost of funds.
Many conduits send rate sheets only once a week, which might lead you to
believe that either A- profit margins are wide enough to absorb weekly price
swings, or that pricing to end investors is not sensitive to daily interest
rate moves. While that might have been true in the past, it is not the case
anymore. Increasingly, A- product is typically originated at margins only
slightly wider than A credit product. It is simply market convention that
subprime rate sheets are sent once weekly. Originators absorb weekly price
fluctuations, earning wide margins one week, and possibly losing money the
next.
Hedging and spread behavior: A- pricing follows the home equity asset-backed
security (ABS) market, which is weakly correlated with the A credit mortgage
market. A- product is not simply an extension of jumbo or Alt-A product.
ABS spreads march to their own drum. Although ABS spreads gapped to historically
wide levels with other spread product in October 1998, they generally have
not been highly correlated with MBS-to-Treasury or jumbo-to-MBS spreads.
ABS spreads have been more stable over time than jumbo-to-MBS spreads largely
because A- prepayment rates are less sensitive to interest-rate volatility.
ABS spreads are driven by risk perception, volatility and supply/demand.
Recently, spread product that is greatly affected by market volatility -
MBS, corporates, agencies - has been very poorly bid. ABS spreads, on the
other hand, have done quite well. They are much less affected by volatility,
and investors have viewed ABS as a safe haven from other spread product.
Current ABS spreads are relatively tight, while MBS spreads are at historically
wide levels.
An originator will incur spread risk and interest-rate risk as it accumulates
product prior to sale. Spread risk cannot be effectively hedged, although
sales can be delayed until the originator perceives spreads as tight. Interest-rate
risk can be hedged, of course, and the 10-year Treasury future was, until
recently, the hedge vehicle of choice.
However, with the recent curve inversion, liquidity and correlation have
moved down the curve, and most investors are using a combination of two-year
and five-year Treasury futures and options. MBS are also an effective choice,
as long as the originator understands that MBS spreads may not correlate
well with ABS.
Trade Execution: Trade execution in the A- market is not immediate. The
conduits typically allow sellers 24 hours to respond to a bid. Most bulk
released deals trade between 102 1/2 and 103 1/2.
Conduits create senior/subordinated securities in the A- market, just as
they do with jumbo and Alternative A (Alt-A) product. Therefore, discount
versus premium note rates greatly affect A- execution.
Conduits tend to create senior securities that are sold at par (100-00).
Residual pieces, created by note rates above par, are generally held by
the conduit. Therefore, demand becomes progressively lower as note rates
move higher than the par rate. The conduits simply do not wish to be stuck
with a volatile residual premium. So it can be quite risky to hold large
amounts of saleable A- product in a declining interest-rate environment,
because execution will suffer as premium compression occurs.
Note rate dispersion greatly affects jumbo and Alt-A execution. The more
widely dispersed the note rates in a jumbo/Alt-A pool, the worse execution
the pool will receive. The same is not true in the A- market. Widely dispersed
note rates will not measurably affect A- execution.
Subordination levels are roughly 4% for jumbos, 7% for Alt-As, and can range
from 13%-17% for A- product. Increased subordination levels drive prices
lower and rates higher for A- loans.
If 87% of a deal is AAA-rated, and trades at a dollar price of 100-00, and
13% of a deal is subordinated and trades at a dollar price of 75-00, the
weighted-average price of the deal will be 96.75. If the subordinate level
were only 4%, the deal would trade at a weighted-average price of 99-00.
Required subordination levels can be significantly affected by levels of
primary MI coverage. MGIC, for example, has been involved in numerous transactions
which have required coverage for LTVs over 75% or even 65%. This is simply
a securitization arbitrage - the cost of the insurance is more than offset
by the economic benefit of a lower subordination levels.
Conduits have an increased appetite for certain jumbo or Alt-A product when
they are trying to fill a forward security sale, and it can often make sense
to delay sales until this bid emerges. However, this phenomenon typically
does not occur in the A- market. Deals are not sold forward. They are underwritten
by securities dealers after the conduit has accumulated product for sale.
Prepayment penalties, usually not a factor in the jumbo and Alt-A markets,
are critical components of most A- deals. In general, the presence of a
prepay penalty can affect price by about one point (100bp).
New, rapidly developing markets offer the greatest promise to competitors
who can move the quickest and the most creatively.
A- lending offers a wealth of opportunity for market participants, ranging
from large investors to the smallest originators. Aggregators can offer
their sellers new product features and innovative risk-based pricing. Correspondents
and brokers can continually scan the investing community for the latest
and most attractive programs and pricing, and do whatever it takes to be
the first to offer those to their borrower base.
In a tough market, where every loan counts, an aggressive A- program can
make all the difference to your bottom line.
Jess Lederman is senior vice president in charge of Ohio Savings
Bank’s national wholesale mortgage banking division. Tom Millon is
vice president, director of capital markets. Stacey Ferguson and Cedric
Lewis are mortgage banking analysts with Ohio Savings.
This article was previously published in the May 2000 Issue of Secondary
Marketing Executive.
