Filed:   January 13, 1998


                    UNITED STATES COURT OF APPEALS

                        FOR THE FOURTH CIRCUIT



                              No. 97-1315
                            (CA-95-1091-K)



Banca Cremi, S.A., etc., et al,

                                              Plaintiffs - Appellants,

           versus

Alex. Brown & Sons, etc., et al,

                                               Defendants - Appellees.




                              O R D E R


     The Court amends its opinion filed December 30, 1997, as

follows:
     On page 2, section 1, line 17 -- Amicus' name in the counsel

listing is corrected to read " PSA THE BOND MARKET . . . ."

                                       For the Court - By Direction



                                             /s/ Patricia S. Connor

                                                       Clerk
PUBLISHED

UNITED STATES COURT OF APPEALS

FOR THE FOURTH CIRCUIT

BANCA CREMI, S.A., Institucion de
Banca Multiple, Grupo Financiero
Cremi; BANCA CREMI GRAND
CAYMAN,
Plaintiffs-Appellants,

v.

                                                                No. 97-1315
ALEX. BROWN & SONS,
INCORPORATED; JOHN ISAAC EPLEY,
Defendants-Appellees.

SECURITIES & EXCHANGE COMMISSION;
PSA THE BOND MARKET TRADE
ASSOCIATION,
Amici Curiae.

Appeal from the United States District Court
for the District of Maryland, at Greenbelt.
Frank A. Kaufman, Senior District Judge.
(CA-95-1091-K)

Argued: September 29, 1997

Decided: December 30, 1997

Before LUTTIG and WILLIAMS, Circuit Judges, and MAGILL,
Senior Circuit Judge of the United States Court of Appeals for the
Eighth Circuit, sitting by designation.

_________________________________________________________________

Affirmed by published opinion. Senior Judge Magill wrote the opin-
ion, in which Judge Luttig and Judge Williams joined.

_________________________________________________________________
COUNSEL

ARGUED: Howard N. Feldman, DICKSTEIN, SHAPIRO, MORIN
& OSHINSKY, L.L.P., Washington, D.C., for Appellants. Susan Sho-
lar McDonald, Senior Litigation Counsel, SECURITIES AND
EXCHANGE COMMISSION, Washington, D.C., for Amicus Curiae
SEC. Michael Roger Klein, WILMER, CUTLER & PICKERING,
Washington, D.C., for Appellees. ON BRIEF: Howard Schiffman,
Woody N. Peterson, Jennifer Tara Holubar, DICKSTEIN, SHAPIRO,
MORIN & OSHINSKY, L.L.P., Washington, D.C., for Appellants.
Richard H. Walker, General Counsel, Jacob H. Stillman, Associate
General Counsel, Susan K. Straus, SECURITIES AND EXCHANGE
COMMISSION, Washington, D.C., for Amicus Curiae SEC. Robert
F. Hoyt, Adam R. Waldman, WILMER, CUTLER & PICKERING,
Washington, D.C., for Appellees. Peter Buscemi, Lloyd H. Feller,
MORGAN, LEWIS & BOCKIUS, L.L.P., Washington, D.C.; Robert
C. Mendelson, Katherine M. Polk, MORGAN, LEWIS & BOCKIUS,
L.L.P., New York, New York; Paul Saltzman, Senior Vice President
and General Counsel, PSA THE BOND MARKET TRADE ASSOCI-
ATION, New York, New York, for Amicus Curiae Association.

_________________________________________________________________

OPINION

MAGILL, Senior Circuit Judge:

Banca Cremi, S.A., Institucion de Banca Multiple, Grupo Finan-
ciero Cremi and Banca Cremi Grand Cayman (together, the Bank)
purchased a number of collateralized mortgage obligations (CMOs)
through John Isaac Epley, a broker with the brokerage firm of Alex.
Brown & Sons, Incorporated (Alex. Brown). Although most of its
CMO purchases were profitable, the Bank lost money on six CMO
purchases after the market in CMOs collapsed in 1994. The Bank
brought suit in the district court against Epley and Alex. Brown,
alleging that Epley and Alex. Brown had committed securities fraud
in violation of § 10(b) of the Securities and Exchange Act of 1934,
15 U.S.C. § 78j(b), and Rule 10b-5, 17 C.F.R. § 240.10b-5, by mak-
ing material misrepresentations and omissions regarding the CMOs,
by selling securities that were unsuitable, and by charging excessive

                    2
markups. The Bank also alleged Texas state common-law tort claims
for fraud, negligence, negligent misrepresentation, and breach of fidu-
ciary duty, and a claim based on the Maryland Securities Act. The
district court granted Epley and Alex. Brown's motion for summary
judgment on all of the Bank's claims,1 and the Bank now appeals. We
affirm.

I.

A.

CMOs, first introduced in 1983, are securities derived from pools
of private home mortgages backed by U.S. government-sponsored
enterprises.2 From 1987 to 1993, U.S. government-sponsored CMO
issuances grew dramatically, from $900 million to $311 billion per
year. The market in CMOs largely collapsed in 1994, and in 1995
new issuances fell to $25.4 billion.

Historically, investments in fixed-rate home mortgages have not
been attractive to institutional investors. Investors in most fixed-rate
securities benefit when interest rates fall. The fixed-rate security then
earns interest at a rate higher than decreased prevailing rates. How-
ever, unlike other fixed-rate investments such as U.S. treasuries,
fixed-rate home mortgages do not benefit from declines in interest
rates. Because home mortgages may be freely prepaid, home owners
frequently refinance their homes to take advantage of a drop in inter-
est rates. When the mortgage is prepaid, the investor's funds are
returned. If the investor seeks to reinvest those funds, as would be the
case with most institutional investors, they must be reinvested at the
low prevailing rate, rather than earning interest at the higher rate of
the original mortgage. This is called the "prepayment risk." If interest
rates rise, home mortgages are generally not refinanced, and they lose
value just like any other fixed-rate security. Thus, investments in
_________________________________________________________________

1 The district court's opinion is recorded at Banca Cremi v. Alex.
Brown, 

(D. Md. 1997).

2 These entities include the Federal National Mortgage Association, the
Federal Home Loan Mortgage Corporation, and the Government
National Mortgage Association.

                    3
home mortgages perform poorly both when interest rates rise and
when they fall.

CMOs concentrate the prepayment risk in some securities in order
to reduce that risk in other securities. In so doing, CMOs were
designed to make home mortgage investments more attractive to insti-
tutional investors, increase the liquidity in the secondary home mort-
gage market, and reduce the interest costs to consumers buying
homes.

A CMO issuer begins with a large pool of home mortgages, often
worth billions of dollars. Each pool of home mortgages generates two
streams of income. The first income stream is the aggregate of all
interest payments made on the underlying mortgages. The second
income stream is the aggregate of all principal payments made on the
underlying mortgages. These income streams are divided into numer-
ous CMO "tranches," which are the securities sold to investors. To
determine what portion of the two income streams are received by an
investor in a CMO tranche, each tranche has two unique formulae:
one that determines the tranche's interest rate, and the other that
determines the tranche's principal repayment priority.

The interest rate on a CMO tranche can be a fixed rate, a floating
rate, or a rate that floats inversely to an index rate. Floating interest
rates can also be leveraged, meaning that the interest rate shifts more
dramatically than the index rate. For example, where a floating rate
CMO is leveraged by a multiplier of two, the CMO's interest rate will
increase by two percent when the index rate increases by one percent.

The tranche's principal repayment priority determines when the
tranche will receive principal payments made on the underlying mort-
gages. Each principal payment is divided among all of the tranches
in a CMO issuance. High priority tranches receive principal payments
first. Support tranches receive principal payments last. Because of
this, support tranches are the most sensitive to "extension risk."
Extension risk is the opposite of prepayment risk: when interest rates
rise, the expected maturity of the support tranche CMO increases,
often dramatically.

CMO tranches are categorized into classes which have similar
properties and risks. The least risky is the planned amortization class

                    4
(PAC). PACs have little prepayment risk, and appeal to institutional
investors for this reason. Two of the riskiest classes of CMOs, inverse
floaters and inverse interest-only strips, are at issue in this litigation.

Inverse floaters have a set principal amount and earn interest at a
rate that moves inversely to a specified floating index rate. Inverse
floaters will often be leveraged, so a small increase in interest rates
causes a dramatic decrease in the inverse floating rate. Usually,
inverse floaters are also support tranches, so an increase in interest
rates causes their maturity date to extend. Inverse floaters earn high
returns if interest rates decline or remain constant, but lose substantial
value if interest rates increase.

Inverse interest-only strips (inverse IOs) do not receive principal
payments. The interest rate for an inverse IO floats inversely to a
specified index rate, like an inverse floater. Interest is calculated by
reference to the outstanding principal amount of another reference
tranche. As the reference tranche is paid off, the principal on which
the inverse IO earns interest decreases accordingly. Like an inverse
floater, a rate increase reduces the inverse IO's floating rate. Accord-
ing to some investors, a rate increase also reduces prepayment of the
reference tranche, extending the maturity of the inverse IO and, ulti-
mately, increasing the total interest payments made on the inverse IO.

