MBNA Credit Card Master Trust

© DNA Training & Consulting

 MBNA Credit Card Master Trust

The asset-backed market began in earnest when mortgage lenders came to appreciate
how dangerous it was to hold long-term assets directly on the balance sheet. Prior to
the volatile Eighties, it had been quite common for US banks and housing lenders to
fund 30-year fixed-rate mortgages with short-term deposits, comfortable in the
knowledge that Regulation Q of the federal banking laws placed an 8% cap on the
interest they could pay on customer deposits. As long as their mortgage assets earned
at least 10% or more, this margin would be sufficient to pay operating expenses and
generate a decent return on shareholder equity.

The spike in short-term interest rates during the ‘79-‘81 period, along with the
elimination of Reg. Q, caught a large number of savings institutions unprepared.
With funding costs in the high teens for several years, it was not long before some six
hundred of them shut their doors and sought the protection of the bankruptcy laws.
Matters were made worse by the “negative convexity” of mortgage portfolios: when
rates were falling and operating margins rising, homeowners were quick to prepay
their debts and refinance them at more attractive levels, thereby eating into the
lender’s net interest spread; while when rates rose, prepayments would slow to a
trickle, just when funding costs were Federal Reserve squeezing the lender most
severely.

The “pass-through” structure was invented in response to this dilemma. Once a large
enough pool of mortgages had been generated, the lender would simply create a trust,
transfer to the trust title to the mortgages against payment of the purchase price, and
hire an underwriter to issue securities into the capital markets on behalf of the trust.
These securities, known as “mortgage-backed securities” or “MBS”, would pass on to
holders all cash received by the trust on account of interest, scheduled principal or
prepayments, minus certain servicing and related costs. The risk that prepayments
would accelerate when rates were low, or would slow down when they were high, was
thus absorbed by a third party and not by the originating institution.

The MBS technology proved valuable in addressing another banking dilemma that
arose a decade or so later. Towards the middle of the Eighties, international banking
regulators became concerned that many banks, throughout the world, were operating
with insufficient equity capital. They feared in particular that competitive pressures
would drive lenders further and further down the credit ladder in search of adequate
yields. They also reasoned that with the financial marketplace becoming increasingly
global, it was imperative for them to coordinate their efforts, in order to ensure that
the rules governing a bank’s minimum capital ratio should not vary from one
jurisdiction to another. To that end, they issued the “Basle” capital regulations, which
stipulated a minimum capital ratio of 8% against a bank’s assets unless these fell in
one of 10 or so “favored” categories that were deemed less risky. The favored
categories included certain forms of sovereign exposures, municipal exposures,
exposures to multilateral organizations, trade-related letters of credit, and loans to
OECD banks. Most other assets, however, would require the full 8% equity support,
including the great majority of commercial and consumer loans.

© DNA Training & Consulting. 2

The Basle rules were severely criticized for their failure to differentiate among loans
on the basis of their underlying risk. Thus a US dollar loan to “AAA”-rated, US-
based Exxon would require the same 8% equity support as one to the Government of
Peru; and an unsecured loan to a high school student was as risky, according to the
regulators, as one to Warren Buffet or Bill Gates. It was not that regulators really
considered all loans in these categories to carry equal risk; they were simply reluctant
to complicate the rules much more by further subdividing the risk categories.

One business that was especially hard hit by the new rules was the credit card
business. Credit card loans were typically unsecured and suffered from a relatively
high write-off rate, somewhere in the order of 4% - 5% per annum. Yet the generous
yields they carried, in the range of 18% - 22%, were enough to cover all write-offs
and still earn the originator an ample return on equity. The biggest US players,
Chase, Citibank and MBNA, boasted an after-tax return on assets from that business,
net of all expenses and write-offs, in excess of 3% -- about triple the typical return on
assets of leading money-center banks. Despite these excellent numbers, the Basle
rules still required banks to allocate the minimum 8% equity against their credit card
portfolios, dragging down considerably the return on equity for that business, and
slowing down lenders’ ability to meet their customers’ growing demand for the
product.

Securitization of credit card loans began in the late Eighties, shortly following a
number of similar transactions involving auto loans and equipment leases -- which
were themselves modeled after the now-commonplace MBS deals. Credit cards
presented a number of novel difficulties, however: they were revolving assets, in the
sense that a cardholder could borrow, repay and borrow again under the same card;
the interest rate charged under a loan could be reduced by the originating bank at any
time, severely hurting the yield generated by a given portfolio; and, as mentioned
earlier, they suffered from a write-off rate that far exceeded the rate for mortgages,
auto loans, and leases – yet the banking regulators had made clear that any direct
recourse to the originator would preclude off-balance sheet treatment. Each of these
issues would need to be addressed before the rating agencies would consider granting
the “AAA” credit rating that was necessary to ensure a wide distribution of the
securities.