Inverse floaters were first introduced in 1986. Inverse IOs were
introduced in 1987. Markets for both of these securities remained
strong in the environment of decreasing or stable interest rates that
predominated between 1986 and the beginning of 1994. On February
4, 1994, the Federal Reserve Board increased short-term interest rates
for the first time in five years. Over the next nine months, short-term
rates increased by a total of 2.5 percent, from 3 percent to 5.5 percent.
In response to the rate increases, a wave of selling hit bond markets
and investors in all types of bonds suffered significant losses.3
_________________________________________________________________

3 Most fixed-rate investments dropped significantly in value. Five-year
bonds had their worst year since 1926. The price on zero-coupon trea-
suries dropped 18.7 percent, while long-term treasury bonds lost 7.5 per-
cent in value.

                     5
CMOs were particularly hard hit, for a variety of reasons. The
jump in rates halted mortgage prepayments. This in turn extended the
average maturity of all CMOs, including, most dramatically, support
tranche CMOs such as inverse floaters. Because of their degree of
leverage, certain CMOs were extremely sensitive to the interest rate
jumps, and their holders flooded the market after the first interest rate
increase. CMO liquidity, which had never been a problem in the sta-
ble or declining interest rate environment that had existed since their
introduction, dried up as all CMO holders tried to sell. The fear of
liquidity problems built on itself, reducing the number of willing pur-
chasers during the critical period after the Federal Reserve Board
increased interest rates. In April 1994 an investment fund which pri-
marily invested in CMOs filed for bankruptcy, reporting near total
losses of its $600 million CMO investment. As a result of these inci-
dents, the market in CMOs virtually collapsed in 1994.

B.

Alex. Brown is a securities brokerage firm, incorporated under the
laws of Maryland, with its principal place of business in Baltimore,
Maryland. Alex. Brown is registered as a broker-dealer under § 15 of
the Securities and Exchange Act of 1934, 15 U.S.C. § 78o. Beginning
in April 1993, John Isaac Epley was employed by Alex. Brown in its
Houston, Texas office as a vice president. Prior to his employment by
Alex. Brown, Epley had worked in Houston as a securities broker for
MMAR Group.

Banca Cremi, S.A., is a credit institution incorporated under the
laws of Mexico. Banca Cremi Grand Cayman is its wholly owned
subsidiary, incorporated under the laws of the Cayman Islands. Both
institutions have their principal place of business in Mexico City,
Mexico. On June 30, 1993, the Bank had assets of nearly $5 billion
and an annual operating income in excess of $36 million.

The Bank's Nuevos Negocios Internacionales (NNI) unit special-
ized in international investment transactions. The unit engaged in
Eurobond issuances, interest rate swaps, investments in Brady Bonds,
and various other esoteric investments. According to the NNI monthly
reports, the NNI unit held investments with a face value of up to $115

                    6
million during 1993 and accumulated income of over $6 million that
year. J.A. at 482.

Three individuals had primary oversight responsibilities for the
NNI unit's operations and investments. The NNI unit director, Jose
Luis Mendez, held a degree in economics, and primarily advised the
Bank on U.S. dollar-denominated investments. The NNI unit subdi-
rector, Armando Aguirre, also held a degree in economics. Prior to
joining the Bank in 1981, Aguirre served as an economics teacher, a
currency investment adviser, and a developer of accounting systems.
Aguirre approved all of the NNI unit's CMO trades. The NNI unit
assistant director, Monica Buentello, held a degree in international
relations, had completed postgraduate course work in international
commerce and analysis, and had participated in seminars on deriva-
tives and CMOs while working at the Bank.

C.

The relationship between Epley and the Bank began in June 1992
when Epley, then an MMAR Group employee, made an unsolicited
phone call to sell CMOs to the Bank. Over the next few months,
Epley discussed CMOs with Buentello and others in the NNI unit.
The Bank allegedly told Epley that it wished to "invest in securities
that: [(1)] had low risk to capital; [(2)] were highly liquid; [(3)] would
be held for short periods (generally 90-180 days); and [(4)] could rea-
sonably be expected to provide a good yield." J.A. at 1537 (Buentello
Aff.). The Bank also allegedly told Epley that it would be "beneficial
if" the investments met the liquidity coefficient requirement of a
Mexican banking regulation, Circular 292.4 

Epley provided the Bank with general background materials
describing the functioning and risks of CMOs. First, Epley provided
the Bank with the MMAR Group Guide to CMO Structures (Group
Guide). The Group Guide dedicated two pages to a description of
inverse floaters and inverse IOs, calling them each volatile. Second,
Epley provided the Bank with the MMAR Group Guide to Inverse
_________________________________________________________________

4 Circular 292 requires that fifteen percent of a Mexican bank's U.S.
dollar-denominated investments be in short-term, highly liquid securities.

                     7
IOs, which described the risks and benefits of inverse IOs.5 Third,
Epley wrote a letter to Buentello dated July 22, 1992 (risk letter), out-
lining four types of risks of inverse floaters: credit risk, coupon risk,
price volatility, and liquidity risk. Describing the coupon risk, the risk
letter stated that in "most cases" the index would need to increase by
six percent before the yield of the inverse floater became zero. See
J.A. at 1437. As for price volatility, the risk letter stated that the price
had an inverse relationship to interest rates, "as with all fixed income
securities." 

liquidity risk, the letter claimed that many firms
would bid on inverse floaters and that demand "currently" far
exceeded supply. 

During the summer of 1992, the Bank's NNI unit independently
investigated the benefits and risks of CMOs. The Bank discussed the
potential investment in CMOs with its counsel and established a
fourteen-step review procedure to be followed prior to each CMO
purchase. The review procedure was later formalized into a written
manual. See J.A. at 1254-79. Buentello, with assistance from Epley,
authored a lengthy analysis entitled "Banca Cremi Investment System
in Inverse Floater." See 

The analysis concluded that
inverse floaters are leveraged "to obtain extraordinary returns" of
around twenty percent. 

The Bank's analysis explained that since the market for inverse
floaters began, high returns and decreasing interest rates had given
rise to "a more and more liquid market." J.A. at 657-58. The analysis
described the interest rate structure of inverse floaters in detail, using
charts and graphs to illustrate the sensitivity of inverse floaters to
shifts in the index rate. Similar to Epley's risk letter, the Bank's anal-
ysis recognized that if the index rate increased by six percent, the
yield of an inverse floater CMO would dwindle to nothing. The analy-
sis noted that, "[l]ike all fixed rate securities, there is an inverse rela-
tionship between interest rates and bond price," and that interest rates
ultimately "influenc[e] returns on the bonds." 

Finally,
while the analysis indicated that investments in CMOs complied with
Circular 292, the Bank also reasoned that if these investments did not
_________________________________________________________________

5 It is not clear exactly when these brochures were provided to the
Bank, but it is undisputed that both were provided prior to the purchase
of the CMOs that are the subject of this suit.

                     8
comply with Circular 292, they would be like any other international
investment. 

659.

Between August 1992 and August 1993, the Bank made additional
efforts to refine its knowledge of CMO investing. In October 1992
NNI unit director Mendez purchased several lengthy treatises that
described mortgage investments and CMOs in extensive detail. These
books were made available to the NNI unit staff. In May 1993 Buen-
tello attended a seminar on derivatives investing. Buentello also
attended a seminar at which CMO investing and pricing methods
were discussed.

Throughout this period, the Bank was courted by other brokerage
houses who sought to obtain the Bank's CMO business, and the Bank
authorized the NNI unit to engage in CMO transactions with these
brokerage houses. Each brokerage house provided the Bank with their
own internal documents describing the benefits and risks of CMO
investing. In January 1993 one of these brokerage houses forwarded
to the Bank an article warning of what could occur to CMO invest-
ments if interest rates were to rise: "[i]f interest rates rise . . . what
investors thought was a safe, secure medium-term maturity can sud-
denly be transformed into a highly risky long-term security." Randall
W. Forsyth, The Pinocchio Security: Here's the Awful Truth About
CMOs, Barron's, Jan. 18, 1993, at 15. The article suggested that, in
these circumstances, a CMO's price would drop ten to twenty percent
"if you can get a bid for it at all." 

to the Bank that
he disagreed with the Barron's article, and wrote in a letter that the
risks of a CMO were "less of a concern" for an institutional investor.
J.A. at 1441. Epley included a different Barron's article by Andrew
Bary, who "specialize[d] in covering the institutional investor's capi-
tal markets." J.A. at 1442.6
_________________________________________________________________

6 The parties have not included this article in the record, which is
referred to in Epley's letter as having been published "one month prior"
to The Pinocchio Security. During that period, Bary authored only two
articles on CMO investing for Barron's, and both were critical of CMOs.
The first article reasoned that "[i]nstitutional demand" had helped the
CMO market grow, despite the fact that "liquidity remains notoriously
poor," even for large institutional investors. Andrew Bary, Capital Mar-
kets: Trading Points, Barron's, Nov. 23, 1992, at 48. In the other, Bary

                   9
The Bank purchased its first CMO in August 1992. For the next
year and a half, the Bank purchased a total of twenty-nine CMOs.
Epley was in continual contact with the Bank during its period of
CMO trading. Epley sent numerous faxes and letters to the Bank
encouraging the Bank to make additional CMO purchases. Most of
these suggested purchases were not pursued by the Bank. Epley intro-
duced the NNI unit to a leading CMO expert and finance professor,
Frank J. Fabozzi. Fabozzi was made available to the Bank for techni-
cal consultation and sent the Bank textbooks he had written which
described CMOs and other financial instruments. Additionally, on
request, Alex. Brown would perform a "portfolio analysis" of the
Bank's holdings.7 Prior to each CMO purchase, Epley provided the
Bank with the yield matrix for that CMO. The yield matrix set forth
the formula of the CMO's floating interest rate. It also provided a
table that indicated how the CMO's yield and average maturity
changed when interest rate and prepayment conditions changed.
Although the yield matrix indicated the average maturity of the CMO,
it did not specify the formula that was used to calculate the precise
maturity of the CMO.
_________________________________________________________________

explained that "[w]hen rates fall, the yield on the inverse floater jumps,
and when rates rise, the yield on the inverse floater plunges . . . resulting
in a sharp drop in the security's price." Andrew Bary, Capital Markets:
Trading Points, Barron's, Dec. 14, 1992, at 54. Bary asked:

        How did an opportunity exist to buy federal-agency mortgage
        securities yielding 20%? Because most portfolio managers . . .
        were afraid of the price volatility of inverse floaters.

that "[i]f a broker offers one, it pays, of course, to be
very careful," in part, because "[e]ven mortgage experts have a hard time
valuing them." 

Bank only requested one such analysis, which was performed by
Alex. Brown in July 1993. Alex. Brown's analysis provided a table that
showed the sensitivity of the price of each CMO the Bank owned to
interest rate changes. The July 1993 analysis indicated that, of the three
types of CMOs the Bank held at that time, the prices of both its inverse
floaters and its inverse IOs were more sensitive to interest rate changes
than its PACs.

                     10
The CMOs initially earned interest at rates as high as twenty-four
percent. The Bank played the CMO market aggressively, trading
CMOs frequently to take advantage of short-term market swings. The
Bank sold eight CMOs within three weeks of purchasing them,
including one trade made within twenty-four hours of the Bank's pur-
chase. The Bank sold a total of twenty-three CMOs, each at a profit,
prior to the CMO market downturn in March 1994. The aggregate
price of these twenty-three CMOs exceeded $96 million. The Bank's
profits exceeded $2 million.

Alex. Brown never charged the Bank a commission for the Bank's
CMO purchases. Rather, Alex. Brown purchased the securities and
then sold them to the Bank with a markup. The Bank never asked
Epley or Alex. Brown what markup it placed on the CMOs, and Alex.
Brown never disclosed the amount of markup. Although Alex. Brown
found purchasers for each of the twenty-three CMOs sold by the
Bank, Alex. Brown never charged the Bank a markdown on the sales.

The Bank held six CMOs at the time of the market downturn in
March 1994.8 These six CMOs are the subject of the instant suit.
After the market downturn in February 1994, the price of the six
CMOs dropped precipitously. The Bank claims losses on the six
CMOs of around $21 million of the original purchase price of around
$40 million.9

The Bank filed a complaint for securities fraud against Epley and
_________________________________________________________________

8 These include: (1) FN 93-169 SC, an inverse floater purchased on
August 31, 1993, for approximately $8 million; (2) FN 93-203 SA, an
inverse floater purchased on September 23 and 28, 1993, for approxi-
mately $10.1 million; (3) FN 93-210 SD, an inverse floater purchased on
December 6, 1993, for approximately $9.1 million; (4) FHLMC 1676 S,
an inverse floater purchased on January 24, 1994, for approximately $4.6
million; (5) FNR 94-29 SD, an inverse floater purchased on February 2,
1994, for approximately $4.9 million; and (6) FHR 1711 S, an inverse
IO purchased on March 4, 1994, for approximately $6.1 million.

9 The Bank's CMO losses were only the beginning of its troubles. Later
in 1994, the Bank's chairman disappeared with $700 million of the
Bank's assets, leading the Bank to be put into receivership by Mexican
banking authorities.

                   11
Alex. Brown in the United States District Court for the District of
Maryland. The Bank claimed that Epley and Alex. Brown violated
§ 10(b) and Rule 10b-5 by: (1) making material misstatements and
omissions regarding the CMOs sold to the Bank; (2) selling the Bank
the CMOs which it knew to be unsuitable investments for the Bank;
and (3) failing to disclose fraudulently excessive markups totaling $2
million on the CMO sales. The Bank also claimed violations of the
Maryland Securities Act, common law breach of fiduciary duty, neg-
ligence, negligent misrepresentation, and fraud.

In support of their claim that Epley and Alex. Brown made numer-
ous material omissions and misstatements, the Bank alleged that
Epley and Alex. Brown failed to provide the Bank with material
information about the functioning of each of the CMOs, including: (1)
the impact of interest rates on CMO price; (2) the precise principal
repayment priority; and (3) the impact of interest rates on CMO
liquidity. The Bank also alleged that the defendants failed in their
duty to provide: (1) information normally included in a CMO
prospectus10; (2) sophisticated computer software to reverse engineer
new issue CMOs to reveal their performance in various market condi-
tions; (3) information regarding the CMOs' qualification under Mexi-
co's Circular 292; and (4) information regarding the CMOs'
qualification under U.S. banking regulations. The Bank also alleged
that Epley and Alex. Brown made material misstatements when they:
(1) downplayed the impact of a change in interest rates on CMO
prices by comparing it to that of other fixed income securities; (2)
stated that demand for CMOs currently far exceeds supply; (3)
claimed that CMOs had relatively low risk levels; (4) claimed that the
inverse IOs sold to the Bank had a self-hedging feature; and (5) told
the Bank that the criticisms in the Barron's Pinocchio Security article
should be less of a concern for institutional investors than for individ-
ual investors.

After discovery, the district court granted summary judgment to
Epley and Alex. Brown on all claims. The district court concluded
that the Bank had failed to raise a genuine issue of material fact as
_________________________________________________________________

10 The Bank does not claim that a prospectus, itself, should have been
provided. In fact, it is unclear whether prospectuses for the CMO tran-
ches existed at the time of their purchase.

                    12
to three of the four elements that must be proved to establish a viola-
tion of § 10(b). First, the district court reasoned that Epley and Alex.
Brown had made no material omissions or misstatements regarding
the risks of the CMOs because they made general disclosures of yield,
price, and liquidity risks. Further, the district court found that the
Bank's allegation that scienter existed because Epley and Alex.
Brown encouraged purchases in order to earn excessive markups to
be based only on speculation. Finally, the district court concluded that
several factors, including the Bank's size and its sophistication and
the expertise of the NNI unit employees, did not allow the Bank to
justifiably rely on any misstatements that Epley and Alex. Brown
might have made.11

The district court rejected the Bank's suitability claim as a subset
of the rejected § 10(b) claim. The district court also rejected the
Bank's claim that the markups were excessive, concluding that most
of the markups were within standards provided by the National Asso-
ciation of Securities Dealers (NASD), and that no evidence had been
submitted to establish that the higher markups were not justified by
special circumstances.

The district court also rejected the Bank's state law claims, holding
that the Maryland Securities Act did not apply to Alex. Brown
because it was a broker-dealer, and that, based on either Texas or
Maryland state law, the Bank's tort claims failed as a matter of law.
The Bank appeals the district court's grant of summary judgment.

II.

We review the district court's grant of summary judgment de novo.
See Myers v. Finkle, 

, 167 (4th Cir. 1991). To defeat the
motion for summary judgment, the Bank must raise a genuine issue
of material fact as to each element essential to its case. See Celotex
Corp. v. Catrett, 

, 322 (1986). A factual issue raised by
the Bank is only genuine if a reasonable jury could return a verdict
for the Bank on each element necessary to its case. See Anderson v.
Liberty Lobby, Inc., 

, 248 (1986).
_________________________________________________________________

11 The district court did not address proximate causation, the final ele-
ment of a § 10(b) claim.

                    13
To establish liability under § 10(b), the Bank must prove that "(1)
the defendant[s] made a false statement or omission of material fact
(2) with scienter (3) upon which the plaintiff justifiably relied (4) that
proximately caused the plaintiff's damages." Cooke v. Manufactured
Homes, Inc., 

, 1260-61 (4th Cir. 1993) (quotations and
citation omitted). In this case, the Bank has failed to present evidence
that would permit a reasonable jury to find that the Bank justifiably
relied on any omissions or misstatements made by Epley and Alex.
Brown.12

A.