*************************

By the mid-Nineties, credit cards had come to constitute the largest asset class in the
US securitization market. Fuelled by the explosion in consumer lending and the
severe shortage of equity capital available to most lenders, the activity came to be
seen as a low-cost, capital-efficient mechanism for constantly replenishing a lender’s
funds
once a portfolio of sufficient size had been generated. A number of players, like
Citibank and MBNA, were able to seize on the technology to grow their business
from some $2 - $5 billion, initially, to over $100 billion in just a few years. The

© DNA Training & Consulting. 3

typical transaction dealt with the high write-off rate by allowing the investors to have
recourse against the portfolio’s “excess spread” – ie the difference between the
portfolio’s gross yield minus the servicing expenses, write-offs, and coupon on the
securities issued – but not against the originator directly. Additional protection to
investors was created by tranching the securities into a senior/subordinated structure:
the trust that purchased the loans would issue a senior tranche of securities,
representing around 90% of its total capital, and also one or more subordinated
tranches, which represented the remaining 10%. If losses on the portfolio should rise
to the point that the excess spread was completely wiped out, the junior tranches
would suffer the first loss, giving the senior tranche additional protection consistent
with the desired AAA credit rating.

The revolving nature of the underlying assets was exploited to create securities whose
final maturity far exceeded the typical maturity of any one loan: when a cardholder
made the required monthly payment, funds in excess of accrued interest on the
securities would simply be reinvested in new eligible loans generated by this, or any
other, cardholder. Only in the last twelve months of the transaction would a
cardholder’s payments be allocated entirely towards repayment of the principal of the
securities. The risk that the net yield on the portfolio would diminish, either due to
competitive pressures or because of a large increase in write-offs or delinquencies,
was addressed by featuring an “early amortization” mechanism, under which
reinvestments would terminate, and the securities paid down immediately, if the
portfolio’s excess yield dropped below a stipulated amount.

More recent transactions have been issued under a “master trust” structure, pursuant
to which the lender’s entire portfolio of loans is pooled together to support several
transactions at once. This permits, among other benefits, for the “cross-
collateralization” of successive transactions: specifically, if the credit support for one
issue proves too low, while for another it is excessive, the excess from the second
issue is “reallocated” to the first, reducing the likelihood of an early amortization or
an outright default. The master trust structure also allows, like an old-fashioned shelf
offering, for the issuance of serial tranches of securities under a single set of legal
documents, each potentially bearing a different tenor, coupon, and level of credit
support. Transactions that previously took months to structure and distribute are now
completed in a few days.

More recently, the master trust technology has been utilized in bringing credit card
securitizations to the investor base of a number of countries that do not have a local
ABS market. Securitization of credit card loans generated in certain parts of the
world has generally proven slow – partly because the size of most portfolios is far
smaller than in the U.S., partly because the laws of some countries frequently prohibit
or restrict the sale of consumer assets, or trigger the payment of VAT or stamp taxes
upon such sale, and partly for other reasons. Yet global investor demand for the
product, after a lengthy gestation period, has become healthy, far outstripping the
supply from local sources. This has opened a window of opportunity for innovative
lenders in many parts of the world: they can issue a tranche from the master trust into
a second trust established overseas, then arrange for a AAA-rated counterparty to

© DNA Training & Consulting. 4

provide a cross-currency swap to the second trust, and finally hire an underwriter to
distribute securities on behalf of the second trust, backed by the cash flows from the
first issue and the swap. This technology has been utilized to issue asset-backed
securities in Euros, Sterling and Yen, among other currencies. It is surely one of the
wonders of modern finance that a US attorney carrying an MBNA credit card today
may be receiving funding from a French farmer.

************************

© DNA Training & Consulting. 5

Questions

• Draw a diagram that shows the sale by MBNA of a $100 portfolio of credit
cards to an SPV, which issues $90 of Senior Notes and $10 of Junior Notes.
Assume the credit cards carry an interest rate of 14%, the Senior Notes 6% and
the Junior Notes 8%. Assume MBNA, which retains the servicing function,
charges 1% per annum for this service. Calculate the portfolio loss rate that
would result in (i) all principal and interest on both the Senior and Junior
Notes being paid, with no excess left; (ii) all principal and interest on the
Senior Notes only being paid.

• The money received annually that exceeds the servicing of the Senior and
Junior Notes and the servicing fee is referred to as the excess spread. Discuss
how the following factors influence the excess spread: (i) higher default rate
on the portfolio; (ii) more people who repay their credit cards in full each
month; (iii) greater competition among issuers of credit cards; and (iv)
movements in interest rates.

• Assume now the SPV issues $90 of Senior Notes, $5 of Mezzanine Notes,
and $5 of Junior Notes, and that MBNA retains the Junior Notes. Knowing
what you might know already about Basle capital adequacy requirements, try
to calculate the amount of “capital relief” that is generated for MBNA by this
transaction.

© DNA Training & Consulting. 6