A dissatisfied investor cannot recover for a poor investment on the
basis of a broker's alleged omission or misstatement where, "through
minimal diligence, the investor should have discovered the truth."
Brown v. E.F. Hutton Group, Inc., 

, 1032 (2d Cir.
1993). Because the justifiable reliance requirement "requir[es] plain-
tiffs to invest carefully," it "promotes the anti-fraud policies" of the
securities acts by making fraud more readily discoverable. Dupuy v.
Dupuy, 

, 1014 (5th Cir. 1977). A plaintiff's failure to
prove that it justifiably relied on a broker's alleged omission or mis-
statement is necessarily fatal to a securities fraud claim. See 

.13

An investor's reliance on a broker's omission or misstatement is
never justified when the "investor's conduct rises to the level of reck-
lessness." 

.14 A plaintiff is reckless if he "in-
_________________________________________________________________

12 Because the Bank cannot show justifiable reliance, its § 10(b) action
fails as a matter of law and we need not address the district court's other
grounds for granting summary judgment against the plaintiff on this
claim.

13 Courts often infer reliance when a plaintiff alleges that the defendant
made a material omission, see Carras v. Burns, 

, 257 (4th
Cir. 1975), requiring the defendant to rebut the inference by proving the
lack of reliance. See 

of reliance exists, however,
"when the plaintiff alleges both nondisclosure and positive misrepresen-
tation instead of only nondisclosure." Cox v. Collins, 

, 395-96
(4th Cir. 1993).

14 Indeed, courts traditionally have held that reliance is not justified
when the investor was merely negligent. See Royal Am. Managers, Inc.

                     14
tentionally refuse[s] to investigate in disregard of a risk known to him
or so obvious that he must be taken to have been aware of it, and so
great as to make it highly probable that harm would follow." 

(quotations and citation omitted). Stated differently,
a plaintiff is reckless if he "possesses information sufficient to call [a
mis-] representation into question," but nevertheless "close[s] his eyes
to a known risk." Teamsters Local 282 Pension Trust Fund v.
Angelos, 

, 530 (7th Cir. 1985) (citation omitted). Thus,
"[i]f the investor knows enough so that the lie or omission still leaves
him cognizant of the risk, then there is no liability." 

weighs eight factors to determine whether an investor
is justified in relying on a material omission or misstatement:

        (1) the sophistication and expertise of the plaintiff in
        financial and securities matters;

        (2) the existence of long standing business or personal
        relationships;

        (3) access to relevant information;

        (4) the existence of a fiduciary relationship;

        (5) concealment of the fraud;

        (6) the opportunity to detect the fraud;

       (7) whether the plaintiff initiated the stock transaction or
       sought to expedite the transaction; and
_________________________________________________________________

v. IRC Holding Corp., 

, 1015 (2d Cir. 1989). Many federal
courts reconsidered the negligence standard after the Supreme Court, in
Ernst & Ernst v. Hochfelder, 

, 193 (1976), held that a per-
son could only be liable under § 10(b) for intentional conduct. See Dupuy
v. Dupuy, 

, 1017-20 (5th Cir. 1977). At least one court has
apparently retained the negligence standard. See Straub v. Vaisman &
Co., 

, 598 (3d Cir. 1976) ("The obligation of due care must
be a flexible one, dependent upon the circumstances of each case. We
require only that the plaintiff act reasonably.").

                     15
        (8) the generality or specificity of the misrepresentations.

(quotations and alteration omitted). No single
factor is dispositive of whether reliance is justified. 

Myers factor, the sophistication of the investor, has long been a criti-
cal element in determining whether an investor was entitled to § 10(b)
relief. See Kohler v. Kohler Co., 

, 642 (7th Cir. 1963)
(considering an investor's "business acumen" and access to "extrinsic
sources of sound business advice" to conclude there was no reliance,
although the transaction might not have been fair if the investor "had
been a novice"); List v. Fashion Park, Inc., 

, 463-64 (2d
Cir. 1965) (no reliance where investor was "an experienced and suc-
cessful investor in securities" who did not ask his broker for informa-
tion regarding the claimed omission). A sophisticated investor
requires less information to call a "[mis-]representation into question"
than would an unsophisticated investor. 

.
Likewise, when material information is omitted, a sophisticated inves-
tor is more likely to "know[ ] enough so that the . . . omission still
leaves him cognizant of the risk." 

investor is an individual, this Court looks to several fac-
tors to determine if the investor is sophisticated, including "wealth[,]
. . . age, education, professional status, investment experience, and
business background." 

. Some of these factors
may not be perfectly suited for application to an institutional investor.
Cf. C. Edward Fletcher, Sophisticated Investors Under the Federal
Securities Laws, 1988 Duke L.J. 1081, 1149-53 (reviewing factors
gauging sophistication of individual investors, and concluding that
there should be a "conclusive presumption" that all institutional inves-
tors are sophisticated). However, the factors which are relevant to an
institution strongly support the sophistication of the Bank. The Bank,
with assets of $5 billion, is unquestionably wealthy. In addition, while
the Bank's investment choices may have been unwise, its investment
experience is extraordinary, and far surpasses most sophisticated indi-
vidual investors. As a business entity, the Bank obviously has a busi-
ness background, and its employees--hired for their business
expertise--had extensive education and experience in economics and
finance.15
_________________________________________________________________

15 The NASD has also promulgated factors specifically designed to
measure the sophistication of institutional investors. Order Approving

                    16
Despite its extensive investment experience and extraordinary
resources, the Bank nevertheless contends that, while it may be
sophisticated in certain types of investments, it was not sophisticated
in CMO investments. See, e.g., McAnally v. Gildersleeve, 

, 1500 (8th Cir. 1994) (recognizing that an individual's sophisti-
cation in "stocks and bonds" did not necessarily suggest sophistica-
tion in commodities futures options); Order Approving NASD
Suitability Interpretation, 61 Fed. Reg. 44,100, 44,112 (1996) (NASD
Fair Practice Rules) (in approving NASD fair practice rules, SEC rec-
ognized that even a sophisticated institutional investor may not be
capable of understanding a "particular investment risk"). The Bank
argues that deposition testimony of its employees and an expert wit-
ness that the Bank was unsophisticated in CMO investments created
a genuine issue of fact.

We reject this argument. The Bank's NNI unit, whose function was
to invest Bank funds in dollar-denominated investments, employed
three well-educated investment professionals to select a sound, but
profitable, investment strategy. Mendez, Aguirre, and Buentello con-
ducted a thorough, independent investigation of the benefits and risks
of CMO investments by attending seminars, purchasing treatises on
the subject, and developing a multi-step review process for each
CMO investment. Rather than blindly relying on Epley and Alex.
Brown, the record shows that the Bank rejected Epley's suggested
investments far more often than it accepted them. Indeed, the Bank
consulted with five other brokerage houses regarding CMO invest-
ments, and each of these brokerage houses gave the Bank detailed
information describing the benefits and the risks of CMO invest-
ments. After a year of trading in CMOs, the Bank displayed a knowl-
edge and an aggressiveness that belie its current claim that it did not
understand CMO investments. See J.A. at 447-48 (indicating dramatic
price changes over short time periods for many of the Bank's profit-
_________________________________________________________________

NASD Suitability Interpretation, 61 Fed. Reg. 44,100, 44,112 (1996).
The two most important factors are an institutional investor's "capability
to evaluate investment risk independently, and the extent to which the
[investor] is exercising independent judgment." 

In light of
the undisputed record, it is clear that the Bank would also qualify as a
highly sophisticated institutional investor under the NASD standards.

                    17
able CMO trades: FNMA 92 112 SC was sold after one day at a profit
reflecting an annual increase of over 350%; FN 93-115 SB was sold
after two weeks at a profit reflecting an annual increase of around
58%; FNMA 1992 162 SB was sold after two weeks at a profit
reflecting an annual increase of around 38%); NASD Fair Practice
Rules, 61 Fed. Reg. at 44,105 n.20 ("[An institution] who initially
needed help understanding a potential investment may ultimately
develop an understanding and make an independent investment deci-
sion."). Accordingly, we agree with the district court that the Bank
was a "sophisticated investor" for the purposes of this case. See Banca
Cremi v. Alex. Brown, 

, 515 (D. Md. 1997).

The second Myers factor also lends no support to the Bank's claim
of justifiable reliance. There is no evidence in the record that the
Bank enjoyed a long-standing business or personal relationship with
Epley or Alex. Brown, and the undisputed evidence strongly supports
the lack of any such relationship. The Bank began purchasing CMOs
a mere two months after first being "cold called" by Epley, and the
conduct of the Bank--consulting with competing brokerage houses
and rejecting most of Epley's recommendations--suggests that the
Bank dealt with Epley and Alex. Brown at arm's length. Accordingly,
this is not a situation where the defendants could have "exploited the
business relationship [with the plaintiff] knowing that [the plaintiff]
was not likely to investigate the merits of [the defendants'] recom-
mendation." Straub v. Vaisman & Co., 

, 598 (3d Cir.
1976) (quotations omitted).16

The third, fifth, and sixth Myers factors look to whether the Bank
_________________________________________________________________

16 Relying on innuendo and veiled suggestions, the Bank apparently
contends that Buentello and Epley had a sexual relationship, and that
Epley and Alex. Brown exploited that relationship. We first note that an
appellate brief is not a paperback romance novel, where tawdry goings-
on can be hinted at with a smirk and a wink; if the Bank had a point to
make with its innuendo, we would have preferred that it were clear about
it. As there is not a shred of evidence in the record to support the Bank's
salacious suggestions, and even less evidence to suggest that this hinted-
at relationship had any effect on Buentello's business decisions, we find
it unfortunate that the Bank found it necessary to tarnish the reputations
of the opposing party and its own employee.

                    18
had access to the relevant information on CMOs, whether Epley and
Alex. Brown concealed the alleged fraud, and whether the Bank had
an opportunity to detect the fraud. In this case, there is no allegation
that Epley or Alex. Brown concealed specific risks of individual
investments, but rather that they misrepresented the risks associated
with an entire field of investment. Clearly, the Bank--through its
independent research, contacts with other brokerage houses, and dis-
cussions with Epley and Alex. Brown--not only had access to, but
actually possessed more than sufficient information to make it aware
of the substantial risks of investing in CMOs. The Bank knew that,
although it could enjoy substantial earnings from CMO investments
if interest rates decreased or remained the same, any increase in inter-
est rates could wreak havoc on its CMO investment strategy. The
Bank also knew that more sophisticated analyses of its CMO portfolio
was available, such as the price analysis performed by Alex. Brown
on the Bank's portfolio in July 1993. The July 1993 price analysis
indicated that both inverse floaters and inverse IOs were more sensi-
tive to interest rate increases than other types of CMOs, not to men-
tion other more conservative fixed-rate investments such as U.S.
Treasury Bonds. Despite possessing this information, the Bank pur-
chased the six CMOs and never requested any analysis either before
or after purchase.

The fourth Myers factor, whether the defendants owed a fiduciary
duty to the plaintiff, is also not implicated in this case. As will be dis-
cussed below, see infra, § V, Epley and Alex. Brown were not the
agents of the Bank, but rather interacted with the Bank at arm's length
in principal-to-principal dealings, and no common law fiduciary duty
was ever created.

The seventh Myers factor looks to whether the Bank initiated or
sought to expedite the transaction. While the Bank had sufficient time
to review each of its CMO purchases, and evidenced independent
decision-making by rejecting numerous suggested purchases by Epley
and employing elaborate procedures to review each suggested pur-
chase, it was Epley who initially suggested these investments to the
Bank. Accordingly, we agree with the Bank that this factor lends
some support to its argument for justifiable reliance. But see Chance
v. F.N. Wolf & Co., No. 93-2390, 

, at *6 (4th Cir.
Sept. 30, 1994) (unpublished, table decision reported at 

)

                     19
(defendant entitled to judgment as a matter of law on justifiable reli-
ance despite fact that defendant "initiated all of the stock transac-
tions").

The final Myers factor looks to whether the misrepresentations
were general or specific. The Bank argues, and we agree, that the
defendants' alleged misstatements were general. Contrary to the
Bank, however, we conclude that a general statement creates less jus-
tifiable reliance than would a specific statement. See Hillson Partners
Ltd. Partnership v. Adage Corp., 

, 215 (4th Cir. 1994)
(explaining that investor is more justified in relying on specific pre-
dictions); Zobrist v. Coal-X, Inc., 

, 1518 (10th Cir.
1983) (noting that there is no "valid reason" to rely on general misrep-
resentations as to risk when more specific warnings have been pro-
vided); Hughes v. Dempsey-Tegeler & Co., 

, 177 (9th
Cir. 1976) (sophisticated investor was not justified in relying on gen-
eral positive comments regarding investment risks). Epley's general
positive statements concerning CMOs did not justify reliance when
the Bank possessed a variety of resources, including investment-
specific yield and average life tables, scholarly works, and an article
published in a popular business journal explaining in great detail the
workings and risks of CMOs.

In sum, in this case the Bank had access to an extraordinary wealth
of information regarding CMOs. With few exceptions, the depth and
breadth of this information illustrated one overriding point: invest-
ments in CMOs, while potentially very profitable, were undoubtedly
highly risky. As a sophisticated business entity handling five billion
dollars of other people's money, the Bank had the advice of its own
employees and a horde of the defendants' competitors. Nevertheless,
the Bank invested in CMOs through arm's length dealings with the
defendants. While the vast majority of these investments were profit-
able for the Bank, a half-dozen proved disastrously timed,17 and the
_________________________________________________________________

17 The Bank has never alleged that the defendants had any reason to
believe that the CMO market would collapse when it did. Indeed, in a
meeting at which Buentello discussed its CMO losses, the Bank con-
cluded that the CMO price drop was caused by the "totally unpredictable
situation [of an increased estimate of U.S. GNP] and the unprecedented

                    20
Bank now alleges that its misfortune resulted from its justifiable
naivete in listening to the defendants' purported lies.

As in any "action[ ] for fraud, reliance on false statements must be
accompanied by a right to rely." Foremost Guar. Corp. v. Meritor
Sav. Bank, 

, 125 (4th Cir. 1990). Here, the Bank lost its
right to rely by its own recklessness. The Bank continued to purchase
CMOs after it had sufficient information, given its sophistication, to
be well apprised of the risks it would face were interest rates to rise.
Given that the Bank was aware of the risks involved in investing in
CMOs, the Bank was not justified in relying on Epley and Alex.
Brown's alleged omissions and misstatements. Accordingly, we
affirm the district court's grant of summary judgment against the
Bank on this claim.

B.

The Bank next contends that Epley and Alex. Brown learned that
CMO investments did not comply with the requirements of Mexican
Circular 292 and U.S. banking regulations, but failed to inform the
Bank of this. Because the defendants allegedly had a duty to the Bank
to inform them of this understanding of Mexican and United States
law, the Bank contends that the Bank justifiably relied on this omis-
sion. Because the Bank contends that it would not have invested in
the CMOs had it possessed this information, the Bank contends that
the omission was material and caused its investment losses.

While the plaintiff has presented a very creative claim, it is not a
meritorious one. Any non-reckless Mexican bank attempting to inter-
pret Mexican banking regulations would turn to Mexican lawyers or
_________________________________________________________________

move by the Fed to increase short-term rates in a preventative fashion."
J.A. at 458. In light of this recognition by the plaintiffs that their losses
were caused by general market forces rather than the defendants' alleged
misdeeds, it is puzzling at best on what basis the plaintiffs hoped to
prove liability under § 10(b). Cf. Bastian v. Petran Resources Corp., 

, 685 (7th Cir. 1990) (noting that "[d]efrauders are a bad lot and
should be punished, but Rule 10b-5 does not make them insurers against
national economic calamities").

                     21
Mexican government officials for assistance, rather than relying on an
American salesman's opinion. Furthermore, while the requirements of
Circular 292 are not discussed in detail by the parties, it appears
undisputed that to comply, a security must have a stated maturity of
less than one year. As many home mortgages have maturities of sev-
eral decades, we believe that only a reckless investor would fail to
recognize that investments in CMOs might not comply with Mexican
law. Accordingly, even if the Bank had some legitimate, or even
coherent, reason for eschewing appropriate legal assistance in inter-
preting Circular 292, any reliance on the defendants' silence was not
justifiable. See 

(no justifiable reliance where
a plaintiff "intentionally refuse[s] to investigate in disregard of a risk
known to him or so obvious that he must be taken to have been aware
of it, and so great as to make it highly probable that harm would fol-
low" (quotations and citation omitted)).

III.

The Bank next claims that Epley and Alex. Brown committed
§ 10(b) fraud by selling securities to the Bank that the defendants
knew were not suited for the Bank's investment goals. We disagree.

While this Court has never considered an unsuitability claim under
§ 10(b), several courts have recognized an unsuitability claim in cer-
tain circumstances. See, e.g., O'Connor v. R.F. Lafferty & Co., 

, 898 (10th Cir. 1992) (recognizing two types of unsuitability
claims, one based on § 10(b) fraud and one similar to churning
claim); Craighead v. E.F. Hutton & Co., 

, 493 (6th Cir.
1990) (recognizing unsuitability claim as a type of fraud claim);
Lefkowitz v. Smith Barney, Harris Upham & Co., 

, 155
(1st Cir. 1986) (per curiam) (same); Clark v. John Lamula Investors,
Inc., 

, 600-01 (2d Cir. 1978) (recognizing unsuitability
claim).

A § 10(b) unsuitability claim has five elements:

        (1) that the securities purchased were unsuited to the
        buyer's needs;

                    22
        (2) that the defendant knew or reasonably believed the
        securities were unsuited to the buyer's needs;

        (3) that the defendant recommended or purchased the
        unsuitable securities for the buyer anyway;

        (4) that, with scienter, the defendant made material mis-
        representations (or, owing a duty to the buyer, failed to dis-
        close material information) relating to the suitability of the
        securities; and

        (5) that the buyer justifiably relied to its detriment on the
        defendant's fraudulent conduct.

Brown v. E.F. Hutton Group, Inc., 

, 1031 (2d Cir.
1993) (emphasis added). A claim for § 10(b) suitability fraud "is a
subset of the ordinary § 10(b) fraud claim." Id.; see also 

(Court recognizing that this type of suitability claim
could be analyzed "simply as a misrepresentation or failure to dis-
close a material fact. In such a case, the broker has omitted telling the
investor the recommendation is unstable for the investor's interests.
The court may then use traditional laws concerning omission to exam-
ine the claim." (citation omitted)).

Because the Bank's suitability claim is a "subset" of the Bank's
§ 10(b) claim, we conclude that it must fail for the same reason: lack
of justifiable reliance. The Bank had sufficient information concern-
ing the risks of CMOs to render unjustified any reliance on a recom-
mendation that the securities were suitable investments. Yield tables
apprised the Bank that inverse floaters' yield decreased and average
maturity increased dramatically if interest rates increased. In light of
its sophistication, the Bank must be presumed to have been aware
that, were interest rates to increase, there would be substantial "risk
to capital," contrary to one of its stated goals. 18
_________________________________________________________________

18 For example, the yield table for one of the six CMOs, FN 93-210 SD,
indicates that a one point increase in interest (accompanied, as the table
suggests, by a reduction in prepayments) would convert a security matur-
ing in 3.8 years and yielding 14.8 percent interest into a security matur-
ing in 16.11 years and yielding only 7.42 percent interest. Naturally, an
investor as sophisticated as the Bank would understand that the result of
this change would be a dramatic fluctuation in the market price of the
security.

                    23
Similarly, the Bank knew that CMOs were a new product whose
demand had grown because the market remained strong, and knew
that a Barron's article had warned that an interest rate increase might
lead to liquidity problems. Thus, given the Bank's sophistication, it
must have been aware that if interest rates increased, its CMOs might
not remain "highly liquid," contrary to another of the Bank's stated
goals.

Finally, the Bank was apprised in the yield tables that the average
maturity for each of the CMOs was well in excess of the "90-180
days" it claims as its investment goal. Given the Bank's sophistica-
tion, it must have understood that it might have to hold the security
for over 180 days if it did not want to accept the going market price
for the CMO. See J.A. at 447-48 (Bank earned a substantial profit in
selling three CMOs after holding them for over 180 days). In sum, the
Bank appeared to have been most interested in its final stated goal:
"good yield." The only reasonable conclusion is that the Bank itself
chose its CMO investment strategy by balancing CMO risks and ben-
efits against its four competing goals.19 The Bank cannot now claim
that its investment strategy was the result of it justifiably relying on
the recommendations of Epley and Alex. Brown.

IV.

Next, the Bank contends that Epley and Alex. Brown fraudulently
charged excessive markups for the CMOs purchased by the Bank, and
that the district court erred in granting summary judgment on this
claim. We disagree.

"A securities broker's mark-up equals the price charged to the cus-
tomer minus the prevailing market price." Bank of Lexington & Trust
Co. v. Vining-Sparks Sec., Inc., 

, 613 (6th Cir. 1992). For
_________________________________________________________________

19 The Tenth Circuit also recognized a distinct type of suitability claim
which arises when a broker's act of purchasing a stock is fraudulent.
O'Connor v. R.F. Lafferty & Co., 

, 898 (10th Cir. 1992).
One element of this claim is that "the broker exercised control over the
investor's account." 

Bank does not claim that Alex.
Brown controlled the Bank's CMO account, this type of suitability claim
is unavailable to the Bank.

                    24
example, where the securities purchased have a market value of
$100,000 and the broker charges its customers $105,000, the markup
is $5000, or five percent. Where, as in the instant case, the broker
charges no commission, the markup represents the sole compensation
received by the broker for the transaction.

"A mark-up is excessive when it bears no reasonable relation to the
prevailing market price." 

in determining reason-
ableness include:

        (1) the type of security involved; (2) the availability of the
        security in the market; (3) the price of the security; (4) the
        amount of money involved in a transaction; (5) disclosure;
        (6) the pattern of markups or markdowns; and (7) the nature
        of the member's business.

3C Harold S. Bloomenthal, Securities and Federal Corporate Law,
App. 12.13 (Apr. 1, 1992). Although each case requires an indepen-
dent analysis for determining whether a given markup is reasonable,
the Board of Governors of the NASD has determined that, as a gen-
eral guideline, a five percent markup is reasonable. See id.20

Although securities brokers are required to disclose markups in
equity securities, see 17 C.F.R. § 240.10b-10(a)(2)(ii)(A) & (B), there
is no requirement for a broker to disclose the amount of markup
charged for a debt security in a riskless transaction. See 

once considered promulgating regulations requiring such a disclosure.
See Release No. 34-15220, 15 S.E.C. Docket 1260, 

(S.E.C.) (Oct. 6, 1978), at *1. However, the SEC withdrew its pro-
posed regulations "because [the SEC] has concluded that [the regula-
tions] would not achieve the purposes of the proposal at an acceptable
cost and that there are alternative ways of achieving the same goal
with fewer adverse side effects." Release No. 34-18987, 25 S.E.C.
Docket 1223, 

(S.E.C.) (Aug. 20, 1982), at *2.
_________________________________________________________________

20 Because the reasonableness of markups is situation-specific, the
NASD guideline recognizes that markups of under five percent may be
excessive. See 3C Harold S. Bloomenthal, Securities and Federal Corpo-
rate Law, App. 12.13 (Apr. 1, 1992).

                    25
The SEC has brought administrative actions for fraud under 17
C.F.R. § 240.10b-5 against securities brokers who have allegedly
charged excessive markups to their debt securities customers without
disclosing the markups. See, e.g., Ettinger v. Merrill Lynch, Pierce,
Fenner & Smith, Inc., 

, 1033 (3d Cir. 1987) (citing
cases). The SEC has explained that

        [i]nherent in the relationship between a dealer and his cus-
        tomer is the vital representation that the customer will be
        dealt with fairly, and in accordance with the standards of the
        profession. It is neither fair dealing nor in accordance with
        such standards to exploit trust and ignorance for profits far
        higher than might be realized from an informed customer.
        It is fraud to exact such profits through the purchase or sale
        of securities while the representation on which the relation-
        ship is based is knowingly false. This fraud is avoided only
        by charging a price which bears a reasonable relation to the
        prevailing price or disclosing such information as will per-
        mit the customer to make an informed judgment upon
        whether or not he will complete the transaction.

Trost & Co., 12 S.E.C. 531, 535 (1942). Because the securities dealer
creates this implied duty to disclose excessive markups by "hang[ing]
out its professional shingle," this theory of liability is referred to as
the "shingle theory" by the SEC. See Reply Br. of the Amicus Curiae
SEC at 2.

In Ettinger, the Third Circuit recognized a private cause of action
for a violation of this implied duty to disclose certain markups. 

, 1036 (failure to disclose excessive markup action-
able under Rule 10b-5 despite dealer's compliance with Rule 10b-10).
To date, it appears that only the Sixth Circuit has also recognized
such a private cause of action. See Bank of 

(affirming denial of liability for alleged fraud, and concluding that
"[a]lthough an undisclosed mark-up of 5 percent on municipal bonds
might sometimes violate Rule 10b-5, the Bank fails to convince us
that the district court clearly erred when it found that a 5 percent
mark-up on the bonds sold to the Bank, without more, was not so
excessive as to require disclosure"). As with all allegations of fraud
under § 10(b), a plaintiff alleging a "shingle theory" failure to disclose

                    26
an excessive markup must present evidence to satisfy four elements:
(1) a misrepresentation or omission of a material fact; (2) made with
scienter; (3) upon which the plaintiff justifiably relied; and (4) which
proximately caused the plaintiff's damages. See 

(describing elements of fraud).

In the instant case, the Bank alleged that Epley and Alex. Brown
charged undisclosed and excessive markups on nineteen CMO
transactions.21 These included two markups of 5.25 percent, one
markup of 4.99 percent, seven markups of between 3.1 percent and
3.77 percent, seven markups of between 2.4 percent and 2.84 percent,
and two markups of 2.06 percent and 1.78 percent, respectively. In
brokering over $100,000,000 in CMOs to the Bank, Epley and Alex.
Brown received some $2,000,000 in markups.

Although all but two of the markups in the instant case were below
the NASD's five percent guideline, the SEC contends that "there is
no safe harbor of five percent for markups on any securities," Reply
Br. of Amicus Curiae SEC at 11, and the Bank alleges that all nine-
teen of the markups were excessive. To support its allegation of
excess, the Bank proffered a well-compensated expert witness who
testified that any markup of more than one percent was excessive. See
J.A. at 1758 (James I. Midanek Dep. (Sept. 12, 1996) at 205).22

By presenting expert testimony that Epley and Alex. Brown's
markups between 1.77 percent and 5.25 percent were excessive, the
Bank contends that it has met its burden of resisting a summary judg-
ment motion. Because the Bank has presented a "prima facie" case of
excessive markups, the Bank asserts that "the burden of proof shifts
_________________________________________________________________

21 It appears that some of these transactions may have taken place while
Epley was with his previous employer, MMAR Group. Because we con-
clude that no fraudulent markups have been proven by the Bank, we need
not distinguish between those transactions involving both defendants and
those transactions involving only Epley.

22 Mr. Midanek, who holds a bachelor's degree in accounting, acknowl-
edged that he received $450 per hour for his testimony on behalf of the
Bank. See J.A. at 1593 (James I. Midanek Report at 4).

                    27
to the broker to show that the markups were reasonable." Appellants'
Br. at 45.23

Because the markups in the instant case were undisclosed and
allegedly excessive, the SEC, as amicus, contends that these omis-
sions were necessarily material. See Br. of Amicus Curiae SEC at 17
("Because a reasonable investor in making his investment decision
would consider it important that he was being charged an excessive
markup, the Commission has long held that such a markup is material
as a matter of law."). Further, because this case involves an omission
rather than a misstatement, the Bank contends that reliance can be
presumed. See Appellants' Reply Br. at 21; see also Edens v. Good-
year Tire & Rubber Co., 

, 207 (4th Cir. 1988) ("where
fraudulent conduct involves primarily a failure to disclose, positive
proof of reliance is not a prerequisite to recovery" (quotations and
citation omitted)). Finally, the Bank contends that the scienter
requirement is satisfied "[w]hen a dealer knows the prevailing market
price and the price it is charging the customer and knows or recklessly
disregards the fact that this markup is excessive." Appellants' Reply
Br. at 21 n.33 (quoting Meyer Blinder, 50 S.E.C. 1215, 1230 (1992)).24

In summary, the Bank argues that securities dealers have an
inferred duty to disclose certain markups, despite the SEC's specific
decision not to require such disclosures through regulations. Dissatis-
_________________________________________________________________

23 The Bank cites First Honolulu Sec., Inc., 51 S.E.C. 695 (1993), in
which the SEC declared:

        The NASD, as proponent of the issue, had the burden of intro-
        ducing prima facie evidence of the excessiveness of the markups.
        The NASD met this burden by presenting evidence that the
        transactions at issue existed, the size of the transactions, the
        nature of the securities, the prices paid by Applicants contempo-
        raneously, and the prices charged to the customers. Once the
        NASD presented evidence of the markups, the burden shifted to
        Applicants to refute this evidence.

n.23.

24 The element of proximate cause and damages is also automatically
satisfied in a "shingles theory" claim because the funds that went to the
dealer for the allegedly excessive markup would have been retained by
the customer had the customer known that the markup was excessive.

                    28
fied customers who feel that an undisclosed markup is excessive,
including a markup of well under the NASD five percent guideline,
may bring a private cause of action for fraud based on a dealer's fail-
ure to meet this inferred duty to disclose. Once a plaintiff presents an
expert witness's testimony that the markup charged is excessive, the
burden shifts to the defendant to prove that he did not commit fraud.
To the extent of defeating a summary judgment motion, every ele-
ment of fraud--materiality, reliance, scienter, and proximate cause of
damages--is inferred or can be presumed. In other words, once a
markup exceeds the amount that a paid expert witness, after the fact,
declares excessive, a defendant must proceed to trial to prove that he
did not commit fraud.

We are acutely uncomfortable with this scheme. If the state of the
law were actually as the Bank and the SEC contend, it is unthinkable
that any dealer would ever fail to disclose any markup, no matter how
minimal, and thereby risk a lawsuit that would inevitably lead to the
expense and notoriety of a jury trial. Indeed, the SEC acknowledges
that this is the only practical alternative that dealers would have. See
Reply Br. of Amicus Curiae SEC at 11 ("a broker-dealer in doubt
over whether a markup is proper may protect itself by making such
a disclosure"). It is puzzling why, if Rule 10b-5 always imposed this
requirement to disclose on dealers, the SEC considered amending
Rule 10b-10 to impose a preexisting requirement. See Release No. 34-
15220, 15 S.E.C. Docket 1260, 

(S.E.C.), at *1. It is
even more puzzling why the SEC, having once abandoned an effort
to administratively require such disclosures, should now seek to judi-
cially impose the identical requirements on dealers. See Release No.
34-18987, 25 S.E.C. Docket 1223, 

(S.E.C.), at *2.

Assuming that a private cause of action is available to the Bank,
we must reject the Bank's suggestion that the burden of proof in this
case should shift to the defendant. As the facts of the instant case
demonstrate, it is very easy to accuse someone of fraud, and it is clear
that the mere accusation of fraud can be damaging to a defendant's
reputation. A plaintiff alleging fraud has both a heavy burden of
pleading fraud with particularity, see Fed. R. Civ. P. 9(b),25 and in
_________________________________________________________________

25 As was explained by our sister circuit in Parnes v. Gateway 2000,
Inc., there are three reasons for Federal Rule of Civil Procedure 9(b)'s
heightened burden of pleading in fraud cases:

                    29
proving each element of the cause of action. See, e.g., White v.
National Steel Corp., 

, 490 (4th Cir. 1991) (In case of
fraud, "plaintiffs carry an unquestionably heavy burden of proof. The
existence of actual fraud is not deducible from facts and circum-
stances which would be equally consistent with honest intentions. In
sum, a presumption always exists in favor of innocence and honesty
in a given transaction and the burden is upon one who alleges fraud
to prove it by clear and distinct evidence." (quotations, citations, and
alteration omitted) (applying West Virginia law)). In the instant case,
the plaintiff has failed to meet this burden.

While reliance can often be inferred in a failure-to-disclose case,
see 

, that inference has been rebutted in the
instant case. See Carras v. Burns, 

, 257 (4th Cir. 1975)
(inference of reliance is rebuttable). An economist for the Bank, NNI
unit director Mendez, explained why the defendants' markup was
unimportant to the Bank:

        Q[uestion by counsel]: Was Alex Brown the counter party
        in these transactions?

        A[nswer by Mendez]: Yes, of course.

        Q. Did you know what the difference was between their
        purchase price and their sales price?

       A. No.
_________________________________________________________________

        First, it deters the use of complaints as a pretext for fishing expe-
        ditions of unknown wrongs designed to compel in terrorem set-
        tlements. Second, it protects against damage to professional
        reputations resulting from allegations of moral turpitude. Third,
        it ensures that a defendant is given sufficient notice of the allega-
        tions against him to permit the preparation of an effective
        defense.

, 549 (8th Cir. 1997) (quotations and citations omitted).
These same principles support our decision that a plaintiff alleging fraud
should retain the burden of proof.

                    30
        Q. And you never asked?

        A. No.

        Q. Because as is true of all these other investments made
        during this period of time, that's not consistent with busi-
        ness practice?

        A. That's correct.

        Q. Because you satisfy yourself with respect to the cost to
        you?

        A. To the price that we were buying it for from Alex
        Brown.

J.A. at 272-73 (Mendez Dep. (Feb. 29, 1996) at 56-57 (emphasis
added)). NNI unit subdirector Aguirre, also an economist, agreed with
director Mendez's appraisal:

        Q[uestion by counsel]: [J]ust to be clear, Banca Cremi was
        accustomed to dealing with some transactions in which the
        intermediary made a profit on the purchase and sale instead
        of charging a commission?

        A[nswer by Aguirre]: I believe that this is something that
        would not be valid for only an institution like Cremi, but for
        any type of institution.

        Q. That in some markets people act as principals for prof-
        its, rather than agents for commissions?

        A. You see it daily. That's a daily activity in any exchange
        market.

        Q. In that circumstance, is Banca Cremi's interest in see-
        ing that the price it pays, if it's buying, or the price it sells
        for, if it's selling, is a price it believes is reasonable?

                     31
        A. Well, I find that to be actually a very complex question.
        You rest, and you have to take into consideration the exper-
        tise within each area and each field where your operators are
        operating. I couldn't ask whether the price is good, whether
        it was bad, and I couldn't do any exchange operation. There
        are hundreds of them that are transacted daily.

        Q. And the experience and expertise of the person you're
        dealing with is also germane to what they charge. Correct?

        A. Again, let me tell you: This is something--it's a daily
        activity in the markets, whether the profit was large or
        whether it was small.

J.A. at 241-42 (Aguirre Dep. (Feb. 28, 1996) at 85-86). The district
court accurately summarized this testimony in noting that

        as a matter of general policy, the Bank did not inquire about,
        or express any concern about the amount of any markups.
        According to the deposition testimony of Aguirre and
        Mendez, so long as the Bank believed the total price of an
        investment transaction was acceptable, the amount of the
        broker's markup was not a consideration in its decision to
        invest.

Banca 

(citing Aguirre Dep. at 84-86;
Mendez Dep. at 56-57).

In this case, it is clear that the Bank did not rely on any shingle-
created presumption that the defendants were charging an undefined
"fair" rate of markup. Rather, the Bank was uninterested in the defen-
dants' income, and instead based its trading decisions solely on the
purchase prices of the CMOs. Given the state of the record and the
utter lack of proof that a disclosed markup would have mattered, we
conclude that the Bank's excessive markup claim must fail.26
_________________________________________________________________

26 We also note that the alleged presumption of scienter in the instant
case is rather difficult to square with the uncontested fact that Epley and
Alex. Brown stayed within the NASD five percent guideline on all of the

                    32
V.

Finally, the Bank argues that the Bank's state law statutory and tort
claims for breach of fiduciary duty, fraud, negligence, and negligent
misrepresentation necessarily raise factual questions for a jury to
decide, and that the district court erred in granting summary judg-
ment. We conclude that the Bank's state law claims fail as a matter
of law, and accordingly affirm the district court's grant of summary
judgment.

We review the district court's interpretation of state law de novo.
See Salve Regina College v. Russell, 

, 231 (1991). This
Court must follow the forum state's choice of law rules to determine
which state's substantive law to apply. See Klaxon Co. v. Stentor
Elec. Mfg. Co., 

, 496 (1941). In this case, Maryland, the
forum state, adheres to the lex loci delictus rule--the place of the
wrong--for determining which state's tort law should be applied. See
Chambco v. Urban Masonry Corp., 

, 299 (M.D. App.
1995). Because the alleged wrongdoing of the defendants occurred in
Houston, Texas, we conclude that a Maryland court would apply the
law of Texas to the Bank's tort claims. See Farwell v. Un, 902 F.2d
_________________________________________________________________

markups with only two exceptions, and in each of these charged only .25
percent above the guideline range. Scienter exists "if the defendant knew
the statement was misleading or knew of the existence of facts which, if
disclosed, would have shown it to be misleading." Carras v. Burns, 

, 256 (4th Cir. 1975) (quotations and citations omitted). In light
of the defendants' adherence to an industry-wide guideline, the plaintiff
has rather a heavy burden to persuade that the defendants nevertheless
knew that their markups were fraudulently excessive. Cf. Decision by
National Business Conduct Committee of NASD, District Bus. Conduct
Comm. v. MMAR Group, Inc., (Oct. 22, 1996) at 14, reprinted in Appel-
lees' Add. at 33 ("Generally, in litigated cases, the [SEC] has not found
mark-ups or mark-downs of less than 4% to be excessive, and has not
found mark-ups or mark-downs of less than 7% to be fraudulent, regard-
less of the size of the transaction. . . . We do not suggest that we there-
fore may never find mark-ups below 4% to be excessive or mark-ups
below 7% to be fraudulent. Rather, we believe that these factors should
be considered in light of respondents' arguments regarding lack of notice
in the particular circumstances of this case.").

                    33
282, 286-87 (4th Cir. 1990) (applying Maryland choice of law, and
concluding that Delaware substantive law applied where allegedly
negligent psychiatric care occurred in Delaware and suicide allegedly
caused by negligent care occurred in Pennsylvania); Sacra v. Sacra,

, 9 (M.D. App. 1981) (applying substantive law of Dela-
ware in wrongful death case where automobile collision in Delaware
propelled vehicle over state-line into Maryland where impact with
utility pole caused death).

Under Texas law, for Epley and Alex. Brown to be liable in tort
for a breach of fiduciary duty, they must first have owed a fiduciary
duty to the Bank. See Duncan v. Lichtenberger, 

, 953
(Tex. App. 1984). "Fiduciary relationships arise when a party occu-
pies a position of confidence toward another." Miller-Rogaska, Inc.
v. Bank One, 

, 663 (Tex. App. 1996). "A fiduciary
duty is an extraordinary one and will not lightly be created. . . .
Although a fiduciary duty may be imposed on relations outside the
traditional ones, the exact same requirements necessary to establish a
traditional fiduciary relation must be met to establish an informal
fiduciary relation." Gillum v. Republic Health Corp., 

,
567 (Tex. App. 1989). Because "the relationship between agent and
principal is a fiduciary relationship," Magnum Corp. v. Lehman Bros.
Kuhn Loeb, Inc., 

, 200 (5th Cir. 1986) (quotations, cita-
tion, and alteration omitted), a securities broker can be a fiduciary to
a customer where "[t]he relationship between a securities broker and
its customer is that of principal and agent." 

law).

In the instant case, the Bank's employees have made it abundantly
clear that the defendants did not act as agents for the Bank, but rather
conducted their business at arm's length in a principal-to-principal
relationship. See J.A. at 272-73 (Mendez Dep. at 56-57); 

(Aguirre Dep. at 85-86). There was accordingly no formal relation-
ship giving rise to a fiduciary duty, and the record reveals no informal
relationship which could allow the imposition of such a duty. See,
e.g., Farah v. Mafrige & Kormanik, P.C., 

, 675 (Tex.
App. 1996) ("[T]he fact a business relationship has been cordial and
of long duration is not by itself evidence of a confidential relation-
ship. The fact one businessman trusts another and relies upon another
to perform a contract does not rise to a confidential relationship. Sub-
jective trust is not enough to transform arms-length dealing into a

                    34
fiduciary relationship." (citations omitted)). Because Epley and Alex.
Brown were not fiduciaries, they cannot be liable for the alleged
breach of a fiduciary duty.

Nor has the Bank presented sufficient evidence to allow its allega-
tion of negligence to go to the jury. The essential elements of the tort
of negligence include the existence of a duty owed by the defendant
to the plaintiff, a breach of that duty, and damages to the plaintiff that
are proximately caused by the breach. See Greater Houston Transp.
Co. v. Phillips, 

, 525 (Tex. 1990). In this case, the
Bank alleges that the defendants "owed a duty to use reasonable care
in selling securities to Plaintiffs and in advising Plaintiffs to purchase
securities[,]" Compl. (Apr. 11, 1995) at 24,¶ 87, reprinted in J.A. at
34, and breached that duty "[b]y recommending and selling to Plain-
tiffs securities which they knew or should have known were highly
speculative and risky, and were unsuitable investments for Plaintiffs
in light of their financial condition, risk profile and investment objec-
tives." 

88.

While a securities broker acting as an agent has a "duty to disclose
to the customer information that is material and relevant to the
order[,]" Magnum 

, Epley and Alex. Brown
were not, as we have discussed above, acting as the Bank's agents.
The Bank has directed us to no authority that would support a broad
general duty of a principal selling a security to warn another principal
that an investment may be imprudent or may not meet all of a buyer's
investment goals, and we rather doubt that Texas would impose such
a duty. In any event, it is apparent that, in light of the undisputed facts
of this case, Epley and Alex. Brown met any duty they may have had
by adequately informing the Bank of the risks and advantages of
CMO investment.

Similar to the federal action for fraud under § 10(b), the Texas
common law tort actions of fraud and negligent misrepresentation
contain the element of reliance. See Trenholm v. Ratcliff, 

, 930 (Tex. 1983) (elements of fraud); Federal Land Bank Ass'n
v. Sloane, 

, 442 (Tex. 1992) (elements of negligent
misrepresentation). As we have explained in part II.A. of this opinion,
the Bank cannot show justifiable reliance in light of the undisputed

                     35
facts of this case. Accordingly, the district court correctly granted
summary judgment against the Bank on these claims.

Finally, the Bank contends that Alex. Brown, which was incorpo-
rated under the laws of Maryland, should be liable for a violation of
the Maryland Securities Act, Md. Code Ann., Corps. & Ass'ns §§ 11-
101 to 908. Specifically, the Bank alleges that the defendants violated
§ 11-302(c) of the Act by failing to advise the Bank of the risks
involved with investing in CMOs.

Section 11-302(c) provides:

        (c) Misrepresentations.--In the solicitation of or in dealings
        with advisory clients, it is unlawful for any person know-
        ingly to make any untrue statement of a material fact, or
        omit to state a material fact necessary in order to make the
        statements made, in light of the circumstances under which
        they are made, not misleading.

Md. Code Ann., Corps. & Ass'ns § 11-302(c) (emphasis added). The
Act specifies that an "`[i]nvestment adviser' does not include . . . [a]
broker-dealer or its agent whose performance of[investment advisory
services] is solely incidental to the conduct of business as a broker-
dealer and who receives no special compensation for them." Md.
Code Ann., Corps. & Ass'ns § 11-101(h)(2)(iv). In this case, it is
clear that, to the extent that Epley and Alex. Brown provided "invest-
ment advisory services," such services were "solely incidental to the
conduct of business as a broker-dealer." 

Bank was not
an "advisory client" of the defendants, § 11-302(2)(c) does not apply,
and the Bank cannot pursue an action under this statute.

For the foregoing reasons, the district court's grant of summary
judgment is affirmed.

AFFIRMED

                    36